Prius Battery Replacement Kit (GenII/GenIII) with NEW custom cells

Is this torque for obd2 that you mentioned? I have it and an adapter. If so, what modules do I need in your opinion? I want to start tinkering.

 

Yes. Obdii Bluetooth device. Torque Pro app. (not free version). Android phone.

Set up a custom screen similar to my picture from the post.

 

The pro version came with my adapter.

 

Touchy point for me. Anything I DO sell or throw away, I end up regretting and fretting over for years... (7.5A Variac, broken AAA-D-cell charger with extremely good contactmechanism, etc)

@2k1Toaster: I do not need a pack yet, but could you have a look at what shipping to the Netherlands would cost? Just to give a rough idea? Customs will have something to add I fear...
And how much does it weigh? I will probably visit the US in the next couple of months, maybe I can smuggle it past customs...

 

You could try private messaging him by using the "conversation" feature here. It is the envelope next to the alert flag when using the web interface.

 

2k1Toaster.

The problem with cylindrical cells is temperature varation.
The layout of your round cells provide no temperature balancing (which is just as important as voltage balancing).
The prism cells shine here as they are all sandwitched together. Distributing there heat equally. All cells in the orignal hv battery have the same temperature (except the end cells) but the end cell varation is minimal.

I've been racking my brain for 12 hours now trying to figure a way you could temperature balance your cells within the indivual modules. Example; change airflow direction, heat pipes, heatsinks, all cells mounted vertically within each module?

Module to module temperature varation is probably ok. But cells within each module is what needs improvement.
The top row of cells heats up the cooling air passing by the second row of cells. The first and second rows of cells
heat up the airflow to the third row of cells. So the last row of cells always run hotter than the first row of cells. And the
second row cells are always hotter than the first row of cells.

The hotter cells will have a higher internal resistance compared to the cooler row of cells. This higher internal resistance (cells) will have less amp/hour capacity than the cooler cells.

The Charging current acrossed the higher resistance cells will cause higher voltage drops (higher power distribution) acrossed the hotter cells.

The hotter cells along with there decreasing capacity will getter hotter and hotter with each charge/discharge cycle.
eventually (in a couple of years or maybe only one summer) exploding.

ERICBECKY you can configure the torque voltage gauges to show more decimal points of resolution. But Displaying internal battery module resistance is more useful than cell voltage. Cell resistance is the same as cell capacity just different units (voltage/resistance).

Ericbecky you've neglected to show hv battery temperatures in your pictures of the torque apk screens.
Please setup module resistance gauges on the next torque page and share with us.
The standard torque apk gauges show only .1 decimal point of resolution The individual gauges can be configured for as many decimal places as the battery ecu is capable of. 1 decimal place is not good enough for the higher resolution that the ecu outputs.

2k1toaster
Perhaps your module could be redesigned in such a way that they could be inverted periodicly (yearly, bi-yearly)
to exchange the hotter temperature cells with the cooler temperature cells. Or (this is crazy) change the polarity of the cooling fan periodically to change air flow direction. Putting a suction on the top row of cells and airpressure/flow to the bottom cells (probably not efficent at all).
whatever nessary to extend module cell life?

I would especially like to know what is the maxium current displayed with torque apk durring hard regen braking and hard throttle conditions.
My hv battery with 95k miles and 8 las vegas summers peak at 92 amps regen braking and 105 amps heavy throttle.
What is the amp flow under these conditions with your cylinderical battery?

These peak current readings (on my prius) have not changed through-out the life of my genII prius. I've had my prius since new with only 8 miles showing on odometer.

GenII airflow is from top to bottom. GenIII airflow is bottom to top. Both a problem for your pack design.

I think smaller diameter longer cells mounted vertically and packed tightly together may be the only way to acheive longivity/temperature-balancing with cylindrical cells.

Thanks for your efforts and stay positive you will be successful and acheive a 10 year battery pack.

 

Call me up and walk me through what you would like and I'll see what I can do. 608-729-4082

Else send me an Android device pre-configured with what you'd like. We can talk through it. Then I can send it back to you when I'm done.

 

Ericbecky
Your existing torque apk screen (shown in your priuschat posts) can be adjusted for more decimal digits of accuracy.
Simply touch and hold any voltage gauge this will bring up a menu for that gauge.
From that menu select display-configure (this brings up a sub menu). At the bottom of this sub-menu is
number-of-decimal-places (option) Uncheck use-default-box then enter 2 or 3 decimal digits.
Some ecu sensors provide 2 decimal digits some give 3 decimal digits of accuracy..
You can add two digits for the battery current gauge (hv current) and two digits for max and minimum module voltages as well as the individual battery module voltages.

From the torque apk display screen you can scrowl (wipe) up and down to other display screens.
You should choose an unused screen (or delete unwanted gauges from unused screen) and add hv temperature 1,2 and 3 and also add internal-resistance-r01 (module 1) or as many battery modules as you like. Up to internal-resistance-r14 (add 3 decimal digits to the internal-resistance-guages (giving milliohm accuracy .001) probably .020-.029 would be acceptable on a heated up module). The lower the internal restance the better.

There are three battery sensors (thermistors) in the hv battery pack. These sensors would be ideal if they could be placed on a first row battery string, middle row battery string, and third row battery string.

Prefereable all on the same module (the sensor wire lengths might not allow all to be placed within the same module). But if placed on different modules would probably be ok.

And perhaps you could share your readings on hv battery current during peak regen braking. Normal is about 92 amps provided temperature sensors are all below 103 degrees F.

