Having trouble charging with Meanwell RSP-750-27

[Roberto Aguilar]

Hello,

I'm hoping someone can help me debug an issue I'm having with the Meanwell RSP-750-27.  Even though the ElectroDacus shows that the charging FET is on, I don't see the batteries getting charged.

I made two adjustments to the connector the charger's board:

First, I plug pin 14 to EXTIO4+ and pin 13 to EXTIO4- so that the ElectroDacus can turn the charger on and off.

I also wanted to reduce the current from the charger so I connected pin 7 to an Arduino and set it to output 2V.  I wanted to confirm the pin connection in the diagram below from the power supply's PDF [1]; I assumed that pins 7 and 11 are connected via a pull down resistor, so I put a 10k resistor between them.  Does this sound correct?

I watched the charger charge the batteries once and then it didn't charge again once the batteries were drained again.  They are currently at 3.0V and the charger will not send voltage.

I can hear that the fans are on.  I can also see the LED on the charger is green when I have the breaker on.  If I turn the breaker off the LED on the charger goes red.  Visually, everything looks okay.

Anything that I can check to debug further?

Thank you,

-Roberto.

[1] https://www.meanwell.com/Upload/PDF/RSP-750/RSP-750-SPEC.PDF

[Nils]

Hi Roberto,

I don't use this charger in particular. But every time something didn't work with the EXIOs for me it was because I had the to be controlled devices connected with the wrong polarity. As far as I understand the SMBS0 uses opto couplers to switch the remote devices. Those only will switch your control signal if polarity is correct (it's a diode).

It seems unlikely, though, as you said it worked once. Unless you fiddled around with the cable config since then... ;)

Hope it helps. 

Best,

Nils

[Roberto Aguilar]

Good idea; I will check this.  There's been all sorts of fiddling, so this is possible.  I will report back.

Thank you!

[Dacian Todea]

The pin 7 and 11 require an external DC voltage not a resistor.

A photo of the SBMS main screen may be useful to see if there is any other problem that will not allow charger to be enabled.

Maybe grid power was down or you removed the grid connection then reinsert. From what I understand the grid needs to be connected first then the charger will need to connect to battery else charging will not be enabled.

This charger is more useful for occasional charging from the grid. It will not make sense to charge the battery from grid all the time as if you have access to grid you can directly use the grid much more efficient than grid to battery then battery to inverter maybe as low as 60 or 70% efficiency depending on the inverter and charger efficiency.

[Craig Mckenzie]

good morning

I have the Meanwell Rsp 750-15.

i have had the same problem using mine. 120vac to power to unit, fans come on but no output.

i found a guy on youtube that used a watt meter to get the Meanwell to work.

so i bought a 150 a watt meter from amazon and hooked it up and run with no battery connected. see up my voltage to 15v in my case.

them hooked it up inline with the battery cables. worked perfectly getting 50 amps 

Works as expected. talking with the tech from digi - key he said it needs a load to start the charging. can't figure out why my lifepo4 560 ah battery wouldn't have created enough of a load to start the unit but it wouldn't. soon has i used the watt meter inline it worked perfectly.

Jason from everlanders on you tube shows you how to install and install a potentiometer to control the charge rate and have the Sbms0 control the Meanwell charger.Here is a link .

https://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&cad=rja&uact=8&ved=2ahUKEwiSz6SDqpn4AhVOiI4IHT5cC7IQtwJ6BAgCEAI&url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DA_BT6XLMGFM&usg=AOvVaw0JSY2SlkJMAeoeoG2PHe1A

[Dacian Todea]

Is that a DC or AC wattmeter ? What model exactly. It will not make sense for a wattmeter to do anything as all you are adding is in case of a DC wattmeter a shunt so just a bit of resistance. If that is what works you can do that by adding a bit of a longer cable that is just thick enough to deal with the current. So maybe 1m of 10AWG cable. What cable where you using and how long before installing that wattmeter ?

[Craig Mckenzie]

Just in case the other link doesn't work to Jason from Everlanders is here. 
https://youtu.be/A_BT6XLMGFM

Jason from Everlanders has great ideas with fantastic video content. 

[Dacian Todea]

Yes it seems using the build in 6V reference you can set the max charge current.   He used a potentiometer. Someone I think asked about arduino controlling this and that can be done with maybe a few small SSR/optoisolators and a few resistor dividers set for the current values you prefer.

He had the problem of not knowing how much he can ask from the outlets he will find available but if you do not have that problem and outlet can handle 750W (all should be able to handle that) then there is no real reason to reduce the charge current.

