Gen II Prius Individual Battery Module Replacement

Thanks for your help, I will be more observant on my next charge/discharge process to see what voltages are present as the state of charge of the car approaches 100%. The way the car calculates the SOC is a bit of a mystery to me. I assume it keeps adding the charging current up until it calculates 6500 ma (100%) has entered the battery, or maybe it uses voltage and current to calculate the state of charge? I will look for the 342 volts as the criteria for a full charge and not the cars computer number.
Thanks for your interest and help.

Roy from Canada

 

Hi Roy You can use an incandescent light bulb in series with the battery and charger to regulate and limit the current. For instance a 40 watt 220 volt bulb supplied with 220 volts will only pass 0.18 amps a 60 watt passes 0.27 amps. Raising the voltage will raise the current but never into a range that will damage the battery with to much current. The bulbs will blow first.
Monitoring the current you can change the bulb or bulbs till you get the rate of charge you require. For instance you could use 3 120 volt 40 watt bulbs in series to give a maximum current of 0.33 amps at 360 volts or 330 MA just about perfect for your application.
The bulbs resistance goes down as the bulb goes dimmer and up as it gets brighter thus giving a crude but effective regulation to the current.
The battery will have some resistance so it may be necessary to adjust the number of bulbs and there size.
I would not charge the battery to full with the car contactors in as it is unknown as to what the battery ECU would make of this. Just monitor the voltage into the battery then discharge finally to your 60% level.
As "valde3" above states the idea is to gently overcharge the battery holding the current at around 350 ma with the battery at around 342/345 volts for about 15 minutes. This is only around 0.5 watts per cell. Lower charge rate is better than higher, but if say 50% lower time will need to be increased.

John (Britprius)

 

Roy I think you have a typo in your last post that should read 342 volts not 242.

John (Britprius)

 

I think Prius uses both voltage and current to calculate battery stage of charge. Probably uses voltage to determine “low” (lowest voltage it lets it go before charging it on normal use) and “high” (highest voltage it will charge it in normal use) and then calculates up and down from those by current.

 

The Prius uses voltage and coulomb counting for state of charge and discharge, but charging from an external source directly to the battery will not be calculated.

John (Britprius)

 

Thanks John - the idea of the light bulbs changing resistance thus regulating the current is a good idea. Do you use that method to control your current when charging? How well does it regulate?

 

John - I think the computer in the 2001 prius is seeing the charging current just as it would from the cars generator system, the computer in the car is showing me an accumulation and use in techstream while charging/discharging. This probably because the connector built into the car it does not go directly to the battery as may be the way you charge your newer cars, When I get my spare battery case I will trace the wiring from that connector. I suspect when you charge your battery you go directly to the battery terminals and not through the computer so you cannot see the state of charge go up and down as I do. Is that possible? The 2001 to 2003 were sold with chargers in some cases and connectors to charge in the trunk for all of them I think.

Roy

 

Yes John it should be 342 volts. Thanks for the correction, it will help others to understand my post.

Roy

 

Initially I did use bulbs and they work very well once sizes are worked out. Switches can be put across the bulbs to take them out of circuit. I mounted three bulbs on a board with switches below. It looked Heath Robinson, but was very effective.
I am now using a voltage regulator chip configured as a fixed current regulator. I am not happy with the circuit at the moment because it can only have a maximum voltage across it of 40 volts . Anything above this blows the chip so I have to be extra careful how I connect and disconnect the charger bearing in mind for the gen2 I need 250 volts to charge the battery "90 volts less than the gen1" so this is work in progress.
To give an idea of the resistance change in a bulb. I have just measured the resistance of a car 50 watt headlamp bulb. Cold it measures 0.5 ohms meaning supplied with 12 volts at start up it would take 24 amps when hot it would 50 watts divided by 12 volts =4.16 amps or a resistance of 3 ohms.

John (Britprius)

 

I will rig up a similar system and see how well it regulates for me.

Thanks

Roy

 

If you wish you can put the lamps in the input side of your variac but bear in mind your voltage hear is only 120 with an output three times this value so the bulbs or bulb will need to be rated at perhaps a total of 150 watts. Being much safer to handle 120v ac compared with 340v dc. Try three 60 watt bulbs in parallel. you can put them in or remove them or change the wattage to give the output you require.

John (Britprius)

 

Yes would be safer to regulate the output of the variac before it is converted through the dc bridge. Will try that.

Thanks
Roy

 

Perhaps surprisingly it will also work on the input side of the variac. Here because of the near 1 to 3 step up of voltage in the variac the current on the input is 3 times that of the output. On the input you can put your 120v lamps in parallel and mix the wattages in any way you wish. On the output the 120v bulbs will need to be in series "3 of them", and all of the same wattage to handle 340v.
The variac will still work in the same way, but the current through it will be limited. Even shorting the output would do no harm the bulbs would just light at full brightness.

John (Britprius)

 

I will report back later in the week if my daughter brings me the car.

 

For those who want to build their own 234-240V 1.2A Constant Current Charger (including over current protection, over voltage protection) from less then a dosen of electronic components (SMPS board plus few other things) under the $300 budget are welcome to ask for a Design Pack (Schematics and BOM).

 

Hi Kiwi --- That is a nice thing to do, many will probably take you up on your offer.

Roy from Canada

 

So i'm about 600 miles on my rebuilt/rebalanced battery.

I ended up replacing 6 modules total. 4 were just toast (sub 4000 mah capacity) and 2 were sub 5000 mah. I paired them up by delta-volt and capacity the best I could, but ended up popping 2 codes recently: P3014 and P3015.

Upon scanning I realized that these codes are "block voltage is weak," which leads me to believe that I mispaired these or the weaker modules I paired them up with was too weak. Of course, i cleared the codes and they go away. My question is this though, how does the ECU detect if a block voltage is weak? is it based on the deviation from the rest of the modules under load? Or is there a low voltage threshold that it cannot go under?

Because if that is the latter case i'll just buy two modules to replace the weak ones in those two blocks, rebalance and call it done. However, if it the code is triggered based on the performance of the other modules, I will need to rethink my strategy of how to keep the ECU happy.

Anyone have any thoughts or ideas?

 

The DTC is triggered by a voltage difference in the block pairs I believe of 0.3 volts. Because of the flat nature of the Nimh voltage discharge curve this discrepancy is most likely under load where the internal resistance of the cells plays a bigger part. Did you load test the modules to find there internal resistance before rebuilding the battery?

John (Britprius)

 

High John - Roy from giving my 2 cents worth re voltage differences. On my 2001 prius cars I see voltage differences of .3v regularly and .5 sometimes and the cars run with no codes. These readings are when the battery is not charging or under heavy load while driving. When the car is charging the battery the voltages get close to .2v difference.
However I did have one battery that was off by 1 volt and It triggered the code and the triangle of death, my techstream identified the module pair and replaced a module.

 

Roy is this a difference of 0.5 volts across all of the pairs or a difference between one pair compared with the next lowest.
I hope you understand my meaning there.

John (Britprius)