Running all four controllers in series would be the most efficient IF YOU CAN MAKE IT WORK!! Getting active loads like power converters to operate in unison would be a serious challenge indeed. BUT it would be worth a few hours of analysis to investigate.
And I am relieved that "Pedant" who evidently does design switchers, has verified that it is not simple.

To make this work you would need access to the charger control loop so as to make the 'sharing' a part of the control system of each charger. This is a notoriously difficult thing to get right and given that the individual chargers operating conditions are a function of the state of the battery connected to each charger there does not seem much likelihood of it being possible to get them all to balance all the time. Like I said, unless I am missing something, this is not a good idea. 4 individual converters, one for each charger is a way better idea and means far more manageable power levels for each charger. Don't be thinking it will be that much easier though. It will still be a significant challenge.

 

The batteries are protected by BMS.

As I understand the BMS on most lithium batteries they are concerned with making minor voltage corrections to each cell to ensure equal charge goes into each cell. A BMS can also protect from overloads I think, but have you checked that the BMS in question will disconnect the battery to protect from a short circuit or overload?
I should also point out that 48V with battery backing is kinda at the tipping point for being able to sustain an arc. So be sure to check whatever protections you use (fuses for example) that they can break the expected fault current and will operate with the voltages you have. Neither specification should ever actually be assumed. A sand filled ceramic body fuse will likely be the go but check, check and recheck. It is a critical component after all.

 

This is an interesting thread
Your commitment is amazing.
You implied you learnt to program in a similar way.
Not enough people learn now days, just wanting to get the code on line.
One big thing though @Deepfried , is the difference in risk,
You get a bit of code wrong, the first that can happen is you crash the PC.
These sort of currents and batteries are just dammed dangerous.
Short the wrong bit out , and the spark can blind you,
The heat can cause pcbs to catch fire ,
The batteries can explode , yes even those with "protection" ,
I think I can say that we are all worried for your and others safety,
Switch mode supplies are basically big radio transmitters. That sort of amps, could well block most all radio and phones in a large area,
That's why we keep pointing at off the shelf ideas, and ideas to let you build up to what you want.
For instance , you mentioned about parallel power supplies , you can't just join the outputs of small switchers to make a big one .
You also talked about pcb for the 150A , have you looked at what size pcb trace you would need for that ?
There are many things in this that can injure you or others, and it worries us.

Running 150A in pcb traces is no easy thing. Even if you went to the extreme of 13oz Cu it is just impractical most of the time. What I would recommend is run a trace as a guide but bolt a copper bar over the trace and connected at each end to whatever the trace runs to so that the bar carries the current, not the trace.

 

As I understand the BMS on most lithium batteries they are concerned with making minor voltage corrections to each cell to ensure equal charge goes into each cell. A BMS can also protect from overloads I think, but have you checked that the BMS in question will disconnect the battery to protect from a short circuit or overload?

Most BMSs will provide several functions, including the basic protective ones:

Charging
Balancing—as you have mentioned
OVP (over voltage protection)—to avoid over-charging
Over Discharge Protection—to avoid attempting to charge a cell that is in danger of having internal shorts.

Discharging
OCP (over current protection)—shuts off output from the pack if a limit is exceeded
Over Discharge Protection—shuts off output from the pack is a minimum voltage is detected.

 

Most BMSs will provide several functions, including the basic protective ones:

Charging
Balancing—as you have mentioned
OVP (over voltage protection)—to avoid over-charging
Over Discharge Protection—to avoid attempting to charge a cell that is in danger of having internal shorts.

Discharging
OCP (over current protection)—shuts off output from the pack if a limit is exceeded
Over Discharge Protection—shuts off output from the pack is a minimum voltage is detected.

Thanks for that. I was not sure because some of the lithium battery bms I have looked at did not look to have a heavy current connection, only light balancing connections. The bms I am used to don't have the high current switches either. They will raise alarms etc but were typically telephone exchange power plant at 48V and up to 7,500A. The bms could call for lower rectifier voltage but nothing can disconnect the batteries or rectifiers en mass from the exchange other than a fuse in case of catastrophic failure on the distribution bus. These batteries had fault currents up around 100kA and huge ceramic fuses behind blast proof shields.
Not the same thing and not the same application, although lithium batteries can be quite... incendiary.

 

These batteries had fault currents up around 100kA and huge ceramic fuses behind blast proof shields.

Yes and pulling those fuses while disabling the alarms for maintenance is one of those "lessons learned" things for AT&T from the 1970s...

But small scale BMSs are all in one affairs. Since LiPo pouch cells almost always come with a "protection board" that uses the ubiquitous DW-01 chip. Those boards only do the safety functions, no charging management—but that do have a MOSFET to shut down the cell, and they can be used in conjunction with a constant current supply to safely charge the cell since they will cut off at 4.2V. "Protected" cylindrical Li-Ion cells have the same sort of board in a different form factor.

 

IF YOU CAN MAKE IT WORK!!

If frogs had wings…

You cannot put active circuits in series! What happens when one battery has fully charged before the others? How is that going to work?

 

If frogs had wings…

You cannot put active circuits in series! What happens when one battery has fully charged before the others? How is that going to work?

I added that entirely accurate qualifier because it was, and is, clear to me that just because I see no way to make something work as desired does not mean that another individual with more experience and skill in that area may be able to produce a satisfactory scheme. I had thought that my qualifier:
"Getting active loads like power converters to operate in unison would be a serious challenge indeed." was adequate to describe the complexity of the challenge.
Just because I see no way to do it does not mean that I believe that somebody else may be able to do it. I did not go into the many details because certainly there are a whole lot of individual things that would need to be addressed to make such a scheme work.

 

If frogs had wings…

You cannot put active circuits in series! What happens when one battery has fully charged before the others? How is that going to work?

The batteries get discharged together in parallel. Their charges shouldn’t ever be too far off.

But if need can sense voltage / current and once one is charged stop them all.

From what I’ve gathered so far, my first attempt will be to run 4 separate loads in series.

 

The batteries get discharged together in parallel. Their charges shouldn’t ever be too far off.

That's where your first surprise will be.

 

I NEVER said it would be simple. AND I never said it could not be done.
AND I have seen applications with big batteries in parallel-series. And one battery of the parallel pair explodes, under a few hundred amps load. the load is an array of 1000 watt spotlights, usually 32 of them

 

I’ve been working on my plan of execution and I encountered my first issue lol… if I were to do the loads in series wouldn’t it require input of 48v at 150A to give each load the 12v 150A power they want?

If that’s the case then I will certainly step back from that… 48v at 150A isn’t something I’m willing to toy with.

Unless I’m mistaken a 48v to 12v SMPS is my only viable option.

 

I’ve been working on my plan of execution and I encountered my first issue lol… if I were to do the loads in series wouldn’t it require input of 48v at 150A to give each load the 12v 150A power they want?

If that’s the case then I will certainly step back from that… 48v at 150A isn’t something I’m willing to toy with.

Unless I’m mistaken a 48v to 12v SMPS is my only viable option.

If you need 150A for each of four 12V batteries, that is 150 x 12 x 4 = 7200W.

To get 7200W from a 48V battery that is 7200/48 = 150A.

Doesn’t matter whether they are series or parallel.

Thank you for confirming everything we have been saying in this thread.

 

OK, one thing is that if they are in series, the CURRENT will not be multiplied by four, only the voltage and the total power will be multiplied. THAT was my only reason for suggesting a series scheme.
If I were needing to charge four similar 12 volt battery packs from a 48 volt source, it would start with a very reliable current regulator used to avoid destruction. Then each battery would have a voltage monitor that could bypass the battery to keep from excess charging. AND of course the current regulator will need to be able to control the current no matter what the voltage output is. That might require reducing the current, which is probable because one battery will always finish first.
so none of the blocks have an impossible function.

BUT THEN, if those same batteries need to power a single load in parallel, it does get rather tedious and complex.

 

If you need 150A for each of four 12V batteries, that is 150 x 12 x 4 = 7200W.

To get 7200W from a 48V battery that is 7200/48 = 150A.

Doesn’t matter whether they are series or parallel.

Thank you for confirming everything we have been saying in this thread.

Not once was it mentioned I would need 48v at 150A in the thread. But yeah I do agree 48v at 150A is nothing to meddle with on a single board.

Back to my original plan of a 48v to 12v board.. That way the 48v side of things will be <=40A. Could anyone suggest the latest and greatest MOSFETs they’d recommend for this project?

I’m going to begin with getting something working then make it as efficient as I can with the little I know.

 

Not once was it mentioned I would need 48v at 150A in the thread.

Yes, it was, and by you:

if I were to do the loads in series wouldn’t it require input of 48v at 150A to give each load the 12v 150A power they want?

Four chargers... it all comes back to power:

If you need 150A for each of four 12V batteries, that is 150 x 12 x 4 = 7200W.
To get 7200W from a 48V battery that is 7200/48 = 150A.

 

Yes, it was, and by you:



Four chargers... it all comes back to power:

Yeah I meant prior to me mentioning it.. But yeah feel free to suggest components I should use.

 

OK, now for the new question: Do you REALLY want to charge those 12 volt batteries at 150Amps??? I don't recall the size of the packs in amp-hours, but it seems that is going to put some stress on the batteries and reduce the useful life a bit.
I suggested a different option in my post #74, which is an alternative approach, not even requiring any switcher supplies to change the voltage, but rather a sort-of-smart charge current regulator.
Probably it also will need to be a switcher, but only one of them would be needed.

 

A compromise approach is multiple, off-the-shelf, high-density DC/DC converters. Fully-isolated or not, it reduces the project to one of interconnects, thermals, and control, which are hard enough at these power levels. My personal bests are non-isolated 6 kW at 360 Vdc out and fully-isolated 4 kW at 5.1 Vdc out. Note that these were 20-25 years ago; the range of component options is way better now.

ak

 

I think there is confusion among us over what the TS means by putting the loads in series.

Here is his original idea:

The loads are 12v charge controllers. All 4 will have identical constant power draw while switched on.

He is not talking about putting batteries in series, he is talking about putting battery chargers is series.

And has the false impression that all of them will draw the same current always. This cannot work.

He is also has the false impression that you can charge a 12V battery with 12V.