[Craig Mckenzie]

i added a #4 cable to get enough resistance to start the dc charger to work but that did not work. but when i used a dc watt meter inline with the battery cable it worked as expected.Very strange? i thought their would be enough of a load with the battery cable. but seeing online a lot of people have had issues getting the unit to start. Anyway this watt meter was enough of a load to get the unit to work every time.

[Roberto Aguilar]

I want to reduce the charge current to more slowly charge the batteries.  At the moment without reducing the charger's current I would be charging the batteries at about 1Amp and I would like to keep it below 500 milliamps to treat the batteries a little nicer.

And, you're right, the outlet I have the charger can easily handle it at full current.

Thank you,

-Roberto.

[Dacian Todea]

#4 is overkill for 50 or 60A and #6 will be more than sufficient but the length of the cable is also relevant. If cable is to short and thick resistance may be to low and with the LiFePO4 super low internal resistance there may be some overcurrent at power up that is exceeded.

It can not be about the load of the wattmeter just about the resistance and that5 wattmeter clearly has much thinner wires plus the shunt resistance.  So is very likely a problem of oversized and short cables between battery and charger.

I think that power meter only has 12AWG wires.  So about 1m of 10AWG wire will likely be enough to do the same thing and cost way less than the unnecessary power meter.

You can not reduce the current from that charger to 1A. In spec they say minim 40% so  60A *0.4 = 24A but I think in the everlander video he mentioned 6A minimum so 10% of max using that potentiometer.

Battery will prefer the max current and stop charging when full rather than very small current. As long as battery can handle 60A so you have at least a 250Ah battery then ideal will be 60A not 6A or 1A as 1A charging will be more detrimental to battery than 60A.

20A will be fine but 1A charging will be way to low so battery will be forced charged much more. 

[Roberto Aguilar]

This video is great, thank you!  The one thing I noticed is that he's using a pot to control the voltage from the supply itself, while my voltage is external.  I'm going to re-inspect all my connections when I have a chance later today and report back.

Thanks again!

[tobi]

I use a dssr20 for connection of the meanwell to the battery. Since then, all startup problems are gone.

Tobi

[tobi]

I forgot to mention if you do not limit the current you have to use 2xDSSR20.

[Roberto Aguilar]

Thanks, everyone, for chiming in and helping me with this.  I have some good news and a bit of a somewhat anti-climactic solution to my problem.

The good news is that the charger is working again!  🙌🏽

On the less satisfying side, the problem fixed itself when I adjusted the power cables connected to the power supply.  The lugs holes on the wires I used might simply be too big for a good contact.  After adjusting them and clamping them down a second time, the power supply magically worked.

I did clean up a couple of other things while debugging this.  Even though I saw the RSP-750 charge my batteries once, I did have a wiring issue with my current control wiring.  Per Dacian's suggestion (sorry, I accidentally private messaged him), pin 7 goes to my device's voltage source and pin 11 goes to my device's ground.  I'm using an Arduino to send 2V so that it lowers the charge current for the time being since my battery pack isn't very big.  Once I add more batteries I will adjust the voltage accordingly.  Thank you, Dacian!

I still don't fully understand the role wire thickness plays here.  One of the suggestions is that my wire thickness + short length may play a role in preventing the charger from kicking on.  I am having trouble conceptualizing why.  That said, I will replace my 6AWG wires with 8 AWG wires that I have lying around and in the process make sure the lug connector has a better fitting hole.  Sounds like 10 AWG is even better in this case, but I don't have any handy.

Thank you again, everyone for the help; I'm learning quite a bit in this process.

-Roberto.

[Dacian Todea]

Roberto,

I do not have this charger but a few people that have this have reported that it will not start charging if the charger is first connected to battery then connected to grid.

And someone recently mentioned that by adding a DC power meter between battery and charger solved the problem of charger not starting when connected to AC supply. If that is true it means that when connected to AC while already connected to battery there is some sort of surge current that will deactivate the charger and the only thing that wattmeter was doing is reduce the resistance as the wattmeter had 12AWG wires and a series current shunt internally thus just extra resistance added to the circuit.

Since a wattmeter is more expensive than a bit of wire the solution for those that need to make the connection in this order is to have a cable with sufficient resistance for the surge current not to  be a problem and deactivate the charger.

For your case where max charge current is 28A any 10AWG cable will work and even 12AWG if is silicone insulated cable.  

Probably this supply was designed for Lead Acid batteries that have much higher internal impedance than LiFePO4 and going also with a very short and overkill cable thickness means some sort of inrush current limit is exceeded as it works when extra resistance is added like in the case of adding that DC wattmeter in series.

