DIY Electric Car Forums banner

1 - 19 of 19 Posts

·
Registered
Joined
·
61 Posts
Discussion Starter #1
I have attached pictures of the section screens as we have programmed them. We keep getting an error in the 3I phase, or 3I-3V transition, or 3V phase. We see the voltage rise to 380.7X then it faults out. "SDT Batt VoltHi : ERR" Please take a look and suggest changes to either diagnosis or just point out our logic error...
In the pictures the status lights were blinking but are not always visible here, so we are using them for diagnosis.
120180
section1.jpg
section2.jpg
section3.jpg
section4.jpg
section5.jpg
Monitor clear 3-I Phase.jpg
Monitor error 3-I Phase.jpg
Monitor error page.jpg
120181
120182
120183
120184
120185
120186
120187
 

·
Registered
Joined
·
1,317 Posts
I've been using multiple NLG513 for 8+ years, I have various programs I've used etc. I'm trying to wrap my head around the logic, especially the redundancy of the OR statements of above a min or above 360 min. Also why the multiple stages stepping up the voltage, I get stage one as a safety check, but what is stage 2 for? Stage 5 looks like you want to overcharge the battery, or let it "float". I assume this is a lithium pack? Give me some details on the battery, and what you are trying to accomplish. The stage 3 issue could just be the above 380v as it's just not required because you have a charge voltage of 384v and a condition to move on when current drops below 5A, This is all you really need. I find the Brusa works a lot better if you keep the programming simple.
 

·
Registered
Joined
·
61 Posts
Discussion Starter #3
image.png


I am charging a 2013 nissan leaf 24kWh pack to 4.0V*96=384V
Section 1 is connect and check voltage for safety before proceeding. The minimum 1 minute requirements are to ensure I can see the lights flashing indicating which section it is in rather than skipping through sections making diagnostics more difficult.
Section 2 is stage one in the above image.
Section 3 is stage 2 in the above image, bring it up a little more and saturate. The 360+ minute numbers are to turn off based on too long a time.
Section 4 is stage 3, shutdown, BUT we have a DC-dc convertor that charges the 12V system to run the car side and if we turn of entirely the DC-dc will deplete the 400V system until we get out there to turn the whole system off. Thus the 0.5A to keep things topped.
Section 5 is stage 4 to bring it back up if the DC-dc system depletes too much.

I would be interested in a schematic of your high and low voltage cross connection system if you have one.

We considered a Porsche 914, 928, 944, but with family opted for something that could carry 4. lopified.blogspot.com
 

·
Registered
Joined
·
1,317 Posts
I get what you are going for, however I'm not sure if that's the best way to deal with the issue.
What is your 12v drain? The brusa will actually put out 14v @ 0.5A on the 12v input line if it has AC power, this is based on the manual, I've never actually checked.

For section 1 to be a safety check, I would put the charge current at 0A, a safety check should be automatic and not rely on you being there to watch lights. If voltage is below 288 and current is set to 0A it will just hang out indefinitely, with the charge current set to 1A it will just charge it up from whatever voltage it's sitting at, even if the pack was discharged dangerously low.

I live in Canada, and my first stage is a battery temperature check. If the battery is below 3 degrees Celcius it will just pause until the condition is met. There is an external heating system that tries to keep the battery at 15 degrees C when plugged into AC.
Stage 2 charges to 1v less than my target voltage at full current. Transitions to Stage 3 once the pack reaches a certain voltage. (This stage isn't even required, I was playing with the idea of having an external toggle switch that would override stage 3 making this the final stage so that I could charge to 70% or something like that, but still have the option to charge to 90-95% at the flick of a switch)
Stage 3 charges to target voltage at full current with the termination condition of the current dropping below 2A, or a time limit. This is technically the only Stage required to charge a lithium battery. Anything beyond this should have a specific purpose (like your first stage voltage check) but it shouldn't pass a safety stage automatically with a timer.

I also have a temperature sensor connected which will turn off charging if the pack gets too hot. I think this is important for lithium, the brusa is capable, why not do it. This overtemp setting doesn't rely on being in a certain stage, it's simply a global setting that will turn off the charger no matter what it's doing.

The reason the brusa has such flexible stages was for lead acid. Lithium doesn't require it, and if you overcomplicate the logic it could become less reliable, not more reliable.

I don't like the possible sacrifice of the high voltage pack just to keep a 12v battery alive. We had a Fiat 500e in recently that discharged the high voltage pack down to 60v while sitting in an attempt to keep the 12v battery alive. I think this is retarded logic, lets save a $100 battery by destroying a battery worth thousands. The Ioniq hybrid for example has a 12v restore button which turns on the dc/dc converter using a button on the dash if the 12v battery dies. I did much the same in my Porsche 944 by having a small 3S 18650 pack that could be turned on to apply pack voltage to the dc/dc converter to "jumpstart" my car if the main 12v battery was too low to turn on the dc/dc converter. This used a momentary pushbutton to briefly connect the dc/dc but once it was powered it supplied 12v to input of the main contactor turning on the dc/dc with the car. Since the drain on the 18650's was simply coil current on the contactor for about 1-2 seconds I never had to charge it, it would probably be good for 1000 "jumpstarts". When I come across a deal I'll be building my 12v battery out of LTO cells so that there is no issue with a drain to 0v.
 

·
Registered
Joined
·
1,317 Posts

·
Registered
Joined
·
61 Posts
Discussion Starter #6
Responses inline.

I get what you are going for, however I'm not sure if that's the best way to deal with the issue.
What is your 12v drain? The brusa will actually put out 14v @ 0.5A on the 12v input line if it has AC power, this is based on the manual, I've never actually checked.

The Brusa does output 12V when it gets 220V input, that is what we use to turn it on.
We have several pieces partially built. The drain is the Nissan leaf BMS and the DMOC adapter that interfaces with the Nissan BMS (and Nissan battery temperature control and heaters), DMOC, VW ECU, Brusa Charger, Vicor DC-dc, and it controls the Nissan Leaf precharge and main relays. All of these will communicate on CAN.
Currently the BMS operates on its own to maintain cell voltages. It always runs.
The DMOC adapter runs all the time unless we turn it off under the hood.
The current wiring requires us to flip a switch in the cabin to turn on the mains relays. When the mains come on the DC-dc primary stage and fans turn on, but no output until we turn it on. The problem being that when charging is complete the DC-dc primary stage, BMS and DMOC adapter are still on until we turn the system off with the switch.


For section 1 to be a safety check, I would put the charge current at 0A, a safety check should be automatic and not rely on you being there to watch lights. If voltage is below 288 and current is set to 0A it will just hang out indefinitely, with the charge current set to 1A it will just charge it up from whatever voltage it's sitting at, even if the pack was discharged dangerously low.
Good idea, will alter.

I live in Canada, and my first stage is a battery temperature check. If the battery is below 3 degrees Celcius it will just pause until the condition is met. There is an external heating system that tries to keep the battery at 15 degrees C when plugged into AC.
I am in the central U.S. My garage is not heated, but usually stays above 0C, 15C seems warm to me? I did install the Nissan pack heaters, they are controlled by the BMS, there are 3 temp sensors, one in each pack, but all report to the BMS and therefore the DMOC adapter which will tell the charger to charge or not eventually. None report directly to the charger.
Stage 2 charges to 1v less than my target voltage at full current. Transitions to Stage 3 once the pack reaches a certain voltage. (This stage isn't even required, I was playing with the idea of having an external toggle switch that would override stage 3 making this the final stage so that I could charge to 70% or something like that, but still have the option to charge to 90-95% at the flick of a switch)
Another neat idea, I think we can eventually do that through the CAN input.

Stage 3 charges to target voltage at full current with the termination condition of the current dropping below 2A, or a time limit. This is technically the only Stage required to charge a lithium battery. Anything beyond this should have a specific purpose (like your first stage voltage check) but it shouldn't pass a safety stage automatically with a timer.

