diy solar

diy solar

Daly BMS SOC nonsense

Bsjoling

New Member
Joined
Feb 20, 2021
Messages
23
I've never really understood the daly soc... I've seen it swing from 2% to 100%, it definately isn't telling me what I expect to be seeing regarding the volts... 13.00 is 30% for Lifepo4, yet the soc dial will often indicate 87%.

4s2p 200ah setup, 400ah in total, 8 cells.

I plugged in the pc and ran the sinowealth bms tool.

1639990414607.png


by setting the design & full charge capacity up to 990000mAh, it brought the soc reading down to 37% @ 13.05v, which I'm happy enough with.
However, this mornings readings show we're now down to 12.844v, which to me should be about 18%,
I've also set the


dftterCur down to 100ma. I also run a drok coulomb counter on the dash, and I would like to make sure that I can see parasitic draw, got quite alot of electronics on the RV.

Now in the pic below there is a range of soc vs volts, which look like they have been incorrectly set.... I was wondering if anyone has played with these, or figured out why they are not reading correctly? I had a little fiddle with the figures, didn't seem to move the current reading of 35%

1639990647377.png

this is my little table of soc to volts, interpolated values are utter guesstimates
1639994722848.png


adjusting the % - mV values seems to write ok, but doesn't seem to change anything.


1639990838582.png

I hope this may be of use to some of you diy pioneers, hope too that someone better experienced than I can help me resolve this mess!

Later today I'll set the soc-mV values as to what I think they should be, and put the Ah back to 400 and report my findings.
 

Attachments

  • 1639989337100.png
    1639989337100.png
    46.9 KB · Views: 5
  • 1639989513500.png
    1639989513500.png
    29.3 KB · Views: 5
  • 1639993213532.png
    1639993213532.png
    7.3 KB · Views: 6
LFP voltage as a measure of SoC is almost as useless as lead-acid. The best way to get confused with SoC is to assume that voltage has a strong correlation to SoC. It does not.

I have a group of CALB cells that settle to 3.30V @ 100% SoC. That would put the 4S @ 13.2V, where the voltage would suggest it's at 70%. I consider a 30% error to be significant, and it's not just me. I'm not using a measure that may introduce a 30% error.

The SoC voltage list in the DALY and many others is for CHARGED voltage, i.e., if you charge to the stated voltage, that's the approximate SoC. These are comparison values - sanity checks if you will.

First, you have a 400Ah battery, so your Design and Full charge capacity should be 400,000mAh. Period. End of story. ANY other value in here GUARANTEES inaccurate SoC. Period. FIX THIS.

The BMS computes SoC based on the full charge capacity value and actual amount used, i.e., it COUNTS the current used and computes the SoC, e.g.,


20ah used since last full charge

SoC = (400-20)/400 = 380/400 = 95%

It doesn't give a shit about the voltage. It trusts that you've correctly entered the values, and the battery meets that spec.

The SoC/Voltage values are used periodically to tweak the SoC, particularly after the unit has been reset and the SoC has been lost.

Have you confirmed that all the cells are actually 200Ah? If not, then you can't assume they meet that spec.

Have you confirmed that the DALY is measuring the correct current? If not, then you can't assume the DALY is correctly computing SoC based on use.


If your DALY is not computing SoC correctly, most likely your battery doesn't meet capacity, or the DALY isn't counting current correctly.
 
My Daly BMS 12V current measurement is bad. Everything below 1A it is not seeing. Result is that after a week or 2 of not charging in my camper the SOC is way of.
I don't use it at all, just for checking if it is working. The battery monitor is doing the real monitoring work.
 
My Daly BMS 12V current measurement is bad. Everything below 1A it is not seeing. Result is that after a week or 2 of not charging in my camper the SOC is way of.
I don't use it at all, just for checking if it is working. The battery monitor is doing the real monitoring work.

This is good advice that I should have given as well, but I got into the weeds. The SoC reported by any BMS is never particularly accurate. Dedicated battery monitors are typically much more reliable.
 
Thanks sunshine-eggy, appreciate a good reply! Some stuff to think about.
I run a drok coulomb counter too, tho got to admit it isn't the easiest thing to set up!
 
Hmm.. not quite sure what I did with it, but it seems to be giving some discharge info for the first time ever...
ANyone got any tips on tweaks? for some reason it seems to run 0.1-0.2v lower than the daly
 

Attachments

  • 71t2bF0munL._AC_SY355_.jpg
    71t2bF0munL._AC_SY355_.jpg
    29 KB · Views: 6
Hmm.. not quite sure what I did with it, but it seems to be giving some discharge info for the first time ever...
ANyone got any tips on tweaks? for some reason it seems to run 0.1-0.2v lower than the daly

I'm pretty sure I have that unit. Personally, I wouldn't trust either without confirmation of accuracy with a known accurate tool. I suspect if you put a voltmeter on it and compare all three, you'll get a 3rd value for voltage. :)
 
I'm pretty sure I have that unit. Personally, I wouldn't trust either without confirmation of accuracy with a known accurate tool. I suspect if you put a voltmeter on it and compare all three, you'll get a 3rd value for voltage. :)
yes, noticed that one already, I have about 7 digital meters from cheap to expensive.
 
Just as I was beginning to get excited at the thought of getting useful data, I ran the system on charge for a little while, Daly went almost straight from 80 up to 100%, drok unit went from 82 -> 86% (which seemed more realistic)
 
Have you charged your battery to at least 14.4V yet? Measured at the battery? Until you do, your BMS SoC will likely be erratic. The battery must be fully charged before the DALY or any BMS will "reliably" track SoC.
 
Have you charged your battery to at least 14.4V yet? Measured at the battery? Until you do, your BMS SoC will likely be erratic. The battery must be fully charged before the DALY or any BMS will "reliably" track SoC.
No, my high volt protect is set to 3.50/14.00.
So how does one set a bank up so that it runs from 20->90%? I would like to maximise the life of these cells, and Lifepo4 love to cycle between these values, however I have noticed that having gentler high-volt-cutoffs (like 3.35/13.4) seem to cycle the battery down such that I've run flat a couple of times. 14.4 seems quite an aggressive charge voltage when 13.4 is fully (99%) charged... I would love to understand...
I'll up the cell charge by 0.1v and see what happens.
 
It's clearly not sinking in as you keep spitting it back out at me. As we've already discussed voltage is a terrible way to determine state of charge, so you need to quit clinging to it, so...

13.4V is not 99% charged. Get this out of your head.

I'm assuming you're using this:

1640535767470.png

Does anything stand out to you? How about a 50% difference in SoC between 13.1V and 13.3V... a 0.2V difference. More evidence of a poor indicator of SoC.

The ONLY time the above chart is pseudo-accurate is when the battery has been resting for awhile.

An additional concept you may not be aware of: Just like 12V lead acid needs to be charged to higher voltages (14.X+) to attain full charge, so does LFP.

The BMS is a safety device. It shouldn't be used to manage routine operations. To that end, there's no point or benefit to having HVD at 3.50. Set HVD to 3.65 and LVD to 2.5V.

You can't target a charge voltage and hit 90%. Your charging equipment should be setup to manage charging to your desired limits

Your loads should be managed to prevent discharge beyond your desired limits.

Most recommend 3.0V/cell as a discharge floor, which will leave you something like 10-25% depending on the load when you hit 3.0V.

The only way you can charge to 90% is as follows:

1) fully charge your battery to ensure the BMS knows where true 100% is.
2) discharge no higher than 80% SoC.
3) charge using a constant current source - if you're using solar, this is going to less reliable.
4) once the BMS indicates 90%, note voltage AND current.

For future charging, you need to terminate charge when you've reached that voltage and that current.
 
Thank you. I appreciate what you're saying. I'm not stupid, I hold a degree in electronics and am not a stranger to issues that fall outside theory.

I understand there is a certain voltage push that is required to get the chemistry to work. There is a lot of bad information about, and I'm seeking to understand via those that have better experience than myself in this field. I'm studying how to set up an optimal motorhome system that doesn't break the bank. I'm sure you tear your hair out seeing the sales methods and promises of 'drop in' solutions. I would like to see if I can get good and proper results from readily available products before going nuts with a custom arduino control.

If you have any links to decent articles, please do share... I'm happy to learn to a molecular level the ins and outs of using lithium effectively. Before I get to the point of charge conditioning, I just need decent info on what I'm dealing with regarding the Daly units, then the drok vac coulomb counters. it's not straightforward measuring voltage to high degrees of accuracy especially when combined with current ebbs and flows.
 
LOL

Wow. To be frank, it's like I'm dealing with Jekyll and Hyde here. All prior posts suggest you're a complete noob with no education or fundamental understanding whatsoever. Had they been otherwise, my time invested would have been much lower.

You seemed slave to erroneous information and did nonsensical things to make nonsensical things make more sense nonsensically - not something I would expect from someone with the claimed credentials.

Now that I know your goal, RV setup for max cycle life:

Shore/solar charging where long slow charges are practical:
13.8V absorption, 13.6V float
The longer slower charge at a lower voltage will get the battery to 98%+ and is less stressful to the cells (charging is worse than discharging) and maximizes cycle life).

For generator charging (lower voltages run the generator longer):
14.4-14.6V absorption, 13.6V float
Faster charge to run the generator only as long as needed.

BUT! Typical RV wiring is typically rubbish. If you have excessive voltage drop between source and battery, it's going to throw a wrench in the works.
 
Lithium didn't exist when I was at uni, computers still ran on punch cards, and now we are in such an era of misinformation that it is difficult to seperate crap from reality.
Thanks for your time so far, I'll get back to you next time I've got a window for experiamenting :)
 
I've never really understood the daly soc... I've seen it swing from 2% to 100%, it definately isn't telling me what I expect to be seeing regarding the volts... 13.00 is 30% for Lifepo4, yet the soc dial will often indicate 87%.

4s2p 200ah setup, 400ah in total, 8 cells.

I plugged in the pc and ran the sinowealth bms tool.

View attachment 76474


by setting the design & full charge capacity up to 990000mAh, it brought the soc reading down to 37% @ 13.05v, which I'm happy enough with.
However, this mornings readings show we're now down to 12.844v, which to me should be about 18%,
I've also set the


dftterCur down to 100ma. I also run a drok coulomb counter on the dash, and I would like to make sure that I can see parasitic draw, got quite alot of electronics on the RV.

Now in the pic below there is a range of soc vs volts, which look like they have been incorrectly set.... I was wondering if anyone has played with these, or figured out why they are not reading correctly? I had a little fiddle with the figures, didn't seem to move the current reading of 35%

View attachment 76475

this is my little table of soc to volts, interpolated values are utter guesstimates
View attachment 76478


adjusting the % - mV values seems to write ok, but doesn't seem to change anything.


View attachment 76476

I hope this may be of use to some of you diy pioneers, hope too that someone better experienced than I can help me resolve this mess!

Later today I'll set the soc-mV values as to what I think they should be, and put the Ah back to 400 and report my findings.
The way that I set it is with the following formula:

Assume the following for a LiFePO4 battery. The lowest voltage is 2.50v and the top is 3.65v. The numbers between these 2 is your 100% or 1.15v. Eg. assume your current voltage of the pack is 3.45v average. Then 3.65 - 3.45 = .2 volts. 1.15v - .2v = .95v. So .95 / 1.15 = 82.6% SOC.

This seems to work for me.
 
Back
Top