Fechter's Mini Meanwell Limiter Board

[jana]

Ok, thanks

[Degull]

The version 4.2.5, with the three pots, was there ever a professionally made board for this design? I have a good relationship with a circuit board manufacturer locally. If there is any interest I can have them made and distribute them here for other members. The meanwell chargers are a great cheap option for newbies like myself and I think it would be a worth while project for everyone.

[hillzofvalp]

That would be nice

[NeilP]

Yes there was versions of boards for three pot 4.2.5 4.2.6...Gary Goodrum at tppacks sold them ..They are no longer on the site, as there were supposed problems with them...but

I have got three of them and two are up and running on both my meanwell builds.
once the issue with gettting the componets in the correct way around was sorted, they worked fine.

one of the transistors...or FETS..can't remember had slightly diffferent versions..and the pinout was changed..BS107A and then BS107B or something similar.

If you read up you see Richard mentionin ghtat there were problems, but as I said, above that issue with the pinout they work fine.

And again..another BUT....
i am not using mine for current limiting...I have done that with the mod on the board..R37. The issue was supposed to be that these boards caused a buzz...but these things tend to buzz a bit anyway when current limtied .I only use these 3 pot boards as end of charge monitor and for the LED that changes from orange to green when done charging

Speak to Gaary at tppacks he may sell you them

[wineboyrider]

I've got one from Fechter that I didn't have a chance to put together, because my Meanwell failed just before I got the boards in.

[Degull]

I've got one from Fechter that I didn't have a chance to put together, because my Meanwell failed just before I got the boards in.

I have a 48 volt 500 watt Meanwell that would love to meet that board. I'll take it if you want to sell it!

[NeilP]

I do have a picture of the PCB for the three pot version..but I am not going to post that up just yet I am afraid. It is not my design..and I can't go sharing something like that without the OK of Gary or Richard first..it is there design and they do have businesses to run, if Richard can OK, it then I can pass it on.

but not till I know I am not going to upset anyone

[Degull]

I do have a picture of the PCB for the three pot version..but I am not going to post that up just yet I am afraid. It is not my design..and I can't go sharing something like that without the OK of Gary or Richard first..it is there design and they do have businesses to run, if Richard can OK, it then I can pass it on.

but not till I know I am not going to upset anyone

Totally understandable. I'm offering to have the boards made but I don't want to step on anyone's toes.
I'm sure that I'm not the only one that would love to get a few these boards. The Meanwell power supplies are well documented here on the forum and they come at a great price. I've been reading about them for a couple of weeks now. These limiter boards basically sum up a bunch of knowledge shared here already. It is much appreciated.

[fechter]

It's OK to post it.

[NeilP]

OK, here they are
There are three images: A front image, a rear , and then a third that is a composite, so you can see where the tracks coincide as it is a double sided board

But I would check with Gary at tppacks first, before you go spending money and making a load. He may still have some ready to sell. They are not on the site anymore, but he may still have some

[wineboyrider]

I've got one from Fechter that I didn't have a chance to put together, because my Meanwell failed just before I got the boards in.

I have a 48 volt 500 watt Meanwell that would love to meet that board. I'll take it if you want to sell it!

PM sent.

[fechter]

I'm working with Degull to try to get more of these boards made. I first need to review the layouts and make any needed corrections. Hopefully we can make these available again soon.

[hillzofvalp]

[cor]

I'm working with Degull to try to get more of these boards made. I first need to review the layouts and make any needed corrections. Hopefully we can make these available again soon.

Hi Richard,
Let me know if you like help - for example requesting a local (Sunnyvale) PCB house for a quote to produce & populate these. That would allow making them SMT and selling them complete without hassle about populating boards or making kits.

BTW, what was the issue or issues with the 3-pot board? I could see that the float control was iffy in that it had a control range of 5 mV while you probably also need to deal with noise and the opamp offset which is 3mV worst case, so more than half the control range. The noise can be dealt with by careful design and the control range vs input offset just means that the control range should be a little larger and also allow going negative. Or use an opamp with better spec.

[NeilP]

Hi Cor

As mentioned on the other thread about parallel operation of these Meanwells, and your comment

Hi Neil,
With the access to the voltage control on these power supplies, it is rather simple to add a circuit to allow them to share current.
The basic principle is that a supply has a current sharing input and a sharing output. If the sharing input is not connected then the supply behaves as voltage source and represents the amount of current drawn on its sharing output.
When the sharing input is connected, then the supply compares the sharing input to its own amplified current sense signal and adjusts the output voltage to match the current that is represented at its input. In other words, all connected supplies will behave as current source.
Such a board would look a lot like a mini current limiter.

If there is another run of these boards..would a mod to them be possible to have a parallel sharing option as well?

[nicobie]

I'm working with Degull to try to get more of these boards made. I first need to review the layouts and make any needed corrections. Hopefully we can make these available again soon.

This is good news indeed!

[Degull]

Hello everyone. I'm a newb in the Ebike world and I've been lurking around the forum for quite sometime. At the moment I'm working on building my own a123 48v battery and I plan on feeding it with a Meanwell PS. These limiter boards are perfect for people like me that are looking to get into the hobby. My motivation behind getting the limiter boards manufactured is to make them available to everyone again. More importantly, not having to build one on a bread board. I'm not looking to profit, I'm just giving back to the community by fronting the cash to have the boards manufactured then distribute them.

Hi Richard,
Let me know if you like help - for example requesting a local (Sunnyvale) PCB house for a quote to produce & populate these. That would allow making them SMT and selling them complete without hassle about populating boards or making kits.

BTW, what was the issue or issues with the 3-pot board? I could see that the float control was iffy in that it had a control range of 5 mV while you probably also need to deal with noise and the opamp offset which is 3mV worst case, so more than half the control range. The noise can be dealt with by careful design and the control range vs input offset just means that the control range should be a little larger and also allow going negative. Or use an opamp with better spec.

A complete SMT solution would be the best no doubt but that might be far off if we have to rework the design and go through a testing phase. At the moment Fetcher has a couple of working designs. The simple limiter version and the 3 pot version. Both of these boards are working and I was going to propose that we have both of them made. If you have anything that we could simply add to the versions that already exist then please your thoughts are welcome! I would love to get my hands one of these pronto!

If there is another run of these boards..would a mod to them be possible to have a parallel sharing option as well?

That would be amazing addition as well but again that will slow down the whole process for those of us that do not have a limiter already. It looks as though there are no more of these boards floating around for sale and I for one would like to have one now. Unless you want to part with the pretty one you took the pics of

[cor]

If there is another run of these boards..would a mod to them be possible to have a parallel sharing option as well?

Hi Neil,
I do not know how much interest there is for paralleling power supplies to get higher current charging.
If there is and Richard is interested then I can share a quick sketch I made, it would add another dual opamp to the current limiter board.
He said that he will review/redesign the 3-pot board anyway, but I will let Richard decide if he wants to try to include the current sharing thing.
I have a couple S-240-48 on which I could experiment with current control and sharing while charging my battery pack.

[cor]

A complete SMT solution would be the best no doubt but that might be far off if we have to rework the design and go through a testing phase.

Hi Degull, welcome!
I actually joined here recently, I had an EV for a while but came here to see what others are doing to their eBikes. I bought an almost unused, 12yo bike with bad batteries and am about to install its first Lithium pack.
The SMT was just an idea, it would cut down on assembly time and errors and frustration in getting parts - once the layout is tested to work.
Another interesting option is PCBfabExpress who have a "bare bones" offer that gives you two-sided FR4 boards with plated-through holes, which seems to be something that Richard was already using for the limiter boards. They have a $40 lot charge and $0.60 per square inch, so when ordering 50 boards of 2x2 inch (I estimate that is the size of the limiter) then the cost is $3.20 per PCB if my math is correct.
SMT boards will typically be more expensive, especially if populated, but at higher quantities this quickly comes down and becomes competitive with through-hole boards, especially if you take the cost for your own time into account.

[NeilP]

Another interesting option is PCBfabExpress who have a "bare bones" offer that gives you two-sided FR4 boards with plated-through holes, which seems to be something that Richard was already using for the limiter boards.

