Bulk charging via digital DC-DC boost Minghe BST-900W

Overclocker

10 kW
Joined
Mar 17, 2014
Messages
794
Location
Philippines
LaGUrWs.jpg


tcJvumW.png


DjHJTLS.jpg


$20 at banggood

OEM seems to be Minghe BST900

anyone tried this? it seems like you could simply punch in your volts and amps and you're off bulk charging your high voltage pack from a 24v power supply
 
I've used one on test only.

I wanted to use a 12V solar panel to trickle charge a 16S LiFePO4 pack - it works but you need a very low impedance supply. I had to put 50,000mfd capacitors across the panel to keep it stable.

It works better in the car for charging a folding bike, but again low source impedance is a must, so thick input cables / close to battery is important.

Dermot
 
That will work fine - it's pretty much what the output side of every charger is.

The problem with solar panels is that if you ask for even a bit more current than they can source, the voltage collapses. They're not stable without a solar charge controller designed to deal with that.
 
Biggest problem is no charge termination so you'd have to keep a careful eye on it. Li-ions hate trickle charging and it leads to plating of metallic lithium which negatively effects the safety of the cell. Maybe there is some kind of cheap circuit board that can cut the current off that you can combine it with?
 
redilast said:
Maybe there is some kind of cheap circuit board that can cut the current off that you can combine it with?
Like a BMS (should)? ;)
 
amberwolf said:
redilast said:
Maybe there is some kind of cheap circuit board that can cut the current off that you can combine it with?
Like a BMS (should)? ;)


li-ion charging is CC/CV WITH cutoff once the current dips below a certain level, otherwise the charging current would just keep tapering to all the way to zero which as redilast said leads to plating

this cutoff function is provided by the charger, not the BMS
 
Constant voltage charging is not trickle charging.

Trickle charging is a constant current charging technique for non lithium cells that charges at a fixed low current and allows the voltage to creep up, a constant current mode. This trickle charging is not suitable for lithium as the voltages raise to dangerous levels for lithium. Lithium chargers must control the final charge voltage precisely.

Plating is caused by voltage exceeding the proper termination voltage. As long as the voltage it correct (for the temperature) and re it doesn't hurt to let it sit for a reasonably short time as the current decays toward zero. It is essentially the same as leaving the battery sitting fully charged.

You can always make a current sensitive dropout relay (or use a timer) if you want, but I have done manual charge termination for years without problems. Just don't leave it for days, as the termination voltage causes slow degradation of the electrolyte (whether plugged into a constant voltage or that voltage is supplied by the cells themselves). It is best to delay, charge just in time and then immediately discharge, leaving the pack at a 50-60% storage level most of the time.
 
Alan B said:
Plating is caused by voltage exceeding the proper termination voltage. As long as the voltage it correct (for the temperature) and re it doesn't hurt to let it sit for a short time as the current decays toward zero.

You can always make a current sensitive dropout relay if you want, but I have done manual charge termination for years without problems. Just don't leave it for days, as the termination voltage causes slow degradation of the electrolyte. It is best to delay, charge just in time and then immediately discharge, leaving the pack at 50-60% most of the time.


yeah i guess i'll just keep an eye on it. it's not like i'll be leaving it charging while i go on holidays :mrgreen:

but i think the plating happens even at the "proper" voltage if you do a very low trickle charge. actually the 4.20v is an arbitrary number if you think about it (i.e. why not 4.189572v?). it's just a nice round number for industry standardization
 
Overclocker said:
this cutoff function is provided by the charger, not the BMS
Depends on the BMS. If the BMS is designed to cut off charge input once cells have all reached their "full" level (and perhaps some in balancing mode), then the BMS itself can provide this function.

At least some of them definitely do this (otherwise they have no reason to have a charge-input-side FET); probably many of them do.

IF the BMS is "smart" then it can keep the pack cut off from charge input after this point, and not re-enable charging until the charger has been disconnected and reconnected if the cells drop just a little bit below that "full" point.

I don't know if any of them do that or not, but they could.
 
BMS Cutoff voltages are generally outside of the range you want to work with, they are a "safety net" rather than an optimum value. For example they may not cut off till over 4.3 and under 3.0, the charger generally sets the real cutoff value for normal charging rather than the BMS.

I'm not a battery chemist (though I know some), but plating is generally a very voltage sensitive process. Stay below the voltage threshold and it won't occur.

The typical behavior for a lot of chargers is to continue to output the specified max voltage after declaring "charge complete". During this time the BMS does balancing. Charge complete is declared when the current drops below the threshold, but the charger continues to output.

This is not "trickle charging" because the voltage is constant, in trickle charging the voltage is allowed to creep up while a small current is output, generally for lead-acid chemistries. The voltage increase is the problem for lithium cycles.
 
Alan B said:
BMS Cutoff voltages are generally outside of the range you want to work with, they are a "safety net" rather than an optimum value. For example they may not cut off till over 4.3 and under 3.0,
Then perhaps that ought to change. :)

THen even the dumbest of chargers would be safe to use with a pack like that, as long as it still has the right CC/CV setup for those cells, and has at least sufficient final voltage.
 
Let's not forget that many 18650 cells are rated for 4.35V...

