This is a quick calc I did … based on preloaded BMS values.
These intermediate voltages are used by @Louisvdw’s driver to control the charge current magnitudes at various points in the charging cycle. Ideally these voltages approximate the charging curve of a cell. They are hard-coded in the driver but can be adjusted to your own values, depending on your particular cell capacity.
@Louisvdw 's driver as standard has low charging rates at low and high voltages and pumps it a bit in the mid-range.
Just a correction. My driver use the battery SOC to drive the charge limits. It will be influenced by these cell voltages, but I don’t use the cell voltages directly.
2 Likes
OK, so it is a combo of coulomb counting and cell voltage?
yes. that would be the BMS’s logic for SOC.
Methinks a BMS should work on the average volts of all the cells (much more accurate than a lead-acid bank), with coulomb counting to get the average SOC, until one cell shoots out “dangerously”, then the BMS logic must change and go into “protection” mode.
I wonder how all this will work 1-2-5 years from now, with more banks out there, the limitations currently in place.
Or like Victron makes the BMV that it can link per cell, how would that affect the SOC versus the BMS SOC calculation.
Cause it seems to me, no-one has the answer on why they chose those voltages for SOC indication.
If you do that, then the moment you hit a high cell in a not-perfectly-balanced battery, the SOC stops short of 100%. Then the customer calls the help line and asks why. That costs you money.
So yes, it’s a bit of a cheeky shortcut, but that’s one reason why the BMS simply goes to 100% when any one cell spikes. Cause 9 out of 10 times (when the cells behave) that’s spot on, and the rest of the time it avoids unnecessary time spent explaining it to an irate customer who insists it should be different 
Everything is not always about irate customers costing moeny. 
Here we have no irate customers, just DIY’ers asking questions around more knowledgeable people, based on what they see/experience BECAUSE of a bank out of balance … this SOC and volts and all that has gotten me thinking since the first day I opened the app on my phone.
And I may be the first it seems, to ask these questions here.
It also may be a good exercise in the end, because we can alter the BMS settings, that once we have this figured out, it would become 2nd nature to help newbies down the line same as what we had on that other forum, when people asked about settings for their BMV’s and lead-acid batteries, cause it jumped prematurely to 100% SOC.
Just saying.
1 Like
1 is the BMS SOC and volts prepogrammed.
2 is quick non-clever-oak Excel calc on a 4.566% increase on the volts from 2.8 to get to 3.5v, as example.
Why the DIFF column in 1?
At the extremes (high/low) it spikes and is not flat.
Could you, me being not too clever, explain the above sentence with the (1) diffs in my picture?
If I understand your question correctly then you are referring to the 0.05V in Diff around the 40/60 SOC range while farther out it use 0.1V Diff.
These are 3.2V cells and they keep close to that level for most of their capacity. When they get empty the start drop their voltage faster. And when they get full their voltage spike up faster. So around the middle is all flat meaning small voltage changes against SOC changes.
1 Like
Excellent … that makes sense then!
Thank you for the patience … I’m getting there.
Yeah I get it, I’m just explaining why BMS makers tend to do what they do… because the “normal” audience are not DIYers. Of course there is nothing wrong with having a battery that stops at some SOC less than 100%, and gives you a true idea of the real SOC…
But then, we already spent a lot of time explaining top and bottom balancing, and what constitutes the “real” SOC is pretty much dependent on how you decided to manage that part
@Louisvdw, regarding the charging rates you have hard-coded. On what recommendation (documentation) did you base a low charging rate at a low state of charge?
I can see easing off the juice at the top end of the SOC. I also have seen that being recommended in many a source.
I haven’t seen any rationale for doing it at the other end of the charge curve.
Not saying there isn’t a valid reason, just wondering what this rationale is?
My own thinking would be that if I am right down at a low SOC, say after a prolonged overcast period. And the sun popped out for an hour, I’d want to get some charge into those batteries as fast as was safe.
I wouldn’t intend to operate around those low levels of SOC very often, but when I was, I’d like to know what I was sacrificing in terms of longevity to get the bank up to a mid-range SOC pdq?
I think you misunderstand.
The charge limit is only at the top end.
At the low end there is only a discharge limit. So when your battery is running low this will limit a large current draw that will pull your cells below their low limit, but rather using smaller current draw which mean you get more out of the battery before the cells get to the low limit cut of.
1 Like
Ok, that makes sense, thanks, and indeed I see that when I look at your code again.
Getting there … 
Have you calibrated the current shunt in the BMS yet?
Not yet … I am getting the guts together to do that entire exercise.
If you have any salient TTT-safe methods of doing that, please share … or forever hold your peace.
I bought the App for the iPad. It looks more “finessed” than the free version enhanced App for Android.
1 Like