Power supply current limiting question

[fubgumfaw]

Hi,
This is an EV project I've had in the back of my mind for a while now. I want to build a single PSU based universal charger that works in set voltage increments so I can charge most any EV from my monster charger.

1. I have 8 Meanwell SE-600 PSU's with their outputs all in series. The are rated for 25 amps with fully isolated outputs.
2. The first 2 PSU's are set to 24.6v each for 49.2v (12S lion) total. This is my starting output voltage.
3. The 3rd through 8th PSU's are set to 16.4v(4S LION voltage increments) each.
4. Since they are all in series, I add the 16.4v increments to the base 49.2v to get the voltage increment that the specific EV needs.
5. With 8 PSU's in series, I get voltage increments of 49.2v(12S), 66v(16S), 82v(20S), 98.4v(24S), 114.8v(28S), 131.2v(32S) and 147.6v(36S).
6. Battery packs vary widely. Not all have the same internal resistance or capacities. Their discharged and charge voltages depend on the series cell counts.
7. In the past, I have used 3 of these PSUs to make a 66v charger and 4 PSU's to make an 82v charger.
8. I added resistors in parallel to "sneak up to" the 25 amp limit. Not very scientific...shrug!
9. I want to limit charge current to 25 amps regardless of the voltage output I use.
10. The individual PSU output voltages must stay at their set voltages. Without this I loose my 4S increments.
11.  I want to use a single resistor bank for all 7 output voltages and add or subtract resistors to it for the output voltage selected.
12. I have an 8 position rotary switch that activates 3PDT relays to select the output voltage I want. The relays apply 110v AC to only the PSU's needed for the selected output voltage. The 3PDT relays have a free pole on them.

Given: At 82v I used 7, 3 ohm (.43 ohms total) resistors in parallel to limit current to about 25 amps. When charge current drops low enough a switch across the resistor bank bypasses it.

I want to add or subtract parallel resistors to my total resistor bank dependent on the output voltage so I always maintain that 25 amp maximum regardless of voltage selected. How do I calculate the resistance needed from this one data point for the other 6 voltages?

[fubgumfaw]

I googled around some, but I didn't find a circuit that used mosfets that could do adjustable current limiting. I'm sure it exists...

[pqass]

I'm not sure I should be entertaining this because I think we have an X-Y problem https://en.wikipedia.org/wiki/XY_problem here.

As I understand your requirement, you wish to have a power supply to selectively produce an output beginning with 49.2V then in 16.4V (ie. 4S) increments, up to 147.6V.  And you wish to attach, in series, resistor(s) to limit the current through the circuit to at most 25A.

Please don't use this approach to charge Li-ion batteries as charging these batteries is more complicated than other types and is unforgiving.   See: https://batteryuniversity.com/learn/article/charging_lithium_ion_batteries.

Besides, at an output of 147.6V @ 25A, you'll need to supply >3.6KW which is more than a single 120V mains circuit will allow.
Also, there is a limit to how high the floating supply can be above ground (earth). Check the supply spec.
And then there's the unpowered supplies whose output is still in-series with the rest of the circuit. Pretty sure that won't work.
And, of course, the $$$ for 7 supplies!

However, to answer your specific question about a constant current limiter, you'll need a much beefier version of these:
see Figure 1 here: https://www.electronicdesign.com/power-management/article/21801456/current-limiter-offers-circuit-protection-with-low-voltage-drop   
or:  https://electronics.stackexchange.com/questions/202326/low-overhead-constant-current-led-driver

[fubgumfaw]

I'm not sure I should be entertaining this because I think we have an X-Y problem https://en.wikipedia.org/wiki/XY_problem here.

As I understand your requirement, you wish to have a power supply to selectively produce an output beginning with 49.2V then in 16.4V (ie. 4S) increments, up to 147.6V.  And you wish to attach, in series, resistor(s) to limit the current through the circuit to at most 25A.

Please don't use this approach to charge Li-ion batteries as charging these batteries is more complicated than other types and is unforgiving.   See: https://batteryuniversity.com/learn/article/charging_lithium_ion_batteries.

Besides, at an output of 147.6V @ 25A, you'll need to supply >3.6KW which is more than a single 120V mains circuit will allow.
Also, there is a limit to how high the floating supply can be above ground (earth). Check the supply spec.
And then there's the unpowered supplies whose output is still in-series with the rest of the circuit. Pretty sure that won't work.
And, of course, the $$$ for 7 supplies!

However, to answer your specific question about a constant current limiter, you'll need a much beefier version of these:
see Figure 1 here: https://www.electronicdesign.com/power-management/article/21801456/current-limiter-offers-circuit-protection-with-low-voltage-drop   
or:  https://electronics.stackexchange.com/questions/202326/low-overhead-constant-current-led-driver

The PSU's are already in use in this format. This is done and working. The rotary switch selects the right relay to get the correct PSU output and to power the correct number of PSU's for that output. I want a more compact/better way to limit current regardless of output voltage to 25 amps.

