Even Newer 4 to 24-cell Battery Management System (BMS)

[fechter]

A cheezy way might be to place a hefty diode(s) in series with the charger output. Depending on the diode used, it should give about 0.6v less per diode. The diodes will get pretty hot depending on the current, so would need a heat sink. Some chargers won't work with diodes in series with the output also.

Have you ever opened up your charger? I don't know about that model, but many of them have a trimmer pot on the board to adjust the voltage.

 

[Kingfish]

A cheezy way might be to place a hefty diode(s) in series with the charger output. Depending on the diode used, it should give about 0.6v less per diode. The diodes will get pretty hot depending on the current, so would need a heat sink. Some chargers won't work with diodes in series with the output also.

Have you ever opened up your charger? I don't know about that model, but many of them have a trimmer pot on the board to adjust the voltage.

Richard, I would be keenly interested in having a diode (or diodes) spec’d out which would eliminate the spark that is developed by the battery during reverse-charging of the filter caps on my Meanwell setup when the two units are connected.

The assembly can output from 42 to 84V at 1kW. There is no room on the Meanwell circuit board before the terminal block so I would need to have something inline.
Forgive me; electrical engineering is not my forte

But hey – I can write an application that makes fish go across the screen! :wink:

 

[mwkeefer]

Kingfish,

When you connect your 2 meanwells to your pack... I am assuming you are not powering the meanwells before connecting the pack, it's the only way you would get a cap charge spark in reverse.

When I charge with a meanwell, I always power the (cluster) of meanwells up first then connect the pack.... This allows me to check the charger voltage is still okay before connecting the pack (important eh?) and also since it pre-charges the output there is ZERO spark or arcing from charging the cap (it's already charged) via the pack.

The meanwells take about 200-500ms to match output voltage (maybe longer but atleast that long) and begin flowing current into a pack - that's why connecting to a powered meanwell won't cause a spark... think of it as a soft start circuit (I think that's it's purpose!).

If your dead set on doing things as you have been, just calculate a startup resistor (spark arrestor) and use that to precharge the caps in the meanwells from the battery pack before connecting for charge - the same as you would with an eBike controller having the identical issue

Between Richard, Jeremy, Gary and myself - post pics of the charger your using we can probably get it set lower or figure out where and how to mod it.

I would not suggest a diode to acheive the required voltage drop, I am fairly certain we can just mod a resistor or two to get the output voltage to a proper multiple of cell voltage desired.

Regards,
Mike

** EDIT ***
PS:
I know your charger now (had to look it up) internally they are nearly identical to the 1010B+ iCharger, just the thunder power branded model instead... the unit you have has a USB port right? Also could you power on and check your chargers firmware version and revision...

One trick would be to set it for LiIon which will terminate at 4.1v per cell, much more conservative but perhaps only 400mah lost from top end capacity... I know I drop from 4.16 to < 4 within .600mah on a 5AH pack so... you get the idea

 

[rscamp]

I know your charger now (had to look it up) internally they are nearly identical to the 1010B+ iCharger, just the thunder power branded model instead... the unit you have has a USB port right? Also could you power on and check your chargers firmware version and revision...

One trick would be to set it for LiIon which will terminate at 4.1v per cell, much more conservative but perhaps only 400mah lost from top end capacity... I know I drop from 4.16 to < 4 within .600mah on a 5AH pack so... you get the idea

Unfortunately, the latest firmware for the TP1010C is V2.5 and this doesn't have a LiIion setting - just LiFe and LiPo. I updated it to V2.5 some time ago...

 

[Hinse]

Hi everyone,
I have a 4-24 - Cell Battery Management System built from the PCB that TPPacks sold me last year. I would like to build another one for a 4P22S A123 M1 pack.
Does anyone know where we can get the PCB board?
TPPacks and rechargeablelithiumpower have stopped selling them.

Thanks,
Pierre

 

[fechter]

We sold the last of the Ver. 2.x boards a while back.
We are hoping to relase the new version soon, but we've been saying that for nearly a year now. There have been many design issues and changes that have delayed things.

 

[AndyH]

Hi everyone,
I have a 4-24 - Cell Battery Management System built from the PCB that TPPacks sold me last year. I would like to build another one for a 4P22S A123 M1 pack.
Does anyone know where we can get the PCB board?
TPPacks and rechargeablelithiumpower have stopped selling them.

Thanks,
Pierre

I still have 24 channel V2.6C boards, Pierre.

Andy
www.rechargeablelithiumpower.com

 

[GGoodrum]

This is the candidate "replacement" for the v2.x-series of BMS boards:

View attachment 3x8s BMS-v4.1.4d.png

It is 3.95" x 13.00" and has the latest charge controller, with the new layout. It still needs a bit of testing, to verify some resistor values, related to the HVC opto turn-on point, but everything else is pretty solid. I'm up past my ears, in the CellLog-related stuff, so I'm not sure when these will be ready, but we'll see.

-- Gary

 

[Gregb]

Make it sooner rather than later, pretty pretty please........
awaiting eagerly.
Greg

 

[GGoodrum]

Make it sooner rather than later, pretty pretty please........
awaiting eagerly.
Greg

Andy is going to do the testing on these, this coming week. The boards are ordered, and are being sent directly to him.

I'm going to be tied up the rest of this week, testing the CellLog-based setups.

-- Gary

 

[AndyH]

Andy is going to do the testing on these, this coming week. The boards are ordered, and are being sent directly to him.

I'm going to be tied up the rest of this week, testing the CellLog-based setups.

-- Gary

I'm looking forward to it, Gary. I'll start on the bench with a few cells, then build it out to 21 channels and abuse the board with my experienced TS pack. Between one weak cell and the Paktrakr load on 9 of the 21 cells, we'll see how the BMS works and how it can handle out of balance packs. Any ole BMS can handle your perfectly balanced LiPo packs. :wink: :lol:

Andy

 

[Anonymous]

I could do with a 16s bms.... *fingers crossed* thank you for your marvellous work

Andy is going to do the testing on these, this coming week. The boards are ordered, and are being sent directly to him.

I'm going to be tied up the rest of this week, testing the CellLog-based setups.

-- Gary

I'm looking forward to it, Gary. I'll start on the bench with a few cells, then build it out to 21 channels and abuse the board with my experienced TS pack. Between one weak cell and the Paktrakr load on 9 of the 21 cells, we'll see how the BMS works and how it can handle out of balance packs. Any ole BMS can handle your perfectly balanced LiPo packs. :wink: :lol:

Andy

 

[AndyH]

Gary - the BMS V4.1.5 boards arrived and the first control section is built. I'll build a couple of channels and then put it to work on the bench. If that goes well, I'll build out 21 channels and get it on the scooter for more fun.

Andy
View attachment 615control.jpg

 

[wookey]

This is all looking very nice chaps. I'd just like to second indubitably's suggestions of an idiot's guide. As both this analogue BMS and the cell-log based setup appear to be coming to fruition at much the same time the number of possible options has grown rather, and I know I'm not sure which way to jump. Something describing the pros/cons, pyhsical arrangements, layouts and connectors of the various options would be marvellous. I realise that probably needs to wait until the final boards and BOMs are actually done - there are only so many hours in the day. I think the cell-log-based scheme is probably most pleasing, especially as I can use the celllogs to make poor-man's manual balancing less painful in the meantime. (Do celllogs come with a connector pigtail or not?)

I know I am getting very bored of manual-balancing on a weekly basis (even if I have taught the wife to do it now, as she's using the bike , so the tantalising nearness of actual product is great. I've been holding off buying connectors for the packs, waiting for it to become clearer how things fit together. I'm still a bit vague about that: It looks like fitting 5-way JST-XH female connectors to my (4 off 4S) packs is likely to be compatible, except that I'm not sure if 4S/5way connectors will go into the 9-way connectors on the 8-Channel LVC + 4p Parallel Adapter for example? Or if I'd need to either have a Y-cable for 2x 4S-> 8S , or change the male connectors on the board to be 5-way, one row offset to the left end, one to the right end, or fit offset 9-way connectors to each 4S pack so they plug nicely in to the 8-channel boards. How is this normally done? And are the connector options the same if I go the analogue route? It's all very hard to work out without the boards in front of me, but I'd really like to get some connectors on my packs now as it would make manual balancing a lot less fiddly.

