Skip navigation
1 2 3 Previous Next

RoadTests & Reviews

139 posts

During my review of the Megger  MIT420/2MIT420/2 insulation tester in Blog 9, I tested a 120MW Generator rotor winding that actually failed the insulation resistance test. Following this issue, the bedplate heaters for the generator were reconnected and placed back into service. Around 4 weeks later, I returned to the generator to carry out a further insulation test with the expectation that the heat would have driven out the moisture from the winding and a healthy insulation reading would be obtained.

 

Generator Bedplate Heater

 

With the heaters being disconnected during the last test, the winding temperature was taken as being the same as the ambient temperature. With the heaters in service, a higher winding temperature was expected and this was verified by measuring the resistance of the embedded winding temperature sensors. These are PT100 devices and tables are easily obtainable to cross reference the resistance obtained to a temperature value.

 

The resistance measured cross referenced to 21 degrees Centigrade which is circa 10 degrees above the ambient conditions. To maintain the winding in preservation, the recommendations are to maintain the winding temperature 5 degrees above the ambient and the humidity below 30% RH.

 

Initially the resistance was measured with the resistance function on the Megger  MIT420/2MIT420/2 However it was found that the resistance value in excess of 100 ohms was set to a scale without any decimal places This introduces a slight inaccuracy when using the insulation tester for this test For the purposes of an insulation test this isn't an issue but when measuring the winding resistance the corrected value can be affected due to the inaccurate temperature correction as detailed in the video below(lasts 1m 45s

 

 

A comparison of the rotor winding resistances obtained can also be seen in the table below.

 

Winding resistance corrected comparison

 

For this reason the winding temperature sensor resistance was measured with the Keysight  U1461AU1461A to obtain a more accurate value In fairness to the Megger I don't think it was ever designed to replace the resistance function on a multi-meter Even the Keysight  U1461AU1461A only displayed one decimal place for the scale so a good multi-meter would generally offer an extra digit over this as well

 

The 1 minute insulation resistance test was conducted at 250V and then 365V using the variable voltage function of the Megger  MIT420/2MIT420/2 This setting was used to match the working voltage rating of the rotor winding A final test was then recorded at 500V

 

One minute IR test values

When carrying out multiple insulation tests at different voltages, the expectation would be to see a drop in insulation resistance as the voltage is increased. The results obtained didn't display this trait and this can sometime happen with large inductive apparatus when carrying out a succession of tests, as stored charges can sometimes remain and affect the next test result. This is resolved by allowing for longer discharge times between tests.

 

However, the values obtained during this set of tests were much better than the previous tests as shown in the comparison table below.

 

Insulation test comparison

 

The DAR values were found to be above the minimum value of 1.2 for both the 365V and 500V tests, where as during the previous test they were around 1.00, which would be considered too low. The video of the instrument set-up on the variable voltage and the full DAR test is below and lasts circa 3 minutes.

 

 

As such a good 1 minute insulation test value was obtained, the polarisation index test was conducted at 365V and then again at 500V to see if an increase in voltage would have an adverse effect on the insulation reading obtained.

 

Polarisation Index Test ResultsPolarisation index test plot

 

A good polarisation index value was obtained for both test voltages, 3.05 for the 365V test and 3.03 for the 500V test, well above the minimum value of 2.00. The insulation resistance plot seen above shows the classic response for a polarisation index test. This time the 500V test produces a slightly lower reading than the 365V test as expected. The drop however, is not significant enough to cause concerns.

 

The video below is the polarisation index test at 365V. I have sped up the test to reduce it down to around 4 minutes.

 

 

Overall the  MIT420/2MIT420/2 continues to perform very well and I am happy to continue using it It has done around 300 tests to date of various types and the battery voltage drops down to 50 during the heavier current tests so appears to have a reasonable life

WarrenW

Digilent OpenScope MZ

Posted by WarrenW Dec 17, 2017

Hi All,

Wow. Fast delivery. The email stated ETA Wednesday and  already have it. Big Ups to UPS and E14.

So excited to have a good play with this thing. I am hoping to eventually link it to my cell phone hotspot and use the OTG to power the device and end up with a fully portable

scope and logic analyzer.

 

I will be taking it away on my annual leave over Xmas and having a good delve into what it can do and starting to write up my findings for the roadtest review.

 

Thanks Digilent and E14 for the chance to test this amazing little device.

Warren

 

Introduction

I love to build stuff using manual and power tools. I’ve not moved up to 3D printing yet, so I still tend to use materials like wood, metals and sheet plastic, hand-marked and cut. For small items, handheld rotary tools are extremely useful, and I’ve been meaning to upgrade my aging tools for quite a while. Two of the major brands are Proxxon and Dremel, and the Proxxon ones are lower cost in Europe. I decided to purchase a couple of Proxxon rotary tools, and this blog post documents my findings, and explores their usage for drilling, milling, cutting and polishing operations.

 

Why Proxxon?

The Proxxon tools are quite a bit lower cost than Dremel here in the UK, so the temptation was strong to purchase a couple of Proxxon tools instead of a single Dremel. I had encountered Proxxon tools before; at a couple of work labs there are Proxxon table saws. I purchased a Proxxon sander a while back, and it functions well.

 

In the past, I’ve used the Bohler Minitool system a lot. The system uses a 12V mains transformer and various tools like drills and table saws can plug into it.

 

The particular Bohler Minitool drill shown here is not like the Proxxon or Dremel tools, because it has internal gearing and the top speed is low. This tool is ideal for slow speed drilling into wood and plastics, and the chuck accommodates up to 6mm diameter drill bits. It is getting hard to find Bohler Minitool products at any reasonable prices in the UK. The Proxxon and Dremel tools run at higher speeds, and for many use-cases the high speed is quite important.

 

So, all things considered, I decided to try the Proxxon rotary tools for now.

 

Proxxon Rotary Tools Range

Some of the tools are powered directly from the mains, and others operate from a mains to 12V transformer much like the Bohler Minitool transformer shown above.

 

In the 12V category, there are two main series; the Micromot 50Micromot 50 series and the FBS 12 series. The products in the range can have suffixes like E, and EF. The E indicates that there is a variable speed control dial on the product. The F indicates that a chuck is supplied fitted to the tool. Collets are available to purchase too, so you don’t lose out on that upgrade capability if the chuck version is initially purchased. The Micromot 50 is lower power (40W) and is lightweight and thin. The FBS 12 is more powerful (100W) but is larger. In the directly mains powered range, there is the FBS 115 and FBS 240FBS 240 series that otherwise looks identical in specification to the FBS 12. Finally there is the IBS series, which is the more precision tool because it has a (partially) aluminium body. All the others have plastic bodies.

 

For the 12V supplies, there are three different mains transformers available. The largest one offers up to 5A current, and the two smaller ones (NG 2NG 2 series) offer 2A.

 

I tried two rotary tools; the Micromot 50/EFMicromot 50/EF and the FBS 12/EF, both with built-in speed controls. I used the NG 2/E mains transformer which offers 2A capability and has a built-in speed control.

 

Micromot 50/EF Overview

The tool has an interesting design, it has a long thin neck up to the drill chuck, making it possible to hold like a pen. The body is made of plastic and it feels rigid. The speed control is at the rear, and it doubles as the power switch too when turned to the minimum position and clicked into the off position. This is annoying, I’d have preferred a separate switch which the Micromot 50 (non-E version) offers.

 

Another strange feature is the chuck ‘key’; it is integrated into the neck of the tool, as a small metal plunger. Other rotary tools often have a hole for inserting a metal rod (or the shaft end of a drill bit if the metal rod is lost) temporarily while the chuck is loosened/tightened. The integrated plunger has the benefit that it won’t get lost, but it is in an awkward position where you might want to hold the tool. I’m in two minds about it, but overall, I think I prefer the traditional hole method. I can see that others may prefer the integrated plunger method however.

 

A couple of nice features relate to the collar, and the tool end. The collar of the tool is metal (likely steel; it is magnetic) and it allows for clamping the tool for certain operations if desired. There are attachments available (such as a drill stand) or a custom one could be fashioned. The other nice feature is that the chuck head can be unscrewed and replaced with colletscollets. This is great for PCB drilling or other operations where precision is needed. Collets grip the bits better and higher chance of them being fully centered. A set of collets can be purchased at low cost. The chuck head is unscrewed as if a drill bit is about to be removed, but by continuing to unscrew, the chuck head comes off the threaded shaft. The threaded shaft is hollow and has a slight internal taper, designed for the collet to be inserted inside the shaft.

 

I also really liked that it had an integrated speed control dial, marked with speeds, and fully variable. With the Dremel tools, that option is only on the 4000 series4000 series and above, which is more than double the cost of the Micromot 50. As mentioned earlier I just didn’t like that the Micromot 50/EF has the integrated switch at the lowest dial setting; I’d rather have a separate on/off power switch.

 

Still, the Micromot 50/EF is really compact and slender; there likely isn’t a lot of space to fit a switch.

 

Although I didn’t open it, it was possible to find a photo online at the reprap.org website showing the insides; a shaft bearing is visible at the end of the neck of the tool:

 

FBS 12/EF Overview

The FBS range is incredibly compact too. It is about the same length as the Micromot 50 series, but a bit fatter. It can still be held in a nearly pen-like form and there is a soft grip area to do so. With both tools (and especially the Micromot 50), the figertips can be just a few centimetres away from the work surface, particularly if collets are used instead of the chuck. The FBS range has similarities with the Micromot 50, such as the integrated plunger for unscrewing the chuck or collets, and the rotary speed control (although it only goes up to 15,000 RPM for the FBS series).

 

The major difference is the power. The difference is very noticeable (more on that later). I also really liked that the FBS models have a separate power switch, not integrated into the speed dial. The build quality is excellent, and the tool made reassuring noises even at top speed. Nothing screamed or resonated at the high speeds.

 

For a more detailed photo of the insides of the mains-powered version online, see this website (the photo is taken from there, but is reduced in quality).

 

NG 2/E Mains Transformer Power Supply

The mains transformer/power supply has a single output (I’d initially thought it had a couple of outputs; had I known, I would have purchased the larger NG 5/E model. The attached mains cable is a good length; 1.8m.

 

The tools use a connector that has (approximately) 2mm diameter pins. In fact 2mm banana plugs can fit into the power supply, although it isn't intended for it.

