Planing CanView and BMS+ conversion with lithium

Here are some pics. Still trying to use my DMM to log the lithium battery pack current. It has been interesting as to how the lithium pack // NiMH has interacted with the two Enginer converters.

The loop connections on my miniBMS slave modules are failing one by one as they are becoming open circuit permanently. Cannot figure out how or why?

I have used 100A/415V service fuses (council fuses) that are used for household switchboards in Australia.

 

Thank you for the pictures. So you really did the first two seat 12.5+1.3 kWh Prius on the world

Just a "few" questions.

Where is the problem ? Electronic (opto ?) ? Connectors ? Definitively, even after a "reset" ?
Is it during driving ? Charging ?
What said dimitri about this ?

Do you have some bms wire too close and // with power wires ?
So you could have high inductive current ?

How many amp. did you read on your dmm just when connecting Lithium on Nimh ?

 

I am still setting up my software for PClink on Windows 7 on my notebook. The lithium pack is not supplying too much current as they are being assisted by my 2 Enginer kits. Surprisingly my Lithium pack was full when I got to work as it was top balancing - red LEDs. The lithium pack acts as a storage buffer to do only the heavy loads when the required current is too much for the Enginer kits to supply.
I have not got around to inspect what is wrong with the loop connections on my miniBMS slave boards but they are definitely permanently open circuit. Just as well my whole 230v battery pack is top balanced. I will probably replace the normal loop wires with screened cable connected to ground to remove any induced voltages.

 

..

 

Interesting.

So, if you don't deplete A123 and your bms don't control all the cells, so you don't know if one could goes in overvoltage condition.

I always use, say 1-2Ah, in ev mode at the beginning of my travel so there is enough place for a hard braking.

Why not drive enginer kit deconnected and use 1-2Ah until your bms being 100% ok ?

It is even possible that your 2 enginer kit charge the A123 cells ? What is the enginer end voltage ?

 

Yep, I will be reducing my Enginer converter operating voltage from 242volts to 236 volts to avoid overcharging cells. 236/70=3.37 volts. I can also reduce the voltage at which the Canview relay switches off my converters. Once my EV display unit arrives I will know how full the lithium battery is. My Enginer kits charges both NiMh // lithium packs. It has taken me a while to figure out how to disable power saving mode on my notebook computer and DMM so they keep logging indefinitely.

A bit of over voltage is not too damaging as compared to overcharging and the two are separate conditions which some users cannot differentiate. I am more worried by taking the pack charge too low that a cell could become reversed voltaged and die.

I have now set my second converter setpoint to an CCL variable of 110Amps. This seems to keep the converter from running at a high NiMH//lithium SOC. Tonight I drove home from work 20kms doing it in OutOfGas mode easily. I did not see the HV battery voltage fall below 228volts. The two converters pretty much run most of the time supplying about 30amps to both NiMH//lithium or directly to electric motor inverter.

 

I will re-arrange my 5 lithium battery banks so the BMS faces out so I can visually check if they are top balancing whilst driving on my rearvision mirror.

 

I finally managed to log and plot the lihium pack battery current using my DMM and notebook computer. It shows a peaK discharge current of about 75 amps. I did a short drive in outofGas mode to maximise discharge current to 125 amps. I will post a picture of my graph on monday when I have fast internet access. Hard to say what the converters were putting out in combination with the NiMH pack.

75mV = 200 amps

 

My last slave board went today. Each day in the morning I have found the miniBMS head end board buzzer going off. The buzzer was going off because one of my slave boards had opened circuited and one just measured 10ohms but that was enough to make it buzz. Anyhow I traced the PCB and found that each cell drives a optoMOS relay with a normally open contact. Each one connected in series to make the slave loop circuit. Anyhow on four of my slave boards one of these optoMOS relays has a blown output. The input is a diode with a Vf of 1.2volts which I measured and confirmed to be good. The input and output have an isolation of 1500Vrms. However the CPC1014N - optoMOS relay has a MOSFET output and according to the data sheet it does not have a voltage rating across the output. Anyhow the noise - induced EMF has destroyed at least one output of 4 slave boards and the remaining have a resistance in the order of 1.2 ohms. Normally they have a resistance of 0ohms when output is closed. It appears that I need to replace all the optoMOS relays on each board. I have asked Dimitri for advice in an email.

