AC watts, DC watts, VoltAmps, Kwh, battery runtime, huh?! I've confused myself, newbie needs a math lesson please!!

[MrSparkle]

@cyberfed
Im glad we can help and I am sure others also are.

But wait stop!
I see a warning sign!

You mention 2V-3.5V and 2.5V-3.65V this is "all over the place".

robbob calculated in post #15 that you probably have LEV60F, Li-FePO4 chemistry-type cells with nominal 3.2V
the one you are linking in your latest post
https://batteryhookup.com/products/new-case-with-6x-3-2v-74ah-lifepo4-prismatic-cells
Are these EXACTLY your cells or do you assume those are your cells?
It is important because different Lithium chemistries have distinct different voltage working ranges!

(Example, Li-NMC usually have nominals around 3.6-3.7 and working range ~3.0-4.2V)

So, the very first important step, before you do ANYTHING else, is to establish the chemistry of your cells.

If it is Li-FePO4, then examine the chart Steve made in:
https://diysolarforum.com/threads/battery-lifespan.88193/#post-1163562

For LiFePO4, 2.5V to 3.65 is the absolute outer edges, notice the red zone.
2.8V to 3.5V is very safe "outer limits".
But for daily use 2.9V (or even 3.0) up to 3.4V recommended (to 3.45 for balancing)

So... do you have Li-FePO4 cells or Li-NMC ? (There are also other chems, but these two are most common.)

 

[cyberfed]

@MrSparkle
They were sold to me as LFP Lithium Iron Phosphate. Battery hookup has them listed as LFP as does Jag35 (though they are out of stock). I have seen other spec sheets too. I think I may have one on my computer that matches the 2.0 - 3.5 range. Yeah I do same as the one posted here.

They are NOT LIION aka NMC cells. I have played with those many a times. Here's the only way I can prove they are LFP besides going by what sellers are saying, when I top balanced a different pack of the same cells, (16x. in parallel with 3.65V 30A which took foooorever to finish) but as the batteries approached 3.65 the current output began to drop towards zero at the 3.65V mark. There's no writing on the battery that would indicate its chemistry I just checked. They look 100% the same as alll the pictures shared and they were bought from jag 35 https://jag35.com/products/clearance-lev60-74ah-lifepo4-prismatic-cells

I've set the BMS to shutoff at 3V and charge to 3.55. Once they complete they usually stay in the 3500 range for several days if I don't use the pack, if I do it quickly falls into the 3.2 range where it spends most of its life before dipping down fast towards 3.0V and below once low. So it seems to a layman like myself they follow the LifePo4 profile. Powerband being in the 3.2V range. With quick charge up's past 3.4V (towards 3.65) or the opposite starts dropping like a stone around 3V towards 2.x Volts.

I'm a little concerned to drop them down to what these spec sheets are saying of 2V being the bottom out and the 3.5 being the top end. Given this what's the best way to test or ensure they are all balanced? (You have me wondering if I top balanced this inital pack...I know I did others but this one I may have just thrown on the charger (AC) and let it fill up until the BMS cut it at 3.65, and I do not remember what all the other cells were at. I do remember the BMS doing its job and shutting down the charging, waiting like 30 seconds (whatever the default config time was on the BMS to "release and allow charging" then it would charge again until a cell hit 3.65V. So its possible this pack wasn't top balanced because the BMS interferred (as it should).

Recommendations? Parallel them all and charge to 3.65 and see if the current drops to zero. Then they will be balanced?
Given the mysterious #s on the spec sheet I dont know now what to set my upper and lower boundaries to on the BMS.

Help

 

[sunshine_eggo]

I would trust what appears to be a datasheet over batteryhookup. Nothing against them, but they're mostly just a surplus clearing house.

I'm a little concerned to drop them down to what these spec sheets are saying of 2V being the bottom out and the 3.5 being the top end.

That's why I recommended 2.5V.

