diy solar

diy solar

Upgrading 15s Battery to 16s

Well be aware some of these units cannot charge the batteries and invert at the same time. Some inverters can pass through your generator power and supplement with battery, but not all. Additionally, some generators are too "dirty" for the inverter to even use.

I'd be tempted to hook your generator to a chargeverter (once back in stock) to directly charge batteries, and then use the inverter to run the loads. Granted that does cost more, but it may be worth it to do what it sounds like you're trying to do.
 
Do you have a suggestion on where to purchase from? I had been planning on signature solar because I've heard good things about their support.

Edit: also not sure that I really need as much as 5k anyway, though if it was the same price as I'm looking at I would go for it. My plan is to use this to run my home server rack (about 350 watts peak), with the ability to provide backup power for my furnace (250 watts while running) and one refrigerator and one chest freezer (100 watts each running, though the refrigerator defrost cycle is approximately 600 watts for 30 minutes once a day).

It will be set up in such a way that I can use it to bolster my small generator by using it to load level while the generator is running, allowing the generator to provide a constant output (its about a 1200 watt unit) to charge the batteries and use the battery/inverter system to handle all the sudden load changes that my generator struggles with.
I purchased directly from SRNE, delivered in under 2 weeks. Powmr and SGP are relabeling them as well.

Make note that generator charging will likely pass through loads while charging the battery in any of these "cheaper units"
 
Do you have a suggestion on where to purchase from? I had been planning on signature solar because I've heard good things about their support.

Edit: also not sure that I really need as much as 5k anyway, though if it was the same price as I'm looking at I would go for it. My plan is to use this to run my home server rack (about 350 watts peak), with the ability to provide backup power for my furnace (250 watts while running) and one refrigerator and one chest freezer (100 watts each running, though the refrigerator defrost cycle is approximately 600 watts for 30 minutes once a day).

It will be set up in such a way that I can use it to bolster my small generator by using it to load level while the generator is running, allowing the generator to provide a constant output (its about a 1200 watt unit) to charge the batteries and use the battery/inverter system to handle all the sudden load changes that my generator struggles with.
We are here if you need us! If you have any questions or concerns, please don't hesitate to reach out to me at engagement@signaturesolar.com!
 
Well be aware some of these units cannot charge the batteries and invert at the same time. Some inverters can pass through your generator power and supplement with battery, but not all. Additionally, some generators are too "dirty" for the inverter to even use.

I'd be tempted to hook your generator to a chargeverter (once back in stock) to directly charge batteries, and then use the inverter to run the loads. Granted that does cost more, but it may be worth it to do what it sounds like you're trying to do.
Thanks for the heads up. I just got off the phone with Signature Solar. I think what I'm going to do is get a EG4 Chargeverter to keep the batteries topped off, and go 100% inverter powered on the circuits I wish to back up.

As I see it this does two things for me. First, it will let me hold the batteries at around 50% charge at all times so that way I retain about 5 kWh of battery backup in my system. Second, it gives me full deniability on anyone complaining that I am interfacing with the electrical grid as it will be DC coupled and completely incapable of providing AC power anywhere near the grid. That will hopefully save me from any issues with local inspectors/utility companies or whatever.
 
That is allegedly the actual JK BMS store.

That's the only way I would consider using AliExpress.

I would still prefer ordering from a North American source when possible.
Thanks for the link, it is much more clear on the desktop that it is actually from JK.

I think that this is what I'm going to need? Lots of options on that page but this is a 120A BMS which gives me a little headroom and is only a $3 upcharge from the 100 amp model
JK BMS 120A.jpg
 
Note that one only has 0.6A of balancing current. The others have 1A or 2A.
You're thinking I might benefit by going up to the 150 amp BMS to get up to 1 amp balance current? That's a $20 jump which isn't too bad, but the one with 2A balance is a $40 jump so that's getting a little large for my "cheap" project. (as if there's such a thing!)
 
You're thinking I might benefit by going up to the 150 amp BMS to get up to 1 amp balance current? That's a $20 jump which isn't too bad, but the one with 2A balance is a $40 jump so that's getting a little large for my "cheap" project. (as if there's such a thing!)
I would but I'm still running JBDs with passive balancing... I'm sure 0.6A would be leaps and bounds better...
 
Can anyone with more balancing experience chime in on the merits of 0.6/1/2 amp balance current for this effort?

I'm keeping in mind that theoretically there will be 15 cells with identical capacity and then one with presumably higher capacity in series.

The current BMS appears to have 4.7 ohm resistors in line with the leads, so passive balancing I presume?
 
Can anyone with more balancing experience chime in on the merits of 0.6/1/2 amp balance current for this effort?

I'm keeping in mind that theoretically there will be 15 cells with identical capacity and then one with presumably higher capacity in series.

The current BMS appears to have 4.7 ohm resistors in line with the leads, so passive balancing I presume?