[Roberto Aguilar]

Thanks for the additional comment here.  You are spot on.  Turning the breaker between the RSP-750 and the batteries off first, then plugging it into AC power, and finally turning the breaker back on will allow it to operate normally.

While experimenting, I did notice that with the charger unplugged from AC, I saw a 25W power draw that lasted a second or so on the ElectroDacus when turning the breaker on.  At this point when plugging the charger into AC did not allow it to charge.  

There is an LED that is green when the charger is in a good state and goes red when gets into a bad state.  I have seen the LED go red when the breaker is on, the charger is plugged into AC and then the breaker is turned off.  When in this state, unplugging from AC, waiting for the light to turn off (the fans will also keep spinning for a few seconds after unplugging from AC), and then plugging back to AC will put the power supply in working order.  The breaker can then be turned on for charging to take place.

Thanks again.

[Ross Aspinall]

I had the  meanwell RSP-750-15 in a RV Mercedes Sprinter setup while cost effective for the output I returned it for refund.  3 Issues...

1.  you have to power on then connect to battery it wont just charge immediately when connected to battery then turn on the 240vac 

2.  Super noisy like a portable Dyson cleaner LOL the fans stay on ALL the time   

3.  Susceptible to damage from the environments i.e dust and insects as it is all open air to the PCB 

I ended up going for a Renogy 20amp https://au.renogy.com/20a-ac-to-dc-lfp-portable-battery-charger/ totally sealded has its own 2 stage Auto charging 

[DeBommel]

@Dacian: it might be a good idea to delete the Meanwell's as a grid-charger in the SBMS0-manual. They just do not work well with lithium. (I found out the costly way.)

Op donderdag 28 juli 2022 om 04:08:15 UTC+2 schreef causingpo...@gmail.com:

[Dacian Todea]

Unfortunately I do not have one to test myself. Based on description it seems there is a short circuit current limit that is activated when people use extremely thick and short wires to connect to battery and that prevents the power supply to charge if connected first to battery then to grid.

I do not know what that current limit is else I could specify exactly what the resistance of the wires need to be not to trigger that short circuit limit.

Meanwell is a fairly high quality brand for power supplies.

Is this the problem you are having or do you experience something else ?

[Roberto Aguilar]

Can you expand on, "they just do not work well with lithium".  The charger does charge lithium; I have gotten it to charge just fine.  I would like to make some refinements in order to make it more seamless, but making sure it's plugged into AC before plugging it into the batteries solved the mystery for me.

Rather than removing it from the documentation entirely, I would propose updating the documentation so that it explains the order of operations.

Thank you,

-Roberto.

[Roberto Aguilar]

I do have this charger; it's the one I'm using to charge my battery backup.

@Dacian, if you provide me with some wire thicknesses and lengths to try out I can give them a try.  Please note, this may be a slow process as I would have to procure wires.

[Dacian Todea]

Roberto,

What model are you using 15 or 27V and what wires length and thickness are you using now ?

[Roberto Aguilar]

I am using the RSP-750-27; the 27V model.  Using the diagram on page 3 of SBMS0v03d.pdf as reference:

- the positive wire connecting the Meanwell to the PVp breaker is 11 inch (28 cm) #6 gauge

- the negative wire connecting the Meanwell to a busbar is 30 inch (76 cm) #6 gauge

I have #8 gauge wire that I can experiment with, otherwise I'll have to procure any other gauge.  Attached / below are the cable specs that I'm using if it helps.

Thanks,

-Roberto.

[Dacian Todea]

Roberto,

Yes #6 gauge is absolute overkill and wires are also extremely short. This in combination with LiFePO4 that has ultra low internal impedance is the cause of the unit not powering ON if plugged in to AC after it is connected to battery.

Even #8 AWG is to thick about half of #6

You will want something like 10mOhm minimum so 2m of 12AWG (will need cable insulation over 105C or silicone) or 3m of 10AWG. 

It will need 5m of 8AWG to have the same 10mOhm if you want to experiment with that.

The extra cable inductance will likely also be helpful.

When I say 2m of 12AWG I mean total so 1m on the positive and 1m on the negative or some other combination.

The way it is connected now it probably thinks that output capacitors are shorted with such a low resistance and inductance.

You have just about 1m of #6 gauge that's probably around 1.3mOhm and 1mH so depending on your battery that may have maybe 2mOhm so that will mean abut 8kA at power up.

[Roberto Aguilar]

Excellent, thank you for the info.  12AWG silicone insulated automotive wire is on the way.  I'll post an update soon.

Thanks,

-Roberto.