Given this what's the best way to test or ensure they are all balanced? (You have me wondering if I top balanced this inital pack...I know I did others but this one I may have just thrown on the charger (AC) and let it fill up until the BMS cut it at 3.65, and I do not remember what all the other cells were at. I do remember the BMS doing its job and shutting down the charging, waiting like 30 seconds (whatever the default config time was on the BMS to "release and allow charging" then it would charge again until a cell hit 3.65V. So its possible this pack wasn't top balanced because the BMS interferred (as it should).

If all cells were at or above 3.45V when the high cell hit 3.65V at low current, then they should be very nearly fully charged if not fully charged.

Recommendations? Parallel them all and charge to 3.65 and see if the current drops to zero. Then they will be balanced?
Given the mysterious #s on the spec sheet I dont know now what to set my upper and lower boundaries to on the BMS.

Did you try my recommendations to individually charge each cell?

 

[fnnwizard]

Ok I'm not sure if I hit send on my laptop, I was drafting a response that looks similar to this one but I'm not sure now. Out of respect to all of you I wanted to let you know that I am still working on this and I will post the update (if I have one drafted on my personal machine) later today after work. The very short version (which is well recieved here ) is I am taking all the considerations posted to investigate! I didn't want to leave things just hanging out there after so much help was provided. Again I dont recall if my post was the one above. Stupid medications I have to take

Your math is fine for power but for AC can be a bit different.
V*A is for AC power and is used with power factor (PF) to determine real power (Watts) which may not be 1:1 with non resistive loads (inductive and capacitive).

You mentioned 170W-190W fo 9hrs, 50W for 15hrs - power consumption, but also added a fridge at possibly 900W?

Anyway, questions are :
What was plugged into the Killa-Watt Meter when you took measurements?
What was plugged into inverter when you got 4hrs run time?

 

[Supervstech]

remember this.

inverter will drain a constant load the entire time it is turned on. Often around 70W so measure the drain out of the battery with ONLY the inverter running, no loads plugged in.

also, 100W battery out does not EVER equal 100W ac out. the inverter conversion from dc to AC will have a varying amount of waste from 97% down to as low as 70% efficiencyin some loads.

All that said, if you measured 3200Wh over 3 days, i cannot fathom 3700Wh being used in 4 HOURS...

Connect the killawatt meter directly to the output of the inverter and lets see what actual loads are.

 

[cyberfed]

Guys again THANK YOU. We are getting closer to the truth I know it. I think @sunshine_eggo @MrSparkle you are on to something. So more is coming back to memory from this build after cracking it open. First stuff checked/done:

1. Recrimped with Tepco terminal rings and a the appropriate size/pressure for a nice unmovable cold weld.
2. Cables are welding cables from Windynation.com -->no good?
3. Cable are up-sized wherever possible by 1 or 2 AWG ratings if possible (fit)
4. voltages all validated in terms of balance cables, took them off, re did them same values as every meter/bms gives.

HOWEVER,
This is a JBD BMS, it has about 800 million settings you can screw with. Sadly I remember screwing with them...I'm sure that is part of the problem. If I took screen shots could someone point me to some correct values?

I tried to AC charge this pack, it was "full" to begin with. I should say that means voltages of all batteries were around 3.4V. Now I set the cut-off limit of the BMS for voltage to 3.55V and used an AC charger that puts out 7A @48 volts. 15 of the 16 batteries were all tracking pretty closely on their way up (within less than 0.009 V) one battery just one battery #9 was on the struggle bus while everyone else was at 3.55V it was (and still is at 3.379V). Since this BMS doesn't have active balancing there was no load sharing happening if you will.

I am going to manually charge that cell up to whatever voltage the others are at and see if it then tracks in line of if still lags behind. Or would it be smarter to just fully charge it individually and introduce it to the pack once its safe to do so?

The BMS has settings to set the 80% 60% 40% 20% voltages. Those numbers were a hot mess with 40% being higher than 60%! Do I just start at 3.5V (Full) and subtract 20% and throw the new voltage in there? I can post screenshots if need be.

Sorry guys today was a hot mess, speaking of batteries....my g/fs fully electric car took a nose dive and we had to have it towed to the dealer, we wont know for a few days what's going on. It messed up the 12V lead acid battery good, it had vented, you could smell it and it was swollen! Sitting at 10V. So I didn't get to even work a normal day let alone troubleshoot the pack much.