Yes, this is true, it was Richard (Fechter) that put me on to the ExpressSHC ExpressPCB software. He or Gary may still have the ExpressSCH file that they used to make the original run of boards. If they did then that would make getting a new run in the same format very easy I would expect

That would be amazing addition as well but again that will slow down the whole process for those of us that do not have a limiter already. It looks as though there are no more of these boards floating around for sale and I for one would like to have one now.

So you did check with Gary at TPPACKS? I thought he told me he still had plenty of the original boards left but had stopped selling them because of setup documents and component placement issues

Unless you want to part with the pretty one you took the pics of

No sorry, want to hang on to this one for my next build.



But as for not having a limiter, I personally see the limiting function of this board as an almost redundant feature anyway. For me the big feature of the boards is the charge current monitoring and end of charge cut off (voltage drop ) and LED indication when charge complete.

If anyone has the ability to build one of these boards from kit form, then they sure do have the ability to mod the Meanwell to limit current without this board.
To do it on the S series requires soldering two wires to pre drilled holes on the board and wiring the other end of the wires to a potentiometer.
On the NES series, the mod may be a little more involved, as I am not sure if R134 is an SMD or conventional resistor and how it is laid out as I do not have one of these boards anymore to check. I believe it may involve cutting a track also..>Cor will answer I am sure

Be it an S series and the R33/ R37/SVR1 mod or the new NES series and the R134 mod that Cor has worked out, they are both the quickest and easiest way to give you adjustable current limiting compared to the buying, building and setting up a limiter board.
Anyone that has a parts bin in their workshop of old spare electronics components (resistors,pots etc) could have an S series at least, current mod'ed in under an hour, from disassembly to re assembly...One 5 k pot and two lengths of wire..job done. If you wanted to be gash about it you could even do it without removing the board from the case and solder the wires directly to the legs of R33. So 15 mins at the outside.


It has been noted that the NES series Meanwell auto shuts down and has a relay that kicks out when current flow drops below around 20mA. If it could be determined if this 'load monitoring' and the 20mA figure could be made adjustable, then we could do away with needing an extra board at all, as these functions are already built in to the supplies as they are.

[fechter]

When running multiple supplies in parallel, each one will run at its limit until the pack reaches near full charge, at which point they may not share evenly. I don't think this is a problem. Most of the time they will be in CC mode and sharing according to their individual limits. Once a supply hits CV, the pack is essentially done charging.

Most of the issues with the 3 pot board were with the setup and testing instructions. There were also some issues with the parts fitting the holes on the board, both of which are easily corrected. It is still a bit tricky to set the cutoff current as you need some way to get the current at the cutoff point (around 100mA?) to make the adjustment. If everyone wanted the same cutoff current, I could use fixed resistors and eliminate the adjustment, but as cor points out, the error in the amplifiers is significant at the low end of the range so some adjustability is nice.

The simple limiter board was laid out for a particular potentiometer from Mouser that has a tendency to go out of stock, so maybe I can have two sets of holes to fit more than one style of pot. I think I can also add a second set of holes for the shunt resistor to allow currents over 20A.

I'm still on the road now and don't have the files with me. When I get home I'll look at revising them for the lastest changes.

[cor]

When running multiple supplies in parallel, each one will run at its limit until the pack reaches near full charge, at which point they may not share evenly. I don't think this is a problem. Most of the time they will be in CC mode and sharing according to their individual limits. Once a supply hits CV, the pack is essentially done charging.

I understand and the uneven current sharing is not my concern, although this setup requires that every paralleled supply must have its finishing voltage as well as current limit set properly. What I mean is that if any of the paralleled supplies has their supply voltage set too high then you don't notice it during the time that this supply is in current limit, until the moment that all other supplies stop delivering current, then the voltage will rise until the level set at this supply. For lead-acid batteries this is an easy trick to give them a controlled over-charge for balancing purpose (I have actually used this setup in the past) but for Lithium this can spell disaster. When all supplies are forced to share then they will stay at the voltage set for the lowest supply.
Another issue is the turn-off (float) point. If two independent supplies are set for approximately the same voltage then usually one will fall below the threshold while the other is still pumping max Amps and so the first supply will turn off (go to float) and the second will continue charging until that one falls below its threshold and also switch to float, so each supply should be set to the current at which you want the charge to terminate, for example 1A. However, at times it can be that the voltages of both supplies are identical and they both continue to charge until the charge current is 2A and they both switch to float. This is a slim chance, but it can happen and if you parallel more supplies, the effects become more significant: For 4 supplies, they could stop charging at 4A, 3A, 2A or 1A depending on how equal their outputs are. With a current sharing control, this will not happen as all supplies should be regulated to the lowest configured one, with 4 supplies you set the threshold for float to 0.25A on one and they all go to float as soon as the charge current drops to 1A (4 x 0.25)

