Planing CanView and BMS+ conversion with lithium

No individual cell group Voltages? I don't understand the SOC% curve.

 

Hi wb9k,

I have to set all my CellLog8s to do data logging and then collect the data files from all 10 of them. Too much of a task. At this time I just keep a visual on all the CellLog8s. Maybe I will do it later on.

The SoC curve is the SoC of the OEM NiMH HV battery after it has been modified by the BMS+ to make the HV ECU use less ICE. It still does it in forced EV mode as it does not recognise Forced EV mode.

 

Post was reproductive, sorry.

 

Hi wb9k,

I finally just got around to logging the individual cell voltages during my travel to work and the charge at night.




Data files are attached. I have a complete file of all 70 cells into one excel file but it was too big to upload (11MB). I can email it to you if you like, send me an email so then I can email you. My email address is included in my signature.

Let me know what your opinion is.

Thanks,

John

 

I just peeked at the two pics in the body of your message. Looks like you're a long way from getting to 0% SOC in that data. Do you have a curve or set of curves showing a full discharge cycle?

I don't fully understand the curve labels below the full pack level in the second graphic. Are those modules in the pack, or is that individual cell groups in a single module? (IOW, what does the letter "B" represent?)

Forgive me if I'm a little slow to answer for the next few weeks or so. We had a house fire last Monday (actually a garage fire, but smoke and water trashed the whole house), and will be displaced for the next few months. I'm more than a little discombobulated lately. OTOH, this might be just the thing to get my mind off all the spare time I have with nothing to reach for to play with!

Send me that big file at (my moniker)@comcast. net. Two corrections needed there, should be obvious ;-)

See you soon!

 

I only discharge about 23AH on my commute to work. I will have to log a complete discharge if you insist. The last time I did a complete discharge I measured a 38 AH discharge according to my EV display unit before the LVC occurred.

The battery pack is made out of 5 cell groups. Each cell group contains 14 cells which are logged by two CellLoger8s. On each CellLogger cell 8 is not used.
So for example B3.1 means cell group 3 first cellLogger. Cells 1 to 7
So B3.2 means cell group 3 second CellLogger, cells 8 to 14.
B- battery.

Look on the positive side. It could be worse, your house could have been blown over by a tornado.

Your email address will not work for me. I sent you a PM.

 

Hi John,

Nice data collecting! Your nomenclature is a bit different than what I'm used to. We say packs are made of modules which are made of cell groups which are made of cells. So, what you call a "cell group", I would call a "module". What you are calling "cells" I would call "cell groups" (of two individual cells in parallel). Module 1 in the pack would be the negative-most module in the pack. Cell group 1 in any module is the negative-most cell group in that module. We would say your pack is 70S(eries)2P(arallel), comprising of 5 modules that are each 14S2P. This is the terminology I use below. I assume you've arranged your cell groups 1-70 from negative to positive, but I guess it doesn't really matter for our purposes here.

I would suggest getting the pack a little better balanced. Cell groups 64 and 51 are well below the rest of the pack. CG 23, 27, 34, 36, and 66 are also lowish compared to most of the rest of the pack. CG 43 is well above the rest and needs to come down a bit. Ideally, your data would start with the pack fully charged and balanced to within 20 mV or less. 5 mV or less is better. Present spread at the top of your charge cycle is almost 200 mV, which is pretty wide. You would then record a full discharge cycle and the following charge cycle. This will enable you to make good apples-to-apples comparisons of the state of health of each of your cells and see right away if anything is awry. Hope that helps and I look forward to a full data set!

 

OK I can see why you find my jargon that I have picked up by just reading on the web to be confusing. I don't have your industry experience.

Is the word "buddy cell" used in the industry? Or is it just a common laymen's term?

Anyhow I manually balanced the battery pack the best I could before I started using the battery pack and after all this time of top balancing at the end of every charge I would have expected the top balance to improve but it appears that may not be the case. Even though I only deplete the battery pack by 23 AH on my commute to work and back. Anyhow I manage to get 38AH out of each depletion of the battery pack so it cannot be too bad.

I have data harvested a complete depletion of the battery pack tonight and at this moment it is data logging the charge.

So I should be posting data tomorrow night.

PS just incase you do not know my miniBMS boards for this kit start bleeding at 3.45 volts and the HVC is 3.65 volts. My Elcon charger charges a CC of 5.9 amps.

Thanks,

John

 

Hi wb9k,

I have sent you the big 11MB data file to your email address which includes a full PHEV battery pack discharge followed by a charge cycle.

All individual cell log data files are attached in one ZIP file.





Thanks,

john

 

Hi John,

No worries on the terminology. I've never heard the term "buddy cell" outside of the DIY community. There's nothing wrong with one set of terms vs another, as long as everyone knows what is actually being talked about.

I managed to put together the data from the two big files you sent (your "original" second file was fine, I was just confused over what I expected to see....sorry.) I don't see anything really troubling here. I think you could afford to drop your LVC to 2.0 Volts (many of our customers do), especially since charging of the pack starts immediately by the ICE. I like where you have the HVC, but to spot any outlier cells with regard to capacity or performance, you still need to tighten up the balance of the pack. This should be easy to achieve with your plug-in charger. Just let the pack rest for a while after charging and re-plug to initiate another round of balancing. Repeat until your Voltage delta across the whole pack is <5 mV. Might take several times through the routine to achieve that level of balance, but it's worth doing for several reasons. Primary among them is that it will be much easier to see any cells with outlier characteristics. When you're at the top of charge after rest, the cluster of cell traces on the graph should pretty much be right on top of each other.

With balance being what it is now, all I can say is that you don't seem to have any cells that have any real problems and you are doing a good job of protecting the cells from harm by overcharge and overdischarge. Are you getting the mileage you expected from the system?

These files are pretty massive. It's nice to have the resolution, but you can probably afford to dial back the sample rate on the CellLogs a bit, at least during charging.

Keep up the nice work!

 

I am getting 38 AH out of the rated 40AH 70S2P pack. I am happy with that as I only use up about 23AH each day on my commute to work and back, so SoC only goes down to about 42.5% for most pack discharges which should prolong battery life. 38AH capacity is a bit less than what I expected but is close enough for me. After all they are grey-market A123 20AH cells.
Thanks for looking at my data.
John

 

You might manage to squeak out 39+ just by tightening up the balance and allowing LVC at 2.0V . You sacrifice a tiny bit at your HVC (probably a small fraction of an Ah) , but the longevity gains should be worth it, IMO. Keep us posted on how the cells hold up for you.

 

What range (km) do you get from your 38Ahr charge ?

Thanks,

 

Hi SynEco,

I have been on holidays for the last eight weeks so excuse my late responce. I am currently measuring the km range for a full charge/depletion and I will post my result tomorrow.

EDITED: Range for complete battery pack: 66km/38ah.

 

what is the CANVIEW reported A/hr for your 66km range ?

 

Ooopss.... when i read again ..can i assume the 38 A/hr per charge is what CANview reported as actual usage per charge ?