The resistors are in parallel. Some will be in parallel all the time and more will be added or taken away as needed to raise or lower the total overall resistance needed to limit current to 25 amps. What I would have done without any answer here is simply start at 48v and charge a depleted battery pack capable of charging at more than 25 amps. I'd see if I exceed 25 amps or not and add or take away resistors until I got to a little less than the 25 amps. This is exactly what I did a few years ago to get it working at 82v. I'd then switch to the 66v setting and connect a 16S pack and do it again. 20S or 82v is done so next would be 24S or 98.6v...and so on. This is a pretty lame way to do it IMHO. I"m pretty sure there is some variant on ohms law or whatever that can calculate this for me. With that resistance at each output voltage caclulated, repeating for mosfets is pretty easy. That's just a zener and a resistor at the gate to make a voltage divider.

Assume that I know what I'm doing when it comes to charging. I've been using SE-600 Meanwell PSU's to charge for 5 years now. This is pretty well known territory for me. There are several things working to ensure this is safe. I'm not worried about the safety of using PSU's for charging. It works really well! I'm far from alone is using PSU's for a charger.
1. I use smart BMS that are programmed and tested to work as I decide. In an over charge condition, it will shut off current into the pack.
2. The correct output voltage can't overcharge the pack since it is the same voltage as the pack at full charge.
3. The charger has a smart watt meter on it. IF current exceeds 28 amps, it shuts off a relay and disconnects power to everything.
4. The PSU's have their own over load capability and shut down at 30 amps.

On the subject of lithium battery danger...
I've tried my damdest to make lithium cells catch fire. Full charge, drive a nail through a LIPO...smoke, a little sizzling, no fire. Connect a 3S pack to a much larger 4S pack and let it sit there until the 3S cells burst. A little smoke, a POP when the pouches burst open and no fire. I've dead shorted LION cells and LIPO cells and they did get hot and the LIPO cells puffed up. The LION cells leaked a little. I put a half used up 12S/49.2v pack on a 66v charger with no BMS so it had to over charge. The pack got pretty warm, cells puffed up, no fire. I took a fully charged LIPO cell and cut off the bottom of the aluminum pouch so lots of air could get in...no fire, just got warmer. I then cut off the rest of the aluminum pouch and unraveled all the layers inside. Still no fire. I smashed a charged 18650 with a hammer. It leaked some fluid, got warm, no fire. I've been spot welding my own LION packs for several years. I've built a few for others too. In testing out my KWeld spot welder, I experimented to see how much current would blow through the steel cell wall. A little leakage, no fire. I did try...a LOT! Maybe I"m lucky? Lithium fires do happen, but I sure can't make one happen! Anyway, lets move on about the safety factor. I'm not worried.

This charger takes some amps to run. I have done this before. A typical 15 amp circuit pops its breaker with 4 PSU's in parallel and the outputs in series at 82v and making 25 amps. 3 PSU's for 66v out is within the 15 amp limit.  I ran a 40 amp, 110v circuit with a twist lock outlet for its use. At 147v and 25 amps out, I need 30 amps AC. IF I max out the PSU's and make them shut down from overload or 30 amps that's still 36 amps AC.

I already have the PSU's...the money was spent over about 5 years time and it's 8 PSU's not 7 of them. The AC inputs are all in parallel and only if those specific PSU's are powered. The relays disconnect AC to any that are not in use. I disconnect the hot wire to the PSUs. This is all working already. BTW, the minimum isolation voltage for the SE-600 PSU is 3000 volts. I think I'm going to be OK with them in series.

I think you can agree that current follows the path of least resistance. That means into the battery pack. The voltage below an unpowered PSU goes no where into the PSU. You probably know a good bit about switching PSU's. The output of the high frequency transformer goes into a rectifier, inductor and large filter cap at a minimum. That rectifier stops any current flow.

Thanks for the info. I'll do some reading.

[Medved]

I somehow was not able to grasp from your explanation, what is supposed to limit the current.
Is it the current limitation feature within the PSU's? Then you will have big trouble with the tolerance of the current limit and with some of the series connected PSU's getting reversed (their design may tolerate the short circuit, but voltage reversal is another level).

Or you expect a series resistor would do it? Then how do you make the end-of-charge voltage control so it is able to charge the cells as fast as possible even when their voltage approaches the EOC level, yet stay sharp at or below the EOC level?

To me the task you want to accomplish asks more to some ~160V preregulator and then a final stage (again of a SMPS type, likely some buck regulator) with the regulated output voltage and current limit programmable (by your rotary switches,...).


Plus another problem: These PSUs use to be designed as Class-I devices, so requiring grounding for protection Normally (unless explicitely stated otherwise in their manual), they withstand very limited voltage between their output vs GND, usually no more than their rated output voltage, so so their design anticipate to be used by either of their output terminal grounded, but not anything above that. The "3000V isolation" usually means the working input conductors (L and N) vs ground (PE), not the output vs anything else. Again, unless explicitly stated otherwise in their manual (with clear drawing showing what the voltage limit means).
Connecting many of them in series you can easily exceed that limit and blow quite some internal components (there use to be EMC blocking capacitors between the output and PE, usually 63V rated for output rating 24V and below). It may not fail immediately, but over time. And because these use to be ceramic, they fail into turning white hot burning smoking bits of ceramic.

