Chargery BMS News / UPDATE (Nov.25.2020) Ver 4.02 firmware

[cinergi]

don't need precharge again, when DCC cut off charge or discharge, only cut off current flow, the capacitor voltage on charger and loader don't change.

Actually, precharge is always required. Inverter capacitors quickly self-discharge.

 

[Bob B]

Still testing the update, but have made a couple observations.

1. My BMS8T 300A doesn't go from storage to charge til it sees .5A .... Cal is seeing similar results with his logging on another thread.
@Chargery initially reported that it was supposed to change at .3A
2. I am still seeing some SOC / AH tracking numbers I don't understand.
I fully charged my pack with the AH set to 360 and the WH set to 4680. I am discharging to do a capacity test and see if the Chargery tracks correctly. Right now, the Chargery shows 43% which is within a couple % of the other monitor I am using .... but the AH shows 184 and the WH shows 2059.
Neither of those are 43% of full ... So, I'm a little confused but still testing.

 

[Steve_S]

SOC is an issue... I have a multipage page email from Jason to decode... Some suggestion... erf.
Discharge the full pack to 0 SOC at 1C rate, then charge at 1C...
It takes 120VAC @ 25A to run a charger at 75A for a 24V battery pack.
I've got 280's & 175's and to do that means Big Genny which GUZZLES and many hours per battery pack. Let me tell ya, it is NOT CHEAP ! 8 hours run = 25 Litres of gas. My Gen is wired to use the L5:30 120V/30A line for this. I'd have to charge a LOT to switch up to use the 240VAC out and it would be costly too... we are talking some building wire, breakers and much more...

An interesting side-note: The two 280AH packs both had hit 0% SOC & Cutoff in my testing. Applied 75A for 4 hours and then 50A charge for another 4 hours. One came up to say it was 100% but the other said 55%. The cell voltages in both packs were all within 10mv of each other, sitting around 3.390. They Both Zeroed, they should have counted up similarly but did not. The two 175's don't do that but they are used EV Grade cells, they are close in IR at Low Volt & Hi Volt till they reach their edges and then the "runners" appear and cause diconnects. Call me confused or even befuddled.

Above being said there is some KEY POINTS which most folks do miss;
These commodity cells and NOT EV-Grade Cells, these are ESS Grade so the tolerances are not the same, neither are the C-Rates. Higher C-Rates require a higher grade of material. These are NOT Matched & Batched so you will have cells with variable IR rates within a batch, all within spec but varying between one & the other. This inevitably will result in some cells hitting LVD or HVD before the others. THOSE ARE THE LIMITERS ! So ultimately the SOC of a battery pack is limited by the weakest & strongest cells. This has been really been born out with my Thrash Tests with not only the EVE cells but those "other ones" too.

I want to be quite clear a point regarding the Thrash Testing. I have NOT written this up in detail yet, I may do so I may not... I likely should.
I've been at it for 2 weeks and likely another two as I am now just waiting on more gear to get into the next phase of it. As I am offgrid and having to use a Nasty Gasser for controlled push charging, I've dumped over 200 litres of gas into guzzler if not more along with 3 oil changes (every 40 hours and Synthetic to boot). Beasty hasn't run this hard since I was using the Compressor & Nail Gun to build my house... Not doing this lightly or for fun. Gas up here isn't cheap either ! So I'm already out close to $300 on gasoline alone and now more hardware, cells and other gear, by the time I've finished this, there will be yet another $1,000 dumped into it if not more. IF I end up killing Genny (it does have a LOT of hard hours on it, it is several years old) well that's gonna up the cost of my "experiment" significanly (9kw gen not cheap).

Here is a partial description of my Thrash testing extracted from a PM Discussion for anyone interested. Some not in there, some of the equipment I am waiting on now is a replacement IR Tester my YR1030 gave up the ghost and waiting on a YR1035+ as I am having to go back to Cell Level stuff again.

Spoiler: Partial Thrash Testing Info

This is all being done at pack level.
It's in combo's of all 4 packs connected or down to just one.
Charge rates (pushed by me via genset) range from 40A to 75A usually, Have pushed it higher but only as a short test because of my wiring and fusing, I didn't setup my system for that.
When I have Good Solar, I have pushed 125A+/- into them. with solar & gen simultaneously charging. (that's tricky)
Again, all in combos of 1 to 4 packs.
2x 175 & 2x 280.
The 175's are recovered from the ShunBin mess, they are used cells with unknown mileage.