Most of the issues with the 3 pot board were with the setup and testing instructions. There were also some issues with the parts fitting the holes on the board, both of which are easily corrected. It is still a bit tricky to set the cutoff current as you need some way to get the current at the cutoff point (around 100mA?) to make the adjustment. If everyone wanted the same cutoff current, I could use fixed resistors and eliminate the adjustment, but as cor points out, the error in the amplifiers is significant at the low end of the range so some adjustability is nice.

Correct, you need adjustability even if you always want to set a fixed current due to component variations, but with an option to share you actually want to select a lower float current, because it is multiplied by the number of parallel units.

The simple limiter board was laid out for a particular potentiometer from Mouser that has a tendency to go out of stock, so maybe I can have two sets of holes to fit more than one style of pot. I think I can also add a second set of holes for the shunt resistor to allow currents over 20A.

In the schematic you say that the shunt is 5W (I do not know which temperature it will reach at this power level) while I am guessing it can take only 2W because 20A times 0.005 Ohms is 100mV, times 20A is 2W.
I hear you on the component placement/ holes stuffing. It would be good to be able to use different style pots.

I'm still on the road now and don't have the files with me. When I get home I'll look at revising them for the lastest changes.

I will first try a current sharing schematic that I found in a paper, after I designed my own. In essence it will add two opamps plus some passives, so it should work with a single quad opamp or two dual. I have a couple reels of opamps, so I will try to use those in my experiments. Feel free to go ahead with your boards, I think it will take some time to work out the kinks in current sharing setup. As I said, I have a couple S-240-48 that are perfect to test the current sharing while charging my pack.

[fechter]

The current sharing circuit would be useful for the 3 pot setup with float, but for the simple limiter I think it will work fine without it. Each supply would be adjusted for the same CV setting and at end of charge, one supply would tend to take all the load for a short period of time.

Most people are running multiple supplies in series. If you wanted more current, just use lower voltage supplies that supply more current and put them in series to get the desired voltage. In most cases only one of the series supplies needs the limiter circuit.

[NeilP]

Yes, although it was my original suggestion, I do tend to agree that maybe it is work that is not a good use of time.

A few months ago, I would have said different, but now with the appearance of these Alloy shell chargers from E-city/BMS battery, building a high power charger from multiple Meanwells is not as cost effective.

When I originally built my 100volt twin S-350-48 supply. I had ideally wanted a 20 amp supply at 100volts, so that was going to be 6 x S-350-48's
in three parallel strings.
I have just ordered a 2kW Alloy Shell charger from BMS for around $350 dollars delivered to me in UK from China..plug and play..no competition.

buying 6 Meanwells, combined with the hassle of building the system up .. would have made it not worthwhile.
Suppose if someone laready had a stack of Meanwells/ PC server supplies etc going cheap and free it would be a useful idea, but probably a bit limited now, you are right Richard

[cor]

Richard, correct, if the minimal charging voltage of an empty pack is still higher than the lowest voltage that the current limited supply will put out (plus the voltages from the series connected fixed supplies) then only that one supply needs a current limiter.
I used this configuration when I had a bunch of 48V supplies and a Sorensen DCR60 voltage and current regulated supply. I knew my battery pack was never going to drop below 300V and I want to charge up to about 369V, which was 14.2V per (lead acid) battery, so I set my Sorensen to 10A and the highest I could tweak it to which was almost 69V and added a diode to its output so that I could put it on a timer to come on at midnight and shut off by 7AM and have my truck ready for the next day commute and errands (and charge at the low rate night electricity). The 48V supplies together delivered a constant voltage and were pretty low consumption as they were HF switching supplies, unlike the Sorensen which is just a Triac control at line frequency. But the combination of fixed power supplies and one regulated and limited supply worked very well - if you know what you are doing.