I don't think it's the cell being bad honestly (could be I have spares...) I think it was never balanced right and over time it got left behind, I dont know, it cell 9 which is "the other half" of the PCB in terms of connectivity from cells 1=8. I can always try another PCB I have a spare if need be as well. I want to see the result of that battery being filled up first and re-introduced into the pack safely and see if they all get along after that.

Oh and of course my settings that are a hot mess on the JDB BMS that I cant set to factory reset

I plan to also hook up the Kill-a-watt on the inverter side and see what #s I get. Right now. (started at 5:30 PM Saturday) its at 1.1KW consumed, again I didn't get to work today because of the car disaster. Number will jump up tomorrow after a full day of working.

Thank you in advance!

 

[fnnwizard]

Guys again THANK YOU. We are getting closer to the truth I know it. I think @sunshine_eggo @MrSparkle you are on to something. So more is coming back to memory from this build after cracking it open. First stuff checked/done:


I tried to AC charge this pack, it was "full" to begin with. I should say that means voltages of all batteries were around 3.4V. Now I set the cut-off limit of the BMS for voltage to 3.55V and used an AC charger that puts out 7A @48 volts. 15 of the 16 batteries were all tracking pretty closely on their way up (within less than 0.009 V) one battery just one battery #9 was on the struggle bus while everyone else was at 3.55V it was (and still is at 3.379V). Since this BMS doesn't have active balancing there was no load sharing happening if you will.

I am going to manually charge that cell up to whatever voltage the others are at and see if it then tracks in line of if still lags behind. Or would it be smarter to just fully charge it individually and introduce it to the pack once its safe to do so?


Thank you in advance!

Yeah, you have an imbalance issue 20-30%. If all other cells are 3.55V and your low cell at 3.379V, with that charge current, you're likely around 70-80% SOC when other close to 100%, which should still give you 2-3kWhr total though.

Charge the single to same as others, use killa-Watt at inverter to see/verify loads.

You might have your low V cutoff too high if you have been messing with the settings and may have missed it.
BMS basic parameters can be set to:
Low Voltage cutoff per cell 2.85V-3.00V (45.60V-48.00V), 48V being better if your avg loads are low.

It sounds like your pack will be in decent balance once you get the single low one up.

And if so, you can tailor the BMS balancing point to start balancing at 3.45V with cell differential at .20mV to turn on and turn off at 10mV.

 

[cyberfed]

Thank you @fnnwizard @sunshine_eggo @MrSparkle !!

I'm using my benchtop PSU to bring that one cell up to the voltage all the others are at. I can't play with the BMS right now since things are disconnected for charging this single cell. Thank you for the ball park numbers for the BMS, I will set them. I know the BMS was set to shutoff at 3.V There's a few settings in there that I don't know if I borked things up bad, for example I set the 74AH in the BMS it had a section for that. I had my balance set to start at 3.5V at 15mv.

Again ths isn't a smart balancing BMS, it just bleeds high cells down to whatever the pack decides is the average or the lowest cell I dont really know. Didn't come with a manual lol.

I'm concerned there a few settings, this is an EXAMPLE PHOTO NOT MINE....




On the right side Full cell voltage and end cell voltage. I'm not sure if I have cycle and Full capacity, but lets suppose I do, what should these be set to? FullcellVoltage I have set to 3500 and Cell end voltage set to 3000. Then here you can see the 80% 60% and what not, the heck should these be set to exactly?

CellOVP was set to 3.6V with release at 3550.
Cell UVP was set to 2900 with release at 3000
PackOVP I believe was set to 57.6V (16x 3.6V)
PackUVP I dont recall, 47600? maybe somewhere around there.

Balance settings were set to 3500 and 15mv

Given my battery pack specs appear to be 2.0 end and 3.5 full (still don't know about that 2.0V....) if this was your pack how would you set it up?

Once the battery charges up to the rest. I will then hook things back up and then come into this BMS interface to set values. Hoping someone will help me out there really bad!! You can roast me on how dumb I am its ok. I kinda want to replace this BMS for a virtually identical one that has bluetooth because the spot for the BT module isn't there on this one. I'm worried it wont fit in the pack in all the pre defined spots/holes ect..