 

The power supply looks fairly sturdy, although the rotary speed control on it feels a bit toy-like. From my perspective I would really have liked to see a power button on it. There isn’t one, and so the only way of fully powering it off is at the mains connection : (

 

I took it apart (there is one triangular security screw) to see if there was space to fit a switch, but it is tight; I didn’t have any mains switch that would even come close to fitting. Anyway, internally the construction looks good. There is some protection with a varistor on the input mains side. There isn’t much to go wrong here! Perhaps the open trimmers may need replacing one day, but this would be a minor operation. It all looks very serviceable.

 

The output of the supply is a rectified but not filtered output. This is deliberate, so that the speed control can be silicon controlled rectifier (SCR) switched per rectified ‘hump’ in the waveform.

 

At max power (setting ‘5’ on the rotary control) the output looks like this; the output is about 17V RMS:

 

As the control is reduced, the SCR removes part of the waveform, until at the minimum setting it looks like the white shape in the trace here (it is about 14V RMS). At an intermediate setting of ‘4’, the red waveform is visible.

 

With the rotary tool connected and running at full speed in free air, the output amplitude drops slightly, but not much (and some electrical motor noise is of course generated).

 

The top end transformer model is the NG 5/E, and with hindsight I would have purchased that, because it allows for several (three) tools to be left connected to it, offers more power, and has a built-in mains switch.

 

Checking the Speed

The Micromot 50/EF speed control dial is marked in RPM, and it was surprisingly accurate when checked with an RPM meter. Most of the dial covers a useful speed range of 5000-10000 RPM, and then it gets more cramped at the top end to the 20,000 RPM marking.

 

With the power supply control set to maximum, the speed compared to the dial position was accurate to within about 200 RPM across the 5000-10000 RPM range, which is excellent. The top speed was around 24,000 RPM although the dial was marked 20,000 RPM. With a load the speed would decrease of course.

 

With the power supply control set to the minimum speed, then measured rotation speed was about 2000 RPM less than the dial marking, up to about 10,000 RPM, and then the measured speed was about 4000 RPM less.

 

In summary, the speed marked on the tool is fairly accurate up to 10,000 RPM, and beyond that the dial is quite cramped so it would be difficult to be sure, but the maximum unloaded speed is about 24,000 RPM.

 

The FBS 12/EF has a lower top speed of 15,000 RPM but it gets there gracefully. On the other hand the Micromot 50/EF does not make pleasant sounds at its top speed. There is some vibration at speeds beyond 15,000 RPM, and the bearing around the shaft may be contributing to the noise too.

 

Drilling: What Speed to Use?

There are plenty of online resources on drill speeds for various materials, although the numbers vary from site to site, and many of the sites specify speeds in imperial measurements in surface feet per minute (SFM). I took the information from a website PDF document and metric-ized it into a chart. It can be seen that for the drill sizes that can fit in the Proxxon rotary tool (around 3mm and smaller), the tool would be fine for a variety of materials.

 

To use the chart, choose the curve that matches the material, then look up the desired drill hole diameter on the x-axis, and then at the intersection you can read off the guideline speed on the y-axis. The chart is also attached to this blog post as a high-res PDF file; I used that to stick the chart on the wall. The chart is for optimum conditions, so in practice the value should be halved.

 

For PCB drilling, the plastics/composites curve can be used. Also usually a drill press is needed because otherwise there is high risk of drill bits snapping (wear safety glasses! – I literally have half a dozen of them littered around the workshop, for myself and anyone else present).

 

When it comes to feed (plunge) rates into the material, as a rough guideline, for drill bits 3mm or less, with the drill bit running at speeds indicated in the chart above, a rate of about 3mm per second is adequate for most of the materials, but can be reduced to around 2mm per second for tool/die steel, and increased to 4mm per second for plastics, and 16mm per second for aluminium alloys and magnesium alloys. These are under ideal conditions (sharp tool, best coolant, etc), so the figures should be halved. But, if the drill speeds are halved as suggested above, then a quarter of these feed (plunge) rates should be used. The information is based on guidelines from this PDF document.

 

Milling: What Feed to Use?

Although I don’t currently have an intention to do so, in theory the rotary tools can be used for routing/milling operations too. These are operations where the spinning tool can cut a shape into the material. It is an advanced machining task, and care is needed not to plunge the tool too deeply into the material, move the material or the spinning tool in the correct direction, at an appropriate linear rate for the given tool rotational speed, diameter and material type, and supply adequate coolant. For milling a stand is needed and a way to move the tool or the workpiece. For routing by hand, again some stand or jig is usually needed. There is a very high risk of milling bits snapping, inadequately clamped material and clamps flying, and bits of clothing or skin being chewed up.

 

All that aside, there is the question of what linear rate the material should be moving relative to the tool. It is known as the feed rate, and there are tables that can specify it in inches per minute and so on. There are software applications that are usually used to work it out. These are intended for automated machinery usually, but the rotary tool user needs to know this information too. Often the tables do not have the information for the very small tool diameters that the rotary tools use. Based on various bits of information (in particular harveytool.com and the previously mentioned PDF document and mmsonline.com), the chart below was constructed as a guideline. To use it, first ensure that the speed in RPM is correctly selected from the earlier chart above. Then, look up the same tool diameter on the x-axis in the table below, examine the line intersection to see the feed speed on the y-axis in mm per second. If the speed in RPM is halved as discussed earlier, then the result from the chart below needs to be halved too. The chart is for milling tools with four cutting flutes (sharp edges). If the tool you’re using has two flutes, then the feed rate needs to be halved yet again. Two flutes are great for certain materials and tool sizes.

 

You don’t want to go overly slow, because it will just overheat the material and cause a rough finish. But don’t go too fast either, it can cause damage to the tool or the material, or the user!

 

Regarding milling depth, usually half of the tool diameter is feasible, and reduced for the harder materials, but for small diameter bits (3mm or less) then the depth of cut should be reduced further, to prevent the tool snapping. 10% of the diameter could be used as a guideline for bits less than 3mm in diameter.

 

Please note these are all just approximate guidelines, and a lot depends on the specifics of the setup. It goes without saying that the information here is just informational, and newcomers to milling operations should study in depth using material elsewhere; this is just a short blog post and not a complete guide.

 

Cutting Operations

As well as drilling operations, it is possible to use rotary tools for cutting. I had some Dremel accessories, in particular some cutting wheelscutting wheels. They require a high speed to function, so I was looking forward to using the Micromot 50/EF with them. Dremel has the nice ‘EZ SpeedClicEZ SpeedClic’ system, which does away with the screw-and-mandrel and replaces it with a single unique mandrel with a springy end that allows the cutting wheel to be pushed on and turned until it clicks and locks into position. The system is very good but isn’t totally perfect, you can see that the cutting wheel on the right has had its center dislodged.

 

The Micromot 50/EF worked very well with the 0.75mm thick Dremel cutting wheels. I used it to cut a 6mm diameter brass shaft of a potentiometer:

 

The Micromot 50/EF and cutting wheel also worked very well cutting steel screws; it also makes a nice light show.

 

The tool was very competent cutting such items.

 

The cutting wheel that I was using is not designed for plastic, however I did try slicing through some 3mm thick plastic with it. The Micromot 50/EF struggled, the drop in speed was noticeable as the cutting wheel went deep into the plastic. In contrast, the FBS 12/EF had no issue at all.

 

Polishing

Polishing often requires high speeds that are not possible with typical cordless drills. I had a Dremel polishing accessories setDremel polishing accessories set, and I decided to try out the polishing wheel. The Dremel set came with some polishing compound in a tiny pot, but I lost that a long time ago. It is cheaper to just buy a polishing compound block intended for large polishing machines, than the small pots from Dremel. It comes in varieties for ferrous or non-ferrous metals, and in several grades for working up to a mirror finish.

 

The compound is applied to the wheel as it spins, and then the wheel is slowly moved in straight motions across the item to be polished. I decided to test the Micromot 50/EF on a stamped metal (aluminium) container that had a dull surface finish. The speed was set to about 10,000 RPM.

 

It worked really well. The photo shows just half of the aluminium container polished, and the difference is clear.

 

The finish was very mirror-like (reflections were visible).

 

Summary

This was intended to be just a short review, but it turned out there is a lot to cover when evaluating such tools! The tools are constructed well, and their potential use for drilling, milling, cutting and polishing operations was explored.

 

I like both the Micromot 50/EF and the FBS 12/EF, and I will use them for different purposes. The Micromot 50/EF is very lightweight and has plenty of power for most tasks. However, the FBS 12/EF is clearly the tool to use whenever cutting substantial material, or when using certain attachments such as grinding wheels. I liked that both tools have a rotary speed control with markings, so that I can look up the ideal speeds on charts for different materials.

 

If I was to purchase again, I would probably go for the NG 5/EF or the NG 2/S mains transformers, because I didn’t really need the adjustable speed control on the NG 2/E model. The NG 5/EF is the one to really get, because it has a mains power switch built-in.

The Interested Phase

It all started when element14 posted a new road test for a Keysight E3613A power supply.

I thought "that looks interesting", I can always use a good power supply. So I looked up the specs posted in the road test announcement.

They looked great, so I set up a directory to collect documentation on the device, including links to the road test page.

The Research Phase

I then went to the Keysight web site to find more information. I found the document library page for the device and  downloaded and read several documents such as the data sheet and application notes.

The site also led to a Youtube video, so I searched for more Youtube videos and found half a dozen made by various Keysight staff.

The Motivated Stage

After reading through the literature and watching the videos, I was very impressed with the features and how they might improve my capabilities.

Now I really wanted this device, but I was still a long way from deciding to write a proposal.

The Doubting Phase

I always have internal debates about whether to embark on a project like this:

  • this is going to be a lot of work - but it will make my work easier - but it is a big commitment - but I want it .....
  • I don't like the image of scoring all the road test equipment - but I can really use this one - but others may need it more - but it is unfair - but it is a fair competition ....
  • I can't think of a compelling proposal - but I have lots of uses for it - but a features description proposal has little chance - but I want it - but there are lots of talented members writing great proposals - but I can write a proposal and see what happens - but it is a lot of work for a long shot chance ....
  • at your labour rates, you could buy this device with a small fraction of the work of a road test - but there are other benefits ....

These arguments rage internally until I finally decide to hit the submit button and sometimes I write a proposal but don't submit, but I don't let that possibility stop me from proceeding, at least in this case.