So the packs have seen anywhere from 9A charge to 100A (except the 175's, they only got 75A).
They sure as hell charge faster with more amps... LOL
Some warm up faster compared to others at high amp. Biggest culprit cell to busbar connections. Next is the damned IR, it is more susceptible to Amps Received, just like Amps Pulled. I'm going by Chargery's BMS info, I figure 4 should be comparable even if they are all out a bit, the wiring is identical, I made it so. Did cross checking with DVOM and all seems to be ok as far as reporting info.

Divergence points, well, it's weird. I'll leave the 175's out of it cause they are what they are.
The 280's between 3.1- to 3.400 +/- 0.080 are pretty much all happy as clams in perfect waters. Tickle around it and things get interesting. That's when the divergences starts.
- Charging: around 3,385 & up the cells go beyond say 20-25mv (at high amp) and 10-15mv (low amp), they will go off and can hit as much as 250/300mv.
-- Discharge, once below 3.050 they start bouncing a bit, then the lackey will drop and cutoff the pack. The mv difference grows quickly though so I am seeing 6 cells at 3.18, 1 at 3.15 and all of a sudden the "one" will drop to 2.99 or lower, triggering cutoff.

Case in point, and another oddness.
Day before yesterday, I emptied them all, flipped to FLA bank went to bd later, Yesterday, 24 hour genset charge at 75A, all 4 packs. They hit 27.3V @ the BusBars each pack was +/- 0.050 of that.
Shutdown that bit, had enough of listening that to that and it's been crud weather for days so no Sun either. Voltage settled to 26.50 @ the Midnite E-Panel (with Wizbang on Shunt) and same on Inverter Monitor.. They have been napping along more or less and now 24 hrs later (no charge today again due to weather) they sit at 26.37. They'll take another 2 days to hit 0. Maybe 1.5 if I brew a couple of extra coffee pots or something like that. ;-)

*My consumption is quite low, as I sit here writing this I'm using 6.3A from 24V Bank. If the fridge kicks on it goes to 10A for a few a few minutes. Just as an example.

Also, I have to point out... I did not Top Balance these like others each 280 packs set of cell was charged a different way and parallel settled in their sets. 1 was done as 4 cells 12Vx 2, the other as 24v, reordered & 24v again.

With the TekPower 15V/45A benchtop enroute I'll be topping up individual cells and seeing what that does... I may create a harness to do two or 4 cells, but I want to do 3.60 @ 45A on one cell and let it trickle down to a couple of amps. Hope if I start off with a semi decent charge already in there, should not take that much to push a single cell up quick, or even just a pair at a time.

I really wish I had a way of logging the BMS output but at this time I don't. Was relying on getting a Raspberry Pi up & running so it could take the info from the USRIOT N540 which is connected to all 4 BMS' but that is on hold dang it. Using Python from my PC will only read one port and process it too... spent too many hrs futsing with that. Ruckus Stuckus Frickus Fruckus....... and a BamBam too ! LOL.

Thing is quick & dirty, these 280AH cells are fine between 3.150 and 3.400 Above/Below and things get funky, this so far seems to be their happy place.

OH, TEMPS. The powerhouse is kept at 15C/59F, when charging the batteries & inverter warm it up a bit... Tiz below freezing here...

 

[Chargery]

Still testing the update, but have made a couple observations.

1. My BMS8T 300A doesn't go from storage to charge til it sees .5A .... Cal is seeing similar results with his logging on another thread.
@Chargery initially reported that it was supposed to change at .3A
2. I am still seeing some SOC / AH tracking numbers I don't understand.
I fully charged my pack with the AH set to 360 and the WH set to 4680. I am discharging to do a capacity test and see if the Chargery tracks correctly. Right now, the Chargery shows 43% which is within a couple % of the other monitor I am using .... but the AH shows 184 and the WH shows 2059.
Neither of those are 43% of full ... So, I'm a little confused but still testing.

2059/4680*100%=43.9%., 43% SOC reading is correct, please continue discharging, till 0% SOC, check if it is accurate or not. and compare with other monitor reading.
because the current under 0.5A with 300A shunt system is ignored, if any device drain current at under 0.5A when battery is not in using, battery AH and WH will go down, but BMS don't update AH and WH due to 0.0 A current reading, so SOC don't be updated, the errors are always accumulative, if some current don't through shunt, the problem is more serious. so current detection must be very very accurate, even 0.1A for 12 hours, battery capacity will drop 1.2AH, if battery voltage is 12V, 14.4wh will be waseted and BMS don't record this. in one month, if battery is drawn at less than 0.1A for 10 12 hours, the accumulative error would be 12AH or 144wh.
so even with 16bit ADC, 0.1A current can be detected , the BMS is still not "perfect".
any ideas for the problem?
have a nice weekend.

 

[Cal]

so even with 16bit ADC, 0.1A current can be detected , the BMS is still not "perfect".
any ideas for the problem?
have a nice weekend.

I think you can do a lot better than 0.1A resolution with 16 bit adc. Just for the heck of it (for fun), I designed and built a circuit using a 16 bit adc (actually I'm using it as 15 bit adc) that has 31 mA resolution.

Take a look at INA226. That part may get you down to 10 mA. I got some on order and will check them out.

 

[Bob B]

2059/4680*100%=43.9%., 43% SOC reading is correct, please continue discharging, till 0% SOC, check if it is accurate or not. and compare with other monitor reading.
because the current under 0.5A with 300A shunt system is ignored, if any device drain current at under 0.5A when battery is not in using, battery AH and WH will go down, but BMS don't update AH and WH due to 0.0 A current reading, so SOC don't be updated, the errors are always accumulative, if some current don't through shunt, the problem is more serious. so current detection must be very very accurate, even 0.1A for 12 hours, battery capacity will drop 1.2AH, if battery voltage is 12V, 14.4wh will be waseted and BMS don't record this. in one month, if battery is drawn at less than 0.1A for 10 12 hours, the accumulative error would be 12AH or 144wh.
so even with 16bit ADC, 0.1A current can be detected , the BMS is still not "perfect".
any ideas for the problem?
have a nice weekend.

@Chargery .... The % and the WH were within reasonable accuracy .... for 43% 4680 x .43 = 2012 The display showed 2059 ... So, if it was closer to 44% that would be OK.

It is the AH indication that doesn't make sense to me. 360 AH x .43 = 154.8 While the AH reading on the display was 184.
Even at 44%, the AH reading should be 158.4. 184 AH would be 51% of 360.

 

[Bob B]

I did an experiment this evening connecting my 4S pack directly to the external power of the Chargery .... So, the Chargery is being powered directly from the battery instead of thru the cell harness. I connected it as described in the new Chargery manual .... See the diagram below.

This almost completely resolved the issue where cells voltages for cells 1 and 4 read 20-25mv lower on the Chargery than with my meter. I say almost, because for some reason cell #1 is still about 7mv low. Cells 2,3,and 4 are pretty much exactly the same as my meter.

The additional benefit of doing this is that I connected the negative connection on the downstream side of the shunt ..... so any power used by the Chargery ... or balance power will be measured by the shunt.

I added a 5A fuse in the positive feed from the battery .... otherwise connected as shown below. I used these power connectors from Amazon. https://www.amazon.com/gp/product/B01AGQVYQQ/ref=ppx_yo_dt_b_asin_title_o04_s00?ie=UTF8&psc=1

Edit: One negative thing I discovered is that the Chargery completely looses it SOC tracking when powered off and back on. I didn't realize that or I would have waited and done this test later. I was testing the SOC tracking with the new firmware update .... and was fairly happy with it except the AH tracking, which is just wrong. When I powered back up the % went to 100% and the WH and AH readings reset to full value.

 

[Chargery]

thanks Bob for your hard work,
"The additional benefit of doing this is that I connected the negative connection on the downstream side of the shunt ..... so any power used by the Chargery ... or balance power will be measured by the shunt. "
this connection can decrease the actual discharge current, and increase the actual charge current, if the BMS itself consume 0.1A from battery, when discharge at 1A , the BMS will display 1.1A, or when battery is charged at 1A, the BMS will display 0.9A, but the SOC, AH and WH will be more accurate. in future, when BMS can detect under 0.05A, the battery will not be in storage status, because BMS is always powered by battery. only current is zero, battery is in open-circuit, and in storage. at this time press STOP button, BMS can work on sleep mode. if battery is not in storage, BMS cannot work on sleep mode, need find new soluton to solve it.

" One negative thing I discovered is that the Chargery completely looses it SOC tracking when powered off and back on. "
it is fixed on V4.03, Please updat the BMS main unit to V4.03, the new firmware for BMS8T-300 and BMS8T-100 is attached. if your BMS is not 8T 300A or 100A, please let me know.

 

[Chargery]

@Chargery .... The % and the WH were within reasonable accuracy .... for 43% 4680 x .43 = 2012 The display showed 2059 ... So, if it was closer to 44% that would be OK.

It is the AH indication that doesn't make sense to me. 360 AH x .43 = 154.8 While the AH reading on the display was 184.
Even at 44%, the AH reading should be 158.4. 184 AH would be 51% of 360.

soc reading is based on WH, so it is accurate, if the AH setting is not accrate, AH and WH based on AH will have a difference, please updat Ah setting, may be it can be improved. from the tesing result, the AH setting is larger than actual battery capacity.

 

[Tobi17]

Hello,

thank you for all postings. I was not aware that the voltages are that off per cell with the Chargery! Even if external powered the measurements are really not that good. The update v4.02 does not give me any kind of improvement. If I measure the cell voltages directly at the Chargery connector and compare it what the Chargery internally display - there are easy 20mv differences per cell! Imagine the Chargery balance your cells (I have tested this) on this wrong voltage measurements... it will imbalance the cells instead of balance them! Well the voltage reading for a BMS is really very disappointing. I do not think, we do not get any improvement on our old hardware anymore.

One thing I would have expected from a BMS is, to read the voltages from each cell correctly. Maybe this is the most important thing for me.


Bye
Tobi

 

[Bob B]

Hello,

thank you for all postings. I was not aware that the voltages are that off per cell with the Chargery! Even if external powered the measurements are really not that good. The update v4.02 does not give me any kind of improvement. If I measure the cell voltages directly at the Chargery connector and compare it what the Chargery internally display - there are easy 20mv differences per cell! Imagine the Chargery balance your cells (I have tested this) on this wrong voltage measurements... it will imbalance the cells instead of balance them! Well the voltage reading for a BMS is really very disappointing. I do not think, we do not get any improvement on our old hardware anymore.

One thing I would have expected from a BMS is, to read the voltages from each cell correctly. Maybe this is the most important thing for me.


Bye
Tobi

@Tobi17 .... What BMS do you have? What is your pack configuration?

Even when my readings were a little off, the pack was balanced pretty well.

Can you give us the readings of the Chargery Vs your meter for all cells?

 

[Bob B]

thanks Bob for your hard work,
"The additional benefit of doing this is that I connected the negative connection on the downstream side of the shunt ..... so any power used by the Chargery ... or balance power will be measured by the shunt. "
this connection can decrease the actual discharge current, and increase the actual charge current, if the BMS itself consume 0.1A from battery, when discharge at 1A , the BMS will display 1.1A, or when battery is charged at 1A, the BMS will display 0.9A, but the SOC, AH and WH will be more accurate. in future, when BMS can detect under 0.05A, the battery will not be in storage status, because BMS is always powered by battery. only current is zero, battery is in open-circuit, and in storage. at this time press STOP button, BMS can work on sleep mode. if battery is not in storage, BMS cannot work on sleep mode, need find new soluton to solve it.

" One negative thing I discovered is that the Chargery completely looses it SOC tracking when powered off and back on. "
it is fixed on V4.03, Please updat the BMS main unit to V4.03, the new firmware for BMS8T-300 and BMS8T-100 is attached. if your BMS is not 8T 300A or 100A, please let me know.

@Chargery .... I update my BMS8T-300 to the 4.03 update. It does indeed hold the SOC values after a power down now.

I didn't reset to default after doing this update .... is there any need to do that? I am using a different shunt and didn't want to re-do my calibrations.
I am doing ongoing test on SOC and will report some of those results in a few days.

I guess we could start a new thread for 4.03, but that doesn't seem necessary to me.

 

[Chargery]

@Chargery .... I update my BMS8T-300 to the 4.03 update. It does indeed hold the SOC values after a power down now.

I didn't reset to default after doing this update .... is there any need to do that. I am using a different shunt and didn't want to re-do my calibrations.
I am doing ongoing test on SOC and will report some of those results in a few days.

I guess we could start a new thread for 4.03, but that doesn't seem necessary to me.

Don't need reset to default, and don't need calibrate anything. all is ok except the soc can be hold after power on again.

 

[Bob B]

@Chargery ... Thanks for the information.

I have another suggestion for a firmware enhancement based on a conversation from another thread.

Would it be possible to add a parameter to select the BMS itself to be either common port or separate port?
Doing that would be more simple than having a special cable and a different DCC for common or separate. There could be just one type DCC. The Chargery would just command both outputs off any time there was a shutdown event if the parameter is selected to be common port.

I personally don't like the idea of doing a common port configuration, but it would be easier to do that with software than with hardware.

 

[Capt Bill]

thanks Bob for your hard work,
"The additional benefit of doing this is that I connected the negative connection on the downstream side of the shunt ..... so any power used by the Chargery ... or balance power will be measured by the shunt. "
this connection can decrease the actual discharge current, and increase the actual charge current, if the BMS itself consume 0.1A from battery, when discharge at 1A , the BMS will display 1.1A, or when battery is charged at 1A, the BMS will display 0.9A, but the SOC, AH and WH will be more accurate. in future, when BMS can detect under 0.05A, the battery will not be in storage status, because BMS is always powered by battery. only current is zero, battery is in open-circuit, and in storage. at this time press STOP button, BMS can work on sleep mode. if battery is not in storage, BMS cannot work on sleep mode, need find new soluton to solve it.

" One negative thing I discovered is that the Chargery completely looses it SOC tracking when powered off and back on. "
it is fixed on V4.03, Please updat the BMS main unit to V4.03, the new firmware for BMS8T-300 and BMS8T-100 is attached. if your BMS is not 8T 300A or 100A, please let me know.

just curious. I do not see see this new update on the Chargery update page: https://chargery.com/update.asp
Wondring: How come? Why am I not able to find this newer update version 4.03 at this more offical Chargery update web page?? I downloaded one for my BMS8T w 300 Amp shunt; wondering where the 4.03 LCD version is; was just getting ready to do the 4.02 lcd and main unit update; and just found this page with the V4.03 firmware update. I I trust this v 4.03 update enough to try it out? Where did this newer update come from ??? :+)

 

[Bob B]

just curious. I do not see see this new update on the Chargery update page: https://chargery.com/update.asp
Wondring: How come? Why am I not able to find this newer update version 4.03 here, and not on that Chargery Update web page? I download one for my BMS8T w 300 Amp shunt; was getting ready to do the 4.02 lcd and main unit update; and just found this page with the V4.03 firmware update. and wondering if I trust it enough to try it out. :+)

I installed it no problems ... and it resolved the issue where a power down lost SOC tracking data.

No idea why the web page hasn't been updated, but it was pretty amazing how fast they came up with a fix for that problem.

 

[Capt Bill]

I installed it no problems ... and it resolved the issue where a power down lost SOC tracking data.

No idea why the web page hasn't been updated, but it was pretty amazing how fast they came up with a fix for that problem.

Bob, Did you just update the main unit t v 4.03, and your LCD is still on v 4.02? Asking because I never saw a v4.03 LCD update (or assume the one recently found was for the main unit). I am looking for SOC improvements too, and was looking at specifics of what has been listed as recently improvements IS this v 4.03 update indeed just for the main unit update? Is this v 4.03 update kind of a beta update, and we are sort of pre testing it for release ??? those are the question going off in ... :+) Bill

 

[Bob B]

Bob, Did you just update the main unit, and your LCD is on v 4.02?

Yes .... I reported results earlier in this thread.

For me, the 4.02 update resulted in much better .... and smoother .... voltage and current readings. I decided to test connecting the BMS directly to battery power ... as reported above also ... and in the process noticed that the SOC tracking numbers all reverted to 100% after powering down. I reported that and Jason came back with a fix almost immediately.

 

[Chargery]

the latest firmware is posted on our website, please update BMS according to your BMS model, don't need calibrate anything, don't need modify any settings, only update BMS main unit to V4.03, LCD version is still V4.02.
https://chargery.com/update.asp

we need some time to finish all new software for all models, and post on our website, sorry for delay.
good luck.

 

[Steve_S]

LMAO, Jason fixed the naming on the UPDATE page to make the files easier to identify. But not the actual HEX filenames which confuse people a little.

I've also asked @Chargery to post a Clear Answer to the External Power Source questions that are popping up for the BMS, which should be added into future documentation.

To use External Power from Battery Terminals you will require:
1 plug-in adapter: 12V, 5A, 5.5x2.1mm dimension Male Power Jack

For 12V you will require : 1 Shotky Diode (specification) ?
For 24V you will require : ??
For 48V you will require : ??

External Power from a DC Source, (AC to DC) or (DC to DC.)
Power Supply must be capable of providing ## Volts & ## Amps to the BMS
If different for various models, indicated such.