RSP Meanwell power supply Reverse engineer

[Doctorbass]

The Meanwell RSP power supply serie is one of the highest quality and compact charger solution for high power and the RSP serie is great but there is one missing feature we woudl like to have on it: CURRENT LIMIT ADJUST !!!

I am using like 7 of these power supply. Recently i built a 6.6kW charger using these for my Zero motorcycle.

At full power ( CC CVmax) It require 30A at 240V. This is perfect for the most common level 2 charging station of 30A.

However problem is that if i'm charging on a charge station that is on 208V and not 240V it will draw 35A+ and will trip the 30A protection circuit of the charge station.

To solve that i need to be able to adjust the actual current limit to a lower value when desired.. ex: when having 208V input instead of 240V...

I got from my work a deffective RSP-750 (http://www.meanwell.com/search/RSP-750/RSP-750-spec.pdf) witch is perfect for deep investigation on how is the current limit designed and ito know if could install a current adjust feature with the RSP.

I have decided to reverse engineer a part of the circuit but i will some need help on that. What i'm looking for is the proper resistor that i can modify or add a pot to change the actual current limit.

Here is couples pictures of it and of the curent sense section with the shunt resistor on the back of the pcb.

You can see that close to the 3 parallels shunts resistors there is some holes i pointed with teh arrow. I'm sure These holes are for the current sensing to conduct the measured voltage across the parallel resistor to a trace that goes to the vertical daughter smd PCB.

 

[Doctorbass]

Ok i have finished to draw the schematic part that i think is related to the current sensing.

Something curious is that one of the 2 sense traces goes to the vertical daughter board and is in serie with a thermal switch ( see TSW1). it is N-C type so i goess that when the mosfet become too hot on the power section it open and the current sense line see open circuit and trigger teh protection that reduce PWM to 0% duty??...

I am surprized that the two sense connections that are i parallel to the shunt does not goes both to the inverting and non inverting input of the OpAmp...


Teh max current limit is about 36A on this model so this mean that the max voltage across the shunt should be about :
1/((1/0.005)+(1/0.005)+(1/0.003)) = 1.36 miliohm

Vsense max = 36 x 0.00136 = 49mV

Well i feel i might have made a mistake by following the two parallel traces that appeared to be together...

If i take a closer look... the sense trace that i tought was going to the thermal switch (TSW1) might actually be the little trace on the left (on the corner of the yellow transformer base) and might goes to the pin 5 ofthe vertical PCB witch gors to the non inverting input =)

Doc

 

[steveo]

:| I wish i could be of more help, this is not my forte...

Can you use a limiter board with your charger setup?

I bought myself a nice xantrax 0-10amp 0-100v pfc supply.. a bit wider and longer then the rsp you use.. but i love that its variable.. has displays for current.. voltage.. and no bs protection that we get sometimes with some psus.. to bad its only 1kw

Subscribed!

-steveo

 

[dnmun]

you could try increasing the gain on the op amp by changing that 21.52k resistor up to 24.4k and see if the output is reduced by the 15% you need.

 

[teslanv]

Subscribed.

 

[grindz145]

Dumb question: Cant you just bump the current sense resistors up a bit? Will that affect anything else in the supply? If you could switch out a parallel current sense resistor then you could have a couple of different levels.

 

[Doctorbass]

Dumb question: Cant you just bump the current sense resistors up a bit? Will that affect anything else in the supply? If you could switch out a parallel current sense resistor then you could have a couple of different levels.

I already tried that way... but it is really unprecise.. remember these sense resistors are like 5 miliohm so the switch and wiring must have very great contact resistance and low resistance....

this work a bit but not as good as a adjustable current psu.. :?

Doc

 

[Doctorbass]

you could try increasing the gain on the op amp by changing that 21.52k resistor up to 24.4k and see if the output is reduced by the 15% you need.

I will try

Doc

 

[grindz145]

ahh, that's true.... If only there were pots in the miliohm range..

 

[dnmun]

another idea.

your shunt is 1.36mR and has the inverting input to the op amp on the battery side.

if you could add 15% to the 1.36mR, say about .2mR, and then could move the inverting input of the op amp to the outside, the battery side, of the new total shunt then maybe that could add enuff to get it to modulate the power. then your switch would be in a high resistance lead to the op amp and not in the low resistance area of the shunt.

the extra shunt resistors would just eat up a little power when operating in the normal 240V AC regime, then you could have an external selector switch that would switch the op amp inverting input from one side of this added shunt to the other. and you could have a separate negative charger lead that bypassed the added shunt when using it on 240V AC.

i have some 4mR 2512 like that but it would take 20 of them in parallel to get to .2mR but you might find a wire type shunt that was way less than 1mR, or maybe find some 1mR 2512 surface mounts and solder them to some peg board and add the conductor to the ends of them, or use a regular wire shunt that is temp stable at these currents.

 

[Doctorbass]

another idea.

your shunt is 1.36mR and has the inverting input to the op amp on the battery side.

if you could add 15% to the 1.36mR, say about .2mR, and then could move the inverting input of the op amp to the outside, the battery side, of the new total shunt then maybe that could add enuff to get it to modulate the power. then your switch would be in a high resistance lead to the op amp and not in the low resistance area of the shunt.

the extra shunt resistors would just eat up a little power when operating in the normal 240V AC regime, then you could have an external selector switch that would switch the op amp inverting input from one side of this added shunt to the other. and you could have a separate negative charger lead that bypassed the added shunt when using it on 240V AC.

i have some 4mR 2512 like that but it would take 20 of them in parallel to get to .2mR but you might find a wire type shunt that was way less than 1mR, or maybe find some 1mR 2512 surface mounts and solder them to some peg board and add the conductor to the ends of them, or use a regular wire shunt that is temp stable at these currents.

That's a great idea... or I could just add many lower value in serie to the 1.36 equiv actual resistor that I could use a selector to set the non inverting input to the desired node between many serie 0.1 milliohm value ?

let say a 1.36 milliohm resistor in serie with like 6x 0.1 milliohm in serie.. I could go from a range of 1.36 to1.96 milliohm..

or just add a 0.5 milliohm resistor in serie and add a 10 ohm pot in parallel to that 0.5 milliohm and connect the wiper to the non inverting input of the op amp?

Doc

 

[Doctorbass]

ex:

 

[dnmun]

much more elegant. use the trim pot to find the spot where you don't throw the breaker and then measure that and use a fixed resistor on a switch. except the circuit doesn't look right now.

 

[steveo]

did it end up working?

-steveo

 

[Doctorbass]

did it end up working?

-steveo

not tried yet, will do it soon :wink: