DieBieMS v0.8 Batch 3

Been thinking about we’re to locate the difrent parts in the encloser of my next build. BMS all ways seem to be odd sizes don’t fit in nicely with other components. One idea was thinking of sticking the bms below the Battry next to the front wheel in a mettle heatsink. it’s a dead space that quite protected by the front wheels and trucks.

What’s people thort of problems or better ideas? Is this a silly idea as it won’t produce much heat? @jtag This will mean long battry cables to the VESC

I’m looking at the HLG-240H-48A and the GLH-240H-54A. You and @deucesdown both suggest it and I am looking for a cc/cv power supply for the DieBieMS. I’ve got a 10s6p battery but want to keep the current to under 3A, preferrably 2A. would going for the 54A version allow me to charge a 10s and a 12s pack on the DieBieMS without issue? I read that the BMS should prevent over voltage but can’t do anything about over current so if I get the 54A, and just supply the amps I want, the BMS should handle the cutting off of the voltage that it is set to and be fine?

I unfortunately got a Meanwell powersupply but it isnt CC/CV so I gotta find one that will work for my 10s6p build :disappointed_relieved:

https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&ved=2ahUKEwj2pNHC89HfAhXkg-AKHS5KCL4QFjABegQIBBAC&url=http%3A%2F%2Fwww.meanwell.com%2FUpload%2FPDF%2FHLG-240H%2FHLG-240H-SPEC.PDF&usg=AOvVaw1OjFuAkHYHCmA8SvWZ00JU

-48A is nominal 48v, adjustable 44.8-51.2v -54A is nominal 54v, adjustable 50-57v

for current, -48A 2.5-5a -54A 2.23-4.45A

The voltage adjustment range is not that wide, so probably you’ll have to figure out what you want out of it and choose one. You can add one of those ruideng/drok PSUs inline to further control voltage/current. Something like this covers 10s-12s

https://www.aliexpress.com/item/RD-DPS5005-Constant-Voltage-current-Step-down-Programmable-Power-Supply-module-buck-Voltage-converter-color-LCD/32684324154.html

I believe Ruideng is the original mfg.

There’s a buck/boost version, that has wider input range.

You can probably use one of these guys with the meanwell you already have. They have bluetooth and usb options, for remote display/control/logging if you’re into that.

1 Like

Thanks for that! I think I’ll be going with the 48A. I read that you mention it being good for 10s and 12s as well? I think the adjustable Current range will suit me fine at 2.5A at 48v and I can just set the cutoff on the diebiems to about 42v to prevent over voltage. I’m guessing that should be fine? :sweat_smile:

So if I’m going down the right line to charge a 10s4p I need a cc/cv 42v output (to charge the pack) + 3v extra for the BMS min. max at 6amp continues charge for q30 in this config recommended

so I’m presuming it will need to be a 48v was looking at the meanwell led drivers but there’s 4 options in the HLG240H range

48 48A 48B 48C

What’s the difrance what’s recommended?

https://www.mouser.co.uk/Power/Power-Supplies/LED-Power-Supplies/_/N-axgjt?P=1yxt78u&Keyword=HLG-240H&Ns=Output+Voltage-Channel+1|1&FS=True

image

From the datasheet: image you probably just want the A version which is the easiest to adjust, has 2 knobs, 1 for current and 1 for voltage

3 Likes

Standard lithium ion charging is CC/CV with C/10 termination. So for a 10s 5ah pack, that means constant current until you hit 42v, then taper off the current until you get to 0.5a, then stop. At this point you are100% full.

With diebiems, there are some other factors. To get full charge, you need a +0.6v offset for the diode. For the automatic behavior you need another +3v. If you add this +3v, in theory if I understand correctly, the bms still cut charge at 42v (if cells all charged equally) while the charger is set up for 45.6v. This means you never get to the CV portion of the charge, and will never get to 100% full (Which is probably better for your cells in the long run).

10s is 42v, diebems offset (+0.6v, +3v): 45.6v 12s is 50.4v, offset: 54v

Neither 48A nor 54A cover that range.

One way you could go is, If you get the 54A and a DPS/DPH psu, you can get what you need from one PSU. The DPS/DPH are limited to 51v I think… So you can set up the meanwell for 54v, and for 12s plug straight into the meanwell, for 12s go through the DPS/DPH unit to drop it down to 45.6.

In a way this is similar to figuring out how to charge 10s, 12s and 13s. without diebiems. :slight_smile:

BTW I really like putting something like this inline on the output

https://www.aliexpress.com/item/ATORCH-Accurate-Energy-Meter-Voltage-Current-Power-300-100A-AC-Voltmeter-Ammeter-Greem-Backlight-Overload-Alarm/32821731871.html

measures watt-hours transferred, among other things. The numbers are precise but not necessarily accurate. :slight_smile: This particular version is single piece (no separate shunt box), and LARGE. 0-300v, 0-100a.

1 Like

Thanks couldent work out if it was A or C I needed on the note I was reading.

Struggling to fined the min amp on cv mode so it will stay on until all cells are balanced

tho maybe this one would be better? On the bases 1.5a a cell reg charge 4a cell max charge

https://www.mouser.co.uk/ProductDetail/MEAN-WELL/HLG-320H-48A?qs=sGAEpiMZZMt5PRBMPTWcaRTgZe1UVjfl49KWJe2LRtQ%3D

I presuming I’m heading down the right line of thinking hear?

The way CC/CV supplies work, psu will put the set voltage on the output if the current demanded by the load is less than the set current limit. Once the current limit goes below the set limit, the supply will maintain the set voltage. The load determines the actual amps drawn. The supply can set a high limit. there is no low limit (well, 0a is the low limit).

1 Like

I get that but I thort most cc/cv will cut out in cv part of the charge when the load drops below a certain point this stops the bms finishing balancing cells. I’ve seen poste were people over time destroy cells in battery’s because the charger cut out and dident the let the bms balance all the cells because the load was to low.

Of the top of my head I think the diebiems will balance with some thing like 0.4a (I’m trying to think way back for this figure from posts months ago so might be wrong)

Meaning the cv needs a minimum load of less than that to finish balancing all the cells.

That’s the part I’m now struggling to fined in the manuals and looking for help with.

Unless I’m getting confused with a more basic charger.

1 Like

Im not sure if this confused me more or not after reading it a couple of times lol. Thanks for the wealth of info! I should probably just stop bothering with picking a PSU that can do 10s and 12s and simplyify things with picking a pack for my 10s6p haha

This means the 48A should work for me just fine with diebiems plugged directly as I can simply adjust it to the offset, 45.6v and diebiems will supposedly handle the trickle off at 42v(or if i set it lower)? but will not work directly on other BMS because it starts at 44.8v?

I’m really invested in the diebiems now which is why I got in on batch 3 for my backup in case anything goes wrong with the one I’ve got now so I dont wanna screw up my psu all the more as well :sweat_smile:

Edit: You mention that the load determines actual amps drawn, is this during charging or did you mean for discharging? Or I could just be confused with the stuff I’ve read so far lol

I looked up the alieexpress link. I may just get that along with the 48A! My multimeter cant read amps through the probes sadly as it uses a clamp instead for DC current measuring

I get you now, I’m in the same place. It seems to be a fine balancing act between the charge-rate of the bulk charger, the balancing current of the BMS, the pack over-voltage-cutoff, and cell OVC. All indications are that BMS are not balance chargers. They can keep closely balanced packs in balance, but if you have a fast charger or cells are a bit far out of balance, the BMS may not help at all and may even hurt, keeping the full cells at 4.25v for prolonged periods. This is why I like the BMS’s where you can see what the hell’s going on.

@Wraith the in-line watt meters are great to have regardless. The one I posted is nice because it’s huge (stays in place), button to reset watt-hours is not recessed, and has many digits precision. I have another one with a color display, that’s nice because it doesn’t remember accumulated watt-hours, so I don’t have to reset it every charge. Oh and just to make sure, ammeters go in series (even your multimeter one). Most cheaper clamp multimeters can only measure AC current with the clamp.

1 Like

why someone has not simply made an actively balancing bms is fucking beyond me. discharge the highs into the lows and be done with it. yes its complex and full of logic gates but what isn’t these days.

11 Likes

I’ll see about grabbing the watt meter after I get the batch 3 paid out!

would my understanding be correct in the case of DBMS?

Dude. You’re telling me. Coming from the world of building massive LiPo packs that use canbus linked integrated smart balance chargers & a master BMS… Though those systems were not compact. It’s wouldn’t be terribly difficult to do though using an existing solution out there and a nice

5 Likes

Yeah you got it. the -48A is too much for straight 10s. The -42A covers 10s but not 12s. The adjustment range in the various HLG series doesn’t cover enough cases. You should buy one of each :slight_smile:

1 Like

Watch this space. You can buy balancing BMS at the moment, but they aren’t well suited to our application. I am working on a modular solution and make make a post in the coming months.

3 Likes

@JTAG can you make us a tiny balancing BMS/Charger board that can do 100a charge and discharge? I will give you my firstborn child… or like 200USD whichever you prefer.

4 Likes

Haha, tempting, maybe my gf will stop nagging about a child than :nerd_face:.

What dimensions would be acceptable? And what part of charging would you like it to have? Would 100W (USB-PD) charging be fast enough?

7 Likes