Meanwell RSP-1500 as a charger

[andreym]

I recently got RSP-1500-48 from e-bay

so yesterday i finally got it (it usually takes long to ship overseas)
so the first thing i checked is the maximum voltage (it is regulated with a pot to max around 59volts) - that is great for my 16s lifepo4 a123 pack i set it at about 57.6
the next test was to try charge the pack, so i discharged it a bit with about 200w load for half an hour and set it to charge
the current at start was about 39.9amps (according to turnigy wattmeter) and after about 2 minutes it was starting to tapers and ended at 0.5amps when i disconnected the pack (put in a bit less than 2ah in about 3-4minutes)
so my question is:
-is it safe for PSU to hold about 2100w of load? (it is rated to 1500 and i remember Doc was using it at about 1700w)
-can i reduce the current by modifying the shunt or anything else? (had no time to open it and check it out yet)

 

[liveforphysics]

You could file down the shunt area a bit if you like.

If I were you, I would just make sure it's got excellent airflow, maybe add an external fan, then just run it.

 

[andreym]

You could file down the shunt area a bit if you like.

If I were you, I would just make sure it's got excellent airflow, maybe add an external fan, then just run it.

you should hear it! it is already working(sounds) like a vacuum cleaner
I`m not sure i can fit there one more fan,
I`ll try to open it and see what`s inside. because 2+kw charger with only 2.6kg weight is really great thing
the other side is when i charge my a123 prismatic 15ah pack with 30amps it is only 2C but 40amps is a bit high i think, i want them to serve at least several years.
btw, LFP, what is your experience with cell_man`s cells charging? are they able to eat 2-3C charge current while connected together without heating up?
when i tested those capacity i charged them at 20 amps and they were not even warm but 40amps it is about 0.8v vdrop due to IR (about 20milliOhm the whole pack) and about 35watts of pure heat. is it fine for 7.5kg pack that is almost completely sealed and no airflow inside?

 

[liveforphysics]

The highest I was able to charge them was 5c (because I didnt have a bigger current source), and they took it with no problem. Just luke-warm.

I dont see 2c being a problem in any pack design with these cells.

 

[andreym]

The highest I was able to charge them was 5c (because I didnt have a bigger current source), and they took it with no problem. Just luke-warm.

I dont see 2c being a problem in any pack design with these cells.

Thanks for info!
I`ll notice that and maybe i`ll leave the current at current level ))) if i`ll find out how to improve cooling

 

[icecube57]

I dont know about your power supply but i have my regular 350w mean well putting out 450-500w. It hasnt kicked the bucket. Its current limited to 11-12A.

 

[jag]

I`ll try to open it and see what`s inside. because 2+kw charger with only 2.6kg weight is really great thing
the other side is when i charge my a123 prismatic 15ah pack with 30amps it is only 2C but 40amps is a bit high i think, i want them to serve at least several years.

2.6kg for 2kW is really good. (and even the 1.5kW nominal rating) I wish I had that, or even better a 1kW, 1.3kg charger to carry on the bike. The SP320 and S-350 are both about 1kg, and can do near 500W (for a while), so not as good.

I did an ebay search but didn't turn up any more RSP-1500. Mind if I ask how much you payed?

Anyone else found affordable chargers with better than 500W/kg?

 

[andreym]

I did an ebay search but didn't turn up any more RSP-1500. Mind if I ask how much you payed?

Anyone else found affordable chargers with better than 500W/kg?

here is the listing i got it form http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=350387849609&ssPageName=STRK:MEWNX:IT
but it cost me about 50 bucks more to send it overseas from my friend in US
usually they retails at about 500usd+ so i believe i got a very good deal

 

[andreym]

I got some pictures of the parts inside...
and have some questions:
- fisrt of all, Doc, you told that your PSU have fets instead of diodes at the output line, but i did not found any fets there just 4 dual diodes which looks like fets because they have similar package (TO-247 i believe) and spare places for 4 of them more (i think i could solder more of them for lower heat)
- there are 2 big transformers on each board (2 boards total) and each transformer have tf1200(low voltage board) and tf-1198(HV board) ratings, so i assume they are rated to ~1200w each, and 2 of them are for 2400w max total? am i right? or this digits does not related to power?
- why do they use 2 transformers on the high voltage board? is it making 250v from any input voltage and than LV board makes this volts to 48 output? but i wonder why it is required to convert voltage twice? does this mean that the actual efficiency could be even more?