The Brainstorming Phase

I think it is hard to write a compelling proposal for an instrument because they have a specific purpose and all proposals will describe the same features and propose to do similar demonstrations of how they work.  I think it is important to do all that, but to win, the proposal needs to include something different and interesting.

I wasn't coming up with anything interesting, so I started thinking about all the projects I've done where I had problems with power supplies. There were quite a few so I thought through them to see if this power supply could have been used to address any of the issues. It was a bit of a stretch - this supply can replace many defective power supplies, but is too big and expensive to be a solution.

However I did come up with a metal detector circuit that needed multiple supplies, where I couldn't troubleshoot the main circuit because the complex power supply wasn't working properly. This triple supply would be great at powering the circuit so I could troubleshoot the non-power supply issues. This was okay, but not something everyone would be excited about and I didn't feel it was a compelling application, so I kept thinking, trying to pick problems that lots of other people might run into.

What I came up with was the explosion of USB "wall wart" power supplies. There is a huge variety, and many of them have problems powering high-current / voltage-sensitive devices like Raspberry Pi modules. But how does this Keysight power supply solve these types of problems? Well, read the road test, but one thing it can do is help characterize USB cables, which often have significant voltage drops when carrying large currents. Well, now I had a couple of applications although not what I would call a killer app, so I kept thinking, this time about trends in power supplies that might be topical and interesting.

One area immediately stood out - mobile power - batteries. This Keysight power supply can create an ideal charging cycle for Lithium batteries and it can log voltage and current during the whole cycle. It might show how well commercial chargers perform against an ideal cycle and I would like to experiment with that idea. But what about discharge performance? This might be even more interesting to experiment with. A normal power supply typically is not a good programmable load, but perhaps I can find a way to use this very capable power supply in this capacity and use it to log the discharge cycle at the same time.

Now I think I may have enough to write a proposal that is different from everyone else and has enough different applications to be interesting to a reasonably wide range of members.

The De-Risk Phase

Now that I have some hair-brained ideas about what I want to do in the road test, I don't want look like an idiot by publishing them to the world and having them not pan out. (de-risking is not always sufficient )

I also don't want to be successful with the proposal and when it comes time to carry out the road test, I can't complete it because I don't have the right components and materials.

So I do some homework to prove that I have a chance of completing the road test as proposed.

In this case I dug out my problematic metal detector circuit - to verify that I still had it. I also took a look at measuring USB power supplies and cables and realized it wasn't going to be easy to get accurate results, especially without ripping cables apart. USB connectors make it pretty difficult to measure voltage and current, so I searched for USB breakout cards and ordered a selection with various connectors. I couldn't find all the cards I wanted, so I designed some custom USB cards and ordered them, including a card that has banana plugs to plug directly into the Keysight power supply. Hopefully my research into the binding post spacing is accurate.

I also thought through the problem of using a power supply as a programmable load and came up with at least one design that should work with components I have on hand.

At this point I still have not written the proposal and not even decided to submit a proposal, but if I want all this stuff to be ready in time to meet the schedule, it has to be done immediately. Also, all of this is being done knowing that the proposal, if I write it, has a low probability of winning.

I do not mention any of these de-risk activities in the proposal - I don't want the proposal to sound presumptuous.

Contingency

It may sound presumptuous to design solutions and order parts before submitting a proposal or having a clue if the proposal will win, but I often do it and I have never regretted it. I think I gives me an advantage in writing a proposal that makes sense and sounds achievable as well as providing the best chance of succeeding in the road test if the proposal is successful. And the work is not being done in vain - even if I don't win the power supply, I still want to experiment with USB power supplies and cables. Everything else helps improve my knowledge and experience.

The Proposal

For the actual proposal, I broke it into 3 fairly short paragraphs with no pictures:

  1. Product Research and Road Test Motivation
    • In this section I explain exactly what I like about the power supply, what impresses me and how important it is to me.
  2. Road Test Plan
    • In this section I indicate what videos I would produce and what they would cover in terms of device features and performance.
    • I also include short paragraphs describing each of my differentiating applications
  3. Credibility
    • Here I brag unashamedly about my experience, my track record and my previous successes.

 

At this point I re-read the proposal and go through a final round of soul searching before deciding to submit.

 

Summary

The whole proposal is only 890 words, but it took a lot of work and thought to get to there.

This is not exactly a typical proposal for me as my mind can travel down numerous tangents on the way to a conclusion, but it is typical that I put more thought into the proposal than comes across in the text and the actual proposal writing is only a small part of the whole process.

 

Proposal Post Mortem

rscasny posted the main criteria used by Keysight in evaluating proposals. In looking how my proposal matches their criteria, they had to dig a bit to infer that the proposal addressed their criteria, because it didn't explicitly have those categories, but thankfully they found enough linkage. I don't know how close my proposal was to failing, but armed with the new knowledge it should be possible for everyone to improve their proposals.

I didn't want this blog to be too verbose and I'm not sure what I'm leaving out here, so if you have any questions, feel free to ask them in the comments below.

I will edit this and put a link to the road test once I get it posted.

1.0 Review Outline

In part 1, I will first give a general introduction to the RTE1204 and take a detailed look at the physical and software user interface. I am continually learning and finding little quirks here and there, so I will be sure to edit and update part 1 as I post the next sections over the next two weeks. I waited a little while to post this, as my RTB2004 just arrived, and I wanted to add some side-by-side comparisons. It's obviously not fair to compare the performance of the two, but if figured that the interface and usability is fair game.

 

Table 1. Table of Contents for review

Rhode and Schwarz RTE 1204 Review SectionLink
Part 0: Review ProposalRohde and Schwarz RTE1204 2GHz Oscilloscope Review: part 0
Part 1: Introduction and User Interfaceyou are here
Part 2: Analog and FFT Functionalityto be added upon completion
Part 3: Mixed Signal Domainto be added upon completion
Part 4: Conclusion and User Requeststo be added upon completion

 

 

1.1 Introduction to the RTE1204

The RTE family of oscilloscopes from Rohde and Schwarz caters to the mid-range oscilloscope market, with up to 2GHz of analog bandwidth and a 5GS/s sampling rate per channel. The optional 16-channel, 400MHz mixed-signal and decode capability, powerful spectrum and power analysis, and a wide assortment of eletromagnetic/active probes make it clear that this scope is aimed at signal-integrity measurements in modern mixed/embedded systems. Indeed, I was first introduced to this scope when I was in the systems group at a large consumer electronics company, as we had a few in the lab. The RTE series is now a few years old, being released back in 2014, and I would not be surprised if R&S released a minor update to this line with a capacitive touch screen and button/knob interface that matched the RTB and higher-performance RTO families.

 

In their promotional material, R&S mentions the 16-bit mode vertical resolution at length, but to be clear, the actual ADC is an 8-bit device. The RTE "high definition" mode uses standard averaging/digital filtering techniques to trade off resolution with bandwidth, as with other scopes. The vertical enhancement vs filtering trade-off for the RTE is shown in the table below.

 

Table 2. High definition mode bandwidth vs enhanced vertical resolution

Bandwidth

Vertical resolution
in HD mode

(bits)

10KHz-30MHz16
50MHz14
100MHz13
200MHz12
300MHz11
500MHz10

 

 

The RTE1204 has a waveform update rate of 1 million/second and optional increased memory depth up to 50Msamples on 4 channels, 100Msamples on 2 channels, or 200Msamples on 1 channel. There is also an optional 2-channel 100MHz, 500MS/s arbitrary waveform generator, capable of generating differential signals with 0.1 degree phase resolution and <200ps skew. This option also includes an 8-channel 40Mbit/s pattern generator. The unit that I have unfortunately does not have this option, but on paper it seems like a significant step above the 20MHz (Keysight) and 50MHz (Tektronix) arbitrary waveform generators available on comparable oscilloscopes. In my personal opinion, having a built-in function generator is not something I would opt for in this level of oscilloscope, which is intended to be used in a shared lab setting.

 

Table 3. Specifications for comparable oscilloscopes

Rohde & Schwarz

RTE1204

Keysight

MSOX4154AMSOX4154A

Tektronix

MSO5204BMSO5204B

Tektronix
MSO54

Max bandwidth2GHz1.5GHz2GHz2GHz

Sampling rate on all channels

5GS/s5GS/s5GS/s6.25GS/s

Vertical resolution (true ADC bits)

88812
Max effective bits in high-resolution mode16121116
Waveform updates per second1,000,0001,000,000250,000500,000
Standard/max* record length per channel10M/50M4M25M/250M62.5M/125.5M
Approximate
List price (including 16ch MSO)

$26,150

$22,015$24,800$27,800

*with upgrade, not reflected in list price

 

In terms of the major specifications (bandwidth and sample rate), the Rohde and Schwarz RTE oscilloscope family is most easily compared to the Keysight MSOX4000-series and the Tektronix MSO 5000 series (and newer Tektronix 5-series). However, in terms of front-end noise and additional measurement features, the RTE series goes a bit beyond what is offered at least from Keysight here, as it is closer in some ways to their MSO6000x offerings. In particular, the RTE series allows for: eye diagram measurements, histograms, 4 advanced math signals, color grading, customized visualization, as well as a more comprehensive FFT measurement suite. Both Keysight 4000 and 6000 series offer a maximum of 4Mpts of sample memory, which seems oddly low.

 

HDO Oscilloscopes from Teledyne Lecroy are somewhat harder to directly compare to, given their interleaved-ADC architecture described here. Basically, they use more than one ADC with a clock and code offset to increase their effective number of bits. Their marketing department likes to make things confusing on their website by reporting aggregate sampling rate and making it seem like they use data converters with more than 8 bits. In any case, to match the performance of the RTE series, you have to go to the HDO9204, which has a list price of $25800, excluding the mixed-signal option.

 

As a comparison with slightly newer models: the MSRP of a R&S RTO2024 (4 channel 2GHz 20GS/s + 16ch MSO) is $29800, and that of a Keysight MSOS204A (4 Channel 2.5GHz 20GS/s + 16ch MSO) is $30800.