Charge termination would stop the BMS from performing its balance function, which could take several hours.
 
But if the BMS controlled that cutoff, it could reengage it if necessary. Some do this (those with a charge-input-port FET can; how htey do it depends on the BMS).
 
I'm using one of those & it works very well. However I have a lot of old SLAs which are solar charged, so am effectively charging the bike from lead acid batteries not direct from the solar array (the solar array just tops up the SLAs).
What I've learned about controlling such a boost converter:
1) RH button switches output on, LH button switches it off
2) when 'on' RH button toggles between Volts and Amps display
3) up/down buttons alter setpoint for volts or amps (whichever is being displayed)
4) "save" an altered setpoint by a long press on LH button - display should flash once
With large boost ratios (e.g. I'm charging a 60V battery from a 12V supply) the output current available reduces below the setting - I seem to get just 2A charging current.
And yes I keep an eye on things when it's charging. When the arduino based charger controller is finished (soon I hope) this will terminate the charge by opening a series relay when any cell voltage gets up to 3.5V.
 
amberwolf said:
redilast said:
Maybe there is some kind of cheap circuit board that can cut the current off that you can combine it with?
Like a BMS (should)? ;)

I’ve played around with dc-dc’s charging from 12V to over 60V. IIRC 8-10A on the input side barely reached 1.5-2A on the step up side? But it does work.

Adjusting voltages conservatively there’s no trickle charging going on. Once it hit’s the set voltage current stops flowing and doesn’t start up again unless a possibly bad cell leaks off.

Forgive a shameless plug but I started a recent thread about cheap and effective 4S BMS circuits which might be of interest to this topic of bulk charging? - https://endless-sphere.com/forums/viewtopic.php?f=14&t=82262

I’ve been installing these in series up to 16S on RC Lipo using only stock balance wiring. No real protection on discharge other than the controller LVC and maybe a voltage meter or pilot lamp on the charge plug. But charge protection is by far more important than discharge protection, IMO.

Nothing to gain on my end but if interested in better protected bulk charging schemes, read more in that thread.

I’ve bought and now using dozens of these BMS and when charging less than 4-5A, they’re working quite well. I don't mind pushing that much current through Turnigy silicone balance wires but I'd keep it under 3A if using thin PVC (Zippy) packs or similar.
 
bobc said:
I'm using one of those & it works very well. However I have a lot of old SLAs which are solar charged, so am effectively charging the bike from lead acid batteries not direct from the solar array (the solar array just tops up the SLAs).
What I've learned about controlling such a boost converter:
1) RH button switches output on, LH button switches it off
2) when 'on' RH button toggles between Volts and Amps display
3) up/down buttons alter setpoint for volts or amps (whichever is being displayed)
4) "save" an altered setpoint by a long press on LH button - display should flash once
With large boost ratios (e.g. I'm charging a 60V battery from a 12V supply) the output current available reduces below the setting - I seem to get just 2A charging current.
And yes I keep an eye on things when it's charging. When the arduino based charger controller is finished (soon I hope) this will terminate the charge by opening a series relay when any cell voltage gets up to 3.5V.


thanks for the tips! you have this 900w version? i'm planning to start off with a 24v powersupply. the input limit seems to be 15A, so 360w on the input. at 85% efficiency then i should be getting ~300w to the 60v battery, or 5A. sounds a bit low... and probably even lower in actual use

the good news is it just landed in my country! i should be getting it in a few days...
 
Ybjpz7Z.jpg


wIZs2em.png


well it arrived ridiculously fast from china! it worked fine at the default 20v / 5a. when i upped the voltage to around 60v it stopped working and fried my PSU :evil:

upon closer inspection it uses a schottky diode that's only rated 45v! this is a module that's supposed to go up to 120v! i guess that's why it failed at the higher voltages, but then again i'm no switch-mode expert

could anyone who has this unit tell me what part number you have on yours? thank you.


l2lRzvV.gif


here's a simplified diagram of a boost converter
 
I'll check. Mine has been used for days on end at up to 62V. I'm pretty sure I ran it at 80V to check out a lebowski PSU without releasing smoke
I'll get back in a couple of hours...
 
Use one too, put bobc on to them, we both charge lifepo4 from them

I have tried with 12v battery as source
3x 12v server supplies in series = 36v
and a 54v esp120 server supply

great to have as they can charge any voltage higher than the input. Cheap mans satiator.
 
whereswally606 said:
Use one too, put bobc on to them, we both charge lifepo4 from them

I have tried with 12v battery as source
3x 12v server supplies in series = 36v
and a 54v esp120 server supply

great to have as they can charge any voltage higher than the input. Cheap mans satiator.


zQ1lPE5.jpg


do you have it within arm's reach? would you mind taking a peep at the part number? i believe the factory ran out of the correct parts and installed whatever crap they had lying around into my unit :evil:
 
OK mine says STPS30L80CT - looks like an 80V part. I guess I'm lucky I didn't take it over that then!!!
But I might need to one day, I'd better get hold of some 100V shottky's before I do....

It's not assisting china's reputation for quality is it....

That must be the MOSFET on the left - any chance you can read the number on yours? Mine's screwed to a wall & a real pig to read..... (as a low voltage part here would have a similarly amusing effect)
 
Back
Top