See pictures attached

 

[Doctorbass]

Hi, Good deal you had!! about the same price i got for mine 2 years ago!

i'm surprised taht you get 2100W Dc out !!! i can get 1700 max.. but from a normal 120V ac outlet... maybe you connect it on the 220V ?

My 1700W limit is the 18A i draw from the 15A 120Vac circuit breaker of the house grid :lol:

These great charger are more efficient because they have a PFC device. This is a Power factor corrector... and it correct the apparent power and active power ratio to be near 1.00 wich is excellent and allow you to get above 90% efficiency of the AC to DC conversion.

Normally what ameliorate the efficiency of a SMPS are teh PFC and the rectifier desing. instead of using diodes to recover the DC, they use fets. The diode have a V forward voltage higher so it make higher heat lost with high current circuit than if using fets that close and open depending on the phase of the Ac current to to simulate a diode.

The RSP serie are nothing compare than the 350 or 500W meanwell desing... the RSP are the best desing they made. and the highest W per cu inch.

Doc

 

[dnmun]

doc is so right about the diodes. they get super hot too, and they have to put a big low ohm power resistor and cap across the diode to absorb the noise of it slamming shut every time too which also soaks up a lot of energy. i think that's what they call a snubber.

 

[andreym]

Hi, Good deal you had!! about the same price i got for mine 2 years ago!

i'm surprised taht you get 2100W Dc out !!! i can get 1700 max.. but from a normal 120V ac outlet... maybe you connect it on the 220V ?

My 1700W limit is the 18A i draw from the 15A 120Vac circuit breaker of the house grid :lol:

These great charger are more efficient because they have a PFC device. This is a Power factor corrector... and it correct the apparent power and active power ratio to be near 1.00 wich is excellent and allow you to get above 90% efficiency of the AC to DC conversion.

Normally what ameliorate the efficiency of a SMPS are teh PFC and the rectifier desing. instead of using diodes to recover the DC, they use fets. The diode have a V forward voltage higher so it make higher heat lost with high current circuit than if using fets that close and open depending on the phase of the Ac current to to simulate a diode.

The RSP serie are nothing compare than the 350 or 500W meanwell desing... the RSP are the best desing they made. and the highest W per cu inch.

Doc

I know about diodes Vdrop and fets rectifiers but what i`m telling is in this charger there are simple diodes and not the fets
i believe they maybe use fets in lower voltage psu-s(because at higher currents there are really very big losses) but in my 48v version there are doubled diodes in to-247 package (see pictures)

 

[Doctorbass]

Hi, Good deal you had!! about the same price i got for mine 2 years ago!

i'm surprised taht you get 2100W Dc out !!! i can get 1700 max.. but from a normal 120V ac outlet... maybe you connect it on the 220V ?

My 1700W limit is the 18A i draw from the 15A 120Vac circuit breaker of the house grid :lol:

These great charger are more efficient because they have a PFC device. This is a Power factor corrector... and it correct the apparent power and active power ratio to be near 1.00 wich is excellent and allow you to get above 90% efficiency of the AC to DC conversion.

Normally what ameliorate the efficiency of a SMPS are teh PFC and the rectifier desing. instead of using diodes to recover the DC, they use fets. The diode have a V forward voltage higher so it make higher heat lost with high current circuit than if using fets that close and open depending on the phase of the Ac current to to simulate a diode.

The RSP serie are nothing compare than the 350 or 500W meanwell desing... the RSP are the best desing they made. and the highest W per cu inch.

Doc

I know about diodes Vdrop and fets rectifiers but what i`m telling is in this charger there are simple diodes and not the fets
i believe they maybe use fets in lower voltage psu-s(because at higher currents there are really very big losses) but in my 48v version there are doubled diodes in to-247 package (see pictures)

Have you checked in the lower pcb ? ( the PFC that convert from 120Vac to 400V DC)

Sincerly, i never checked for these ideal diodes in my RSP-1500... I was so impressed by the performances and efficiency than i concluded that it probably use Ideal diodes desing as rectifier.

What i concluded about the output power limit is that it is proportional to the input power availlable.. at least for 120V input... The output is CC-CV where the CV is adjusted by the availlable range and the CC depend on the input power with a plateau at 1700W for mine. I never got above 37A out of mine with a 10% soc battery connected to it ( near completly discharged with very low voltage ).. the moment where the 1700W output is the multiplication of lower voltage with higher current..

