NeilP
1 GW
Yea, I have a clone S-350 on its way too
converting meanwell power suply into Lipo charger
- Thread starter wojtek
- Start date
I don't see why removing some material off the shunt would be a bad idea at all.
You really don't want to be running the thing on the stock current limit like that. It wont be easily variable, but with a 350W unit, you would just want it at 350W all the time anyways. Its not like its a 10KW charger you might want to back down at times. Just sand the shunt until its pumping out rated current.
You could mod the fan to be on all the time, but its probably fine around the stock power levels.
You really don't want to be running the thing on the stock current limit like that. It wont be easily variable, but with a 350W unit, you would just want it at 350W all the time anyways. Its not like its a 10KW charger you might want to back down at times. Just sand the shunt until its pumping out rated current.
You could mod the fan to be on all the time, but its probably fine around the stock power levels.
Does anyone know how to mod this S-320-24? This supply faults to 'hickup' mode on overcurrent.
Any ideas if this hickup mode can be disabled?
I can take good photos of the front and back of the board if it helps. This design looks quite a bit different from the current controllable ones.
Any ideas if this hickup mode can be disabled?
I can take good photos of the front and back of the board if it helps. This design looks quite a bit different from the current controllable ones.
NeilP
1 GW
Was it sold as an S-350? if so I reckon you desetve either your money back or a partial refund
I can see one label on it that says S-320
Maybe you can get ( If Richard does build more) a Fecther mini meanwell current limiter board for if
If there is a resistor behind thevoltage adjustment pot, then you should be able to tie in his limiter board to it
If he does not make any more PCB's then if you do a search for the thread about them, then he did publish the schematic for it, build your own on veroboard
Or do the shunt mod, where you clip one of the shunts, or if only one shunt shave some thickness off with a file/ dremel etc. and then make exta sure you remove all shavings before powering up
I can see one label on it that says S-320
Maybe you can get ( If Richard does build more) a Fecther mini meanwell current limiter board for if
If there is a resistor behind thevoltage adjustment pot, then you should be able to tie in his limiter board to it
If he does not make any more PCB's then if you do a search for the thread about them, then he did publish the schematic for it, build your own on veroboard
Or do the shunt mod, where you clip one of the shunts, or if only one shunt shave some thickness off with a file/ dremel etc. and then make exta sure you remove all shavings before powering up
I did note that it says S-320 on the board. The enclosure clearly says S-350-24. I can assure you with a fairly high degree of certainty that this supply was never opened or modded in its life, and this is how it came from the factory.
I don't see a shunt anywhere on the board, definitely not anything close to a standard shunt used on other supplies.
I don't see a shunt anywhere on the board, definitely not anything close to a standard shunt used on other supplies.
NeilP
1 GW
I would get on to the supplier, take photos and also photos of the real S 350 and complain
Looks like it got labelled wrong at ths factory
Looks like it got labelled wrong at ths factory
NeilP
1 GW
Just reposting this here in case any one reading this got the NES...flanders? and others I think sent by mistake by Sure Electronics
Who got an NES meanwell from Sure instead of a S-Sereis Meanwell?
Well I did, and they started off by offering me a $5 USD refund..finally got a $18 USD refund..which considerign it cost me $58 dollars with shipping, I was not impressed with, so I gave negative feedback
A day later, Sure were back on to me with another e-mail asking what else they could do for me if I would remove the negative feedback.
Well I asked for either a full refund or a replacement S-350 if they had one.
They said they just had eight S-350-48 left and they would send me one..if I returned the refund
I then got cheeky and asked for a different model..a S350-12 instead..since I had already bought S-350-48 from someone else..
And guess what...they agreed and have sent it
So if you too ended up with a NES model...tell Sure if it is not sorted, you will give them negative feedback...that will get them moving
Who got an NES meanwell from Sure instead of a S-Sereis Meanwell?
Well I did, and they started off by offering me a $5 USD refund..finally got a $18 USD refund..which considerign it cost me $58 dollars with shipping, I was not impressed with, so I gave negative feedback
A day later, Sure were back on to me with another e-mail asking what else they could do for me if I would remove the negative feedback.
Well I asked for either a full refund or a replacement S-350 if they had one.
They said they just had eight S-350-48 left and they would send me one..if I returned the refund
I then got cheeky and asked for a different model..a S350-12 instead..since I had already bought S-350-48 from someone else..