 

This class of scopes are clearly not marketed towards hobbyists, and the pricing of additional software/hardware options is no exception. Below is a quick summary from a few suppliers:

 

Table 4. Approximate cost for oscilloscope upgrades

Option

R&S RTE

option Cost
(USD)

Equipvalent Keysight
MSO4000 Option Cost
(USD)

Equivalent Tektronix

5-series option cost

(USD)

channel sample memory upgrade$5300 -- 50MN/A$4990 -- 125M
16-channel mixed signal option$2950~$2800$3600
Power analysis option$1700$1530 (but included
in application bundle)

$2300

Arbitrary waveform generator* `

$1020 -- 20MHz
(but included

in application bundle)

$1250 -- 50MHz

Frequency/Spectrum Analysis* `

$1530 (but included

in application bundle)

N/A

Eye diagram/Jitter analysisN/AN/A$4720
Individual trigger/decode options$1550-1675~$1530$1850
Application bundle* `$2750N/A

*waiting for this quote.

`Note: the application bundle does not include this feature

 

I am still waiting for some clarification on the RTE1024 option pricing. I suspect that high-resolution mode and spectrum analysis are now offered by default on the RTE series, but they were formerly upgrades (I certainly hope this is the case). For the RTE1204 under review, the price -- as configured -- is in the neighborhood of $30000. It is clear that Keysight's application bundle offers the best value with respect to software upgrade cost. Tektronix does not seem to offer such a bundle for their 5-series, so trigger/decode options could likely become expensive very quickly.

 

1.2 Physical User Interface

The RTE1204 is shown below alongside the RTB2004. Being a higher-end scope, the build quality is noticeably better, but this feeling may be due in part to the significant heft of the RTE, weighing in at nearly 19 pounds. The front cover for the RTE is solid with a foam insert that conforms to the features of the oscilloscope and uses the rubber bumpers to mount to the scope. This is infinitely better than the cover for the RTB, which gets caught on a lip before attaching to the front panel, as others have noted. I have stopped using the RTB's cover for this reason, and I would not recommend it given its price. The carrying bags are also shown. Neither is anything special, but again the RTE is noticeably higher quality, as would be expected. It does offer both interior and exterior pockets for probes and the MSO components, whereas the RTB bag just has an interior compartment.

 

{gallery:autoplay=false} RTE1204 and RTB2004 Physical Comparison

Side by side of oscilloscopes

RTE1204 and RTB2004 side by side

With Covers

With their covers

Side profiles

Side profile

Rear Shot

Rear shot of the RTE1204

Carrying Bags

Carrying Bags

Image Gallery 1. Physical Comparisons between RTE1204 and RTB2004

 

The front panel of the RTE1204 is shown in image gallery 2. The screen (1) on the RTE scopes is a 10.4" TFT touch screen with a resolution of 1024x728. Truth be told, it works well, and its matte finish does not attract finger prints like the RTB screen does, but it feels dated. TFT screens just do not have the same responsiveness that we have come to expect from capacitive touch screens, and they obviously do not allow for multi-touch gestures (not that an oscilloscope necessarily needs that, but they can be useful). Both the RTB and RTO have 10.1" 1280x800 capacitive touch screens for comparison.

 

 

{gallery:autoplay=false} Rohde and Schwarz Front Panel Comparison

RTE front panel

RTB front panel

RTO front panel

 

Image Gallery 2. RTE, RTB, and RTO Front Panels (images from Rhode and Schwarz user manuals)

 

Upon first glance, the button placement is logical and intuitive. Those familiar with Keysight and Tektronix oscilloscopes will notice that each channel does not have its own dedicated vertical controls, which I do not mind, as the LED channel/color indication works very well. The knobs are all detented, and it is obvious that some oscilloscope users on the internet have very strong feelings about this, which I do not. They work well and are all press-able to change their mode/step size. My only complaint with the layout is that it is not efficient. There is a lot of real estate devoted to buttons on this oscilloscope (real estate that could be going towards a larger screen), and some of these buttons seem redundant or not very useful. For example:

  • In the horizontal options, the "res rec len" and "horizontal" buttons bring up the same dialog box
  • The vertical channels require a separate "signal off" button to turn off a trace, rather than just pressing the same channel button twice
  • The escape button in the navigation does not actually escape from any dialog box. I did not find the navigation buttons, with the exception of the main knob, to be very useful -- this scope cannot be realistically operated without either the touch screen or a mouse, which is understandable but should be noted.

 

The next-generation RTO series front panel does have an updated look with colors that are more inline with the RTB series, and it does seem a little better in its use of space. However, the buttons are not fundamentally different. The RTB series does feel like a refinement, though. Here, the user can turn off channels without needing a dedicated button, and the button choice seems more useful (for example, it includes a force-trigger button that the RTE does not). Overall, I think R&S could benefit from a standardization of all aspects of their oscilloscope front panels. They seem to be moving in that direction, but there are still major differences between the RTO, RTM, RTB and especially the HMO (formerly Hameg) lines.

 

Given that the GUI of this scope is so customizable, it is a little disappointing that there are not more user-programmed options for hard keys. Only the print button has some degree of customization, allowing for different screenshot/report saving options. I would happily trade some buttons on the left hand side for macro-programmable buttons. The horizontal and vertical knobs can be switched from position to offset/ref point, and the navigation knob acceleration can be changed. It is nice that the LCD backlight intensity can be changed (unlike for the RTB), and the LED brightness can be turned fully off.

 

{gallery:autoplay=false} Frontpanel software configurability

LCD Screen and LED Brightness control

Print button control

Knob control

Image Gallery 3. Front panel software controls

 

Booting the machine, as shown in the video below, takes approximately two minutes. You will also notice that the fans and relays get very loud during the self-test phase -- luckily it returns to a dull hum within another minute, and in all of my testing so far, I have never had it reach the same volume as it does while booting.

 

Video 1. Boot time and noise of the RTE1204.

 

1.3 Graphical User Interface

Having used a number of R&S instruments (mostly vector network analyzers), I always find it interesting to compare their interface with Keysight's. My opinion -- and feel free to correct me if I am wrong -- is that you get a lot more customizability with R&S. From which side your signal menu "bar" shows up on, to the font size, to the color of almost everything, you can control it. Of course, you can configure it to be totally overwhelming and user-unfriendly, so there is a bit of a learning curve.  Keysight, by comparison, feels like a more controlled (and restrictive) experience. This is neither a good nor a bad thing -- it's just an observation.

 

I'll start with my favorite feature about this oscilloscope's interface: the "Smart Grid". It is a phenomenally different way of using an oscilloscope, and it lends itself extremely well to a touch screen. In a nutshell, "SmartGrid" allows you to drag and drop channels measurements, math functions, etc. into different sections of the screen. It's not just limited to quadrants -- you can stack up to 4 sub-windows in both x and y directions, and you can also sub-divide each window until the screen is jam-packed with information. Each tab can be annotated with a little menu that allows an easy way to close it, or you can alternatively "dock" your waveform back on the signal bar to be displayed later. It is possible to drag each window to individually control the size as well.  See below for some screen-captured videos of these features.

 

Video 2. Demonstration of the "SmartGrid" -- moving traces/functions/FFTs and resizing the windows

 

 

I became so accustomed to this feature that I was very disappointed to find that my RTB scope only has a very primitive version of the Smart Grid (despite what the marketing material claims). In particular, the RTB scope does not allow a user to drag and drop different channels to different sub windows -- that would make for an impressive firmware update!

 

The two menu bars that are most accessible are those on the top and the bottom. The top toolbar is user-configurable, and this is what I use most often to get to the core features of this scope. To use any of the pinned actions, you simply click the icon and then click the trace that you want to apply the action to. The bottom toolbar mostly overlaps with hard-key functionality and it would be nice if you could hide it.

 

Toolbar

Screenshot 1. Top Tool bar customization options

 

Having the "show signal bar" paired to the top toolbar comes in very handy if you need to free up some real-estate when there are a lot of tabs open on the smart grid. You can switch this signal bar position to either the left or the right, control the transparency, or even enable "auto-hide" after a select-able amount of time. It is also easy to scroll through all of the available signals, measurements, and math functions.

 

Video 3. Signal Bar options on left/right and auto-hide

 

The next notable feature is the zoom. This option, when paired to the top toolbar, actually has 4 modes, selectable via a drag-down menu, as shown. Note-able differences are that standard zoom keeps the original waveform and settings in a top tab, and opens a separate zoomed tab. It is possible to create multiple tabs with different zoom locations of the same waveform and couple these zoomed regions either by position or by size. Adjusting the size (and to a lesser extent the position) of the zoomed regions can be somewhat difficult with the touch screen, and the navigation knob helps a lot. Hardware zoom changes the vertical resolution and time-base on the current window, and fingertip zoom just gives a snapshot of the signal below your finger.

 

Video 4. Normal Zoom

 

Video 5. Hardware Zoom

 

Video 6. Coupled Zoom

Video 7. Fingertip Zoom

 

With the SmartGrid, it is very nice to be able to display cursors on different tabs simultaneously. Luckily R&S thought of this, and I've shown how it is implemented in video 8. Unfortunately, you are limited to only two sets of cursors, so if you have all 4 (or more) channels up on separate tabs, you will have to pick and choose. It is also nice that you can pin the cursor measurement results to a separate area of the SmartGrid.

 

Video 8. Using separate cursors on different SmartGrid tabs

 

As you would expect, the same "pinning" can be done using both the quick and normal measurement functions. The quick measurements can be configured such that the user's the most desired measurements all come up with a single click. Normal measurements work the same way, and statistics are available with both, as shown in videos 9 and 10. In Video 10, you can see an issue that I came across frequently -- during a single trigger, the scope seems to display many waveforms overlapped for any trace that is not set as the active trigger. Fiddling with the horizontal position fixes this, but it gets a bit annoying, as that also resets the statistics.

 

Video 9. Quick measurements with statistics (also showing a single-trigger bug)

 

Video 10. Normal measurements

 

The last measurement I will cover in part one is the histogram (I'll save masks, search, FFT, etc. for later). You have the option of choosing up to 4 vertical or horizontal histograms. With multiple traces in a single window, you can only display one at a time, but again using the SmartGrid allows simultaneous histograms to be shown. Changing their relative size requires going into the menu -- you cannot simply drag it. I also found that occasionally the histogram icon in the top toolbar became grayed out for reasons I cannot explain. I could still go into the histogram menu and enable them via the bottom toolbar, but not the top.

Video 11. Histograms

 

I'll skip adding a bunch of additional videos (unless there are specific requests) about how to add labels to traces, change diagram names, etc. Suffice it to say that the user has a lot of control when it comes to customizing how it looks and how traces are displayed. You cannot draw on the screen as you can with the RTB -- it would be interesting to know if that feature has been added for the RTO scopes.