When battery temperatures hit 103-104F current limiting is imposed from the ecu. (battery cooling fan starts running at 96F)
So what are the peak regen braking amps on a cold or normal temp (below 100F) battery pack. As viewed from your torque apk. I see in your posts you already have the battery current gauge installed.

Thanks for all your testing and keep it up..

 

Here is the set up.
For some reason block 6 IR is not available from the drop down.

Have you got the pid handy so I can program it?

 

One more quick screenshot while parked.
I added the temp.

This is way too much info to be useful for day to day monitoring. Looks like there may be a bit of lag in the data.

But hopefully someone finds this helpful for now.

 

And here are two screenshots.
One with hard acceleration up a hill.
And hard braking going down the hill.

Again some lag in the data. But hopefully it is useful.

Edit: I notice that the battery temp reading is slightly cut off when I rotated my phone. I've adjusted the layout so it is fully visible for the future. Sorry about that.

 

I'm curious about the IR. My typical IR using techstream on a prismatic battery pack is around .019 on a newly installed cool battery, which then can gradually increase to ~0.021 when fully warm. The numbers you're showing are typically the point where I see significant capacity problems with a module. I've had screen captures of Imax 150+ amps (174 iirc) during heavy accell and ~105 on heavy braking. I understand different designs probably have different characteristics, so I wouldn't consider a different IR to be necessarily unexpected. I wonder if the higher IR creates an artificially low current max. Even if it did, I'd imagine it wouldn't be significantly noticeable during normal operation. I still have Battery "A" installed for experimenting in my 2007. When I get home, I'll throw the laptop at it and see what the blocks show for comparison. It's all original except for having the prolong treatment performed several weeks ago.

Sometimes it can be pretty tough getting screen shots while putting the hammer down, lol. You did a pretty good job. I'll cheat and sit in the passenger seat with the laptop while my son gets to abuse my car from the driver seat. Same thing when I test a battery after installing it in a customers car. They abuse it while I monitor. It can be fun, especially when you have a enthusiastic owner. I've been thrown against some seatbelts pretty hard a few times during the braking tests. I think forum member J.N. has the regen record for me with a 115. I may be wrong about that, tho, as my eyes went blurry and almost popped out during that brake episode. These little cars are amazing sometimes.

 

Ok
Last shot of the day.

I filled in the space intended for IR 6 with IR 7 because I didn't like the uneven look of my screen. And I was too lazy to look up the PID for IR 6.

 

What do you think about the IR difference mentioned in post 253?

 

VERY NICE Ericbecky!

Internal resistance r05
pid:21d0 Unit:Ohm Max/Min: 10.0/0.0
Equation: 0.001*t
Internal resistance r06
pid:21d0 Unit:Ohm Max/Min: 10.0/0.0
Equation: 0.001*u
Internal resistance r07
pid:21d0 Unit:Ohm Max/Min: 10.0/0.0
Equation: 0.001*v

Strange you don't have r06 hummm.

I'm not sure how many gauges are safe to use on one screen. I keep my cheep chinese $6.00 blutooth odb loaded with only 12 gauges per screen.

Large screen android device with lots of gauges programmed might eventually exceed obd blutooth data throughput ?
There is a menu in torque apk that shows data errors when data throughput is exceeded.

Your hv battery current can also be expanded one more decimal place.
nice work Eric.

 

+1

Just as one cell HAS to be the runt of a module, one module HAS to be the runt of the pack. In an ideal battery the difference between the best and worst would be zero. In a very good battery the spread would be as small as possible. Eric's load and charge snapshots show that block 12 is the runt and block 10 is the bully of this pack.

This should not be taken as passing judgement on this style of pack. It's just what the numbers show. Block 12 has the widest voltage swing and highest IR. Block 10 has the smallest voltage swing and lowest IR.

I always see .019 on a newly installed pack. I think this is just Techstreams default reading until it has enough data to calculate an IR.

It's great to see some real world test numbers from this new style pack.

Brad

 

Eric those battery readings indicate cells over heating. take control of the hv battery cooling fan using torque app
and the following pid's (try to cool batterys down to .025 ohms or lower) leave the fan speed on continously if nessary.

Set Battery Cooling Fan Speed 1
pid: 30810001 unit:No reply req'd Max/Min:0.0/0.0
Equation A

Set Battery Cooling Fan Speed 2
pid: 30810002 unit:No reply req'd Max/Min:0.0/0.0
Equation A

Set Battery Cooling Fan Speed 3
pid: 30810003 unit:No reply req'd Max/Min:0.0/0.0
Equation A

Set Battery Cooling Fan Speed 4
pid: 30810004 unit:No reply req'd Max/Min:0.0/0.0
Equation A

Set Battery Cooling Fan Speed 5
pid: 30810005 unit:No reply req'd Max/Min:0.0/0.0
Equation A

Set Battery Cooling Fan Speed 6
pid: 30810006 unit:No reply req'd Max/Min:0.0/0.0
Equation A

I run fan speed 4 all summer long and speed 3 all winter long (for the past 4 years) .
I have duplicate android screens with the only differences between duplicate screens are my fan speed choices.

Since editing torque pid' can get troublesome I'de recomend entering only a couple of fan speed settings.
speed 4,5 and maybe 6.

A manual speed setting from torque app will over-ride the ecu's ability to control fan speed.

example you set fan at manual setting 1. The batterys start overheating ecu trys to increase fan speed but fails.
Speed is locked to the slowest speed till new torque screen is selected.
A torque screen with no fan gauge selected reverts back to automatic ecu controlled fan speed