10 Replies Latest reply on Sep 25, 2019 3:42 AM by michaelkellett

    Lead acid protection circuit

    zagene

      Hi everyone,

      I'm always looking for alternatives for a DC supply, as we have a constant problem with mains supply failure.

      Add to this that I am a Ham operator who like to keep things up and running, even on DC power if required.

      For DC supply I have the normal AC to DC switch mode supplies but the problem is with a suitable protection for my (very expensive according to me) AGM 12VDC batteries.

      If you fail to constantly monitor the AGM battery output voltage or with swapping batteries accidentally (and it do happens if you are in a hurry at night) connect it reverse polarity you have a problem.

      I have read many articles on battery protection especially with RPP and current flow back prevention as well as a battery low voltage cut-off so I'm starting a project that is compact and cost efficient enough to add to all my SLA's and AGM (and also all flooded cells for that matter) 12V batteries.

      So my circuit is more a combination of several other circuits combining an efficient RPP without to many heat losses as well as combining it with a ATTiny ADC to read the voltage to cut off the battery at a preset low voltage, nothing very extravagant or to large or difficult to populate on a single sided PCB and fit it on the battery output.

      My requirements would be to feed a radio or other load with DC at a maximum of 30A @ 13.8VDC (as required by my radio when transmitting.

      The circuit area I'm not very sure of is the cut-off feed from the ATTiny45 via the 2N7000 that will cut the ground from the main P-ch MOSFET as a high side switch with a maximum current of 55A, Maximum gate voltage of 20V and a RDS(on) of 20mOhm at Vgs of -10VDC.

      The second problem may be an Inrush current needed by the load which I do not totally want to limit as in certain instances my radio may require but I have to protect the MOSFET.

      The other thing maybe to place a uni directional TVS at the battery poles of the circuit e.g. 1N6377 to clamp high voltage peaks etc.

      Please advise me on the following circuit: (its available on PDF format here https://drive.google.com/drive/folders/1jG2ML2XMi52CiVaMMP-QXK-w0ViSZoYC?usp=sharing ) as well as the IRF4905 datasheet

        • Re: Lead acid protection circuit
          jw0752

          Hi Eugene,

           

          You may want to check and verify the polarity of the IRF4905 as in its current config the internal Schottky will conduct current battery to load. As far as the rest of the design I will defer to some of the other guys with more design experience than me.

           

          John

          1 of 1 people found this helpful
          • Re: Lead acid protection circuit
            dougw

            Interesting application.

            You could also use something like an LTC4352 to create an ideal diode which also has under voltage protection.

            An alternative is LM74700.

            4 of 4 people found this helpful
              • Re: Lead acid protection circuit
                zagene

                Hi Douglas

                I researched quite a lot on the ideal diode controller bu because of reasons mentioned below I could not continue, but I stumbled on another similar controller that is locally available at a much more competitive price range but some things is not very clear. Can you please help me here.

                Here is part of my circuit of the ZXGD3112N7TC ideal diode controller (SOIC package 127mil in stead of 65mil or 50mil that would be much better to prototype by hand) see https://www.diodes.com/assets/Datasheets/ZXGD3112N7.pdf

                 

                From the datasheet I gather that the PWR GND and GND is referenced to the same source ground to probably accommodate the driving of the N-channel MOSFET. It differs quite a lot from the Ideal Diode controllers of Linear Technology (the documents of LT also better explanatory than the one of Diodes Inc) see Figure 4 on page 6 of the above document.

                As from this figure the VCC should be from another resource probably a linear regulator if the source voltage is higher than the maximum VCC of the device which is 25V, but still referenced to the device ground which is the source voltage and seemingly not the system ground or input ground.

                Since the probability of a DC 12V supply (14.4VDC max) is not above the maximum rating of VCC, then the zener is most probably not necessary, but again, referenced to what ground? I'm a bit confused here.

                Ive looked at other similar devices and all of them only have a 100n cap between the VCC and device ground (the devices with a ground referenced to the source level). What about a charge pump for the gate drive. Is it not needed.

                LTC4357's IN pin is maybe similar to this devices GND pin but with the LTC4357 the device GND is protected by a bidirectional transzorb between the source and system Ground with a Dclamp zener (Uni Directional transzorb) between drain and the device ground to limit the overall device voltage to a VDD level lower than the operating supply range. Would the ZXGD3112N7TC circuit then be similar than the circuit of the LTC4357 on page 10 marked Reverse Input Protection. I'm not entirely sure about the ZXGD3112N7TC's VCC supply level, or maybe I should just use the 100n cap in this instance but what about unexpected transient spikes. This could aid me quite a lot with multiple outputs ORing.

                (See https://www.analog.com/media/en/technical-documentation/data-sheets/4357fd.pdf on page 10)

              • Re: Lead acid protection circuit
                jw0752

                Hi Eugene,

                 

                After reviewing the polarity protection application I believe I was wrong to question the polarity of the IRF 4905. It is correct as you have it. Sorry for the confusion but I have a little difficulty analyzing the current flow and logic in the P Channel MOSFETS. If it wasn't correct as you have it the internal Schottky would allow reversed polarity current to flow.

                 

                John

                • Re: Lead acid protection circuit
                  ldjohn

                  Hi there Eugene, one scenario is when you try to switch off the load, the internal diode will conduct inside the mosfet, thereby continuing to conduct despite your efforts to stop it.

                  1) consider placing back to back mosfets in your design to isolate the load, because of the diodes.

                  2) consider adding another circuit in to allow a PSU to be connected as well as managed.

                  3) in certain scenarios where you have equipment that fails ( radio, charger, PSU) it may be good to allow under and over voltage protection. Sometiems a charger gets stuck and takes the battery over 16V. In your scenario, simply add another channel to the A/D to read input voltage for PSU and another set of back to back mosfets to control the output.

                  4) during high transient cut over currents, it may be pertinent to investigate limiting of inrush current by adapting the resistive load between Pchan gate and transistor pulling the circuit to ground for activation. This will create a small ramp time. Once engaged, no more ramp needed so thats good.

                  5) make sure that you put protection on the inputs of the A/D converters. i usually place a 5v1 Zener Diode (assuming 5v logic) between the A/D input and ground so that it limits maximum input to the A/D. This is also a great trick as if things go really wrong, the A/D pin is protected from over-voltage as well as limits negative voltage to around -0.5V which should be in the micro's safe area.

                  Keep going. You have the makings of a great project.

                   

                  On choice of mosfet.... under full conduction at 30A, the 0.02R resistance will add about 18W worth  of dissipation and what looks like 0.6V drop over the device. Not bad at all

                  1) put more in parallel, making it a great deal cooler ( by putting in parallel, you can reduce power dissipation

                  2) put more in parallel, making the total on resistance better

                  2) put a decent strip of heat-sink on the devices. Aluminium U-channel / whatever you have will be great. If you can use an aluminium enclosure, that will work as well. Remember to put a small fan on it to keep it cool. Don't waste too much energy.

                   

                  Regards

                  John

                  3 of 3 people found this helpful
                  • Re: Lead acid protection circuit
                    zagene

                    Hi Everyone,

                    I noted all of your comments and suggestions and this resulted a storm in my mind, thus my late reply.

                    John, TY for your suggestions. Regarding the Ideal diode controller (mosfet driver),

                    I've researched the component quite a lot in the past few weeks, designed some circuits etc but here is the problem:

                    The the LTC4357 or LTC4352 is only locally available trough RS Components and id on my local currency extremely expensive although it is a very nice component and I would certainly like to build my designs around it. The footprint is also very fine and to etch a custom PCB and build it by hand would also present some difficulty. The price of it also make prototyping to expensive as I have learned that just merely calculating values and laying out a PCB do differ from real world conditions. So I just have to do without it (Im also not sure if the gate voltage would drive the available high current MOSFETs in saturation so that may be another hurdle making the design and layout more complex)

                    As I went along with this design my ideas became more complex.

                    My further question, and quite a nagging one, is would this prevent current flowback from the Battery and the "Charger" if I place a circuit like this between it. I recently lost a power supply that I used for battery maintenance charging at 13.5VDC because (it seems to be) the current flowed back into the PSU when the mains supply failed destroying the PSU or will I have to design another circuit similar to an ideal diode circuit to prevent this?

                    My idea is to design a complete battery maintenance system that would utilize inputs from a PV charge controller (maybe an hybrid one for PV and Wind generator supply) to keep the battery charged with the input of the mains charger if the inputs from the DC charge controllers is too low, and thus using the battery system like a DC ups with inputs from mains supply and PV/Wind charger. I hope this make sense.

                    We have a huge problem in South Africa with the mains supply (failing and not very constant level) and my (and fellow HAMS) main objective is to keep the communications network running if everything else fails.

                    My knowledge is also limited to radio and electronic Hobbyist and DIY'er as I never had the chance to gain expert knowledge in anything.

                    John, another thing, limiting the inrush current especially during the radio transmission periods which may require an instant current of up to 20A, would it be something similar like this circuit with Q2 as the 2N22222N2222 or must it be over the combination of the 2N22222N2222 and 2n7000 switch

                    What would the influence be on the requirements of the radio during TX as if the supply current and voltage is not adequate it may damage the final PA and or not TX at all.

                    Pardon for all this many questions

                    Regards

                    1 of 1 people found this helpful
                      • Re: Lead acid protection circuit
                        dougw

                        If you need to prevent reverse current from flowing through the body diode of a FET, you can use an N-channel FET in series with a P-channel FET.

                          • Re: Lead acid protection circuit
                            zagene

                            Would something like this work then (I changed the p-channel mosfet to FDS4435BZ for a smaller package for mounting on a PCB that is not wider than a 7A/h SLA wich have a limited output current). The current flow would be then only in one direction preventing current flowback if the input fail.

                             

                              • Re: Lead acid protection circuit
                                michaelkellett

                                Forget the ZXGD3112 - it's totally weird, for example fig 4, page 6, needs two additional isolated floating 10V supplies and still ORs the source supplies via the MOSFET drain/source body diode.

                                 

                                 

                                I can't quite decide if your circuit will work, when Input voltage is present U1 will conduct (due to the body diode) so the mid nodeof the MOSFETs will go up to nearly Input voltage.

                                If Q2 is on the gate of U1 ispulled low so it turns on and shorts the body diode.

                                So the source of U4 is at Input voltage and it's gate is low so it should turn on.

                                 

                                I would reduce R2 to 10k and increase the 10Rs to 100R.

                                 

                                To get a good idea of how it will work in practice why don't you use a simulator - LTSpice is free and can do this - don't worry about getting models for the MOSFETs  - just use something similar that's already in the library.

                                 

                                There are lots of other free simulators.

                                 

                                MK

                                2 of 2 people found this helpful
                          • Re: Lead acid protection circuit
                            colporteur

                            Your post mentioned switching batteries. My commentary is not on the circuit but, on the batteries.

                             

                            My daughter has gutted and is retrofiring a 1983, 14ft travel trailer. I am currently building from the ground up, the electrical system, converter, fuse panel, wiring and battery.  I have discovered that using 2 X 6VDC batteries to achieve 12VDC is preferred over a single 12VDC battery.

                             

                            The information I have is anecdotal. Both the travel trailer vendor and a battery supply house recommends 6VDC batteries used in golf carts. They indicated the performance of recover time, number of charge cycles and the length they hold a charge is much better. The 6VDC battery has a demanding market, that the battery companies are focusing on.

                             

                            The other benefit they suggest is the reduce cost of replacing a battery. I would think you would replace both batteries if you were doing a replacement but that is just me.

                             

                            I have taken them up on their suggestion to use 6VDC batteries. I am still working on the trailer so I can't give you any field data that suggests the recommendation has some merit. When you think about it, golf carts seem to have a lot of getup and go and last a few beers at least going around the course.

                             

                            Sean

                            2 of 2 people found this helpful