[fubgumfaw]

I somehow was not able to grasp from your explanation, what is supposed to limit the current.
Is it the current limitation feature within the PSU's? Then you will have big trouble with the tolerance of the current limit and with some of the series connected PSU's getting reversed (their design may tolerate the short circuit, but voltage reversal is another level).

Or you expect a series resistor would do it? Then how do you make the end-of-charge voltage control so it is able to charge the cells as fast as possible even when their voltage approaches the EOC level, yet stay sharp at or below the EOC level?

To me the task you want to accomplish asks more to some ~160V preregulator and then a final stage (again of a SMPS type, likely some buck regulator) with the regulated output voltage and current limit programmable (by your rotary switches,...).


Plus another problem: These PSUs use to be designed as Class-I devices, so requiring grounding for protection Normally (unless explicitely stated otherwise in their manual), they withstand very limited voltage between their output vs GND, usually no more than their rated output voltage, so so their design anticipate to be used by either of their output terminal grounded, but not anything above that. The "3000V isolation" usually means the working input conductors (L and N) vs ground (PE), not the output vs anything else. Again, unless explicitly stated otherwise in their manual (with clear drawing showing what the voltage limit means).
Connecting many of them in series you can easily exceed that limit and blow quite some internal components (there use to be EMC blocking capacitors between the output and PE, usually 63V rated for output rating 24V and below). It may not fail immediately, but over time. And because these use to be ceramic, they fail into turning white hot burning smoking bits of ceramic.

The SE-600 PSU's can handle 3000v differential between ground and the output. I've been using them with their outputs in series for a very long time now. This isn't an issue. My 82v charger has been used for 5 years and the 66v version for 3 years. IF those little components were going to become molten ceramic, they would have done so by now. There is zero evidence that any components in the PSU's are even getting warm. Anything not current limiting related is extraneous and already done and working long term. No smoke, no fire, no smoldering components...lol.

I have been using 100w resistors to limit current on the 66v and 82v versions. I've since taken the 66v version apart and added those 3 PSU's plus one more to the under side of the 82v charger to make the 8 PSU version. The original 82v charger is still 100% working as I built it 5 years ago. All I did was add another 4 PSU's underneath it.  I can reuse the same resistors for the 8 PSU version. I suspect I'll have enough, but maybe not. It seems more appropriate to just use 2 or 3 mosfets on a heat sink to do the same thing as a big bank of resistors. Either way I'll make a good bit of heat. This can't be helped. The difference is the resistors are fixed values and the mosfets are adjustable. In either case in series with the output from the PSUs will be some amount of resistance to "slow down" current flow to 25 amps so I never exceed the PSU current limits.

The charger has a smart watt meter on it. It is programmable based on various triggers. It has 2 logic level outputs that can be used to control a relay or a mosfet. I can use one of them and the current trigger function in the watt meter. This signal with a relay can bypass the resistors at some low current threshold. It could also turn on the current limiting mosfets to full. This is a threshold based function and current draw will drop off considerably before the watt meter triggers the output. The watt meter control signal is not my preferred method of current control. It is more automatic than flipping a switch when current draw drops below some threshold level so there is that.

This schematic looks promising and adapting it for 36v to 150v is probably doable. It has feedback from the shunt and differential op-amp to adjust the mosfet gate so it stays at a current limit. As charge progresses, it should be able to slowly turn on the mosfets a little more and a little more to account for nearing close to full charge and the lower current draw this creates. This is the better way to do it and not use the watt meter threshold functions at all for this. It's infinitely adjustable and always tracking at the best possible current draw level.

https://www.electronicdesign.com/power-management/article/21801456/current-limiter-offers-circuit-protection-with-low-voltage-drop

All that I really need is to adapt some schematic to be able to work at a wide voltage range (36v to 150v) and to always keep current draw from exceeding 25 amps.

Probably the next question is why 36v to 150v? 48v(12S) discharged is 3v per cell or 12 cells x 3v = 36v. This is the lowest voltage the charger output will ever see. 36S at full charge is 147.6v so that's the upper limit. At 3v per cell and 36S that's 108v. This is the highest voltage the current limiter will ever have across it. 150v is just playing it safe.

[fubgumfaw]

Previous PSU based chargers I have built and still use.
https://ibb.co/album/D9NJrf

If you know how to make the images show up in the post that would be great. I already tried the typical "bracket img bracket" stuff that usually works in forums to no avail. Maybe the free imgbb account I'm using doesn't allow this?

1. I've built and sold to friends probably 10 Lenovo based chargers. I use 4 Lenovo 170 watt laptop PSU's with their outputs in series. It makes for a nice 9 amp, 82v charger that is super reliable, silent and cheap. Not one of them has ever failed to work. No smoke, no fire, no burnt components. I started with the smaller 90 watt Lenovo PSU's and made 2 of them for 82v. I later discovered the larger 170 watt PSU and sold off the 90 watt chargers. Recently I found a bunch of HP 230 watt laptop PSUs for super cheap so I'll be making a 12 amp version from them too. The HP PSU's are easy to take apart and get to the board so I can adjust the outputs as needed. The Lenovo PSU's are a lot of work to take apart and so they are useful in 20v increments only.  I use a 170w Lenovo PSU based 82v charger pretty regularly.

2. In that same album is the 66v charger based on 3 SE-600 Meanwell PSU's. I later replaced the 10w resistors with a lot fewer 100w resistors. The little rocker switch bypasses the resistor bank.

3. My 82v charger is based on 4 SE-600 PSU's in series. You can see 7 100w, 3 ohm resistors in parallel along the right side to limit current. The rocker switch bypasses them once current draw is low enough.

4. The new 8 PSU Super Charger is in there too from before I got it wired together. the old 82v charger has 4 more PSU's stuck underneath it

5. There's a schematic I put together to conceptually work through 7 PSU's in series.

6. The other schematic is something I drew up, but never actually built that I thought could work for current limiting. Since it is all low side control there is no need for opto isolation between the op-amp and the mosfet gate. I should add a zener at the gate. Gain is probably way too high and I wouldn't use an SUP8510. The ACS current sensor would provide the feedback loop to the op-amp. What do you think?

[fubgumfaw]

Thinking about this more...
I need high side current limiting. My circuit does low side which is great for not needing anything special for driving the mosfet, but this also artificially raises the voltage of the pack as it is now part of a voltage divider with a large portion of "division" under the pack. I'm not sure what the BMS will do with this. Anyway, 0v referenced, the BMS won't have issues with this. Hence using high side current limiting...

I'm also somewhat skeptical that a single current limiting circuit can handle the 36v to 150v range I need. Maybe if I have an array of resistors I can switch in and out to accommodate the changes the charge voltage make? So far this is the best thing I have seen. It is high side and gets feedback from a shunt. I think I might try an ACS current sensor instead as that will give me electrical isolation for the op-amp. It was designed around 12v and I need it to operate at 150v.

https://www.electronicdesign.com/power-management/article/21801456/current-limiter-offers-circuit-protection-with-low-voltage-drop


Looking on ebay, I found this nice little AC&768 based board. It includes an op-amp so that you can set an adjustable limit and trigger the op-amp output. No idea if it is a logic level or a voltage range...
https://www.ebay.com/itm/ACS758LCB-050B-ACS758-AC-DC-Detection-Over-Current-Protection-Module-TOP/183835493209

[fubgumfaw]

I bought a couple of these current sensor boards to try out.

https://www.ebay.com/itm/NEW-ACS758LCB-050B-AC-DC-detection-over-current-protection-module-L2KD/184443760162

It's rare this happens, but they were cheaper on ebay than on aliexpress.

[ledtester]

fwiw, you might be interested in these aliexpress listings:


"Professional Custom 36S lithium battery charger automatic 151.2V 6A 36 cell li ion battery charger for lithium ion battery pack"

https://www.aliexpress.com/item/33005503550.html

"Custom professional 36S lithium battery pack charger automatic 151.2V 8A li ion battery charger for 36 cell lithium battery pack"

https://www.aliexpress.com/item/33008206820.html


It seems they can custom build a charger to your specs.

[IanB]

The resistors are in parallel. Some will be in parallel all the time and more will be added or taken away as needed to raise or lower the total overall resistance needed to limit current to 25 amps.

This is dumb. But since almost this entire thread consists of you saying you know what you are doing and you don't care what anyone else says to the contrary, I don't see any reason for anyone to respond to this thread. Please keep talking to yourself while we watch.

[fubgumfaw]

fwiw, you might be interested in these aliexpress listings:


"Professional Custom 36S lithium battery charger automatic 151.2V 6A 36 cell li ion battery charger for lithium ion battery pack"

https://www.aliexpress.com/item/33005503550.html

"Custom professional 36S lithium battery pack charger automatic 151.2V 8A li ion battery charger for 36 cell lithium battery pack"

https://www.aliexpress.com/item/33008206820.html


It seems they can custom build a charger to your specs.

I think I looked at those listings a while ago. To get 30 amps was pretty expensive and the chargers above are obviously 6 and 8 amps. I believe these are single output voltage since I see no external way to adjust them.

My charger, the cost has accumulated over time. I have been collecting the PSU's for a few years and so any significant cost has been incurred a little at a time. Also, by the time I came up with making this super charger, I already had 7 of the meanwell PSU's and was using them in 3 and 4 PSU's in series for 66v and 82v chargers. At this point, getting an 8th PSU was logical and only a little added new cost to create a 30 amp charger that can charge at 12S to 36S depending on which PSU output is selected.

For someone looking for a single fixed output voltage charger at low amperage, this might be a good choice for them. For someone who has all the PSU's like I do, not so much. 

[fubgumfaw]

I read what people post and I respond back in kind.
I have yet to see PSU's in series like I am using them be a problem. I am far from alone in using them like this. The output isolation is high enough that I won't exceed this limit. It's a good point and I appreciate that it was brought up. However, I'm not seeing it be a problem...

I do this with Lenovo 90w  (4.5a) and 170w (8.5a) PSU's as well and it too works really well with no fires or danger of fires. I've continuously had and used Lenovo PSU chargers since 2015. 3 makes a little over 60v and 4 makes a little more than 80v. All my EV's run on 16S or 20S and they make nice portable and silent chargers for them! I recently found 230w (11.5a) laptop PSU's from HP. I bought several to use in the same way I have been using Lenovo PSU's for the past 6 years. I think they will be OK too in this use case. The DC output is fully isolated from the AC input.

I have an idea...I have probably 20 of the 90w Lenovo PSU's. I had them given to me by an IT guy from an old job. I should stack like 10 of them in series and then once my DL24M arrives putting them under 3 amps load at 200v. This is just at the 600w limit of the load tester so it should be fine. AND that will be solid proof of what I already think is true...that nothing will explode or catch fire. Worst case, the circuit breaker pops open and I blow up some PSU's I don't care about.

[fubgumfaw]

The resistors are in parallel. Some will be in parallel all the time and more will be added or taken away as needed to raise or lower the total overall resistance needed to limit current to 25 amps.

This is dumb. But since almost this entire thread consists of you saying you know what you are doing and you don't care what anyone else says to the contrary, I don't see any reason for anyone to respond to this thread. Please keep talking to yourself while we watch.

I think you are focusing on the wrong thing. I think it's healthy to question "the PSU's in series" and the implications involved of doing this and that has happened. I've also responded back that I'm already doing this without issues. How this is me saying "I don't care what anyone else says" is a mystery to me! Quite the contrary! I responded with my results and that it works like I describe with SE-600 Meanwell PSU's. I've also done the same with Lenovo Laptop PSU's without issues. Where is the "I don't care what anyone else says" in there? I'm completely missing your perspective!

To point.... this thread is about current limiting and so far I have gotten a few people who actually talked about that while the many have questioned "PSU's in series". I posted that board I found with an ACS current sensor on it and how I hope to implement it for current monitoring/limiting. What I was hoping for is a ready made solution that can handle the voltage and current limits that I mentioned in my first post. I'd LOVE to have something like that so I don't have to create it from scratch! A mosfet in it's linear zone creates adjustable resistance and heat losses. I'll probably end up creating something myself from mosfets for current limiting. I did get a few posts for schematics that I'll try to use. Thank you to the people that contributed on topic suggestions about current limiting.

[Vovk_Z]

The resistors are in parallel. Some will be in parallel all the time and more will be added or taken away as needed to raise or lower the total overall resistance needed to limit current to 25 amps.

This is dumb. But since almost this entire thread consists of you saying you know what you are doing and you don't care what anyone else says to the contrary, I don't see any reason for anyone to respond to this thread. Please keep talking to yourself while we watch.

It isn't dumb. It is un-efficient.

[Vovk_Z]

Here is a schematics for S-350-24 (I can't find SE-600-24). I can't say exactly what to do, but I'll play with R33, R34, R48. And I would read how to do current limiting with LM494.

[fubgumfaw]

The resistors are in parallel. Some will be in parallel all the time and more will be added or taken away as needed to raise or lower the total overall resistance needed to limit current to 25 amps.

This is dumb. But since almost this entire thread consists of you saying you know what you are doing and you don't care what anyone else says to the contrary, I don't see any reason for anyone to respond to this thread. Please keep talking to yourself while we watch.

It isn't dumb. It is un-efficient.

LOL...
There's 2 ways to skin the cat and with my set up I have to choose the inefficient one.
1. Adjust the PSU voltages to track just slightly higher than the pack voltage so that 25 amp draw is maintained. This is how CC charging works.
2. PSU voltage is not adjusted and apply a resistance after the PSU to limit or impede current flow to 25 amps. This method produces waste heat.

[IanB]

LOL...
There's 2 ways to skin the cat and with my set up I have to choose the inefficient one.
1. Adjust the PSU voltages to track just slightly higher than the pack voltage so that 25 amp draw is maintained. This is how CC charging works.
2. PSU voltage is not adjusted and apply a resistance after the PSU to limit or impede current flow to 25 amps. This method produces waste heat.

In your posts above you implied you were putting the resistors in parallel with the power supplies:

8. I added resistors in parallel to "sneak up to" the 25 amp limit. Not very scientific...shrug!

The resistors are in parallel.

If in fact you are putting the resistors in series with the power supplies and not in parallel with the power supplies as you seemed to say, then it doesn't sound so crazy.

[tooki]

The resistors are in parallel. Some will be in parallel all the time and more will be added or taken away as needed to raise or lower the total overall resistance needed to limit current to 25 amps. […]

Dude, you suck at explaining things.

Your resistors are in parallel with each other, but the resistor bank is in series with the output. You didn’t bother mentioning this critical point at any time.

Assume that I know what I'm doing when it comes to charging. I've been using SE-600 Meanwell PSU's to charge for 5 years now. This is pretty well known territory for me...

Your questions, technical approach, and attitude suggest you don’t really know what you’re doing and don’t really appreciate all of the risks involved.

The fact that you’ve gotten away with a bad approach for 5 years doesn’t make it a good approach.

I think you can agree that current follows the path of least resistance.

No, because it does not.

Well, it does, in that the current always takes all available paths, which the path of least resistance is one of. The various resistances simply determine how the current is divided amongst all the paths.