I also have a temperature sensor connected which will turn off charging if the pack gets too hot. I think this is important for lithium, the brusa is capable, why not do it. This overtemp setting doesn't rely on being in a certain stage, it's simply a global setting that will turn off the charger no matter what it's doing.
Can you point me to appropriate temp sensors? These would be glued to the outside of the battery boxes as the boxes are already sealed and installed so I would just reduce the required activation temp.

The reason the brusa has such flexible stages was for lead acid. Lithium doesn't require it, and if you overcomplicate the logic it could become less reliable, not more reliable.

I don't like the possible sacrifice of the high voltage pack just to keep a 12v battery alive. We had a Fiat 500e in recently that discharged the high voltage pack down to 60v while sitting in an attempt to keep the 12v battery alive. I think this is retarded logic, lets save a $100 battery by destroying a battery worth thousands. The Ioniq hybrid for example has a 12v restore button which turns on the dc/dc converter using a button on the dash if the 12v battery dies. I did much the same in my Porsche 944 by having a small 3S 18650 pack that could be turned on to apply pack voltage to the dc/dc converter to "jumpstart" my car if the main 12v battery was too low to turn on the dc/dc converter. This used a momentary pushbutton to briefly connect the dc/dc but once it was powered it supplied 12v to input of the main contactor turning on the dc/dc with the car. Since the drain on the 18650's was simply coil current on the contactor for about 1-2 seconds I never had to charge it, it would probably be good for 1000 "jumpstarts". When I come across a deal I'll be building my 12v battery out of LTO cells so that there is no issue with a drain to 0v.
Completely agree about saving the 12V for the 400V... Just have not figured out the way around it yet.
The original configuration had the DC-dc connected direct to the 12V and always on when the mains relays were on, BUT it lead to the mains relays getting weilded shut as the DC-dc would pull current during the precharge closure and mains closure. We also learned that the Vicor DC-dc although listed to be capable of 12V 99A output is actually capable of 13.2V and HIGH amp output. Thus if the lead acid 12v battery was at 11.5V the DC-dc would output as much power as required to bring the battery up to 13.2V. This is why the BMS and DMOC adapter run from the battery and the entire car runs from the DC-dc until we have a solution to connect the DC-dc to the lead acid. Your solution of using a supplemental battery to momentarily hold the relays closed sounds good and we are thinking of ways we can use the "start" rotation of the ignition to caused similar momentary closure.
 

·
Registered
Joined
·
1,317 Posts
How big is your lead acid battery? How much does your BMS and DMOC adapter draw? What DMOC adapter are you referring to? I have a GEVCU v5.2 and it turns on and off with the ignition.
 

·
Registered
Joined
·
61 Posts
Discussion Starter #8
Lead acid battery for this car.
We need to measure the drain, BMS, dmoc adapter, 3 main relays, dmoc, anything the dmoc adapter turn on.

We are running the Wolftronics dmoc adapter.
 

·
Registered
Joined
·
1,317 Posts
I used the Wolftronix adapter for a while, good device, I managed to break it. I got away with a tiny beat-up 7ah UPS battery as my 12v for a couple of years, the drain with the car off was minimal. Just the radio as my Porsche 944 has minimal 12v systems. I'm using the Gevcu now and it turns on and off with the ignition, I don't recall but doesn't the Wolftronix do that as well? What I'm getting at is that if you have minimal draw when the car is off you shouldn't have huge currents trying to charge the battery back up when you turn on the ignition.
 

·
Registered
Joined
·
25 Posts
Hi @zepol_wube, I've been using the Brusa chargers for over ten years and used to correspond with the manufacturers quite regularly in the early days.

Your problem with the error is caused by your setting of the "switch off immediately" settings which are not described in the manual.

I looked up a very old 2010 Email and a copy of the text is as below.

About the SDT_BattVoltHi: This is a crowbar function to protect attached gear (like inverters) from overvoltage caused by the charger, e.g. due to load drop.
Cause it has to react very quickly, it’s not very precise, nor filtered, thus it’s prone to any kind of spikes or disturbances etc.
BattVoltHi is NOT suitable to protect LiIon batteries against a precise maximum voltage threshold.