I am going to hook up the kill-a-watt to the inverter once this is done and see what I get , I'll measure current leaving the battery to the inverter and after the inverter to my load. I have a heavy duty extension cord I peeled back the outter sheath so I could access the 3 AC wires inside (sheathed) so I could measure AC current as well. First time I tried measuring AC current I couldn't understand why I was getting 0A duh I had it over all the wires


Thanks guys, hoping I can get a bit more help to fine tune this thing!

 

[cyberfed]

Ok the battery is equal to all the others, verified with benchtop MM as well as the BMS itself. All within .00x Volts.

I will mess with the actual BMS settings after work to re-configure those.

I also used some 0000 steel wool to remove what I would call "gunk" similar to the stuff that builds up on a keyboard or mouse that was around bottom portion of the PCB that makes contact with each cell. So those are now clean, shiny and gunk free. I made sure to clear all metal bits from the wool before reinstalling. So right now the pack is resting with all cells at 3.55V Pack is not on so the BMS is not doing anything, if there's any balancing happening its happening from the batteries themselves being interconnected in the 16S config.

So I had a possible 20-30% imbalance/loss a 30% imbalance would be about 5.7 hrs runtime assuming the 17 hrs listed previously. Add in whatever we lose to the inverter plus possible (likely) misconfigurations in the BMS due to me. Maybe that gets us down to the 4+ hrs range? I got nothin..

 

[Supervstech]

Ok the battery is equal to all the others, verified with benchtop MM as well as the BMS itself. All within .00x Volts.

I will mess with the actual BMS settings after work to re-configure those.

I also used some 0000 steel wool to remove what I would call "gunk" similar to the stuff that builds up on a keyboard or mouse that was around bottom portion of the PCB that makes contact with each cell. So those are now clean, shiny and gunk free. I made sure to clear all metal bits from the wool before reinstalling. So right now the pack is resting with all cells at 3.55V Pack is not on so the BMS is not doing anything, if there's any balancing happening its happening from the batteries themselves being interconnected in the 16S config.

So I had a possible 20-30% imbalance/loss a 30% imbalance would be about 5.7 hrs runtime assuming the 17 hrs listed previously. Add in whatever we lose to the inverter plus possible (likely) misconfigurations in the BMS due to me. Maybe that gets us down to the 4+ hrs range? I got nothin..

Fingers crossed for ya.

 

[sunshine_eggo]

Given my battery pack specs appear to be 2.0 end and 3.5 full (still don't know about that 2.0V....) if this was your pack how would you set it up?

Change only:

UVP 2500 release 2700
pack UVP 40000, release 43200
bal to 3400, 20mv

I don't see it, but if there's an option to balance only during charging, disable it. You want to balance even if it's just sitting there.

Changes may be warranted to the above, but more data is needed.

Again, set inverter to cut off at no lower than 10.5. Honestly, since I'm concerned the BMS might force 0% if UVP is reached, you might want to set it higher. I want to avoid a potential UVP event that's not observed because it's very important to confirm the BMS reported SoC and Ah consumed BEFORE UVP.

 

[fnnwizard]

I had my balance set to start at 3.5V at 15mv.

The 3.5V is actually set too high. The differential setting of 15mV is the much more important number. This is from my own data over a decade under testing conditions and actual usage.

When batteries are within 10% balanced, we rarely see anything greater than 20mV delta between hi and low cell within the V range that corresponds to 20%-80% SOC, UNLESS there is/are connection resistance issue(s), the low/hi cell is actually bad, or measuring device issue.

Normally, we don't wan't to start balancing below 3.410V due to the relatively flat curves + circuit resistance at charge rates.
Which simply means, the hi cell at 3.380V may not be the high cell when over 3.410V, again especially under charging rates.

The easy recommendation has always been to wait til the cell gets near 3.450V due to the exponential rise in V as V approaches upper limit and the natural current taper so that V differential is more from cell and less from circuitry resistance.