And guess what...they agreed and have sent it
So if you too ended up with a NES model...tell Sure if it is not sorted, you will give them negative feedback...that will get them moving
flanders
100 mW
Hi Neil, nice to see you got something out of them
I did complain after the supply was sent, they did not even offer me anything, just said it was a better one.
I have just sent them another email, after all I ended up paying for a s350 clone after they sent me the nes model which I could not mod.
I did complain after the supply was sent, they did not even offer me anything, just said it was a better one.
I have just sent them another email, after all I ended up paying for a s350 clone after they sent me the nes model which I could not mod.
NeilP
1 GW
Well do what I did and post negative feedback on e bay and they then get back to you asking you to remove it, and what it will take for you to do so
Right. My Meanwell S-350-48 clone has arrived. I've bought it here. You can see there how it looks on the outside.
On the inside, the board looks like this:
It was nicely put together, no loose blobs of solder in the case, but few of them stuck onto compoments. After initial checkup, I've fired it up, and measured voltage range, which was about 41 to 55V. As that's nowhere enough for 14S LiPo charging I tried moding R25, but it stuck to 55 no matter what. So, I've removed ZD1 (for now), and now voltage goes nicely to 59 or so volts.
As there is only one shunt, I've tried R33 mod. At first it behaved strangely, with loud buzzing from PSU whenever I tried to limit current, and I couldn't make current go past 4A no matter what I did. But after I've removed ZD1 not only voltage rose, current also increased dramatically. Unlimited, 14S pack would pull 10A easily. And now when I dial it down with pot in parallel with R33 it works nicely and doesn't produce any loud buzzing.
All in all, so far so good. If the weather holds up, I'll try cycling my pack few times tomorrow, just to see how this clone holds up at 5A current limit.
On the inside, the board looks like this:
It was nicely put together, no loose blobs of solder in the case, but few of them stuck onto compoments. After initial checkup, I've fired it up, and measured voltage range, which was about 41 to 55V. As that's nowhere enough for 14S LiPo charging I tried moding R25, but it stuck to 55 no matter what. So, I've removed ZD1 (for now), and now voltage goes nicely to 59 or so volts.
As there is only one shunt, I've tried R33 mod. At first it behaved strangely, with loud buzzing from PSU whenever I tried to limit current, and I couldn't make current go past 4A no matter what I did. But after I've removed ZD1 not only voltage rose, current also increased dramatically. Unlimited, 14S pack would pull 10A easily. And now when I dial it down with pot in parallel with R33 it works nicely and doesn't produce any loud buzzing.
All in all, so far so good. If the weather holds up, I'll try cycling my pack few times tomorrow, just to see how this clone holds up at 5A current limit.
amigafan2003
1 kW
Quick update - I ordered a NES-350-48 to charge a 2s2p lipo pack contructed of four 5000mah Zippy 6s lipo packs.
Shaved about 2/3rds off the corner of the shunt and set voltage to 49.9v and hooked up to the pack - Turnigy watt meter shows max current draw of 7.74amps and 383 watts so happy with that - only 9% above spec. Amps started to tail off when the pack voltage reached 49.5 and the power supply switched off (well the amps stopped with an audible click) when the amp draw dropped to 20ma. I left the pack connected for 3hrs constantly checking the watt meter and the powere supply didn't try feeding any more amps. Fan kicked in and out when the psu got warm (not to warm - could still touch it) and the amp draw stayed level which is an improvement on the S-350-48.
Very happy with this PSU as a bulk charging solution- lets see how long it lasts.
Few photos (I've modded a volt meter into the top cover).
Shaved about 2/3rds off the corner of the shunt and set voltage to 49.9v and hooked up to the pack - Turnigy watt meter shows max current draw of 7.74amps and 383 watts so happy with that - only 9% above spec. Amps started to tail off when the pack voltage reached 49.5 and the power supply switched off (well the amps stopped with an audible click) when the amp draw dropped to 20ma. I left the pack connected for 3hrs constantly checking the watt meter and the powere supply didn't try feeding any more amps. Fan kicked in and out when the psu got warm (not to warm - could still touch it) and the amp draw stayed level which is an improvement on the S-350-48.
Very happy with this PSU as a bulk charging solution- lets see how long it lasts.
Few photos (I've modded a volt meter into the top cover).
NeilP
1 GW
amigafan2003 said:
Good that it works well....but ...something of minor importance, but worth getting correct
If you have two 6s packs in series then the correct notation is 12s not 2s
so your pack is a 12s2p pack
If i had 4 10cell nano tech packs connected so i have two blocks 2 of in parallel and the blocks of 2 series'd so i have 20 cells in series from 2 packs I would have a 20s2p pack. Not a 2s2p pack.