You should set its level approx 10-15% above real end of charge voltage in order to prevent nuisance tripping.

I had the problem for ages when it would shut down mid charge. Just make the voltage setting higher than you have set it and it will work correctly.
 

·
Registered
Joined
·
61 Posts
Discussion Starter #11
Rusty, I just got this email this morning from Brusa.
Hi,

nice project you have there, pretty sure to learn a lot and have a unique car afterwards :)

I think you have done everything right basically, congrats you have come so far without support. What you probably didn’t know (and which is not described clearly I admit) is that the "SDT Batt VoltHi: ERR" is produced by a crowbar like circuitry, which measures output voltage very fast but not very accurately. This is intended to avoid damage to attached circuitry in case the battery suddenly disconnects and causes a load drop to the charger. The resulting voltage overswing might damage sensitive electronics like DCDC-converters, amp-hour counters etc., so it must be avoided by fast shutdown of the power stage.

Now: As I said, this fast measurement circuit is neither precise nor is it well filtered, so any peak / disturbance / noise etc will trigger the shutdown and throw the error.
The value which controls this crowbar/shutdown circuitry is the second value in the orange/yellow fields, which says «switch off immediately if Battery voltage above [V]»

And this value is set to 390V in your case, which is very close to the 384V you intend to reach. So remedy is pretty easy: Add a healthy headroom to the «switch off immediately if Battery voltage above [V]» value, let’s say you set the limit to 460V, and you’re done. Required headroom depends on various influencing parameters, however with a margin of 20% you’ll probably never see any nuisance tripping of this crowbar again.

I hope this information is useful, and wish you all the best with the project!

Best Regards,


PROBLEM SOLVED
 

·
Registered
Joined
·
61 Posts
Discussion Starter #12
I used the Wolftronix adapter for a while, good device, I managed to break it. I got away with a tiny beat-up 7ah UPS battery as my 12v for a couple of years, the drain with the car off was minimal. Just the radio as my Porsche 944 has minimal 12v systems. I'm using the Gevcu now and it turns on and off with the ignition, I don't recall but doesn't the Wolftronix do that as well? What I'm getting at is that if you have minimal draw when the car is off you shouldn't have huge currents trying to charge the battery back up when you turn on the ignition.
Yes, the Wolftronix unit can turn on off with the ignition also, it was that way to start, but we have not figured out what was causing the weilding of the main contactors so we are playing it cautious at the moment.

How / What did you break on the Wolftronix unit such that we do not repeat your error? If you are done with it, we can always use spare parts on the shelf just in case...
 

·
Registered
Joined
·
1,317 Posts
Yes, the Wolftronix unit can turn on off with the ignition also, it was that way to start, but we have not figured out what was causing the weilding of the main contactors so we are playing it cautious at the moment.

How / What did you break on the Wolftronix unit such that we do not repeat your error? If you are done with it, we can always use spare parts on the shelf just in case...
The Wolftronix unit handles precharge much better than the GEVCU (which is what I'm using now), by actually measuring voltage and closing contactors at the appropriate time. I'm surprised you are having issues with welded contactors, that's one thing that was flawless for me. I had a relay (brake light maybe) that I forgot to add a protection diode to, and it eventually destroyed the power supply in the Wolftronix, there was no built in protection for that. Otherwise, I think the Wolftronix is a far superior device, much more feature-complete with cruise control, eco/performance modes, methods of detecting a shorted or open throttle, gauge drivers and all that fun stuff. I sent back the mainboard, I don't think I have anything left but the case.

I just looked at the schematic for the Wolftronix, it mentions closing the negative contactor with pre-charge OR Positive, however if you take that literally and only connect it to the positive contactor control it wouldn't have charged the DMOC. It needs the diodes and to be connected to both precharge and positive for the system to work properly. How do you have the system wired?