Ultimately, they hasten the top balancing process if using active balancing (vs. wiring cells in parallel) and can compensate for SLIGHTLY different cell characteristics. For normal operation with a healthy battery, even 0.05A passive balancing current is enough.
 
Ultimately, they hasten the top balancing process if using active balancing (vs. wiring cells in parallel) and can compensate for SLIGHTLY different cell characteristics. For normal operation with a healthy battery, even 0.05A passive balancing current is enough.
Thanks!

Since this pack has the 15 original cells welded together, do we think that the 1A balancing current in the BMS would be sufficient to balance in the 16th cell assuming I get the voltages as close to each other as possible first?

My thought is to fully charge the 15 cells to whatever voltage the original BMS will allow, then charge the new cell to match that voltage before putting it inline. I'm wondering if I can avoid doing 32x total individual top balancing charges since I can't separate the existing packs to parallel the cells, so I'll have to clip onto the busbars one cell at a time on both packs if that is the only reasonable way forward.
 
My only concern is the reliability of the 15S BMS. If it will cut off, great. If not... potentially cooked cells.
I have my power supply set to 55V so that should safeguard against cooked cells assuming they stay in balance, so far so good on that front. Still only up to 51.3V while charging at 5 A presently. BMS is now up to 40% SOC so depending on how accurate that is I have about 12 hours of charge time left, if the current holds which of course it will slow down as it gets near the top.

I'm not charging unattended with this current setup, so maybe I'll have results on the first pack by the weekend.
 
I have my power supply set to 55V so that should safeguard against cooked cells assuming they stay in balance, so far so good on that front. Still only up to 51.3V while charging at 5 A presently. BMS is now up to 40% SOC so depending on how accurate that is I have about 12 hours of charge time left, if the current holds which of course it will slow down as it gets near the top.

I'm not charging unattended with this current setup, so maybe I'll have results on the first pack by the weekend.
51.3V with 15s cells, correct?
That's 3.42V / cell. Things will start happening quickly. I'd be watching them closely.
 
I have my power supply set to 55V so that should safeguard against cooked cells assuming they stay in balance

This is how cells get bloated and ruined. I would not exceed 51.75V

so far so good on that front. Still only up to 51.3V while charging at 5 A presently. BMS is now up to 40% SOC so depending on how accurate that is I have about 12 hours of charge time left, if the current holds which of course it will slow down as it gets near the top.

You can't trust the BMS. It's completely inaccurate and out of sync.


51.3/15S = 3.42V/cell

at 5A, that's at a high state of charge.

I'm not charging unattended with this current setup, so maybe I'll have results on the first pack by the weekend.
 
do we think that the 1A balancing current in the BMS would be sufficient to balance in the 16th cell assuming I get the voltages as close to each other as possible first?
I guess this is what you meant, but for clarity, close voltages is only good enough if you're above the upper knee. So, "close" at 3.63V would be OK, but "close" at say 3.35V would not be any good.
 
@42OhmsPA @sunshine_eggo

Well you guys were right, the end went really fast and the crappy BMS could not be trusted. Unfortunately I didn't read your replies in time and half fell asleep on the couch so it went 2 hours past my post there.

Final cell voltages got higher than I intended in a couple places. Final numbers at me shutting it off were:
3.613, 3.703, 3.797, 3.758, 3.657, 3.618, 3.598, 3.647, 3.731, 3.659, 3.561, 3.666, 3.667, 3.673, 3.733

This morning I didn't take pictures of the display but everything has settled back to that 3.4ish range within about 25 mV

@SeaGal Thanks, I plan on bringing everything up towards that 3.63 range as balanced. Last night's experience shows I can't trust the stock BMS though so I guess I'll be doing it manually per cell on my second pack.
 
@42OhmsPA @sunshine_eggo

Well you guys were right, the end went really fast and the crappy BMS could not be trusted. Unfortunately I didn't read your replies in time and half fell asleep on the couch so it went 2 hours past my post there.

Final cell voltages got higher than I intended in a couple places. Final numbers at me shutting it off were:
3.613, 3.703, 3.797, 3.758, 3.657, 3.618, 3.598, 3.647, 3.731, 3.659, 3.561, 3.666, 3.667, 3.673, 3.733

They're surprisingly balanced. I wouldn't call it good, but it's not horrible. Better to evaluate when the peak cell is at 3.65V. "Dumb" BMS tend to cut off at 3.75, and you didn't exceed that by much. Some of the PACE BMS used in server racks default to 3.9V for over-voltage protection, so it may be operational.

You should be fine.

This morning I didn't take pictures of the display but everything has settled back to that 3.4ish range within about 25 mV

At rest, even imbalanced cells will read very close. The 160mV at higher voltage is far more relevant.
 
Good to know, thanks. Glad I probably didn't do damage.

For the second pack I'll series charge with the still functional BMS installed but I'll cap the voltage to 51.75 as you recommended. I'll then target that 3.45 volts to charge the new 100 Ah cell that's going in and let the 0.6A balance on the new JK BMS take it from there.
 
Back
Top