 

We've got a few dozen Enginer systems using Mini-BMS and they all work great...the occasions where we thought we had problems had turned out to be our fault...and Clean Power Auto was prompt in helping educate us and resolve the issues related to how their BMS works. Dimitri should be able to help you also...

 

Thanks Paul. Dimitri has informed me that it probably has nothing to do with EMF. I agree as the output has a blocking voltage of 60 volts. It is a problem with the component quality - bad batch which does not get picked up during the one hour bench test. Should just be a matter of weeding out the bad ones. I have been using a miniBMS with my Enginer kit and it has always worked great. Dimitri has offered to repair boards or pay to recover cost of optorelay replacements if I do my own repair.

Dimitri also informed me that these optorelays are bidirectional and it is normal for them to measure around 1.2 ohms across the output in the closed or on position. The head end board will tolerate 1000ohms across the loop connections. Interesting.

 

Here some are more photos. I drove to work in pure EV mode with max speed 75Km in OutOfGas mode for 24km. Notice the low HV battery NiMH//lithium internal resistance value of 80mOHM.

I was getting a FE of 1.3L/100km just using two Enginer kits without Lithium 4.6KWH pack.

 

I have now congfirmed that you get a ERR CELL VOLT alarm due to the mod. for even current draw. I am also using 10 CellLoggers for datalogging function with my miniBMS. I would like to see a photo of your BMS setup. You could also use a miniBMS head end board (has ignition and buzzer/external buzzer and voltage protection relays) with your 10 CellLog8S transistor outputs. I am thinking of using my CellLog8S with my 75mV shunt to log the lithium pack battery current, just need to find a way.

 

Your 0,2 L/100km is superb.

Internal resistance seems low.

Here is a test i done.

1 pressing accelerator pedal,
2 release accelerator and press brake pedal
3 release brake pedal.



These values come from prius can bus so both nimh and lithium //.

Internal resistance is about 0.1 - 0.11 ohms.

 

I have tried using my serial output from my Canview and my CanDecoder program will not work, it just gives me an empty excel file even though the binary file has data. I think the CanDecoder program is not compatible with Windows 7. I have tried setting file compatibility to Windows XP and it still does the same. I will probably have to go back to using my old windows XP laptop.

 

I guess the two DC Converters change the internal resistance value shown on my Canview.

 

I have made a video of me driving doing OutOfGas mode on the motorway doing 84kph. 84kph is the maximum speed I can get. The DTC was cleared straight after invoking OutOfGas mode by using my modification. The EV display unit on the left is for my DC Converter input current. The one on the right is for my lithium pack output current. The Prius will not go back to normal until I come to a dead stop and cycle the power button. If I install the ECM module available from PIS I should be able to start ICE in highway speed if and when load is too high. It would be nice to be able to use the cruise control. It is a real pleasure to do EV mode at highway speed and the PHEV system is not working that hard to sustain 84kph. I did notice all my miniBMS slave boards red LEDs all light up during regen braking from decelerating from 80kph to 40kph when taking the exit ramp.
My two Enginer DC Converters do most of the work whilst the lithium 230v 20ah pack does the heavy lifting when the load is too high. So my lithium pack pretty much stays highly charged by my two Enginer PHEV kits.
YouTube:

 

It turns out that noise from my DC Converter is causing my miniBMS LVC signal on the loop wire not to work. With both DC Converters not running the miniBMS LVC signal works OK. With the converters running and the loop wire disconnected I measured a negative voltage of 5.05Vdc on the return loop wire. With the converters switched off I measured 250mV on the return loop wire. Of the two cell loop terminals one feeds 12Vdc. I am thinking of wiring a 1kohm pot in series to drop the induced voltage. The same noise may have caused my optorelays to faulter. My whole miniBMS is turning to s... due to noise from my two Enginer converters.

 

I am going to try to isolate the converter noise by the use of a normally open relay with a 12v 145mA coil with a power diode across the coil. The coil current has to be limited to a maximum value of 400mA (1amp peak) which is the output load rating of the optoMOS relay. I connected the cell loop series string of 5 slaveboards into the negative (ground) side of the coil. This worked so this is my solution so far.

Since installing a relay to isolate the Converter noise the optoMOS relays have stopped failing.

There is no way that the converter noise can be strong enough to energize a 12v 145mA coil.

This is a lousy way to spend on the day of Xmas eve.