Get it?
As for dropping current more..well you could, but there are people that run these constantly at over 100%..so i would not worry too much. You could maybe do the Fan mod to...short out the transistor that turns the fan on..
You can see the pic I did on Shelbys page
https://sites.google.com/site/shelbyelectro/Mods-and-technical/batteries/bulk-charging-with-a-meanwell
amigafan2003
1 kW
NeilP said:
Got it - cheers
yeah - not worried at all really - they are dirt cheap - not like I'm going to lose alot of money if it dies. I still have an iCharger 106b to charge the pack if the Meanwell goes pop so wont be stuck either.
I'll have a nosey
NeilP
1 GW
Not sure what happens to the current with the 'shunt shaving ' mod, but with the R37 parallel resistor mod, doing that fan mod results in the current going up a bit. Not much, half an amp or less from what I remember
cor
100 W
Power supplies typically work in the same way, whether they have controllable outputs or not. I have modded many fixed power supplies to give them voltage control and current limit. The voltage control mod is typically just an addition of a potmeter, replacing the fixed voltage divider for the feedback.ZOMGVTEK said:
I typically added current limit by either using a series resistor that was already in the supply for surge limit to measure current, or by adding a small resistor and then adding limiting using a single transistor and a few resistors.
That current limit is not extremely precise and it is temp dependent, but that is actually a way of protecting the unit: the hotter the supply gets, the lower the current limit, due to the falling B-E diode drop in the transistor.
I have even gone as far as making the current limit output voltage dependent, to deliver constant power iso constant current.
All that is needed is the schematic of the feedback loop, often easily re-engineering from a picture or PCB.
NeilP
1 GW
Well I had a NES Meanwell, and I cooked the output rectifier when I tried making it current limitinf. Fetcher looked at the PCB and suggested where the current monitoring/ resistor divider network mya have been.. I tried and mod...and let the smoke out.
I will post the pics of the NEW PCB up here later...or maybe I alreayd have further up this thread..anyway no time to chec k now, I am at work.
But Cor, if you can have a look when I do post them, maybe you can offer suggestions of how to mod the NES 350 series supplies.
Cheers
Neil
edit , had time to look, go back to page 14 and mainly 15 to my posts...you will see a selecion of composites of the NES board
I will post the pics of the NEW PCB up here later...or maybe I alreayd have further up this thread..anyway no time to chec k now, I am at work.
But Cor, if you can have a look when I do post them, maybe you can offer suggestions of how to mod the NES 350 series supplies.
Cheers
Neil
edit , had time to look, go back to page 14 and mainly 15 to my posts...you will see a selecion of composites of the NES board
cor
100 W
Hi Neil,
Normally I am staying near Fechter (I am in SF South Bay area) but currently I am staying with family in The Netherlands. I don't know if you are shipping the blown one to Fechter, it may be cheaper to ship to NL than to California - I can take it back with me.
Regarding the remarks about the SCR1 (Silicon Controlled Rectifier, another name for Thyristor): that is usually part of the OVP circuit, to shutdown the supply so modding that is not needed unless you want to increase the max voltage above the point where the OVP bites.
The Isense is not the proper place to connect the current limit, because Isense is used to shutdown the power supply when the current through the switching transistor gets too high, that is why you get hiccups from the supply when it is overloaded.
My preferred way to add current limit is to detect the current, then at limit I like to pull the voltage feedback, so that you get a linear decrease of the voltage output to maintain the current at limit.
I have looked at the pics but it is not clear where the shunt is, whether it is the small metal bracket with the folded edge, or the "nail" next to the 3 output elcos.
Normally I am staying near Fechter (I am in SF South Bay area) but currently I am staying with family in The Netherlands. I don't know if you are shipping the blown one to Fechter, it may be cheaper to ship to NL than to California - I can take it back with me.
Regarding the remarks about the SCR1 (Silicon Controlled Rectifier, another name for Thyristor): that is usually part of the OVP circuit, to shutdown the supply so modding that is not needed unless you want to increase the max voltage above the point where the OVP bites.
The Isense is not the proper place to connect the current limit, because Isense is used to shutdown the power supply when the current through the switching transistor gets too high, that is why you get hiccups from the supply when it is overloaded.
My preferred way to add current limit is to detect the current, then at limit I like to pull the voltage feedback, so that you get a linear decrease of the voltage output to maintain the current at limit.
I have looked at the pics but it is not clear where the shunt is, whether it is the small metal bracket with the folded edge, or the "nail" next to the 3 output elcos.
NeilP
1 GW
Edited:
Posted a reply at least three times but they never appeared. so a less detailed reply was tried ..and it appeared
shunt is LHS near the three electrolytic caps..near R111 just visible under the white glue
Would rather see if I can repair the unit if I can.. i have bought a new output rectifier...the part that has blown
Cor, when are you heading back to the USA?
Posted a reply at least three times but they never appeared. so a less detailed reply was tried ..and it appeared
shunt is LHS near the three electrolytic caps..near R111 just visible under the white glue
Would rather see if I can repair the unit if I can.. i have bought a new output rectifier...the part that has blown
Cor, when are you heading back to the USA?
amigafan2003
1 kW
NeilP said:
The shunt on mine is totally free of glue - I just had to bend R111 to one side gently so I could get a dremel in there - I'll take a photo later. Component number on the board is J100.
NeilP
1 GW
amigafan2003 said:
Wish I had done the shunt mod too rather than tried the R33 mod
cor
100 W
Hi Neil,NeilP said:
I will fly back Jan 18th.
If the output rectifier blows then there is also a good chance that other things have blown, it can be as simple as a resistor or even a trace (at 20A output) but it could also be the primary switching transistor or the controller, so first thing is to see if the supply is still functioning somewhat - is the controller still receiving power supply and trying to regulate (but not able due to for example the optos not giving the right feedback) or are the silicon pieces all blown?
If the controller is still running and the switching transistor still works, is there AC coming out of the transformer or did that get damaged? That is also a piece that would be difficult to replace (you can in theory rewind a blown transformer, but it is not easy)...
So, please post what you see this supply still do.
NeilP
1 GW
Ok, will do that tomorrow
cor
100 W
Note that the primary side of the power supply is connected to the AC mains, so take care in not touching it, if necessary apply an isolation transformer to give you less change of touching lethal voltage, be warned. If you don't know how to work safely around high voltage then do not attempt (legal disclaimer).
You can check first that the main high voltage caps still charge to ~340V DC (in 230V country) or else you simply have a blown fuse or mains rectifier.
If that is still OK then you can check that the TL3845 Vcc is above 8V and the Vref 5V.
The Vfb should be around 2.5V when the supply is running, so probably at this moment it isn't.
The Isense should be below 1V or else the controller will shut down (typically the Isense receives pulses, so a scope is a better instrument than a DVM for this)
Further, the two optos can be checked: the OVP should not be active (otherwise the SCR1 will keep the supply disabled) while the other opto-coupler for the voltage feedback should be active (LED voltage around 1.2V and the opto output in the circuit for the voltage feedback)
I do not have a schematic for the whole feedback loop and I would need to become intimate with this power supply before I can give very specific advice, so all this is based off the pictures posted and the datasheet of the UC/TL384x PWM controllers.
BTW, these controllers are found in many different architectures, some even without opto-couplers where they sacrifice precise regulation for the sake of cheap implementation, see for example the UC3843 datasheet with a proposal schematic for an off-line flyback regulator. The Meanwell is certainly of higher class, but I have also found this PWM controller in my charger for the original 24V lead-acid pack in my EVG bike.
You can check first that the main high voltage caps still charge to ~340V DC (in 230V country) or else you simply have a blown fuse or mains rectifier.
If that is still OK then you can check that the TL3845 Vcc is above 8V and the Vref 5V.
The Vfb should be around 2.5V when the supply is running, so probably at this moment it isn't.
The Isense should be below 1V or else the controller will shut down (typically the Isense receives pulses, so a scope is a better instrument than a DVM for this)
Further, the two optos can be checked: the OVP should not be active (otherwise the SCR1 will keep the supply disabled) while the other opto-coupler for the voltage feedback should be active (LED voltage around 1.2V and the opto output in the circuit for the voltage feedback)
I do not have a schematic for the whole feedback loop and I would need to become intimate with this power supply before I can give very specific advice, so all this is based off the pictures posted and the datasheet of the UC/TL384x PWM controllers.
BTW, these controllers are found in many different architectures, some even without opto-couplers where they sacrifice precise regulation for the sake of cheap implementation, see for example the UC3843 datasheet with a proposal schematic for an off-line flyback regulator. The Meanwell is certainly of higher class, but I have also found this PWM controller in my charger for the original 24V lead-acid pack in my EVG bike.
Similar threads