Glad to hear the Brusa is working. It's an amazing device, but can be finicky!
 

·
Registered
Joined
·
25 Posts
@zepol_wube I have found the Brusa chargers really great pieces of kit but the interface is very poor and to my knowledge was never improved.

The charging profile is also confusing with so many choices. Pity they didn't at least have a few examples and I have found that the simplest one from CC to CV works just as well.

I did get some other interesting info from them and have a few pdf's I will attach along with the latest manual.
 

Attachments

·
Registered
Joined
·
61 Posts
Discussion Starter #15
The Wolftronix unit handles precharge much better than the GEVCU (which is what I'm using now), by actually measuring voltage and closing contactors at the appropriate time.
I see this as a timed event now, not a measured voltage. I would be interested to see this code and schematic.
I'm surprised you are having issues with welded contactors, that's one thing that was flawless for me. I had a relay (brake light maybe) that I forgot to add a protection diode to, and it eventually destroyed the power supply in the Wolftronix, there was no built in protection for that.
Yes, that issue has been addressed on our board.
Otherwise, I think the Wolftronix is a far superior device, much more feature-complete with cruise control, eco/performance modes, methods of detecting a shorted or open throttle, gauge drivers and all that fun stuff.
We have still to implement the cruise and gauge drivers, anything you have in the way of schematics or code would be helpful. We hope to attack the power steering today and the cruise later this week.
I sent back the mainboard, I don't think I have anything left but the case.

I just looked at the schematic for the Wolftronix, it mentions closing the negative contactor with pre-charge OR Positive, however if you take that literally and only connect it to the positive contactor control it wouldn't have charged the DMOC. It needs the diodes and to be connected to both precharge and positive for the system to work properly. How do you have the system wired?

The system turns on the precharge relay which also pulls in the neg contactor. When the pos contactor pulls in it mantains the neg contactor as the precharge turns off. We have adjusted the timing on the precharge circuit to allow a bit more time as our 400V pack seems to take a little longer to fill the DMOC caps and DC-dc system. And we placed a purposeful overlap between the pos contactor closing and the precharge releasing to cover any bounce time the pos contactor may have. Our blog lopified.blogspot.com shows wiring.


Glad to hear the Brusa is working. It's an amazing device, but can be finicky!
 

·
Registered
Joined
·
61 Posts
Discussion Starter #16
FYI, we tried your suggestion for stage 1 safety check, 0A 0V. The system will not turn on. 0A 1V, does nto work either, BUT 1A 0V does work, so that is what we have it set to now. Stage 5 has also been changed to hold 384V, 1A such that if we leave it plugged in over night the DC-dc system will not drain the 400V battery. Come winter when the heaters turn on automatically we may need to up the amp value.

Thanks for the help.
 

·
Registered
Joined
·
1,317 Posts
The safety check should be your minimum voltage at 0A IE: 300v and 0A with a next stage criteria of 288V, that way if your pack has been dangerously over-discharged it won't just try to charge without you having to look at it.
 

·
Registered
Joined
·
61 Posts
Discussion Starter #18
We cannot set the amps to 0, the system just wont turn on that way, but the volts are set to 0 so it wont charge anyway. The logic is set to require 288V before passing to the next stage, so we will sit in this 1amp 0volt condition indefinitely if the pack voltage is below 288v. See the very first image in this discussion as a reference for section one safety check. From that image we changed the U phase voltage to 0.
 

·
Registered
Joined
·
1,317 Posts
We cannot set the amps to 0, the system just wont turn on that way, but the volts are set to 0 so it wont charge anyway. The logic is set to require 288V before passing to the next stage, so we will sit in this 1amp 0volt condition indefinitely if the pack voltage is below 288v. See the very first image in this discussion as a reference for section one safety check. From that image we changed the U phase voltage to 0.
As long as it's working that's all that counts, it is kind of bizarre as that's what I've done in the past as a safety and it worked just fine. Maybe it's a different version of the firmware. My charger is the slightly older black version.
 
1 - 19 of 19 Posts
Top