I know they are power limited and current limited.
I would say that it's first power limited ( to protect the entire circuit)... than current limited( to protect the output drive) and finally voltage limited( offered adjustable range)

that might be why you get 2100W with 220V..? is it on 220V input?.. I suspect that this desing vary the max limited output power depending on the input voltage as well...

I think it have also a shunt that measure the current of the 400V DC section as output current sense because it seem to have no shunt on the output...


I have to say that on mine i added a diode in serie to proetct the output against reverse voltage or to protect against overvoltage battery i might connect to the output... ( it already happened to me.. I connect my mongoose setup that was contacted in 100V mode!.. to the 50V output of it!... and i blown the OVP scr and zener near the output stage!
since i did that i never had any problems!. The diode i added is a 40A 100V shootky in serie with the positive output brace. I have cutted that brace in two half inside the psu and have installed the diode directly to it. The brace act also as heatsink and since there is output vent neat the brace it cool down perfectly. I also have an extension cord of dual 10 gauge of 10 ft on the output that end with PP30 connectors. The voltage sensing also have two little 22 gauge wire of 10 ft and i sense the output directly on the PP30 connector so the CC-CV is better because it take account of the voltage drop in the 10ft of the 10Awg cable :wink:

Doc

 

[andreym]

that might be why you get 2100W with 220V..? is it on 220V input?.. I suspect that this desing vary the max limited output power depending on the input voltage as well...

I think it have also a shunt that measure the current of the 400V DC section as output current sense because it seem to have no shunt on the output...


I have to say that on mine i added a diode in serie to proetct the output against reverse voltage or to protect against overvoltage battery i might connect to the output... ( it already happened to me.. I connect my mongoose setup that was contacted in 100V mode!.. to the 50V output of it!... and i blown the OVP scr and zener near the output stage!
since i did that i never had any problems!. The diode i added is a 40A 100V shootky in serie with the positive output brace. I have cutted that brace in two half inside the psu and have installed the diode directly to it. The brace act also as heatsink and since there is output vent neat the brace it cool down perfectly. I also have an extension cord of dual 10 gauge of 10 ft on the output that end with PP30 connectors. The voltage sensing also have two little 22 gauge wire of 10 ft and i sense the output directly on the PP30 connector so the CC-CV is better because it take account of the voltage drop in the 10ft of the 10Awg cable :wink:

Doc

Yes - i live in Russia and the voltage in the outlet is 220v nominal
i believe that with good airflow as advised by Luke it will be ok to handle 2100w continious (i`m still not sure about turnigy watt-meter precision and i`ll check it with CA later)
and i really do not care if the caps will not handle 65khours of continius working because it`ll take only about 20-25 minutes to fill in my 16S1p 15ah cell_man`s cells (i`m planning to double it but a bit later - cells are here i just need time to build the second pack)

I`ll probably also won`t mess with sensing wires as they are only usefull at the end of charge while first seconds of CV phase and after the current drops the wires impedance does not affect final voltage so much.

The next step is to try using eus25-096 bus converters to be able to charge 60v-72v batteries with this monster charger )
btw - Doc, you could probably try them to charge your 24s lipo
they are 1/5 convert ratio, isolated and can handle about 30amps output current (see pdf) and i got some of them on e-bay for cheap some time ago

 

[Doctorbass]

that might be why you get 2100W with 220V..? is it on 220V input?.. I suspect that this desing vary the max limited output power depending on the input voltage as well...

I think it have also a shunt that measure the current of the 400V DC section as output current sense because it seem to have no shunt on the output...


I have to say that on mine i added a diode in serie to proetct the output against reverse voltage or to protect against overvoltage battery i might connect to the output... ( it already happened to me.. I connect my mongoose setup that was contacted in 100V mode!.. to the 50V output of it!... and i blown the OVP scr and zener near the output stage!
since i did that i never had any problems!. The diode i added is a 40A 100V shootky in serie with the positive output brace. I have cutted that brace in two half inside the psu and have installed the diode directly to it. The brace act also as heatsink and since there is output vent neat the brace it cool down perfectly. I also have an extension cord of dual 10 gauge of 10 ft on the output that end with PP30 connectors. The voltage sensing also have two little 22 gauge wire of 10 ft and i sense the output directly on the PP30 connector so the CC-CV is better because it take account of the voltage drop in the 10ft of the 10Awg cable :wink:

Doc

Yes - i live in Russia and the voltage in the outlet is 220v nominal
i believe that with good airflow as advised by Luke it will be ok to handle 2100w continious (i`m still not sure about turnigy watt-meter precision and i`ll check it with CA later)
and i really do not care if the caps will not handle 65khours of continius working because it`ll take only about 20-25 minutes to fill in my 16S1p 15ah cell_man`s cells (i`m planning to double it but a bit later - cells are here i just need time to build the second pack)

I`ll probably also won`t mess with sensing wires as they are only usefull at the end of charge while first seconds of CV phase and after the current drops the wires impedance does not affect final voltage so much.

The next step is to try using eus25-096 bus converters to be able to charge 60v-72v batteries with this monster charger )
btw - Doc, you could probably try them to charge your 24s lipo
they are 1/5 convert ratio, isolated and can handle about 30amps output current (see pdf) and i got some of them on e-bay for cheap some time ago

With the 2 and half years experience i have with the RSP-1500 and the remote sensing connection for charging the Konion 25s packs i can say that it really reduce the charging toime for the CV step!.. seriously!

I'm already using a DC-DC converter ( 3x arthesyn BXB150 dc-dc converter) wired in serie to gain 15V dc at 30A. I just connect the output of the meanwell to each of the 3 input of the DC-DC and than i connect the 3 output in serie with the meanwell in serie with all... so that boost from 55V to 70V if i want.

The arthesyn work really nice!.. I also used 12 of them for a 12s Active Balance charger that charge at 20A with 12 individual 1s output that i can switch for 3.6V or 4.2V each. I modified their over current limit by reverse ingeneering them. so i just need to connect my 1500W meanwell to the 48V input and i charge and balance safely any 12s packs without bleeding current with resistors.

What is the difference between bus converters and DC-DC converter? is it the same?

Doc

 

[andreym]

...
What is the difference between bus converters and DC-DC converter? is it the same?

Doc

The difference is usual dcdc`s have fixed output voltage(with wide input range) and efficiency about 91-92%MAX (usually 85-90 depending on %of load)
and these converters are fixed ratio (i mean if i power them with 50volts i get 10volts out and if input is 55v, the output would be 11v)
also really great efficiency of >94% most of load range and incredibly small size!

 

[Doctorbass]

...
What is the difference between bus converters and DC-DC converter? is it the same?

Doc

The difference is usual dcdc`s have fixed output voltage(with wide input range) and efficiency about 91-92%MAX (usually 85-90 depending on %of load)
and these converters are fixed ratio (i mean if i power them with 50volts i get 10volts out and if input is 55v, the output would be 11v)
also really great efficiency of >94% most of load range and incredibly small size!

I see.. so it dont have the fixed V regulator section... so could we conclude that it have a fixed PWM? ... or also it might be only variable for the constant current section?

Doc

 

[jag]

Here's a useful summary on the different Meanwell series and models:
http://users.tpg.com.au/users/p8king/s100.htm

Much easier to get an overview than on Meanwell's own site. On meanwell's site I have to download individual spec sheets, then make a table myself to compare. Here it's all on one page.

Keep a look out for the new RSP-2000. 42A, 48V. Finally 20 min charging for those of us with cellman A123 pouches! Form factor looks slender, more like the the smaller S and SP series than the other boxy RSP and PSP, so might fit well on a bike rack. Weight is still TBA. I'm hoping for something in the 2kg class like the RSP-1500.

 

[Doctorbass]

Here's a useful summary on the different Meanwell series and models:
http://users.tpg.com.au/users/p8king/s100.htm

Much easier to get an overview than on Meanwell's own site. On meanwell's site I have to download individual spec sheets, then make a table myself to compare. Here it's all on one page.

Keep a look out for the new RSP-2000. 42A, 48V. Finally 20 min charging for those of us with cellman A123 pouches! Form factor looks slender, more like the the smaller S and SP series than the other boxy RSP and PSP, so might fit well on a bike rack. Weight is still TBA. I'm hoping for something in the 2kg class like the RSP-1500.

Excellent usefull info!! Thanks!!

Looking forward about the RSP-2000 if 1U size with 21.4W per cu inch !!!

Doc