But the statement “follows the path of least resistance” generally means “...to the exclusion of higher-resistance paths”, and as such is false.

[fubgumfaw]

LOL...
There's 2 ways to skin the cat and with my set up I have to choose the inefficient one.
1. Adjust the PSU voltages to track just slightly higher than the pack voltage so that 25 amp draw is maintained. This is how CC charging works.
2. PSU voltage is not adjusted and apply a resistance after the PSU to limit or impede current flow to 25 amps. This method produces waste heat.

In your posts above you implied you were putting the resistors in parallel with the power supplies:

8. I added resistors in parallel to "sneak up to" the 25 amp limit. Not very scientific...shrug!

The resistors are in parallel.

If in fact you are putting the resistors in series with the power supplies and not in parallel with the power supplies as you seemed to say, then it doesn't sound so crazy.

The resistors are in parallel with each other to create the resistance needed to limit current to about 25 amps or a bit more.
They are in series with the output of the PSU's. So that may be 2, 3, 4, 5, 6, 7 or 8 PSU's in series and in series with a bunch of parallel resistors. In any set of series PSU's the resistors are only in series with the total number of series PSU's.

When you take small snippets of something someone said you lose context. I'm pretty sure I already said all of this, but you'd never know it based on the snippets you have quoted.

[fubgumfaw]

The resistors are in parallel. Some will be in parallel all the time and more will be added or taken away as needed to raise or lower the total overall resistance needed to limit current to 25 amps. […]

Dude, you suck at explaining things.

Your resistors are in parallel with each other, but the resistor bank is in series with the output. You didn’t bother mentioning this critical point at any time.

Assume that I know what I'm doing when it comes to charging. I've been using SE-600 Meanwell PSU's to charge for 5 years now. This is pretty well known territory for me...

Your questions, technical approach, and attitude suggest you don’t really know what you’re doing and don’t really appreciate all of the risks involved.

The fact that you’ve gotten away with a bad approach for 5 years doesn’t make it a good approach.

I think you can agree that current follows the path of least resistance.

No, because it does not.

Well, it does, in that the current always takes all available paths, which the path of least resistance is one of. The various resistances simply determine how the current is divided amongst all the paths.

But the statement “follows the path of least resistance” generally means “...to the exclusion of higher-resistance paths”, and as such is false.

OK...thanks for the feedback...negative as it may be...
Opinions and scathing comments won't sway anyone. I'm not sure why people bother.
Please present actual facts that show why i could be doing something wrong instead of slamming me as a person which is pointless and offensive.
Saying "someone sucks" is offensive. Saying "I don't understand what you said" is not. Consider being less offensive please.

I do have a couple questions for you.
Have you tried what I have tried so that you know for a fact that it's safe or unsafe?
Are you basing what you said on any factual information or is that just your opinion?

If you do actually have real data to back up your view point, please share it and NOT the attitude!
If you know things like the isolation limits of the SE-600 PSU's being LESS than enough to handle what I'm doing with mine, then this is helpful information.

Ive never claimed to be an electronics expert and there is more I don't know than I do know. I'm more of an experimenter...try it out see what happens. In this case, what happens is that it works and has been since 2015. I think the proof of time is evidence enough to say it's going to keep on working a good while longer. So far I have yet to see evidence to the contrary. This isn't even my idea! I read about it on Endless-Sphere being done with 12v server PSU's. The requirement of isolated outputs was easy to determine so I tried on sevaral of those 90W Lenovo PSU's and it worked! I then bought 4 Meanwell SE-600 PSU's as I knew they too had isolated outputs. This too worked flawlessly and that charger has seen continuous use for many years now. More recently I got around to this building the super charger with 8 PSU's potentially stacked in series. So far so good...seemed a good idea to me so far and is doing what I expected it would do...work. Whether this is the intended use that Meanwell had for the PSU's, well I don't worry about that too much!

[ledtester]

You should bear in mind that electrical engineers are trained in the safe use of electricity. Moreover, they are taught to abide by a professional code of ethics. Two stanzas of the IEEE Code of Ethics stand out in particular:

1. to hold paramount the safety, health, and welfare of the public, to strive to comply with ethical design and sustainable development practices, to protect the privacy of others, and to disclose promptly factors that might endanger the public or the environment;

...

5. to seek, accept, and offer honest criticism of technical work, to acknowledge and correct errors, to be honest and realistic in stating claims or estimates based on available data, and to credit properly the contributions of others;

...

You are dealing with high voltages, high current, high power and a whole lot of stored potential energy. You are using the power supplies in a very unconventional configuration. It is one thing to expose yourself to the risks of this setup, but it's another thing to expose others to that risk - especially if they are less knowledgeable about electronics and how the system works. Just because you've never gotten struck by lightning while standing under a tree during a thunderstorm doesn't mean it's safe to do so.

At the very least I would get separate supplies for your different charging needs (voltages & amps). Safety features that real battery chargers have include automatic current limiting, cut-off for over current and over-voltage events, automatic charge termination, battery temperature monitoring. Note the emphasis on "automatic". If the process requires human intervention that in itself dramatically increases the chance of something going wrong.

[fubgumfaw]

You should bear in mind that electrical engineers are trained in the safe use of electricity. Moreover, they are taught to abide by a professional code of ethics. Two stanzas of the IEEE Code of Ethics stand out in particular:

1. to hold paramount the safety, health, and welfare of the public, to strive to comply with ethical design and sustainable development practices, to protect the privacy of others, and to disclose promptly factors that might endanger the public or the environment;

...

5. to seek, accept, and offer honest criticism of technical work, to acknowledge and correct errors, to be honest and realistic in stating claims or estimates based on available data, and to credit properly the contributions of others;

...

You are dealing with high voltages, high current, high power and a whole lot of stored potential energy. You are using the power supplies in a very unconventional configuration. It is one thing to expose yourself to the risks of this setup, but it's another thing to expose others to that risk - especially if they are less knowledgeable about electronics and how the system works. Just because you've never gotten struck by lightning while standing under a tree during a thunderstorm doesn't mean it's safe to do so.

At the very least I would get separate supplies for your different charging needs (voltages & amps). Safety features that real battery chargers have include automatic current limiting, cut-off for over current and over-voltage events, automatic charge termination, battery temperature monitoring. Note the emphasis on "automatic". If the process requires human intervention that in itself dramatically increases the chance of something going wrong.

These are all good points...
High voltage...maybe. I'd say 380v is "high voltage DC". I've been working with 82v to 220v for many years. I don't really think of this as high voltage. It's pretty easy to manage and to stay away from situations where electrocution can occur.
High current...30 amps...not really no. My latest EV runs at 82v and I see 500 battery amps. That's far from the maximum I've ever seen which is 1000 amps when I used to service industrial 3 phase breakers.

I do agree that Meanwell did not intend their PSU's be used in this fashion, but then neither did Dell or anyone else whose PSU's are getting stacked in series. This is not new and I didn't do it first. Maybe it's a bad idea? Maybe if you don't want to try it for yourself you shouldn't? Does it work and have I ever seen even a hint of a problem? Never. This isn't dismissal, just acknowledgement that this is done by many people including me.

[tooki]

The resistors are in parallel with each other to create the resistance needed to limit current to about 25 amps or a bit more.
They are in series with the output of the PSU's. So that may be 2, 3, 4, 5, 6, 7 or 8 PSU's in series and in series with a bunch of parallel resistors. In any set of series PSU's the resistors are only in series with the total number of series PSU's.

When you take small snippets of something someone said you lose context. I'm pretty sure I already said all of this, but you'd never know it based on the snippets you have quoted.

Nope, you did not. I re-read your posts in their entirety before commenting to make sure I hadn't overlooked it. I hadn't. THAT is why I gave you the feedback regarding your explaining skills.

The resistors are in parallel. Some will be in parallel all the time and more will be added or taken away as needed to raise or lower the total overall resistance needed to limit current to 25 amps. […]

Dude, you suck at explaining things.

Your resistors are in parallel with each other, but the resistor bank is in series with the output. You didn’t bother mentioning this critical point at any time.

Assume that I know what I'm doing when it comes to charging. I've been using SE-600 Meanwell PSU's to charge for 5 years now. This is pretty well known territory for me...

Your questions, technical approach, and attitude suggest you don’t really know what you’re doing and don’t really appreciate all of the risks involved.

The fact that you’ve gotten away with a bad approach for 5 years doesn’t make it a good approach.

I think you can agree that current follows the path of least resistance.

No, because it does not.

Well, it does, in that the current always takes all available paths, which the path of least resistance is one of. The various resistances simply determine how the current is divided amongst all the paths.

But the statement “follows the path of least resistance” generally means “...to the exclusion of higher-resistance paths”, and as such is false.

OK...thanks for the feedback...negative as it may be...
Opinions and scathing comments won't sway anyone. I'm not sure why people bother.
Please present actual facts that show why i could be doing something wrong instead of slamming me as a person which is pointless and offensive.
Saying "someone sucks" is offensive. Saying "I don't understand what you said" is not. Consider being less offensive please.

I didn't say you suck as a person. I said you suck at a particular skill, which is a different thing. (For example, I absolutely suck at team sports. That isn't offensive, it's a statement of fact, and says nothing about me as a person.)

I categorically did not fail at understanding you, in that you omitted critical implementation details. That's entirely on you. As stated above, I verified this omission before commenting.

I do have a couple questions for you.
Have you tried what I have tried so that you know for a fact that it's safe or unsafe?
Are you basing what you said on any factual information or is that just your opinion?

I haven't tried charging Li-Ion this way because it's not the correct way to charge them, and risks overcharging, which is a risk I am not willing to take.

Li-Ion requires CC/CV (that's one thing, not two options) charging. I don't see anything in your setup that would enable it to detect the end of the CC phase and move to the CV phase, nor anything that would provide for the cutoff current detection at the end. Is there any automatic cut-off at all?? (Relying on the protection circuits does not count!!)

As I and others have told you, there's just the issue of risk, which we do not feel you have sufficient respect for. The amounts of energy involved here are enormous. No big deal when everything is going right. But if something goes wrong, it can go very, very wrong. (The biggest risk is fire. Not only is your house burning down a bad thing as such, but lithium fires are damned near impossible to extinguish. Fire departments routinely have to stick around for hours to watch crashed EVs because of how the battery packs can catch fire again spontaneously.)

Ive never claimed to be an electronics expert and there is more I don't know than I do know.

Yet you're stubbornly resistant to input from others.

I'm more of an experimenter...try it out see what happens. In this case, what happens is that it works and has been since 2015. I think the proof of time is evidence enough to say it's going to keep on working a good while longer.

No, absolutely not!

Whenever you're dealing with anything that involves safety (which Li-Ion battery charging absolutely falls under, as do some of the voltages involved here), your design has to be designed to be safe for the worst-case scenario. You cannot design around the best-case scenario. And it's not something where "trying it out and seeing what happens" is a responsible thing to do.

You've gotten away with it for 5 years. But you've got no margin for error, and that makes it unsafe.

Sorry for coming off as harsh, but when it comes to safety, I (and this forum in general) err on the side of caution.

The other issue is caring for the cells. The CC/CV charging regimen also aims to maximize battery capacity. If you were to do only the CC phase, and ending charging altogether when the max voltage is reached, you forgo a substantial chunk of available capacity. OTOH, if you follow the CC phase with a CV phase but have no end current cutoff, you overcharge the cells, which damages them.

So far I have yet to see evidence to the contrary. This isn't even my idea! I read about it on Endless-Sphere being done with 12v server PSU's. The requirement of isolated outputs was easy to determine so I tried on sevaral of those 90W Lenovo PSU's and it worked! I then bought 4 Meanwell SE-600 PSU's as I knew they too had isolated outputs. This too worked flawlessly and that charger has seen continuous use for many years now. More recently I got around to this building the super charger with 8 PSU's potentially stacked in series. So far so good...seemed a good idea to me so far and is doing what I expected it would do...work. Whether this is the intended use that Meanwell had for the PSU's, well I don't worry about that too much!

Just because two or more people have the same bad idea doesn't make it a good idea.

You should bear in mind that electrical engineers are trained in the safe use of electricity. Moreover, they are taught to abide by a professional code of ethics. Two stanzas of the IEEE Code of Ethics stand out in particular:

1. to hold paramount the safety, health, and welfare of the public, to strive to comply with ethical design and sustainable development practices, to protect the privacy of others, and to disclose promptly factors that might endanger the public or the environment;

...

5. to seek, accept, and offer honest criticism of technical work, to acknowledge and correct errors, to be honest and realistic in stating claims or estimates based on available data, and to credit properly the contributions of others;

...

You are dealing with high voltages, high current, high power and a whole lot of stored potential energy. You are using the power supplies in a very unconventional configuration. It is one thing to expose yourself to the risks of this setup, but it's another thing to expose others to that risk - especially if they are less knowledgeable about electronics and how the system works. Just because you've never gotten struck by lightning while standing under a tree during a thunderstorm doesn't mean it's safe to do so.

At the very least I would get separate supplies for your different charging needs (voltages & amps). Safety features that real battery chargers have include automatic current limiting, cut-off for over current and over-voltage events, automatic charge termination, battery temperature monitoring. Note the emphasis on "automatic". If the process requires human intervention that in itself dramatically increases the chance of something going wrong.

These are all good points...
High voltage...maybe. I'd say 380v is "high voltage DC". I've been working with 82v to 220v for many years. I don't really think of this as high voltage. It's pretty easy to manage and to stay away from situations where electrocution can occur.

And yet knowledgeable people manage to get electrocuted every day. Again, you have to design around when things go wrong, and not around people "managing" the situation.

High current...30 amps...not really no. My latest EV runs at 82v and I see 500 battery amps. That's far from the maximum I've ever seen which is 1000 amps when I used to service industrial 3 phase breakers.

For what it's worth, 30A may not be much to electricians, but in the context of electronics, where tiny currents are the norm, 30A is huge (other than the specialization of power electronics, of course).

I do agree that Meanwell did not intend their PSU's be used in this fashion, but then neither did Dell or anyone else whose PSU's are getting stacked in series. This is not new and I didn't do it first. Maybe it's a bad idea? Maybe if you don't want to try it for yourself you shouldn't? Does it work and have I ever seen even a hint of a problem? Never. This isn't dismissal, just acknowledgement that this is done by many people including me.

But it is dismissal of the concerns. You've dismissed every concern anyone here has expressed, which is why we are worried that you do not have adequate respect for the risks involved.

[fubgumfaw]

But it is dismissal of the concerns. You've dismissed every concern anyone here has expressed, which is why we are worried that you do not have adequate respect for the risks involved.

I'm going to respond to only this one thing...
GOSH! I AM SOOO GLAD you are around to tell me my own thoughts! Without you to tell me what I'm thinking, well I'd have to know my own thoughts and know my own experiences and what I have done in the past. So THANK YOU for being a mind reader! I'd be lost without you! I mean between your cheery and positive and humble disposition and your ability to read minds, why aren't we all just doing what you say all the time? I bet that's really frustrating for you! Sort of like beating your head up against a brick wall!