 

1.4 External Display

Normally this would not deserve its own section, but this is the one area where the RTE really falls short. For the life of me, I just could not get an external monitor functioning properly with the RTE1204. I tried different monitor sizes, resolutions, options within the R&S scope application and directly from the Windows control panel. Extending the display worked well enough if I wanted to use the external monitor for something else, but not with the scope application. Duplicating the screen or using only an external display resulted in odd behavior -- the oscilloscope application does not seem to like non-native resolutions. Using only the external display was the closest I got to "working", but then there is an issue with the mouse being offset from where you actually click in the application. Maybe I am doing something horribly wrong, but I would expect this to work without any major changes needed outside of the oscilloscope application.

 

Video 12. External Display Issues

 

1.5 Computer

Because this is a scope with an embedded Windows interface, the computer running the OS also deserves some attention. The specifications are shown below in screenshot 2.

Screenshot 2. Computer specifications

 

While its not very powerful by most modern computer standards, it seemed to have more than enough power to chug through whatever I could throw at it without any signs of slowing down. The oscilloscope application does indeed eat up a lot of power when running many SmartGrid screens at once with measurements and especially FFTs and spectrograms. In video 13, I show a case where I am using a significant amount of processing power to display two simultaneous FFTs and spectrograms.

 

Video 13. Performance when displaying many traces and math functions

 

That concludes part 1. As always, I will be paying attention to the comments, and I will also be actively updating this section (with edit dates and highlights) if I find errors or things to add. Expect part 2 soon. Thanks for reading,

 

James

You will all be thankful to know that I have finally reached my last blog in this road test review of the Megger  MIT420/2MIT420/2 Previous blogs have covered bench reviews and tests a teardown and insulation testing from electrical panels on motors and air circuit breakers In this blog I will cover the testing of generator rotor windings from a 92MVA and a 145MVA turbogenerator.

 

Insulation Tester Models

 

This is the main function of my job and the reason why the road test was of great interest to me. I currently use a Fluke 1555C insulation tester, with a 250V to 10,000V output capability to carry out rotor and stator winding tests on turbogenerators. However, this instrument lacks a 100V output sometimes specified for insulated bearing tests and struggles with voltage compliance at the lower output levels. It also comes in a large padded case and takes up a quarter of the car boot space when travelling. Due to logistics and costs of testing generator stators,I test more rotors and taking the Fluke 1555C around with me is a bit of a burden.

 

Whilst the 1555C has some download capabilities it does not download a complete polarisation index test and I wanted to improve my reports so i opted for a Keysight  U1461AU1461A that on paper seemed to meet most of my needs and would also provide me with meter functionality for RSO and AC Impedance tests Unfortunately the  U1461AU1461A does not appear to have the capacity to test generator rotor windings so has been consigned to use as a multimeter most of the time.

 

This brings me to the road test of the Megger  MIT420/2MIT420/2 It is a smaller more managable unit than the 1555C and has a low enough output voltage for my requirements It does have some multimeter functionality that looks promising enough to meet my needs I am aware that this particular version does not have a results download capability but the next model up in the range the  MIT430/2MIT430/2 does have this capability So this was going to be a learning curve to see if further investment would be beneficial.

 

This will be the most onerous test conditions for the  MIT420/2MIT420/2 that I can offer and will test multiple aspects of its functionality

 

1) The variable insulation test range will be utilised to prevent over stressing of the rotor winding

2) The resistance range will be utilised to verify test connections during recurrent surge oscillograph tests

3) The voltmeter function will be used to measure AC volts during AC Impedance testing

4) The voltmeter function will be used to measure DC volts during winding resistance tests

 

Both the generator rotors tested are of a brushless design, with a rotating diode pack that must be disconnected to allow the insulation tests to be carried out. The rotor winding is, therefore, a DC rated component and considering that it would cost around £600,000 and take 3 months to rewind a rotor, care needs to be taken when applying test voltages to them, especially as the units age and become more contaminated with dirt.

 

Rotor Winding Insulation Test

 

For the 145MVA unit the rotor winding is designed to work at a maximum of 360V DC under short circuit conditions of the stator output. The  MIT420/2MIT420/2 has nominal test voltages of 250V and 500V. If I use 250V then I am not testing the rotor under its worst operating conditions. If I use the 500V range then potentially I am stressing an old winding beyond any voltage level it is likely to see in service. The 92MVA unit has a maximum rotor voltage of 143V. The  MIT420/2MIT420/2 offers nominal ranges of 100V and 250V around this value, with a similar scenario to the larger rotor.

 

The answer is to use the variable voltage function of the  MIT420/2MIT420/2 to set the output voltage to match the maximum in service conditions the rotor will see. This is the safest way I can test an old rotor winding, testing it to its full service voltage without over stressing it and risking damage. The variable voltage function defaults to 10V and is easily set to the required value using the settings function this also prevents inadvertent adjustment of the test voltage during use. Once set the range functions exactly like all the other insulation ranges with timed DAR and PI test setups and the ability to save any results obtained.

 

Throughout the tests on both rotor windings the  MIT420/2MIT420/2 performed without any issues, connections were easily made and the tests conducted. Unfortunately neither rotor was in too good a condition. The 92MVA rotor was tested on day 1 and had a low insulation value with no DAR. Overnight the heaters were turned on and the unit retested again in the morning. A doubling of the insulation resistance was seen but no DAR was obtained. The 145MVA generator was much worse and the insulation tester output voltage could not be maintained on the first test. No heating was available for this unit, so only a marginal improvement was seen the next day probably due to the lower ambient humidity. I hope to return to this rotor in the next two to three weeks after the heaters have been put back into service.

 

Rotor Winding IR Test Results

 

Rotor Winding DC Resistance

 

I do have a micro ohmmeter available to me to test winding resistance at 10A. The instrument however is not the best and will cut-out due to the time taken to allow the reading to stabilise due to the inductive nature to the winding. As an alternative, I have constructed a 100A variable DC supply and utilise a 4 wire methodology to measure the winding resistance at 5 different current values and then take an average as per the IEEE118 standard. The supply contains a 100A to 50mV current shunt that I connect a mV meter to, I then utilise the  MIT420/2MIT420/2 as a DC voltmeter connected directly across the rotor winding

 

DC Winding Resistance Test SetupDC Winding Resistance Test Connections

 

Both rotors tested reasonably well. For comparison on the 92MVA rotor, I repeated the test with the standard multimeter usually used in place of the  MIT420/2MIT420/2.

 

The  MIT420/2MIT420/2 produced a test result to -0.73% of the nominal DC winding resistance corrected to the specified temperature where as the usual test setup produced a result to -0.10% The specification is to be within 2 of the corrected value and so whilst the  MIT420/2MIT420/2 was not quite as accurate as the usual setup it still returned a result within tolerance

 

MIT420/2 DC Resistance Test ResultsU1282A DC Resistance Test Results

 

Recurrent Surge Oscillograph Test (RSO)

 

The RSO test is a specialist test on a generator rotor winding to detect inter-turn faults within the winding. It works on time domain reflectometry. Pulses are injected alternately from the two ends of the windings and the reflected waveforms across impedances connected to the windings are compared. If the two traces overlap, then no fault is present. A fault is seen as a deviation in one of the waveform traces.

 

RSO Test Principles

 

Details of rotor reflectometers and the test can be obtained from Rowdtest Ltd who manufacturer the units and provide the diagram above of the principle of operation.

 

For this test I utilised the  MIT420/2MIT420/2 to measure the resistance of the test connections to the winding and to earth For an ideal RSO test there should be a maximum connection resistance of 1 Ohm across the winding and to earth This would usually be measured using the ohms function of a standard multimeter but I was curious to see if the automatic function of the resistance range on the  MIT420/2MIT420/2 would operate correctly across the inductive load of a large rotor winding As seen in the photo below the instrument function perfectly well recording a 0.26 Ohm resistance reading through the winding whilst sourcing a 200mA current.

 

RSO Connection Test

 

Rotor Winding AC Impedance Test

 

The final test on the rotor winding is an AC Impedance test. This is an alternative test methodology to the RSO test for detecting inter-turn insulation faults. Where as the RSO test only applies 12V, in an AC impedance test, a voltage of up to 110V AC is applied across the winding, therefore applying more voltage across the inter-turn insulation. The test is carried out in the same manner as the DC winding resistance test, with the voltage across the winding and the current through it are measured at 10V steps and then the impedance calculated afterwards.

 

 

A good rotor winding will produce a plot of impedance against voltage with a slight upward trend. An inter-turn fault in a winding is seen as a step change down in the impedance along the plot. There is also the opportunity to carry out a long term historical impedance trend of the winding at a specific voltage, nominated as 50V during my tests.

 

AC Impedance Test Result Plot

 

AC Impedance Trend Plot

 

As can be seen from the plots above for the 92MVA generator comparable test results were achieved using the  MIT420/2MIT420/2 in place of a standard multimeter. Again the instrument has proved that it can be utilised in place of a regular voltmeter for the kind of tests that I carry out increasing its value and flexibility.

Up to this point the on site tests carried out with the  MIT420/2MIT420/2 have been based around motor tests either from a control centre or direct at the motor terminals Unsurprisingly the instrument has performed very well with the only issues really being around making connections to specific items during the earth test on a large motor and when using the supplied test lead to short out the winding after the tests These are all minor issues and relatives easy to overcome

 

This blog introduces the testing400A Air Circuit Breaker of air circuit breakers (ACBs). These are large three phase, low voltage breakers design to switch high currents. Although only rated to 690V AC, due to their robust construction and the types of insulation utilised, we often test them at 1000V, which is a test range that has not been utilised out on site yet with the instrument.

 

The added benefit of the 1000V range on the  MIT420/2MIT420/2 is the it will measure up to 200GOhm resistance value which will be more likely to pickup carbon contamination from arcing as the breaker operates

 

If this was routine planned maintenance being performed on the ACBs then a suite of visual inspections mechanical measurements electrical and timing tests would be completed but as this is mainly for demonstrating the use of the  MIT420/2MIT420/2 I will only be checking that the main contacts have closed using the continuity resistance function and checking the insulation resistance from phase to earth phase to phase and across the open contacts using the 1000V range

 

As the type of insulation and assembly methods differs to that of a motor or a cable, a dielectric absorption or polarisation index is not applicable and only a 1 minute timed test is conducted, the reading will be expected to ramp up to its final resistance value within seconds, and stay there for the duration of the test.