So for instance, even if you had set the V to start balance at 3.350V, the mV delta will be within 10mV in that range unless the SOC is actually over 20% off from Hi /Lo cell, in which case that Hi Cell will still be the hi one when Voltage is > 3.41V.

Balance Voltage to start balancing is a secondary factor and not a good one at that. The delta in mV between cell is the more significant number and when it deviates more than 20mV in the 20-80%, it's better to start the balancing earlier for 2 reasons.

1. To give enough time for low cell to get up to V
2. To lessen the time the high cells stay at upper hi V limit.

See these 2 posts below. Tried to help another user and I stand by every word I wrote in that thread. Got a bit annoyed when a long time member suggested I try to understand LiFePO4 a bit more, lol...

F

Post in thread 'Batteries Issue'

Nov 11, 2024

3.65

3.45

30 mv

30 mv

Not sure, can't check atm

Can't check atm
But op does have the config xml, so should be able to check

Total Voltage Overvoltage Protection : 56.8 v

Ok,

3.65V cell level OVP. This is ok and allows cells max amount of time to see balancing activities. The "max time under balancing" tho could be seconds to minutes vs say 3.55V depending on charge rate).

This is what I would change.

3.45 ->3.40 to get high cell to start discharge earlier. Once the high cell starts to get discharging, there's not a lot of time it can stay at this level since...

• fnnwizard

https://diysolarforum.com/threads/batteries-issue.94002/page-3#post-1260385

On the right side Full cell voltage and end cell voltage. I'm not sure if I have cycle and Full capacity, but lets suppose I do, what should these be set to? FullcellVoltage I have set to 3500 and Cell end voltage set to 3000. Then here you can see the 80% 60% and what not, the heck should these be set to exactly?

SOC vs Voltage is always a moving target since V needs to be specifed with loads. But V will also vary with loads based on cell internal resitance (size) and circuit resistance. You generally can get within 5-10% of SOC for V though if build quality is decent.

CellOVP was set to 3.6V with release at 3550.
Cell UVP was set to 2900 with release at 3000
PackOVP I believe was set to 57.6V (16x 3.6V)
PackUVP I dont recall, 47600? maybe somewhere around there.

The industry standard for OVP, cell level is 3.800V, we use 3.650V. 3.600V is ok
for UVP, 2.500V per cell. I keep mine above 2.875V, with warnings at 2.999V

 

[cyberfed]

So quick question. I made the proposed changes. I think? Lol I got a bit confused but I think I got it.

So even though that these batteries have 2.0V rating we dont set things to say 2300? Or is it just safer and there isn't really any power left at 2500.

The pack was 6mv of delta after charging the one battery to be at the same voltage as the rest. I put on my AC charger just to see since we raised the OV pack value it will charge longer before stopping right? Right now its a trickle with amps since its approaching 57.6V that's when the charger stops. I need to create a custom profile to a higher voltage to meet 16x @3.65 = 58400. But while it charged for a very short time, the delta between the cells never went past 15mv (set to 20 to start balance @ 3.4V. It was consistently the same battery I charged manually to the other cells voltage, cell 9 that was causing the delta to be 15mV or else it would have been around 5.

So it is the lowest voltage cell consistently. Could this be because it's never been to 3.65V ever?

I'm confused now what terminates the charge, the pack voltage?

The charger is done. The delta is at 13mV with most cells around 3601 - 3606 but cell 9 is the only cell (at least right now that is in the 3.5's its at 3594. Since there's no charging going on (or balancing) it seems the pack just slow getting self into the closest delta.

Sorry!!! if I bump up the voltage on the charger to 58.7

 

[Rednecktek]

I don't see it, but if there's an option to balance only during charging, disable it. You want to balance even if it's just sitting there.

This! Turn that setting to OFF, throw it on a charger, and go to bed. Let the BMS have some time to do its thing. As it bleeds off the high cells, it'll kick the charging back on and that'll keep bumping your low cells up.

Leave that setting off and when your battery is back in service it'll have the chance to balance everything out overnight and be all balanced up when the sun comes up in the morning. That'll keep that slow cell in check for a long time.

 

[cyberfed]

just unchecked the charge balance box

 

[fnnwizard]

So quick question. I made the proposed changes. I think? Lol I got a bit confused but I think I got it.

So even though that these batteries have 2.0V rating we dont set things to say 2300? Or is it just safer and there isn't really any power left at 2500.

I did not see sunshine had reply before I did a few posts above and his advice is solid too.

Correct, there's very little useable energy left when you get to 2.500V. The cell starts to get stress going below that as internal resistance starts to ramp up. ( I tried to take a brand new, A cell ( less than 5 cycles from me, and assuming 2-5 cycles from factory) LF280k V3 to 0 volts and here are some pics of the load bank.

It's less than 2 minutes before getting into territory where the cell will expand (measured on digital calipers) more than the norm at >2.500V. with a .2C rate.

The pack was 6mv of delta after charging the one battery to be at the same voltage as the rest. I put on my AC charger just to see since we raised the OV pack value it will run longer before stopping. Right now its a trickle with amps since its approaching 57.6V that's when the charger stops. I need to create a custom profile to higher to meet 16x @3.65 = 58400. But while it charged for a very short time, the delta between the cells never went past 15mv (set to 20 to start balance @ 3.4V. It was consistently the same battery I charged manually to the other cells voltage, cell 9 that was causing the delta to be 15mV or else it would have been around 5.

So it is the lowest voltage cell consistently. Could this be because it's never been to 3.65V ever?

I'm confused now what terminates the charge, the pack voltage?

Sorry!!!

It could be, I usually take a cell to 3.650V ( under the normal .05C) at least once in the beginning to get the bms to reset and see what cell delta mV is. If you are less than 20mV, you are really good and this becomes your baseline for all furture comparison.

But one caveat you have to consider is how well the pack was put together=circuitry (external) resistance. If you read my post in the other thread about how to "artificially" take a cell to hi V and back via C-rate and fully understand it, it will help tremendously to know when a cell is actually acting up or is it due to the external resistance.

I try to have a baseline for just about every pack I have built, but it's hard to do for avg person without an electronic load that can adjust in a fast enough respose time.

In pics below, on a dedicated test computer, the top green line is Volts, bottom is Amps. Rag to wipe hands before touching any terminal. See the time stamps.

 

[cyberfed]

So my charger wont just continue to charge endlessly (my AC) so it shuts itself off. However, with no charging going you can see the values dropping and getting closer, of course #9 always being the outlier


look ok?

 

[sunshine_eggo]

With those voltages, your battery is fully top balanced. Done.

The 2.5V is just in case they're full of shit about 2.0V.

 

[cyberfed]

Nice. I guess I'm going overkill before performing the "real world"test with the inverter and meters to measure. This morning I took the PCB board off and I'm charging each cell to 3.65 manually just to make sure we are g2g.

So if all works well we should get well over 4 hrs unless there's a problem elsewhere yeah? I am going to check with a thermal gun to check for hotspots with the pack running but not closed up. I'll have the inverter with my office load. Along with the Kill-a-watt meter.

Thank you @sunshine_eggo and others you have always been there for me to answer my dumbest of dumb questions, means a lot to me. Tomorrow will be load test day, I'll hook up my office and measuring instruments to the battery at the start of my day and see how long it goes. I'll make sure the Kill a watt meter is reset I recently discovered it saves all the kwh, so I went to check and it was some crazy high number. doh...gosh I'm dumb.

Thank you all! tomorrow shall reveal where the problem may be if there is a problem, hopefully not!!

 

[MrSparkle]

Regardless of what others or you self perceive as a smart or dumb question, the most important is that you are trying to learn. Hence "dumb/smart" becomes moot. You are trying to solve a problem as best you can and ask for advice where your own experience is short. (Compared to lazy people with wreckless mentality.)

From the above. Don't sell your self short. Don't apologize half the time. Be self-critical but also have self-confidence. It is a balance.

I am eager to follow this thread and see what outcome you will get after all the troubleshooting and adjustment.

 

[sunshine_eggo]

Nice. I guess I'm going overkill before performing the "real world"test with the inverter and meters to measure. This morning I took the PCB board off and I'm charging each cell to 3.65 manually just to make sure we are g2g.

This isn't necessary. Your battery is very well balanced based on the numbers you provided.

So if all works well we should get well over 4 hrs unless there's a problem elsewhere yeah? I am going to check with a thermal gun to check for hotspots with the pack running but not closed up. I'll have the inverter with my office load. Along with the Kill-a-watt meter.

Sounds good. Moar data at periodic intervals looking to correlate all data to another instrument.

Thank you @sunshine_eggo and others you have always been there for me to answer my dumbest of dumb questions, means a lot to me. Tomorrow will be load test day, I'll hook up my office and measuring instruments to the battery at the start of my day and see how long it goes. I'll make sure the Kill a watt meter is reset I recently discovered it saves all the kwh, so I went to check and it was some crazy high number. doh...gosh I'm dumb.

Thank you all! tomorrow shall reveal where the problem may be if there is a problem, hopefully not!!

 

[cyberfed]

Thanks guys that's really kind and you have no idea how much that means to me, I'm always to self critical and hard on myself. I feel like I have zero self worth.. working on it and other stuff with my therapist..

Back to batteries!

Ok all batteries were topped off, they all settled down a little as expected, when I put the PCB back on and checked the BMS every voltage matched the MM voltages. So BMS is getting good data from the balance leads. The pack was about 20mA out of what from my "top off" but it balanced out and it was like 6mV. So I cleared the kill a watt meter hooked up the inverted and plugged in my office at 12:20PM Eastern time.

Voltage was hovering at like 58V which I knew would drop quickly since we were at the high end of the batteries.

I'm pulling from 2.4 - 2.8 Amps out of the battery (DC), BMS confirmed the ammeter. So two devices telling me the current out match. I have my laptop plugged into the BMS and its reporting the same values and the pack remains to be in balance at 4ma at the voltages are at 3340 +- 4 to 6mv. Checked battery voltages independently and they match as well. There no point that I measured where I got a different pack voltage. (BMS,PCB,Terminals ect..)

I used my thermal imaging gun, nothing is remotely hot. The BMS comes in as the hottest thing at around 90F.

The hottest thing is the bluetooth adapter for the victron inverter!
Voltage on the kill a watt is bouncing around a lot from 110 watts to 130-140 watts. Amps shows as 1.2A (but changes) on the AC side.

The BT app and the kill-a-watt have a delta of about 10 volts. The kill-a-watt is lower the iphone app is higher. Again the voltage changes pretty much every second but it does seem that the app is off a bit.

Now its 1 PM on the nose, batt voltage: 53.33V
currentt -2.5A
battery delta 7 ma
Normal capacity 74000,aH
Remaining capacity 72170maH
The little graphic of SOC shows 97% full.

Kill a watt shows 0.083KwH

So far so good??

 

[cyberfed]

Approaching 2 hrs (in 20 mins) which before seemed to be the half way mark before death.

Batt: 53.17 V
Current: -2.65
On the LCD BMS screen:
RSOC 94%
ReMain: 69.38AH
FCC: 74AH

On the laptop connected to the BMS hardwired:
Delta 4-7mA
Average Volts: 3.322
Graphic shows 94% full.

Kill-a-watt:
0.214KWh
around 120-130 watts, moves fast..
Amps 1.1 - 1.2

Now its 2:09PM so 10 minutes from the 2 hr mark! I'm thinking the back is holding up yeah?

 

[sunshine_eggo]

2.65 * 53.17 = 141W vs. 125W average output.

About 89% efficient.

1hr 40 min@ 2.65A = 4.4Ah or 69.6Ah remaining vs. 69.38Ah reported.

So far everything is in reasonable agreement.

 

[fnnwizard]

It sounds like everything should be working fine. Most decent computer power supplies have power factors close to .99 so DC watts should be close to AC watts less inverter inefficiencies.

I didn't answer your previous questions about what turns charging off, because it could be both bms or inverter. But here is a visualization of it.

Operating Voltage points are somewhat a personal thing, but there are still best operating procedures one can stick to.