And where does one buy connector parts? Tpacks? hobby-city (They have a lot of prepared JST-XH cables but I don't see pigtails, nor 9-way connectors). Digikey seem to have everything but that way I get to do a lot of dull crimping without the right crimp tool.

Any clarifications would be most welcome.

Sorry for the noobish questions, but I suspect I am not the only one wondering about these things. My credit card remains poised, so here's hoping you won't find any more issues requiring respins.

Oh and one last thing. The meanwells draw about 6W after they have turned down the output current to zero, which is annoying. Your control stuff is only working on the front-end DC, right?, so nothing will actually switch the PSU off completely when charging is done? I'd have to think up some scheme with a relay on the AC side to actually kill the parasitic overnight drain?

 

[flip_normal]

And where does one buy connector parts? Tpacks? hobby-city (They have a lot of prepared JST-XH cables but I don't see pigtails, nor 9-way connectors). Digikey seem to have everything but that way I get to do a lot of dull crimping without the right crimp tool.

Any clarifications would be most welcome.

Answer in 'Resources & EV Related Parts' section.

 

[Anonymous]

Dearest GARY!

I've tried getting your attention, but you're pretty busy i know,
I'm marching forward with my a123 pack and really...I've done the balance charging manually via 3 x imax b6 chargers (el cheapo chargers that take all day to recharge, but I only paid 70dollars for them and they're cross-compatible with lipo for a later upgrade), and now all I need is an LVC?
I've got a friend who might be able to help me with this LVC but I thought I should ask you if you may have anything that strictly does LVC alone.... losing the cells is my biggest worry...I've looked high and low for a LIFEPO4 buzzer alarm that cuts off when each battery reaches 2volts...but i haven't found anything yet.

I appreciate the time it takes you to post, but I could really do with some help as I want my bike finished and on the road before the cold comes in...I've spent the whole summer waiting for parts to arrive. and it's been 3 months without my ebike and I miss the fun it gives.



I used motherboard connectors. "StarTech.com 20 Pin ATX Power Supply to 24 Pin" I marked with a black pen where the connectors went, they slip into one another perfectly, and with consistent usage it becomes easier to plug the balancers into this to recharge.

...

This is all looking very nice chaps. I'd just like to second indubitably's suggestions of an idiot's guide. As both this analogue BMS and the cell-log

 

[GGoodrum]

Dearest GARY!

I've tried getting your attention, but you're pretty busy i know,
I'm marching forward with my a123 pack and really...I've done the balance charging manually via 3 x imax b6 chargers (el cheapo chargers that take all day to recharge, but I only paid 70dollars for them and they're cross-compatible with lipo for a later upgrade), and now all I need is an LVC?
I've got a friend who might be able to help me with this LVC but I thought I should ask you if you may have anything that strictly does LVC alone.... losing the cells is my biggest worry...I've looked high and low for a LIFEPO4 buzzer alarm that cuts off when each battery reaches 2volts...but i haven't found anything yet.

There are 6-channel and 8-channel LVC boards available here: http://www.tppacks.com/products.asp?cat=26

-- Gary

 

[MitchJi]

Hi Gary and Richard.

In case you missed it this seems relevant to configuring your BMS:
http://endless-sphere.com/forums/viewtopic.php?f=14&t=20698&start=0
Excerpt:

I used it to analyze the capacity of a 2300 mAh A123 cell that had been charged to various voltages. The cell was charged with an HP E3614A cc/cv lab supply set to a 6A current limit. The cell was charged until the current fell to 50 mA. It was then removed from the charger and discharged (at around a 1C rate) to 2.5V. Extracted power was:

3.65V -> 2293 mAh
3.55V -> 2282 mAh
3.45V -> 2268 mAh
3.40V -> 2274 mAh
3.35V -> 2157 mAh
3.325V -> 1526 mAh
3.30V -> 953 mAh
3.25V -> 611 mAh

There appears to be no significant difference in capacity when charged above 3.40V. Somewhere between 3.30V and 3.35V cell capacity drops off a cliff. A123 recommends a float charge voltage of 3.45V

Charging to a voltage lower than the standard recommended 3.65V may improve battery life, particularly if you don't trust the accuracy of the cutoff voltage of your charger. A value of 3.50V looks like a good place to set an iffy charger... midway between the recommended charge voltage and where the capacity begins to fall off. Charging at lower voltages takes longer than charging at higher voltages. Those runs at 3.30V and below took forever for the charge current to taper off.

 

[fechter]

Thanks for the heads up. Really good work by texaspyro there.

We can design the cutoff voltages to be about anywhere we want, so we'll definitely be looking at lowering it a bit.

 

[GGoodrum]

I know there isn't much capacity difference, but if you charge at a bit higher voltage, like say, 3.60V, you can shut the charge off when the current is higher that you would normally, and it could end up taking less time to charge than if you charge to 3.45V and let the current drop all the way.

 

[steveo]

any word on when the new boards are avaliabe... they are looking great!... good job guys!

 

[GGoodrum]

Yes, we are zeroing in on this, finally. The new control scheme is working well in the CellLog-based system (see here...), so what we are doing now is a new control section for the full BMS, based on the same circuit used with the CellLog control unit, and then modify the cell circuits so that we have separate set points for the charge/balance point, and the HVC trip point. Initially at least, we will try 4.15V and 4.23V for LiPo and 3.60V and 3.70V for LiFePO4.

-- Gary

 

[MitchJi]

Hi Gary,

I know there isn't much capacity difference, but if you charge at a bit higher voltage, like say, 3.60V, you can shut the charge off when the current is higher that you would normally, and it could end up taking less time to charge than if you charge to 3.45V and let the current drop all the way.

But isn't there a substantial difference in battery cycle life (links with brief excerpts)?:
http://thisweekinbatteries.blogspot.com/2010/02/pull-plug-your-battery-will-thank-you.html

So long story short, stuff (e.g., passive layers and poor kinetics of reactions) happens and things are not as bad as they seem and you can increase the voltage up to 4.2V without bad things really happening. All chargers for Li-ion cells today cut the battery off when it reaches 4.2V. What you have to realize is that at 4.2V, these side reactions are present in finite amounts and start to chemically kill the battery, but its not that dramatic.

Operating to 4.1V makes things better and extends the life, 4.0 V is even better and so on. So why don’t battery manufacturers cut the voltage off at, say, 4 V to get better battery life? Because every time you cut this voltage down you decrease the capacity of the battery and its run time. The 4.2V cutoff is a compromise between good run time and decent (read “not pathetic”) life.

http://thisweekinbatteries.blogspot.com/2010/02/battery-rules.html

In the previous blog post, I told you why you don’t want to charge them too high in voltage (answer: side reactions). Higher the voltage, higher the fade. Turns out that if you swing the state of charge too much (i.e., charge and discharge the battery completely each time) the life decreases significantly.

This is because the battery materials expand and contract on charge/discharge (by as much as 10%). This constant “breathing” results in the particles cracking. As a matter of fact there there is data that shows that if you swing the battery to, say 3% (like in a HEV) you can get 300,000 cycles (yes, you read that right). But if you swing them all the way, you only get 300-1000 cycles. So you can charge and discharge them a lot, but you can't let them swing too wide.

 

[AndyH]

Gary - the BMS V4.1.5 boards arrived and the first control section is built. I'll build a couple of channels and then put it to work on the bench. If that goes well, I'll build out 21 channels and get it on the scooter for more fun.

Andy

Unfortunately, things are pretty 'stuck' at the moment. The channel sections appear to work very well so far in bench testing but the control section does not yet work. There are a couple of board problems, so I've had to stop everything and go back to validate the board layout. Once I can get the control section working I hope to be able to report on how well the new channel circuit work in the 'real world'.

Sorry.

Andy

 

[fechter]

Yes, we're still having issues with the control section.
This has been the hardest part to get right, since the interaction with both the cell circuits and the charger make predicting the behavior very difficult. We're trying a few things now to try and sort this out.

I'm also working on "plan B" which is a relay-based control scheme that we're using with the CellLog setup. This may allow for either analog or CellLog based control to work with the same control circuit. It has a few drawbacks, but many advantages. Still working....