 

I have two types of ACBs available to me to test. One is a relatively modern design and is more compact and includes a built in protection relay. The second one is a much more older design and sits in a much larger metallic frame with greater clearances. The principle of their operation however, is the same, and they are tested with an insulation tester in exactly the same manner.

 

The first aspect of testing an ACB, is to make it safe. Electrically, the ACB is isolated by removing it from its carriage where it can then be placed on the floor or a maintenance trolley. As the ACBs use a powerful spring to close and trip the main contacts, a significant amount of stored mechanical energy remains within the unit that is discharged by closing the breaker. I can then make the breaker safe by inserting a safety pin into the ACB frame to lock the tripping mechanism for carrying out the first set of insulation tests, to earth and between phases, with the contacts closed. This is done so that I am insulation testing the complete primary circuit of each phase and do not have to test each contact individually. Once these tests are complete, I can remove the safety pin and trip the ACB open. This will remove all stored mechanical energy in the ACB and the mechanism will be safe to work on.

 

The following video shows an older style 1600A air circuit breaker being tested. As you will see I had a slight issue with the crocodile clips on the Megger leads that would not grip around the primary contact, so I had to resort to using the leads and crocodile clips from another manufacturer. Other than this slight issue, the tests went well with all results at the 200GOhms maximum reading, showing that the insulation is in excellent condition and there is little contamination.

 

Crocodile Clip Connection Issues

 

A test on a 4000A breaker shows a different story, with a much lower reading being achieved, indicating that there was contamination across the contact in this breaker. With the primary contact having an even bigger diameter than a 1600A contact, I had to resort to the large crocodile clips from my Fluke 1555C insulation tester to make the connections, but the photo below shows an alternative arrangement using tinned copper wire to wrap around the contact and twist tight. The Megger crocodile clip can then be clipped to this which gives a perfectly adequate test connection for a voltage test with a low output current.

 

4000A ACB Insulation Resistance Test

 

This concludes the blog, the instrument has performed perfectly well and continues to be reliable and easy to use.

 

The next blog will be the final installment on testing a generator rotor winding.

In the previous blog some motors that had been refitted after maintenance were tested from their respective control panels I found that the  MIT420/2MIT420/2 had a few quirks whilst using it but nothing that became a major issue or prevented the instrument from being used In the next blog I am off to the other end of the cable and test out some motors at the actual terminal box Actually I am cheating a bit as I have access to motor stores so I tested a new 4kW motor a refurbished 90kW motor and two 132kW motors removed for utilisation at another plant In general testing at the motor terminals is much easier than testing at the motor control centre so I am not expecting any issues with the use of the  MIT420/2MIT420/2

 

The first motor is a 4kW star configured motor. This is e4kW Motor Test at Terminalsasily verified by the shorting strips seen across the three terminals at the top of the block. This means that I will connect the tester to the terminals at the bottom of the block.

 

The same test procedure used during the test from the panel are used. A phase balance test across each pair of phases and then a single insulation resistance test at 500V. As the motor is new and the store is heated, I am not expecting any issues.

 

4kW Motor Test Results

 

As the table above shows, the phase were well balanced and the 1 minute insulation reading was high with a dielectric absorption ration above the 1.2 minimum. With such a high 1 minute insulation value and the fact that the motor is relatively small, cheap and random wound, I would not progress onto carrying out a polarisation index test on this motor.

 

90kW Motor Tests

 

So onto the next motor. This a much larger 90kW motor. Looking at its terminals, it can be seen that the shorting bars are now going across the block indicating that this motor is actually connected up in a delta winding configuration. This is typical for larger motors that are predicted to have long running cycles as the motor runs more efficiently, but takes more power to start. The testing principle is still the same with the phase balance measured across either the top set or bottom set of terminals and then a single insulation resistance test to earth.

 

The insulation test was carried out as both a DAR test and then a PI test to record both values.

 

As this motor is a much larger motor a much lower phase balance reading is obtained unfortunately I got caught out with the retention of the lead null and the initial readings were 0.01Ohms The lead null was reset and then the readings repeated with 0.04Ohms now being obtained to verify the accuracy a phase balance reading was also taken with the Keysight  U1461AU1461A which came out as an average of 0.053 ohms with the lead resistance subtracted Given that this is such a low resistance reading with only a two wire method the results seem comparable to me

 

Again a good 1 minute insulation value was achieved, however, the DAR value was a little low if going by recommended guidelines. Reviewing the PI ratio result, also indicates a low value, which ideally should have been above 2.0. However, as the 1 minute insulation value is high, the DAR and PI values are not as important and this motor is still acceptable for installation should it be required. This motor is also random wound design and may also be impacting on the PI ratio value.

 

The insulation readings were recorded manually at 15 second intervals for the first minute and then at 60 second intervals for the remaining 9 minutes to produce the insulation resistance plot seen next to the results table. This shows a rapid rise of insulation resistance over the first minute that then flattens out and shows very little rise.

 

 

90kW Motor Test Results90KW Motor Insulation Plot

 

 

Music Credit: 'Roadtrip' by Nicolai Heidlas Music

 

The final pair of motors are part of an old vacuum pumping system that has been decommissioned and removed from one site to another for potential installation. The preservation testing of these motors are more critical, as they would have seen deterioration from use due to thermal, electrical and mechanical stresses during operation, with also the potential to have suffered from dirt ingress over the years. These are also in an non-heated store and therefore have the potential for lower resistance values.

 

132kW Motor testsAt 132kW rating, this is a much larger motor would be powered from a star / delta starter and not a direct-on-line starter. As a consequence no shorting bars are installed across the motor terminals, meaning that each phase can be tested individually. There is also the added bonus of an anti-condensation heater, that can also be tested.

 

An expected resistance value for the anti-condensation heater can be calculated based on the rating on the nameplate which was 487.5 Ohms, this can be compared to the measured value and was found to be a good match for both motors. The insulation resistance of the anti-condensation heater is then measured at 250V with all three phases of the motor shorted to earth, as the heater is actually a cable taped to the windings so could short direct to earth, or could short to one of the motor windings. Therefore, shorting the motor windings to earth saves testing the heater against each phase individually.

 

The motor winding resistance is measured across diagonal terminals for each individual phase and was found to be balanced on both motors, but with motor A showing a slightly higher resistance value. In hindsight though, as with the 90kW motor, I may have been caught out by the instrument retaining the lead null setting from previous tests.

 

For me, the retention of the lead null settings is proving a bit troublesome. It is the only meter that retains the setting have changing functions or switching of the instrument. I am not sure what the value of this function is.

 

132kW Motor Phase Resistance

 

The dielectric absorption ratio tests were all initially conducted at 250V as the condition of the motors is unknown, so a lower test voltage is used initially to reduce the likelihood of damage to the winding if it is in a poor condition. As can be seen in the tables below, all the motors had relatively low insulation values and poor DAR.

 

As W Phase on motor B had the best insulation value, it was decided to increase the test voltage on this phase to see if the insulation value reduced significantly which would be a further sign of deteriorating insulation. Not much of a reduction was seen, 60MOhms at 500V in comparison to 61MOhms at 250V.

 

As the motors displayed no DAR ratio it was pointless proceeding to a PI ratio test and it was decided that the best course of action would be to power up the anti-condensation heater on one of the motors and leave it for a couple of weeks before returning to carry out another test and see if an improvement in insulation resistance is obtained I plan to return later November December to recheck the motors to see if the insulation resistance has improved I will also take the Keysight  U1461AU1461A with me to verify the phase resistance readings

 

Vacuum Pump A Test ResultsVacuum Pump B Test Results

This concludes the blog on testing at the motor terminals There were no major concerns found with the use of the  MIT420/2MIT420/2 all tests were conducted safely and effectively It was found that whilst clipping to the motor case to prove the earth connections the crocodile clip would sometime be at its maximum opening and didn't grip very well and I did get caught out with forgetting to reset the lead null function

 

In the next blog I will move on to carrying out insulation tests on air circuit breakers.

As a Roadtester for the Raspberry Pi 3 & MathWorks Learn-to-Program Pack, I've encountered some bumps in the road due to the MathWorks documentation and examples being somewhat out of date.  I'd intended to blog earlier as I expect the other roadtesters have probably also encountered similar issues.  I'll do a full discussion in my review, but in case it might help anyone - the Raspberry Pi support packages for Matlab and Simulink in r2017b do not currently support Raspbian Stretch.  It will give you a build error when you try to create a raspi object.  I ran into this issue because I setup my RPi3 before I tried to install the Matlab software.  The RPi3 hardware in the kit comes with a 16GB micro SD card with NOOBS pre-installed.  All this works great, but when you install Raspbian using NOOBS it will install the latest version of Raspbian which is Stretch.  The Matlab hardware support package allows you to modify an existing OS or to write a new OS image onto an SD card.  Since I had already configured the RPi3 (ssh, wifi, ipaddr, etc), I chose to modify my existing OS.  At that point I couldn't create a raspi object to connect to my RPi3, so I couldn't continue.  I tried building a new OS image and that actually worked because the image was built with Raspbian Jessie Lite.  To their credit, MathWorks support did provide me a workaround that allowed me to run Matlab on Stretch.  I am waiting for a similar workaround for Simulink.

 

I also have encountered a problem specific to my application in that Matlab's IP camera module does not support H264 streaming only MJPEG streaming.  Unfortunately, the specific camera that I'm trying to use only supports H264.  I'm trying to get a third party module, HebiCam, to work.  I'm getting errors rtsp streaming with H264.  I'll post again if I get that working.  Otherwise, I may have to change the project that I was going to do for the roadtest.

 

Good luck to all the other roadtesters.

The bench tests and desktop review have all gone well for the MIT420/2 Insulation Tester. In the next blog, I carried out some phase balance and insulation resistance tests from a low voltage (400V) motor control centre (MCC) for a gas turbine. The motor control centre is a large panel with two main feeders and then split down into multiple direct-on-line starters for various pumps and fans that the gas turbine requires to operate. The motors had been removed for maintenance work and then refitted and reconnected, so a quick test of the motors was required to show that they were safe to reinstate ready for a direction of rotation check.LV Motor Control Centre

 

In all three motors were tested;

  1. A 415V, 30kW Exhaust Ventilation Fan Motor
  2. A 415V, 37kW Cool Air Fan Motor
  3. A 415V, 45kW Exhaust Frame Cooling Fan Motor

 

Each motor has two tests carried out;

  1. A phase balance test using the resistance / continuity function of the MIT420/2
  2. A 1 minute insulation resistance test using the 500V range on the MIT420/2

 

The objective of the phase balance is to ensure that all three phases of the motor have been connected correctly and there is not cable damage or loose terminations.

 

Obviously the insulation test is done to ensure that there are no earth faults on either the cable or the electric motor.

 

As the motors are all fed from direct on line starters, the phase balance test is carried out first to ensure the motor connections are good and then a single insulation resistance test is done, as the phase balance test has shown that the phases are all interconnected.

 

The first motor control centre is a swing-out direct-on-line starter used for smaller motors. All of the controls are mounted onto the door frame and I gain access to the motor contractor by opening up an inner door. This makes measuring the phase resistance a little bit of a challenge as the probes have to be held in one hand across the contractor terminals whilst pushing the MCC apart to gain access. Mean while the instrument is held in the other hand. The meter will read the phase resistance automatically, but pressing the save button was more of a challenge as the meter is held in my right hand, I cannot reach across to the save button. As the panel is 5 foot from ground level I cannot rest the instrument on the ground. I would have really benefited from a magnetic style loop accessory so I could have left the meter hooked up onto the next panel to release my right hand to help take the measurements and operate the save function when required. The instrument looks like it it has a fixture on the back for such an accessory, but I have not been able to find one yet.

 

30kW MCC Swing Out Starter PanelPhase Balance Test in 30kW MCC Starter

 

The 1 minute insulation test went much easier as one of the crocodile clips could be used to clip to the metal chassis of the MCC. As the instrument has the built in dielectric absorption ratio timed function that could be set up before applying the leads, it was an easy one handed operation to operate the instrument. On insulation testing, the reading is also stored on screen at the end of the test, so, unlike the phase balance test, the save button could be pressed without the probes connected. Had I wanted to carry out a polarisation index test that would have required manually writing down the readings every minute, I would have had a very hard time without having an assistant or utilising some other form of connection for the leads.

30kW MCC Starter Insulation Test

 

The test results can be seen in the table below. The phase balance was stable and equal across all three phases. The one minute insulation value was 5.8GOhms which is excellent for a combined motor and cable test of all three phases together.

 

The dielectric absorption ratio was 1.36 which is above the minimum threshold value of 1.2.

 

The tests show that motor was connected correctly, with no earth faults and was ready for a run to prove the direction of rotation.

 

 

30kW Motor Test Results

 

The next MCC panel was a fixed starter panel and slightly easier to work in than the swing-out starter. This type of panels are for slightly larger motors with direct-on-line starters and the largest motors with star-delta starters. The same test procedure is utilised, however ad the contractors in this type of starter have unshrouded connections, the crocodile clips can be utilised to make the connections which removes all of the fumbling around when trying to use the probes in one hand and operate the instrument in the other.

 

37kW MCC Starter Phase Balance Tests37kW Starter Insulation Resistance Test

37kW Starter Motor Test Results

 

The phase balance test results from this motor were unsteady. This sometimes happens when fan motors are tested and the fan is free-wheeling in the air flow turning the motor. This results in small voltages being induced into the motor stator windings that can prevent an ohmmeter from making reliable readings and is not a problem with the instrument. This can be seen in the high U-V reading compared to the other two phases. A bad connection would have given a steady high reading and not a reading cycling from low to high.

 

As with the other motor, the insulation test readings were very high and the dielectric absorption ration above the minimum accepted value.

 

The final motor to be tested was a 45kW motor, this was also a fixed starter type panel and the leftist motor of the three to be tested. For this motor I also decided to carry out a polarisation index test to test this aspect of the MIT420/2 that had not been tested on the other two motors.

 

For this test I tried to make a video, but I apologise for the audio quality, as it was drowned out by the switchroom and panel fans, so I have tried to do some voice over which doesn't explain things quite as well.

 

 

There were no problems carrying out the tests using the MIT420/2 as the crocodile clips could easily be used with the open terminal style contactor installed. At the end of the PI test, it was noted that the instrument had gone over the 100GOhm resistance limit for the 500V range, so internally no PI ratio was recorded. For the purposes of the data table and the insulation resistance plot, I used the 100GOhm figure.

 

45kW Motor Test Results45kW Motor Insulation Resistance Plot

Despite going up into high GOhm resistance, the test results clearly show DAR and PI ratios, this is likely to be due to testing from the MCC that includes the motor and the cable which the latter would generally exhibit a capacitance trait when connected to a DC supply.

 

This completes this blog, the sensible thing now seems to go to the other end of the cable and test out some motors on their own.

Hello and welcome to this tech review of E36312A- A triple output programmable bench power supply from Keysight Technologies.

 

E36312A

 

Recently, I won a triple output bench power supply from Keysight Technologies.

It’s been more than 3 weeks I’m exploring this bench power supply and this blog post reveals my experience with this product so far.

 

Let’s start with the appearance and the front panel of the supply. It has a nice and big colourful LCD, 2 separate control knobs for

voltage and current adjustment, a keypad to enter the desired values of the voltage and current,

colour coded push buttons for each channel along with colour coded outputs (yellow, green and blue) and

few other push buttons for quick actions like tracking and print screen, meter view etc.

It also has a USB port for data logging and screen shot purposes. Overall the build quality is really good.

 

The three outputs are independent of each other, output 1 gives 6V, 5A and other two outputs give 25V, 1 A each.

Each output is colour coded and hence is easy to distinguish between different outputs. Each output has a independent

colour coded buttons to switch on/off and a master switch for all outputs.

 

The unit powered on and the LCD showing the outputs which are colour coded

 

And now the most amazing feature of this power supply- You can combine the outputs of port 2 and 3 in series or parallel just by pressing a button!

It has internal switching which uses relays. You can get a max of 50V output at 1A from output 2 by using series connection and a max of 2A current

at 25V by combining port 2 and 3 in parallel.

This feature is not available in any of the bench power supplies I have used till now.

 

I have used programmable power supplies from multiple vendors in the college labs, this being my first owned bench power supply.

It surely beats all other bench power supplies in the market.

 

Overall, the build quality is great. The LCD visibility is too good. Again a feature not available in most of the bench power supplies in the market.

It also has data logging capabilities and a meter view which displays the voltage, current and power measurement of the device under test.

 

Meter view showing the voltage, current and power consumption of the DUT

 

I would highly recommend this bench power supply to research labs, universities and makers. It is a one time investment but worth it.

Really helpful for testing your electronic designs.

The only downside of this product is that is a bit costly for individuals.

Hi Roadtesters and element14 members,

 

This is a study one of our suppliers is conducting regarding test and measurement equipment.

 

It's a short survey regarding test equipment suppliers you are familiar with. Survey's estimated amount of time to complete is around 15-20 minutes.

 

If you complete the survey, you will be entered for a chance to win one of two $1,000 Amazon e-Gift Card. Sounds like a good deal for a few minutes of your time.

I was told that the information your provide will be STRICTLY CONFIDENTIAL and used only in combination with answers from other respondents.

 

 

Click to go to the survey

 

Sincerely,

 

Randall Scasny

RoadTest Program Manager

The last two blogs covered the bench tests applied to the  MIT420/2MIT420/2 and found that it performed better than expectations with all values obtained significantly within the manufacturer's specified tolerances However working on high energy electrical circuits has many safety implications so before I take the  MIT420/2MIT420/2 out into the field to do some real work I wanted to take a quick look at some of those safety features

 

As standard with most electrical instruments these days the terminals on the  MIT420/2MIT420/2 are 4mm safety socket designs with compatible plugs on the leads to prevent any inadvertent contact with the output of the instrument I also like the location of the voltage function in between the off position and the insulation test functions Merger make it absolutely clear that it does not constitute as an appropriate testing for dead procedure and I fully agree with that But for me it adds an extra safety feature as pausing on the voltage function allows for a last double check to ensure that no live voltages are present before moving to an insulation testing function or that the circuit has been discharged when switching back having completed the insulation testing

 

MIT420/2 Safety SocketsMIT420/2 Function Switch

Some more safety features whilst using the  MIT420/2MIT420/2 for insulation testing are the extra warning beeps before a 1000V insulation test is started two handed operation of the test lock function and the cut-out function that prevents the starting of an insulation test if a live voltage is present These features are demonstrated in the short video below

 

 

The remaining safety feature is the internal 500mA HRC fuse that I looked at during the teardown. There was a PCB track that bypassed the fuse via a 10MOhm resistor so I decided to remove the fuse from the meter and see what the response was.

 

The outcome was impressive, the fuse protects all functions within the tester except for the voltage function which still measures a voltage present even with the fuse removed. On switching to the resistance or capacitance function, a small flashing fuse symbol is displayed and the instrument reads over range.

 

On switching to an insulation testing function nothing is immediately obvious The test button is pressed and the meter initially responds but goes over range and no test voltage is displayed After a couple of seconds the function cuts out and the fuse symbol is again displayed However if a live voltage is present when on an insulation range with the fuse removed the cut out function still worked and the instrument alarmed with the voltage indication Pretty impressive The next video shows the operation of the  MIT420/2MIT420/2 with the fuse removed

 

 

Overall Megger have made an excellent job of adding safety features to the  MIT420/2MIT420/2 it is evident that a lot of thought has gone into the design of this instrument

 

With the instrument review and bench tests all completed, it is time to go and find some work for the instrument to do and see how it preforms out in the field.

Hi Road Test Community,

 

I have a fully-featured Rohde and Schwarz RTE1204 oscilloscope on my hands at the moment, and I thought there may be some interest in reviewing it in the style of an official Road Test. I have been a somewhat passive member of this community for a few years now, so hopefully this will allow me to hone my review skills and provide a unique look at this professional-quality oscilloscope.

 

A bit of background: I had ordered a Rohde and Schwarz RTB2004 COM4 scope back when their amazing launch deal was on, but unfortunately there was an error in the processing of my order. Luckily, the folks at Newark and Rhode sorted it out for me, and in the meantime provided this to me as a loaner scope to allow me to complete an ongoing project. I am not being compensated in any way for this review -- it is all on my own accord, and I will be candid when discussing things I like and do not like. I have used this scope in a professional setting alongside similar Keysight 4000X scopes, so I hope I can bring a realistic user perspective.

 

RTE1204 Front DisplayFig 1. Front view of the RTE1204 Scope, showing one of the two logic probes connected via the rear

 

I propose to complete this review in three initial parts, with a fourth and final section focused on questions and tests requested in the comments.

      • Part 1: Introduction. From a user perspective, I will assess this oscilloscope in terms of the typical metrics:
        • Specs and comparison to other scopes in the same category. Explore what makes this scope unique in the marketplace from its data sheet.
        • Physical and Graphic user interface. What is it like to sit in the drivers seat? Is the button/knob useful and intuitive? Same for the software and GUI -- customization, responsiveness, etc.
        • Misc: noise, form factor, display, build quality and accessories.
      • Part 2: Putting the scope through its paces (analog)
        • Analog channel analysis of various signals (sine, modulated, PRBS, etc.). Measurements, markers, math functions, and different scope settings (including 16 bit "HD" mode, and various built-in filters). I will use some 1+ GHz sources to do eye diagrams and test masks. Different triggering options
        • Spectrum analysis (FFT). Determine the usefulness of the FFT function of this scope compared to an actual spectrum analyzer. Look at THD of an amplifier, spectrogram of a modulated signal (maybe as a signal monitor for a UHF receiver), and an EMI test utilizing the gated FFT functionality.
      • Part 3: Putting the scope through its paces (mixed-signal domain)
        • Break out the logic probes and decode some serial data.
      • Part 4:
        • Remote control and automated measurements
        • Remaining tests suggested in the comments
        • Conclusions

 

I will also post a separate review of my RTB2004 when it arrives, and compare it to its "big brother" RTE1204. It seems like there are a lot of reviews already floating around, so I will hold off on this until after the promised firmware update from R&S.

 

I will be actively checking the comments, and I am happy to update my plan or add to the reviews with any suggestions or ideas (I will surely miss a thing or two). I should mention that I unfortunately do not have access to high-speed active or current probes, which prevents me from testing some of the power analysis features.

 

With that, let the "Rohde Test" begin (sorry... couldn't help it).

 

-James

 

Edit: Dec 5, 2017: Added table 1 to link to other sections

 

Table 1. Table of Contents for review

Rhode and Schwarz RTE 1204 Review SectionLink
Part 0: Review ProposalRohde and Schwarz RTE1204 2GHz Oscilloscope Review: part 0
Part 1: Introduction and User InterfaceRohde and Schwarz RTE1204 2GHz Oscilloscope Review: part 1
Part 2: Analog and FFT Functionalityto be added upon completion
Part 3: Mixed Signal Domainto be added upon completion
Part 4: Conclusion and User Requeststo be added upon completion

Part A of the Bench Tests reviewed the performance of the auxiliary functions on the insulation tester and found them all to be well within the specified tolerances. In this second part of the Bench Tests, I will concentrate exclusively on the insulation testing function.

 

The MIT420/2 has five fixed voltages of 50V, 100V, 250V, 500V and 1000V and then a variable output function from 10V up to 1000V. This is not a voltage ramp function, the output voltage is set in the settings menu to the desired level and the function switch turned around to the variable voltage function to initiate the test. A desired change in the voltage level means that the user has to go back to the settings menu to adjust the voltage and then back to the variable voltage function again to carry out the test. The instrument will remember the last voltage setting even after it has been switched off. However, it reverts back to the default 10V if the batteries are removed.

 

For the purpose of testing an insulation meter, I have a bespoke insulation tester calibration box from Time Electronics. I carry this box around with me when I go to a site to verify that the insulation tester is function correctly before carrying out the tests on generator rotors, mainly due to the value of the rotor and the costs involvMIT420 and Calibratored in carrying out the tests. This instrument has the ability to measure the open circuit voltage, the short circuit current and then apply a range of high voltage resistors across the instrument via 4 decade switches to verify the resistance readings from 10MOhm up to 90GOhm. I also carried out an extra test to verify that each insulation test range could output a 1mA current and maintain the specified voltage level as specified in the BS EN 61557 standard.

 

The calibrator has 4mm safety sockets to plug the insulation tester leads into, but unfortunately due to the ridge on the Megger leads, I had utilise an uninsulated 4mm adapter to allow them to connect to the calibrator. Nothing more than a slight inconvenience, but one example of the disadvantage of the test lead approach by Megger.

 

Test Lead Adapter

Insulation test of leads

 

Despite this, the leads offered by Megger are of excellent quality and to show this I set up the calibrator to an 80GOhm load and tested the insulation tester against this using both the Megger test leads and the test leads I have from a multimeter. These leads are rated at 2000V but are manufactured from PVC and a lower quality.

 

The leads were set up twisted together, which is a poor test set up, but is being done to illustrate the quality of the Megger test lead set. The insulation tester was set up to 1000V for each test.

 

With the Megger test lead set a reading of 80GOhm was achieved and was found to be stable. With the PVC set, the reading was found to be much lower and quite erratic. It ended up as 6.9GOhm which is a significant difference to the 80GOhm setting on the calibrator. This kind of behaviour would seriously impact on the DAR and PI ratio tests the MIT420/2 is capable of carrying out.

 

The outcome is, if you use alternative test leads to the Megger set, ensure they are good quality silicone leads and pay attention to the test set up.

 

On to some insulation testing! The MIT420/2 has a base tolerance of +/-3% +/- 2 digits with an additional % tolerance per MOhm or GOhm dependent upon the resistance value and the range. To test the insulation accuracy each range of the instrument was tested against each decade value. The table below details the results found and the chart summarises the range of deviation seen from the actual resistance applied. All readings found to be significantly better than the base tolerance values with an overall deviation of +1% to -0.5%.

 

Insulation Test ResultsInsulation Test Deviation Range

 

The next few tests were based around determining the performance of the measurement process for the insulation resistance. The first aspect was to measure the open circuit voltage of the instrument for each testing range. To do this, the output of the MIT420/2 was connected to a high voltage probe with a 1GOhm input resistance. This is done to reduce the load on the output of the insulation tester and allow the voltage to rise to its highest. If the output of the MIT420/2 was connected directly to a DC multimeter or bench meter it would have a 10MOhm load applied to it, that would increase the current being drawn and lower the output voltage.

 

The manufacturer specifies the voltage output to be between 0% and +2% + 1 digit. The results in the table show that the voltage output remained within specifications. The first test was carried out using the Time Insulation Calibrator but as the voltmeter on this only has a tolerance of 1%, the tests were repeated with the HV Probe and a bench meter that gave an overall tolerance of 0.15%. It was interesting to see that the instrument drifts further from the nominal voltage values at the lower end, but tends towards a tighter tolerance as the output voltage is increased. This is typical of insulation testers I have used over the years.

 

Open Circuit Voltage Test Results

The circuit voltage tests are then repeated with a 1mA load applied to each range. As can be seen in the table below, the increased load on the output reduces the voltage closer to the nominal values but at no time did the voltage drop below the nominal values.

 

Voltage Output Tests with 1mA Load

The next test was to verify the short circuit current capability of the unit. The standard permits a maximum output current between 1mA and 15mA, typically instruments of the type have an output of around 1.5mA and Megger specify the short circuit current to be between 1mA and 2mA. The tests conducted showed this to be the case. Whilst carrying out the tests, the uA-s-V function switch can be used to cycle the secondary display through to read the output current.

 

MIT420 Short Circuit Test MKT420 Short Circuit Test Results

 

I then carried out a response test of the output by monitoring the voltage with a Picoscope 3404A via a high voltage probe. Due to the input voltage limitation of the Picoscope, the measurement was made via the high voltage probe, utilising a custom probe setup within the Picoscope software to create the correct voltage range. This test was done more out of interest rather than determining if the meter was performing to a specification. The results can be seen in the table below alongside a photo of the setup utilised.

 

MIT420 Output Response Test ResultsMIT420/2 Output Response Test Setup

The rise time of the output varied dependent upon the voltage range, except for the variable output range that was tested at 390V and then 760V with both returning a comparable rise time. The fall time on each range was similar each time, between 20 to 25ms. A sample plot can be seen below on the left that shows the voltage rise as the test button is operated. A small overshoot is occasionally seen, but it is small and causes no concern. The plot on the right shows the ripple of the output by zooming in on the waveform.

 

MIT420/2 Output Voltage PlotMIT420/2 Output Voltage Ripple

 

 

The final element of the test was to verify the accuracy of the timed tests offered by the MIT420/2. Three types of timed test are offered, the first is a simple duration test that is set between 1 to 10 minutes in 1 minute intervals in the settings menu. There is no duration shorter than 1 minute, which could have been useful for testing portable appliances or simple checks on smaller motors and transformers.

 

The second type of timed test is a dielectric absorption ratio (DAR), this is the ratio of the 60 second value divided by the 30 second value and is a representation of the quality of the main insulation as the applied voltage charges the circuit similar to a capacitor. A good insulation system will have a DAR ratio above 1.2.

 

The final timed test is a polarisation index (PI) which is the 10 minute value divided by the 1 minute value. This is more of a measurement of leakage current over the insulation surface mainly affected by moisture and dirt. For a Class F insulation system a ratio value above 2 is acceptable.

 

Not all apparatus will have a DAR or a PI ratio, it depends on the type and construction, hence why Megger have added a starlight timed test function. The IEEE 43 standard for AC and DC dielectric testing also advises that when a 1 minute insulation value exceeds 5GOhm, the DAR and PI values may not be representative of the insulation condition.

 

To measure the time function, I utilised the variable voltage test function, setting it to its minimum of 10V. This was applied across the input of a counter / timer unit set to the interval function. The manufacturer, does not specify a tolerance for the time function, so this was also done more out of interest. The values determined were all within 1% of the nominal value. Interestingly, the timed function gave values less than the set duration where as the DAR function always gave a time duration slightly over the 1 minute. Due to the instrument operation and the test setup on the DAR and PI tests, only the final 1 minute and 10 minute values can be recorded.

 

MIT420/2 Timing Test SetupTiming Test Results

 

This concludes this blog on the insulation resistance bench tests. Throughout, the instrument performed flawlessly, it gave accurate results and had a fast response. The display is crisp and clear to read. Adjustment of voltage output and timing durations was easy to carry out and switching them to the setting menu means that they would not get adjusted easily by mistake.

 

In the next blog, before I take the instrument out into the field, I will review some of the safety features that have attracted my attention.

Filter Blog

By date: By tag: