|Product Performed to Expectations:||10|
|Specifications were sufficient to design with:||10|
|Demo Software was of good quality:||8|
|Product was easy to use:||10|
|Support materials were available:||10|
|The price to performance ratio was good:||10|
|TotalScore:||58 / 60|
RoadTest Review: Keysight Technologies U1461A Insulation Multimeter
In a predictable manner, I must begin by thanking Keysight Technologies and element14 for choosing me to be the RoadTester of another wonderful product. It's not often that I get "back to back" RoadTests, even less so major "meaty" ones, but that also means that it's been extremely busy around here. I prefer to get the review presented as a finished, monolithic "chunk" rather than in parts, so I wasn't able to get this one out earlier and have been beaten to the punch. No shame in that though, as despite it being conceptually simple to test, there is more than meets the eye! And we technically have two months (i.e. for me, until 27th April) to do our review.
As usual, I'll be doing a thorough job of testing and speaking my mind, with the focus being on the meter's adjustable voltage insulation testing and software connection capabilities, as well as verification of the meter's abilities. I hope you enjoy this review, and please feel free to ask questions, leave comments, ratings, likes or visit me at my personal blog.
Please note that careful reading of the EULAs associated with the mobile software appears to forbid use of the software for competitive, or, evaluation purposes without the consent of Keysight Technologies. It is my understanding that their provision of the product under the RoadTest program supersedes the EULA and endorses evaluation of their software for review and comparison purposes.
If the Keysight Technologies branding seems new and unfamiliar to you, don't worry. This is just the new branding for Agilent Technologies. You can expect the same reliable and dependable designs and performance from the equivalent Keysight Technologies branded product - in many cases, it's just a colour scheme and branding change. But since they're "in the middle" of the change, you will find some inconsistencies, especially amongst old inventory and accessories. I will also apologize in advance if I "mix up" the brands during this review, as I'm inevitably going to do at some stage.
The product for this RoadTest is the U1461A Insulation Resistance Testing Multimeter. This multimeter is a combination of an insulation resistance tester (commonly known as a "megger") and a field service technican's multimeter. This unit boasts 4.5 digit read-out, IP67 ingress protection with 3m drop-resistance, a bright OLED display and computer connectivity. Will this be the new standard for field technicians? Let's find out!
Insulation resistance testing is a critical part of ensuring the safety of electrical equipment. Detecting insulation degradation before failure allows for preventative maintenance that could save a lot of money and improve operator safety. Testing of hand-held portable electrical equipment is also a requirement for occupational health and safety reasons. As a result, having this functionality in field service is desirable.
Not having owned any serious insulation resistance testing gear, I will sadly have to compare it with my "value" Tenma Insulation Resistance Tester, which is a very basic unit. However, the U1461A is more than just an insulation resistance tester, and there are many fascinating things to explore - both on paper and in reality.
In summary, the Keysight Technologies U1461A Insulation Multimeter can be thought of as a 4.5 digit multimeter, targeted at field usage for service technicians, coupled with a flexible, configurable, insulation resistance tester. In reality, it is more than that, as the specifications show.
Key features include:
- 50/100/250/500/1000V insulation test voltages, but configurable from 10v to 1.1kV in 1V steps.
- Earth Bond testing from 6 to 60kohms.
- Up to 260 giga-ohms resistance test range.
- 4.5 digit OLED display with 66,000 counts.
- Timed, PI and DAR test.
- Calculation of Cable length by Capacitance.
- Auto discharge of capacitive circuits under test.
- AC/DC Voltage, AC/DC current, Resistance, Continuity, Capacitance, Diode-Test, Temperature features.
- True-RMS AC.
- Low-pass Filter.
- Non-contact AC voltage detection.
- IP67 certification with 3m drop-proof rating.
- -40 to +55 degrees C operating temperature range.
- CAT III 1000V/ CAT IV 600V safety ratings.
- Test inhibit on Live Circuit.
- Remote testing and report generation capability.
- 3 Year Warranty
- 50-60-80 hour battery life at high, medium and low brightness.
- Configurable screen update rates, with slow (1-5Hz) and fast (selected ranges, up to 40Hz).
The list of features on this unit are quite rich. It seems to have been designed with service technicians in mind, in terms of the measurements and ranges offered, and the physical construction. The safety ratings of CAT III 1000V and CAT IV 600V allows for working right up to the power grid. The battery life is a little short, compared to the 270 hours claimed for the LCD-based units, and is a consequence of the OLED display which provides a brighter, wider-viewing-angle display. The incorporation of non-contact voltage detection is a good safety tool, and the fast screen updates make for better visualization of transient events.
However, a feature-rich meter isn't much use if it's not competent at the measurements. A quick summary of the accuracy specifications puts those fears to rest (refer to datasheet for full details):
- DC Voltage (all ranges): 0.09% +/- 1 digit
- AC Voltage (45-65Hz): 1% +/- 3 digits
- DC current (600uA-440mA ranges): 0.2% +/- 1 digit
- AC current (600uA-440mA ranges): 1% +/- 2 digits
- Resistance (600R to 600kR ranges): 0.5% +/- 2 digits
- Resistance (6MR range): 0.8% +/- 2 digits
- Resistance (60MR range): 1.5% +/- 3 digits
- Diode Test (0.21mA): 2% +/- 3 digits
- Capacitance Test (all ranges): 1% +/- 2 digits
- Frequency (99.99Hz to 99.99kHz ranges): 0.02% +/- 1 digit
- Duty Cycle: 0.3% per Khz + 0.3%
- Temperature: 1% +/- 1 degree C
- Insulation Resistance (50v to 250v ranges): 1.5% +/- 5 digits basic accuracy
- Insulation Resistance (500v to 1000v ranges): 1.2% +/- 5 digits basic accuracy
- Earth Bond Resistance (6R range): 0.5% +/- 20 digits
- Earth Bond Resistance (60R to 60kR): 0.5% +/- 2 digits
- Adjustable DC Test Voltage: 0.5% +/- 1 digit.
The DC, temperature and frequency specifications of the meter are superb for a hand-held, having compared specifications a while back because of a desire to upgrade to a better multimeter. The AC specifications and resistance specifications are generally match those claimed by the competition. This makes for a very competent multimeter, even without considering the insulation resistance test capabilities.
When compared to competing insulation resistance testing multimeters, such as the Fluke 1507 and the Fluke 1587, the Keysight U1461A beats them pretty much hands down.
For example, the Fluke 1587 claims:
- AC Voltage: 2% +/- 3 digits
- DC Voltage: 0.09% +/- 2 digits
- AC Current: 1.5% +/- 2 digits
- DC current: 0.2% +/- 2 digits
- Resistance (except 50MR range): 0.9% +/- 2 digits
- Resistance (50MR range): 1.5% +/- 3 digits
- Frequency: 0.1% +/- 1 digit
- Capacitance: 1.2% +/- 2 digits
- Temperature: 1% +/- 1 degree C
- Resistance Test (50v and 100v ranges): 3% +/- 5 digits
- Resistance Test (250v to 1000v ranges): 1.5% +/- 5 digits
Even a dedicated insulation tester, such as the Megger MIT400, only claims an insulation resistance test accuracy of 3% +/- 2 digits in basic accuracy. Therefore, on paper, it seems the U1461A makes for a good unit, as a multimeter, and as an insulation resistance tester unit. It seems to have no compromises in the "combination" of the two units into one which makes for a more convenient compact device, and seems to have a nose ahead of the competition when it comes to accuracy. Price-wise, the unit also sits well when compared with the competition, making it an attractive product.
The unit arrived in a box just slightly bigger than your average shoebox. A warning about lithium batteries was plastered across the outside, indicating that there's some lithium batteries inside (what a spoiler!).
Opening the box revealed a plastic toolbox which is covered by a colour glossy card wrap-around detailing the unit and its features. Notable differences between the U1461A and the more basic U1453A include the lack of low-pass filtered AC voltage, which is useful when dealing with PWM style waveforms, and mV/uA measurements. A label indicating the calibration date and serial number is also attached.
The box is a nice protective plastic hard shell, with the Keysight logo on the side, but (not pictured) the Agilent logo on the slide-out latches. It is a tool-case, but doubles as transport protection.
Inside, the unit is still somewhat "packed" with protective film over the display, and all the probes and cables hidden behind pull out hard-door flaps.
Supplied documentation includes the Calibration Certificate, important notices and supplements, quick start guide, registration reminders, which are included also for the accessories making for a large number of paper-based items which you probably don't need to carry around with you at all times.
A piece of bad news for those in Argentina, Indonesia, Korea, Malaysia, Mexico, Peru, Philippines, Russia and South Africa - you don't get the U1117A IR to Bluetooth adapter due to what appears to be regulatory issues surrounding approval of the Bluetooth module.
The package is not big because it's empty and waiting on a future upgrade - it's filled with awesome accessories right from the get-go. We start with the U1173 USB to IR cable which is used to connect the meter to the PC for data-logging and remote control purposes.
There's the U1117A "Class 1" 100m Bluetooth to IR adapter which allows us to couple the meter to PCs, tablets and mobile phones for remote display, logging and report generation. This unit runs on two AAA batteries, which are included. It's nice to see that they spared no expense here, bundling the best type of batteries (lithium) which optimizes power for longer lifetime, weight and low temperature performance.
A full complement of thin and "fat" probes are provided, along with a remote switch probe adapter, connection cable and test clips. The safety category rating is clearly marked on all components, so do take care to check them before you decide on which combination of items to use.
A thermocouple kit with the regular K-type and J-type thermocouples are provided - both are specified for the same temperature range and resolution, so ensure you choose the right setting on the multimeter.
Of course, then we have the handsome meter itself, which is relatively chunky and solid in build. The unit is covered in a relatively thick and tight fitting orange rubber skin, which is very visible. An integrated robust kick-stand is provided, and direct attachment of the IR cable/adapter is provided without the need for a flimsy plastic adapter. Probe storage is also provided in the back.
The skin needs to be removed to insert the four AA lithium batteries. The batteries go into the top compartment, whereas the bottom compartment is for the fuse. The unit features rubber gaskets on all sections to keep its ingress protection rating.
Overall, the package is very complete, with almost all possible relevant options available packed in the box. You're not likely to be left wanting for anything in particular from the Keysight catalogue. The box does even have some space where you might be able to squeeze additional accessories for your application.
Interestingly, there's a "conflict" of name when the meter is first booted - namely the Keysight branding, but the Agilent logo in the firmware.
I think the package Keysight has thrown together for the U1461A is by far, a very complete solution, and a generous package. In fact, if I just look at most listings online, I wouldn't be aware that I was getting more than the multimeter itself. At the moment, the whole set is available from element14 Australia's site for AU$1088.
So let's play a game and try to price the included options. I tried to match all of the included items and options from the Australian element14 and Keysight Technologies' site, with the exception of an inexact match for the alligator clips included in the bundle. Then we can see what the bare cost that would be expected for the meter.
Catalog No. Description Price, Source
N/A U1461A Insulation Multimeter AU$???? (solve for x!)
N/A U1593A Protective Rubber Sleeve AU$33.00 from Keysight
N/A U1594A Hard carrying case AU$54.00 from Keysight
1440383 U1162A Alligator clips (red and black) AU$11.24 from element14
N/A U1169A Test Probe and Lead Set AU$34.00 from Keysight
19 mm probes (red and black)
4 mm probes (red and black)
Test leads (red and black)
N/A U5403A Remote switch probe and adapter AU$86.00 from Keysight
8760047 Four 1.5 V AA lithium batteries AU$36.75 from element14
2393547 U1173B IR to USB cable AU$33.75 from element14
2420930 U1117A IR to Bluetooth adapter AU$104.00 from element14
1263222 U1180A Temperature Accessory Kit AU$46.53 from element14
Thermocouple adapter (J/K-Type)
Thermocouple bead (J-Type)
Thermocouple bead (K-Type)
If we look at adding all the included options, the total cost of the options comes out to be AU$439.27, meaning the meter body is about AU$648.73. Of course, when it comes to the math of pricing options, it doesn't quite work out like that ... but don't be fooled - if you buy just a bare multimeter of a similar specification (or a multimeter and a separate insulation tester, but I doubt you will get quality) for about AU$650 and you had to buy options separately, you would end up spending the same amount.
Further to that, the Bluetooth remote connectivity for multimeters is a feature which seems to be unique to Keysight at this moment, with none of the major competitors offering such remote control and display facilities to such an extent, without relying on proprietary interfaces or expensive remote units.
I think it's clear that the price itself is actually quite good considering that it is all-inclusive of most options relevant to the meter. You aren't just getting a cardboard box with the meter in it!
The documentation under these pages are quite readable and comprehensive, which involve some application notes to help the less familiar users along with insulation resistance testing concepts. Most importantly, you should read through the User's Guide in the full form to familiarize yourself with the myriad of options and features in the meter, along with the multiple features each pushbutton can engage depending on the mode the meter is in. It is also important to use the technical support search for the options as well, as their documentation is provided separately - e.g. for the U1117A, to make sure you get the most of them.
Aside from this, you will probably also need to refer to any applicable local standards in reference to insulation testing to ensure you are complying with requirements where specified.
The U1461A was shipped with firmware version 1.09. The latest firmware available is version 3.02 which can be found under the Drivers, Firmware & Software tab, with the main change in regards to branding.
End user upgrade of the firmware is possible by downloading the utility and using the USB to IR cable to download the updated software onto the meter. The process begins by turning on the meter, and clipping the head of the USB to IR cable to the body. Then, you need to install the PL2303 USB to Serial cable drivers, followed by the update utility and then invoking it by finding its shortcut. The automatic scan seems to look at the wrong COM port by default, but after it times-out, you can re-select the port where your meter is attached and begin the flashing process.
The process must not be interrupted, otherwise you may cause your meter to be bricked. The process takes longer than most firmware updates I have done, taking several minutes to download the data to the meter and update it. At the end, your meter is automatically rebooted.
You should probably reset your meter settings to default and re-configure it to ensure the settings remain correct, but it appears that even without doing this, your setup preferences are maintained. After the upgrade, if you still have the meter greeting enabled, you will see the new Keysight logo instead of the Agilent logo.
The first thing you would want to do with a multimeter is to use it in "standalone" mode. After all, when it comes to troubleshooting issues in the field, it's probably the first tool you pick up. But not so fast - as with most modern multimeters, there's a few things you need to do first.
Step one is definitely read the manual. There's a lot of guidance and gems in there, which you will otherwise miss. With the assistance of the manual, you can begin the task of configuring the meter to your liking. For example, there are some power-on tests you can do to verify your meter is working properly - I checked my OLED display for dead pixels, and there were none, which was good.
Then you can access the large set-up menu to continue with the configuration. There are 11 pages of menus with options to tweak. This might seem scary at first, but this is actually much improved compared to previous LCD meters which used unintelligible abbreviations for options. Some settings you might want to change include the beep frequency (if you have several meters nearby) and or whether the beep is on or off, as it is very loud. In some cases, it's actually too loud, so I have to settle for it being off. A volume setting would be welcome. Another option you will likely want to change is the precision - as this is a 4.5 digit meter, it comes defaulted to only 4 digit display (which is probably larger and clearer, and more "definite" on value), but you're not seeing all the precision you paid for.
The automatic power off and screen brightness settings are probably also worth mentioning, as they can save battery, but also cause some annoyance. The automatic power off feature turns off the meter after a certain idle time even if the rotary switch is in a test position. If this is set too short, you might be annoyed that you have to reactivate the meter frequently, or you might not have caught the result of a step trip test and have to re-run it. Additionally, I have my brightness set to the lowest level to try and prolong the battery life, as it seems the OLED consumes more power.
The meter greeting settings affect whether the Agilent/Keysight logo and melody are played upon meter start-up. An interesting thing is the User setting which is actually reserved - we don't have tools to customize this. I decided to turn it off, but unfortunately, that doesn't speed up the several second meter start-up procedure. Instead, the meter performs its power on tests with a black screen, which is less intuitive.
In the end, my configuration looked like the following - note that these are not the defaults.
Once you have configured the meter to your liking, exiting the setup menu causes the meter to reboot. This happens even if you're configuring an insulation test time limit or voltage step configuration, which seemed counter-intuitive at first, but is just how the meter is designed.
The rotary selector can be turned to the left for "regular" multimeter functions, or to the right for insulation resistance tester features. The rotary selector has a bit of "sponginess" to it, and takes some force to actuate. Aside from the rotary selector, input is accomplished by a limited number of keys which have multiple features depending on short-press, long-press and mode combinations. This can actually get quite confusing to memorize, and the printed label "hints" sometimes aren't enough - so it's actually necessary to use the meter regularly to get used to it.
The output is via the OLED display, which is quite bright even at its lowest setting. The OLED display is very legible, with the matrix display allowing display flexibility, and a wide viewing angle. The display is also fast to update, with little lag compared to LCD, which is especially helpful when the meter can update quickly and when trying to make sense of the bargraph display at the top. The display also allows for dual readings, which is very nice, in the case of displaying DC and DC+AC readings at the same time.
Using the meter for prolonged periods for logging or other features does consume a decent amount of battery due to the OLED display. Unfortunately, the display also seems to have a tendency to "burn in" with heavily used segments becoming dimmer. A short period of data logging was enough to see this beginning to occur.
For additional safety, the meter comes with an automatic discharge feature, which automatically discharges the connected load at the end of an insulation resistance test. There is also a very sensitive non-contact voltage sensor, which easily detected mains wiring through walls in my limited testing. Remote activation is also possible via the remote switch probe, which allows for easier use of a pair of probes during spot checks - you don't need an assistant to operate the insulation tester, nor do you leave it emitting high voltage when not necessary which is a safety hazard. Another safety feature is jack sensing, which warns the user when the lead is not placed into the expected port for the type of measurement being taken, which should help prevent any major disasters. Of course, there's a relatively expensive fuse as well to protect the current measurement shunt.
There are also convenience features, such as distance measurement by cable capacitance, which I didn't have a chance to try. There is also on-board memory for computer-free data logging and a fuse verification feature which doesn't need you to disassemble the meter. How nifty!
One thing you might feel is lacking is the presence of a high-amps range - this meter only measures current up to 440mA, which is a little surprising for a field-service meter. Furthermore, another thing that might be missing is a low-ohms range below the 600 ohm range.
However, once it is configured to your liking and you get used to it, it's a very nice meter to use. It's a tad on the large and heavy side, and boot times are a little on the long side, but it feels solid with no "give" in the case, and its IP rating allows you to be free of any concerns even in the harshest environments. I grew to love it very quickly.
After being tipped off that there may be issues with the accuracy of basic measurements using the U1461A, I decided to set off in "hot pursuit" to run an experiment to cross correlate readings from the U1461A in 4.5 digit mode with the Keithley 5.5 digit Model 2110 benchtop multimeter. While the magnitude of the reading error from the 5.5 digit is not negligible compared to the 4.5 digits of the U1461A, we can still see just how close the readings are (which should be "very" close). The interesting feature is that while the U1461A is a 4.5 digit multimeter, its ranges are a bit different (6V, 60V, 600V) which results in many cases where the U1461A has an equivalent display resolution to the 2110.
Tests were done on DC volts, AC volts, DC milliamps and Ohms ranges for simplicity, in a manner similar to my comparison of 3.5 digit meters. A total of 630 individual hand-made measurements were used to compile the following graphs.
In the DC volts department, the agreement between the two meters was generally exceptional. Most of the difference remained below 2 counts of difference in the least significant digit, which, being "half" a digit, is expected to vary by 2 counts on each meter (for a total of four counts) even in normal operations. There is an interesting result where at the upper end of the 6V range on the U1461A, the linearity seems a little off, resulting in a maximum discrepancy of 6 counts - still well within expectations.
Working it out as a percentage value, assuming the Keithley to be the "absolute" correct value, the difference is well below 0.03% at the maximum. I have also plotted the 0.012% basic accuracy line (without accounting for digits error) of the Keithley - any differences below this line is almost certainly to be due to the Keithley meter. This is all well below the 0.09% basic accuracy claimed for the U1461A - so we're off to a great start.
Let's try the same with AC voltage, which is a bit trickier and more difficult to measure accurately. It seems that we do have a similar behaviour - there are linearity issues which grow towards the high end of the range, implying this may be calibration related with my unit.
The maximum difference was under 100 counts of difference, which isn't that informative as counts ... so let's try percentage.
As a percentage, we can see that the majority of the error values lie below the Keithley basic accuracy line, meaning that the influence of the Keithley meter cannot be separated. The points above deviate by just over 0.35% but that is still below the 1% claimed basic accuracy. Again, it betters the specification.
Lets now move to DC Current. Generally current measurements are difficult and "error prone", so it will be interesting to see how it fares.
As it turns out, the Keithley and U1461A are both on the same page here, with readings all below 6 digits difference. Slight rises in error towards the high end of the range is seen again, but it sits under 4 digits for most of the time, meaning it's essentially "the same reading".
Looking at the error percentage, we can see that the resulting plot shows a maximum error under 0.025%, which is below half the error contribution due to the Keithley alone. How close are we? Very much identical.
Finally, I'll take a look at resistance using the group of resistors I used for my prior investigation. Let's start with the low-ohms.
At the bottom end of the scale, the error is quite high, above the 0.5% basic accuracy because of the effect of quantization error. Note that this doesn't mean it's out of specification - because I haven't accounted for the +/- digits which would easily eclipse the 0.5% figure. However, it settles down quickly at about 20 ohms and up, meeting the basic accuracy specs alone, which is good news.
At medium resistances, we can see contributions of the Keithley accounting for some of the measured error, and other errors about 0.05% which is still way below the 0.5% claimed basic accuracy. Another win here.
High resistance is particularly challenging, however, it seems we can see the U1461 pretty much beating the claimed basic accuracy levels for all except the last data point. However, because I don't have calibrator-standard quality tools, I think you'll have to agree with me that this "slight" outlier is more likely my fault rather than a fault with the meter itself.
In conclusion, it seems the meter most likely meets and exceeds the specified accuracy in the tested ranges. Occasional "transient" spikes above these are likely due to the test methodology (as the sources I'm using probably have noise and drift over time between manual readings) and quantization errors which I didn't take into account. In DC current, the correlation with the Keithley 2110 was so strong that the values reported were consistently differing by less than the reading error of the Keithley implying very strong agreement. In other ranges, the difference is usually within the "gap" between the basic accuracy of the Keysight and the Keithley, with exception for the low-end of the ohms range where the number-of-digits error is a big contributor to the error.
Testing of electrical equipment and portable appliances in Australia is covered by AS3760:2010, with Amendments 1 and 2. This involves visual inspection, which catches >90% of defects, an earth continuity test and an insulation leakage/resistance test.
Proper testing of handheld tools, for example, may involve the need to engage power switches and cover "contacted" areas with braids or foils. It would be nice if they were included, but because of individual requirements, I can see that we are probably better served by purchasing for our needs. The toolbox does have some additional space for some additional accessories, which is quite thoughtful.
The pertinent number for resistance is generally no less than 1M ohms, at 500VDC, unless it contains MOVs/EMI filtering bridging the insulation where 250VDC is permitted. While the standard doesn't seem to prescribe the time, I believe that 1 minute is the standard.
As a result, basic insulation resistance testing can be done quite simply using the meter in standalone, by selecting timed test, with 1:00 configured for the test time. Hook up the appliance as required, press the test button, and then read the final value once it completes. If it's >1M ohm, then you're good!
There is, however, a slight complication when it comes to testing appliances which have "software" based power buttons as they have to be properly powered during an insulation test according to Australian Standards requirements (to my knowledge), meaning they probably need to be tested with dedicated portable appliance testers instead.
I gave this a try with several appliances around the house, and sadly, the insulation proved to be in tip top shape, so it measured >260G ohms, so there wasn't really anything to show here. For safety, a red LED flashes whenever the output is active.
For preventative maintenance, different tests are provided to try and separate the components of insulation leakage. One of the tests is the Polarization Index (PI) which compares the insulation resistance measured at 10 minutes over the resistance measured at 1 minute. This reflects the polarization current - i.e. the current needed to polarize the "moisture" in the insulation. Readings of below 2 are considered questionable to dangerous, with readings above 2 considered good, and above 4 considered excellent.
Another test is the Dielectric Absorption Ratio (DAR) which compares the reading taken after 60 seconds to that of the reading taken after 30 seconds which also reflects the state of polarization of the dielectric material. Readings above 1.6 are considered good, and below 1.25 considered questionable.
Unfortunately, when you get an "over-range" reading, it doesn't make any sense to compute these figures - so unfortunately, I wasn't able to show this in action.
In insulation testing, the voltage that is produced by the generator is of importance as it can have influences on the measured resistance and could (in the worst case, if out of range) cause false readings or over-stress the insulation during testing. For the best repeatability, and for comparison readings when predictive failure analysis is used, the accuracy of the voltage developed is important.
While the configurability is a big feature of the unit, there is a question just how accurate the voltages output by the unit are. To be sure of the output accuracy, we need to measure the output using another device - so I hooked up the output to the Keithley 5.5 digit Model 2110 multimeter. The meter was set to 10 PLC auto-ranging DC for the highest accuracy, and recordings were taken every 65ms.
My first tests involved measurement of the factory fixed presets of 50V, 100V, 250V, 500V, and 1000V compared to the 250V, 500V and 1000V on my Tenma 72-9400. A timed test of 10-minutes was run, resulting in over 2700 readings at each voltage to be used in computing the statistical parameters.
Surprisingly, the U1461A exhibits more variance in its output voltage compared to the Tenma, but is (mostly) more accurate in its absolute voltage output. However, the specification is for 0.5% +/- 1 digit (which I presume means 1-volt in this case) which it meets easily. Interestingly, the Tenma's output voltage is reversed in convention - the red connector is labelled line, but is outputting a negative voltage relative to the black terminal labelled earth.
Another mode is the step trip test, where it is noted that the steps can be no smaller than 10V, despite the unit claiming to have voltage set points of 1V. During testing, there is a display of the output voltage as well, displayed in 1V increments, which often shows output voltages a volt or two either side of the intended output.
The U1461A was set to do a step trip test, with 1mA current limit, a step number of 100 steps from 10V to 1000V in 10V increments, dwelling for 30 seconds at each step to provide a large number of data points.
The recorded data was processed by discarding the first three samples at every transition to exclude voltage errors going between steps.
The voltage difference between the nominal step voltage and the actual measured voltage was calculated and labelled the delta voltage. This was plotted as a time sequence, and as a histogram.
It seems that the meter, at least in step trip mode, will put out voltages occasionally more than 10V away from the intended set point. However, on an average of the 30 second dwell period, it was only a few volts out. This is fine for a ramp scan, where we're mostly trying to distinguish between 10V steps, but it seems in rare cases, it might just be one step out. This seems to be more severe at lower voltages.
This leaves me rather confused - as the "fixed" factory voltage points before showed very good voltage agreement, the step test seems to show somewhat more relaxed voltage output regulation. Maybe this is something that could be improved, but as it stands, is probably still accurate enough to meet their specifications (mainly, because of the 0.5% +/- 1 digit, in this case, probably meaning 10V step).
One of the features that goes above that offered by basic insulation resistance testers, aside from configurable voltage, is the ability to do a leakage current step/ramp trip test. This involves changing the voltage in steps, dwelling for a specified amount of time, and repeating until either the maximum voltage is reached or a trip level of current flows (e.g. 1mA). This is normally useful for testing voltage-dependent components, such as varistors, transient-voltage-suppression diodes, zener diodes and gas discharge tubes.
Since I didn't have any of them on hand, I decided to do my first few tests by trying to quantify the reverse breakdown voltage of silicon diodes instead. In fact, such a test condition is quite ideal for checking whether your 1N4148's have gone leaky, checking unlabelled components, or for cherrypicking the better ones for voltage multiplier usage (for example). In this case, my five 1N4148's from the same batch are pretty similar - noting that a difference of 10V isn't significant (as per the result above).
Just for kicks, I took home some spent starters and a fluorescent tube from the uni. As these are basically gas discharge tubes with a negative resistance characteristic, it needs a "threshold" voltage to be reached to start the discharge through the gas. As a result, it makes a good candidate to explore the step trip test behaviour in the context of surge protective elements.
I tested two starters - it seems the first starter may have overheated at some point and the gas seal on the capsule may have failed at the lead, letting the gas fill "escape" and be replaced with more insulating air. As a result, that unit saw all 100 steps from 10v to 1000v with no trip. The other unit seems to still be intact and has a pretty nicely defined voltage trip point (as evidenced in the two separate test runs).
A bi-pin 2U style tube was also tested, and found to have a consistent trip point at a higher voltage as expected due to the gas path length.
Then, I remembered I had a pair of left over 36w tubes from ages ago in a corner ... as it turns out, even 1000V isn't enough to get 1mA through those tubes. That can be explained due to the much longer dimension of those tubes.
We can see that with the step/ramp trip test functionality, it is possible to determine the "threshold" point for voltage-sensitive components non-destructively, thus being able to evaluate the condition of metal oxide varistors and gas discharge tubes, commonly used for surge protection. A straight insulation resistance test above the threshold voltage will always show a fail, without actually providing any information about the condition of these elements.
To get started with PC connected data logging, you must first download and install the Keysight Handheld Meter Logger software from their website (linked earlier). As of writing this review, the current version available and used for testing was version 188.8.131.52. Installation was trouble-free by following the installer prompts on both my Windows 7 x64 and Windows 8 x64 based computers.
The next step is to unpack the cable and attach it to the multimeter. This is achieved by aligning the head with the logo face up, and snapping it onto the meter. The head will only mount one way, and this is ensured by slotted-keys, to ensure correct RX/TX alignment. Then, the other side of the cable can be attached to a USB port. The adapter is powered by USB, and appears as a COM port on the computer, using the Prolific driver. By using an optically isolated connection, there is no risk to the computer should the meter's internal insulation catastrophically fail.
The cable itself is probably the same internally as the earlier grey coloured Agilent branded one I tore down in this review of mine. In essence, it seems it is merely a baseband-encoded signal of the serial data at 9600 baud (by default), which is nothing too sophisticated. The commands themselves are just SCPI-like strings sent and received over a serial COM port made over USB.
The software itself is capable of interfacing with multiple meters from the Keysight/Agilent catalogue and performing data logging and memory log downloads. The interface features several modes, including remote display (with multiple-displays supported), graph, raw data listings of logged and manual logged data. This interface is what you see when you engage the regular multimeter style measurements on the left side of the rotary dial.
Logging can easily be started by pressing the start button, and the capture can be zoomed, panned and markers used even during capture. The graph navigation is rather intuitive, with one button presses to scale to fit all data, which is very nice. Unfortunately, manual axis settings are not possible, so the axes will be labelled depending on what sort of signal you have inputted. The saved data can be exported to .xlsx or .csv format for further processing.
The data logging itself occurs at about 1Hz, which is slower than the meter actually updates. This is probably due to the speed of the serial link, which might be improvable by setting the baud rate on the meter to 19200bps (although I didn't try).
The software itself is configurable, thus not only logging is possible, but also alerting and timered control. However, do note that regardless of your regional settings, the date format is in American date format. There is also a remote control tab, although only mode and range can be changed (rotary position needs to be manually physically changed, and no trigger is available).
Global software settings can also be changed, which is where you will find the voice-out feature which "reads" the meter readings every three or more seconds using a text to speech engine. Updates to the software are checked on start-up by default. E-mail configuration can be done also through this screen for the alert feature.
If you place the multimeter in the insulation resistance testing positions, the interface changes dramatically to a job-and-report oriented interface suited for insulation resistance testing.
Templates have been pre-configured for most of the tests you will be interested in running, which allows you to get up and running quickly. Instead of fumbling with the meter, you can enter some important parameters in the screen and they will be uploaded to the meter.
During the test, visual feedback through a graph is shown, which allows you to gauge just what your insulation is doing - for example, an insulation which shows increasing resistance over time is generally good.
The templates themselves can be modified, although the interface to do so is a little cumbersome to get to grips with.
The software allows for exporting test reports, however, even though I had Adobe Acrobat Pro X installed, it insisted that I install a PDF reader. I did find that it had silently spawned Microsoft Word with the report file and tried to make a PDF from it. When I was shutting down the PC, Microsoft Word saved the document for recovery on next opening, so I could see the report (attached). However, I hope they fix this issue - just because I don't have the exact version of reader they expect doesn't mean that I don't have a PDF reader, which hinders the use of the report feature.
Sadly, I couldn't seem to find any way to configure and run a trip test (ramp or step) using the meter logger software, which would simplify the many button presses needed to perform it, and potentially give some insights into the insulation behaviour. I suppose it could be emulated in some way, in software, even if direct access to the hardware configuration is not possible - I really hope Keysight updates the software to include this.
It also doesn't seem that the software is able to log the meter's ambient temperature reading with the reports, which would be rather useful when it comes to compensating for tests done at different temperatures. The meter doesn't appear to have any measurements of humidify either, which is another parameter that can affect insulation resistance readings, however, I haven't met a meter (yet) that has that at this stage.
Preparation of the Bluetooth adapter is fairly simple, with the cover sliding up and revealing a space for two AAA batteries. Installing the batteries, taking note of the polarity, completes the preparation process. The front face of the adapter features a single slide switch with three positions - Off, Setup and On. Two LEDs are provided, one to indicate link status, and the other to indicate low battery condition.
Configuration of the Bluetooth adapter is not necessary, you just slide the switch to the On position, and begin the scanning and pairing process, entering the pre-configured passcode. You will need to look for a device with the default name of Keysight U1117A-XXXXXX and enter the default passcode of 1234.
The Bluetooth adapter supports the Serial Port Profile (SPP) and when paired with your PC's Bluetooth stack, will come up as a COM port. It seems to support another service which is used with Apple iOS devices, which does not install for other targets, but that's not an issue. It is mated by clipping it onto the rear of the multimeter, with no need to use any plastic adapter frames. Opening the Keysight Handheld Meter Logger software, the link will be automatically detected and clicking on connect will detect the meter connected to the adapter. There really is no difference from the software's perspective whether you use the USB to IR cable or the Bluetooth adapter.
Looking at the regulatory labels, it seems that the adapter is made with the Bluegiga Technologies WT11i-A Bluetooth Module, with certification report here.
That being said, you may opt to use the USB to Serial cable whenever you are near a computer because it is more reliable, as it is not subject to potential ISM band interference, and it does not consume any battery. Furthermore, in some radio-sensitive areas where transceivers are not permitted, you are best to stay with the USB to Serial cable, or utilize the on-board logging capabilities of the multimeter and then download the memories when you get back to a PC elsewhere.
The battery life of the adapter is rated at about 20 hours. An indicator LED is provided on the adapter to warn of low battery. Unfortunately, being on the front face of the adapter, it's not visible when looking at the multimeter from the front. When the battery runs out during a data-logging session with the PC, the PC ends up in an endless loop attempting to reconnect with the Bluetooth device. The only way out, aside from changing the batteries to allow the connection to re-establish, is to force close the software which results in the data in the present data-logging run to be lost. As a result, I would recommend using the USB to IR cable where practical.
In all, there are no complicated steps to getting the Bluetooth to work, and from the perspective of the PC, is almost interchangeable. It seems that the Bluetooth adapter expects the meter to be configured for 9600bps though.
If you've tried the previous Agilent Mobile Apps on the Android platform, you might have found the solution a bit lacking. The software had many problems with later versions of Android, and was only advertising support for early 4.0 builds. Luckily, along with the rebranding of the apps from Agilent to Keysight, the apps have been refreshed and they perform much better than before.
To get started with the meter under Android, you need to turn on the Bluetooth radio on your device, and search for the Bluetooth adapters and pair with them by entering the passcode.
You will also need to navigate your way to the Google Play Store to download the three applicable applications for the U1461A - namely Keysight Mobile Meter, Keysight Mobile Logger and Keysight Insulation Tester.
Keysight Mobile Meter
Starting the application shows a full-screen Keysight Technologies logo splash screen. If your Bluetooth is turned off, it will request permission to turn it on. An End User License Agreement (EULA) is presented, which is decently long, with very strange restrictions against evaluation usage, competitive research, etc. I am taking the stance that the RoadTest authorizes me to violate this agreement to evaluate their product and inclusions, and as such, I will be including screenshots and commentary about their mobile apps.
The application has a relatively old "Android 2.x" UI look with the white-on black and grey option pop-up menus, and comes default configured for one meter. Some of the features include remote hosting, which allows your Android phone to become a web server so you can read the readings from any web browser on the same network.
If your device is paired, you can click on the Connect button, choose the device and off you go (provided the Bluetooth hasn't glitched up, which happens occasionally). Readings are displayed in the main window, and are continually updated if the Scan mode is turned on.
The update rate of the screen is much less than that of the meter, which seems to be a common theme of all software interfacing via the IR port on the meter, refreshing at about 1 Hz. The reading seems to be displayed with the loss of some precision as well, only displaying three digits when reading voltage, when the meter was configured for 4.5 digit display.
One major feature is the Voice Out capability, which uses text to speech to read out the meter reading at intervals of three seconds or more, allowing for "hands and eyes free" monitoring of values.
Readings on the screen go red if the connection to the meter is lost, and an automatic reconnection is attempted. Unfortunately, sometimes it takes several goes to succeed, and other times, it does not succeed until the Bluetooth adapter is reset. As a result, I don't advise trying to see the "limits" of coverage, as it could be frustrating to reconnect. No signal strength reading is available (unfortunately) probably due to limitations in the phone OSes, so I would advise staying within about two rooms distance to avoid issues.
Keysight Mobile Logger
As with Keysight Mobile Meter, Keysight Mobile Logger has the same UI theme, along with the EULA and splash screen. The main interface features a top section which is similar to a remote display, and a bottom trend graph section which shows the signal over time as it is being logged. Three meters can be connected as A, B and C, with a math channel available labelled M.
While the top section has the same sort of refresh rate and limited digit precision in the display, the graph itself seems to be much more sensitive and is plotted with as much precision as the meter is capable of. The graph area is responsive to two-fingered zooming and one-fingered panning once logging has been stopped (otherwise the next log refresh overrides your zoom), and the toolbar buttons allow you to change the graph focus. The graph itself is split, with the top section showing the range of the log which is currently plotted. Unfortunately as the grid lines are black on grey, they are quite difficult to see on the phone screen - if Keysight had a change of colour scheme, that would fix it.
The graph axes are fixed, similarly to their other software, so the axis range will depend on the signal you are measuring at a given time.
Logs are recorded to an SQLite3 database, which can be exported to CSV or backed up to external storage and or Dropbox cloud storage, making this very useful for making records of measurements. Manual logs are stored in a separate area, every time the manual log button is pushed. For ease of back-up, the whole database can be backed up as well in one step.
Keysight Insulation Tester
This software is specifically intended for use with the Insulation Resistance Testing multimeters, and is very similar to the software above when the multimeter is placed in basic measurements mode (voltage, current, temperature).
When the rotary switch is placed in the insulation resistance test mode selection, the interface is changed to a specific "job oriented" interface, offering the choice of test modes and test parameters. Job numbers and remarks can be set directly from the screen, although other details need to be set via editing the record which seems counter-intuitive. Reports can be exported from the interface as well, direct from your mobile device.
Timed test modes are very common for measuring the insulation resistance. You could even use the pass/fail mode to do portable appliance testing, and reduce the output to a pass/fail - no "reading and interpreting" a value. It doesn't seem that Step/Ramp trip test is available from the mode selection, and the Memory selection seems to be perplexing to me.
A live plot of the insulation resistance is shown at the bottom of the main screen, and allows you to understand the health of the insulation based on the trend of the graph which reassures you of the test progress, especially during long 10-minute polarization index tests.
While the data is recorded into the database, it doesn't seem the temperature is recorded. Further to this, entering in DUT information doesn't seem straightforward, requiring the use of either an edit to the database record after a test, or just using the Remark field only and writing in the details after printing out the report.
I am fairly impressed with where the Android applications are at. Compared to the mess they were before, their usability has improved dramatically, and there was no significant major issues unlike before where I couldn't even get Mobile Logger to work at all. The graphing, logging and displaying were as responsive as expected, although increasing the precision of the display would probably be a welcome change and doing a few colour-scheme/theme updates to keep with modern Android UI concepts would also be nice.
There could be more improvements to the Insulation Tester app to make it easier to enter DUT data, make it possible to record test temperature, and use ramp/step trip tests.
Getting started with the iOS device was easy, beginning with pairing the Bluetooth adapter with the device, and entering the passcode.
Interestingly, the device was able to automatically detect that the "required" apps to use the accessory have not been installed, and helpfully steers you to the app store where they can be downloaded free of charge. Just like the Android counterpart, Keysight has three applicable mobile applications for this meter - Keysight Mobile Meter, Keysight Mobile Logger and Keysight Insulation Tester.
All apps start with the Keysight Technologies logo splash screen, and on first launch, presents the user with an End User License Agreement (EULA) which must be agreed to in order to utilize the software. The EULA is extensive, and features unusual statements in regards with evaluation, and competitive research and comparison.
Keysight Mobile Meter
Testing began with Keysight Mobile Meter which allows for remote screen display of up to three separate multimeters. It offers the same features such as remote IP hosting for even more remote viewing of the measurements.
Getting started involves trying to get a Bluetooth connection by re-arranging your meter into either one of the top three positions.
Then, we should be able to log as usual, similarly to the Android program. Unfortunately, I could not get this to operate reliably, and while the adapter was detected, the measurements were not displayed on the screen but did seem to be displayed via remote IP access. Clicking on the share button resulted in the instant force closure of the program. This may be due to iOS 8.3 compatibility issues. As a result, using a mixture of meters including the U1461A on iOS in remote display mode may be a problem at this time.
For performance reasons, you are warned if any other Keysight Mobile apps are running, and to close them to avoid performance issues.
Keysight Mobile Logger
The Keysight Mobile Logger app however, had no major issues with remote display and data logging. The data is shown using an identical graphing module which supports pinching to zoom and swiping to pan, making it relatively intuitive to use. One small bug is the battery indicator which does not seem to show the battery status under iOS.
I suppose that if you were to use a mixture of devices on iOS, your best bet would be to stick with the Mobile Logger app. Voice Out is available across all the iOS applications, and sounds pretty much the same as Voice Out on the desktop or on the Android programs.
Keysight Insulation Tester
The Keysight Insulation Tester app is probably the most applicable program to use under iOS if you're intending just to use a single device - i.e. the U1461A with your iOS device. Depending on the position of the rotary switch, the interface changes.
With the rotary switch in the basic measurements (voltage, amps, temperature), you will see a screen that's almost identical to that of the Mobile Logger app, along with similar features. The issue with the no-battery-level display is still here with the Insulation Tester app.
However, if you place the meter rotary switch in the Insulation Testing positions, the interface is changed to a job-oriented interface designed for Insulation Testing purposes. This allows you to set an ID for the test, set the test method, set any required test parameters and run the test. Afterwards, you can export the results by e-mail as a PDF report, and view your previous test results.
It does seem that the ambient temperature is not logged during the test, which might be a bit lacking. You will manually have to put it into the Remark field, reading from the meter's screen, which defeats the purpose somewhat.
The report itself does look a little bare and fuzzy (low resolution graphics). The other reason is that I didn't bother to enter any major bits of information, as it doesn't seem straightforward to do it - via editing existing records in the database, and only the Remark feature is directly accessible.
Some unusual experiences were had under Test Method, with the User selection resulting in unknown behaviour (often crashes). I couldn't find any step/ramp trip test either, which would be nice, as it would save fumbling with the meter's buttons.
Under iOS, it seems that the Keysight Mobile app suite can do with a few bug fixes here and there. However, as it stands, the Keysight Insulation Tester app seems to perform core functionality of data logging, insulation testing and remote display correctly. I would like to see Keysight Mobile Meter fixed so that it connects properly with the U1461A, the share button crash fixed, and also make entering DUT information in the Insulation Tester app more straightforward to save time.
Aside from the small bugs, it seems the test methods under the Insulation Tester app can do with the Step/Ramp Trip test, as it doesn't seem to be available. Recording of ambient temperature into the data reports would be very welcome as well.
Keysight currently offer two main Bluetooth adapters - the U1117A bundled in with this product, and the U1177A which is a lower powered Bluetooth adapter which is cheaper. The Class 2 transciever is normally advertised with a 10m range, whereas the Class 1 claims a 100m range.
One difference between the two adapters is that the U1177A can be configured to customize some of its name and PIN code, and the U1177A does not support iOS devices (they will not come up on Bluetooth scans and cannot be pared due to only supporting SPP profile). The U1117A, owing to the higher powered radio, offers a shorter battery life of 20 hours compared to 30 hours on the U1177A.
I did an informal testing of range in an indoors environment, and it seems that the range may be a bit optimistic especially where the ISM band is noisy, or the device is paired with a Bluetooth radio in a phone or tablet which may only be Class 2. Regardless, the U1117A achieved superior range, with the U1177A dropping out at a distance of about 6m through a few walls, whereas the U1117A maintained a solid connection with me walking outside my house for a distance of about 25m.
If you are considering which adapter to purchase for an Agilent/Keysight meter you already own, the U1117A may be a little more expensive, but the range and iOS compatibility may be something important to you and make it worthwhile.
One of the advertised features is the possibility to use several meters connected to the same computer, phone or tablet. Under the PC, it seems you can connect up to ten units by USB or five units by Bluetooth, whereas under the mobile applications, you can connect up to three. For various reasons, the numbers of devices that can be connected to Android devices may be limited by your particular mobile device to two.
I tried to combine the U1461A with my U1241B on the Keysight Handheld Meter Logger software on the desktop, but to no avail. Regardless of connection order, or selection of display, I would get an error message which seems to be due to the mixture of multimeter with insulation testing multimeter. I think this is something Keysight can probably fix quite easily, and shouldn't really be an issue for basic measurements (e.g. Volts, Amps, Temperature).
That being said, my experiences with the mobile applications under Android seem to be more favourable, with no significant issues encountered in pairing the U1461A with U1117A and the U1241B with U1177A to the same phone. Under Keysight Mobile Meter and Keysight Mobile Logger, both meters were available for use, although occasionally if the connection was lost to the Bluetooth unit, a hard power cycle of the Bluetooth adapter is required. I'm not sure if this is a problem due to the Bluetooth hardware in the phone used for the testing.
Testing of multiple meters under iOS was not possible, as the U1177A which I already own is not compatible with iOS, and does not come up under a Bluetooth device scan.
The Keysight Technologies U1461A is a very solidly built meter, backed up with a generous serving of accessories and a three year warranty. On paper, it has very competitive specifications which meet or exceed those of the competition, while not sacrificing anything in the combination of insulation resistance tester and multimeter into the one relatively compact unit.
In testing, the unit was a little difficult to use at first, owing to the multiple-functions allocated to buttons, however, this was easily overcome with the help of comprehensive and well-laid out documentation and some practice. The unit was found to likely better the specifications when tested for multimeter functions, and has adequate voltage regulation for single voltage insulation resistance testing. However, for stepped voltage tests, it seems that it may be out by one step in rare situations, which still does meet specifications.
The unit becomes much more appealing once the computer-control facilities are considered, via the included USB to IR cable, or via the Bluetooth to IR adapter. This also permits the meter to be connected to iOS and Android phones or tablets. Mobile remote display, text to speech readout, data logging and insulation resistance testing are possibilities, leveraging a device you already have. While this is not entirely bug free, it was generally functional enough to accomplish most tasks including field recording of results and report generation. Some improvements here to fix bugs, update colour scheme, and fix interoperability between regular multimeters and insulation resistance testing multimeters would be welcome.
Of course, I didn't have the time to explore every feature in the meter, but it was certainly quite impressive how much they managed to fit within a "handheld" unit. I suppose they got it right when they defined the following in their application's XML files: <Model Number="U1461A" Name="Unicorn">, as this unit is pretty much the mythical beast in reality.
It's been another big effort, and a bit of a crazy ride with back to back RoadTests, and I have to say that I'm thoroughly "reviewed-out" for now. I'll have to take a break from RoadTesting and reviewing for a while because of other commitments to my study, but I hope to be reviewing more gear in the future. I hope you've enjoyed this review, which I have tried to cover some "different" aspects of the U1461A compared to others. In order to get a well rounded "feel" of how the U1461A stacks up, I encourage you to read reviews from other fellow RoadTesters back at the main page. Feel free to see what I'm up to at my personal blog at http://goughlui.com