-
Howdy! we're looking for donations to finish custom knowledgebase software for this forum. Please see our Funding drive thread
Server power supplies in series!
- Thread starter auraslip
- Start date
auraslip
10 MW
- Joined
- Mar 5, 2010
- Messages
- 3,535
Well I got it done in it's hopefully final form for a bit. Put two s-150-27 units in parallel and then in series. Had to set each to the same voltage and current limiting. About 10a. I used to run them at 6.5a in series to charge to 16s ping, but that is above their 5.5a rating. I actively cooled them with a fan, and this worked well for 6 months.
The s-48-350 now works well charging the ping. It still makes the hiss when it's current limiting above 5a though.
Any ways, better video now.
Also, I kept shocking myself while I was working on the units. Don't know why. The line is broken, so it much of been coming through the neutral and ground. That's stupid. I need to install a DPST switch to break the L and the N.
[youtube]fNyiekUMTfU[/youtube]
The s-48-350 now works well charging the ping. It still makes the hiss when it's current limiting above 5a though.
Any ways, better video now.
Also, I kept shocking myself while I was working on the units. Don't know why. The line is broken, so it much of been coming through the neutral and ground. That's stupid. I need to install a DPST switch to break the L and the N.
[youtube]fNyiekUMTfU[/youtube]
auraslip
10 MW
- Joined
- Mar 5, 2010
- Messages
- 3,535
my god. what a colossal waste of time and money. the s-27-150 units shut down when the battery pack is completely discharged. I guess they can't take dropping 11 volts.
I could just have bought two standalone chargers by now, and used the extra time I spent working on this to find a job.
I could just have bought two standalone chargers by now, and used the extra time I spent working on this to find a job.
dennyt
10 W
Here are some photos of the Dell DPS-500-CB chargers I'm using. 12V 41A.
I connected the pins in a pattern like this:
N N N N N N
N N N N Y N
N N N Y N N
N N Y N N N
I isolated the -12V from chassis ground by cutting two jumpers, J920 and J921 on the upper & lower halves of the case, near the output end.
I connected the pins in a pattern like this:
N N N N N N
N N N N Y N
N N N Y N N
N N Y N N N
I isolated the -12V from chassis ground by cutting two jumpers, J920 and J921 on the upper & lower halves of the case, near the output end.
auraslip
10 MW
- Joined
- Mar 5, 2010
- Messages
- 3,535
I finally got around to writing an article about this on my site https://sites.google.com/site/shelbyelectro/Mods-and-technical/batteries/server-psu-charger
ZOMGVTEK - I used some of your pictures. I hope it's ok. If not let me know. I figured they were a much better example than mine!
ZOMGVTEK - I used some of your pictures. I hope it's ok. If not let me know. I figured they were a much better example than mine!
auraslip
10 MW
- Joined
- Mar 5, 2010
- Messages
- 3,535
Well the stack has worked great for many months, but it's developed a problem.
It'll push current for 20 seconds, and then one unit shuts down and flicks on and off for about 10 seconds before coming back online. Don't know what's causing it. It still works, but is just a slower charge.
Luckily, I have a spare unmodded unit, and another unused unit in the stack, so I just need to isolate the bad unit and swap it out.
It'll push current for 20 seconds, and then one unit shuts down and flicks on and off for about 10 seconds before coming back online. Don't know what's causing it. It still works, but is just a slower charge.
Luckily, I have a spare unmodded unit, and another unused unit in the stack, so I just need to isolate the bad unit and swap it out.
auraslip
10 MW
- Joined
- Mar 5, 2010
- Messages
- 3,535
Actually that's a good point. Before I was pushing the meanwell to it's max....
but on my new bike I made a mistake and made the charge wires from 22awg. And since it's like 6 feet long (through the rear rack and bag where it's stored all the way down to the triangle and then back up inside it) They were getting warm, so I turned the meanwell down to 4.5a. Maybe that's not enough power to make it stay on.
OTOH, if that were the case It'd be shutting down at the end of the charge cycle.
So to answer your question, no. These are rated at 12v60a, but I've only done 8a max.
but on my new bike I made a mistake and made the charge wires from 22awg. And since it's like 6 feet long (through the rear rack and bag where it's stored all the way down to the triangle and then back up inside it) They were getting warm, so I turned the meanwell down to 4.5a. Maybe that's not enough power to make it stay on.
OTOH, if that were the case It'd be shutting down at the end of the charge cycle.
So to answer your question, no. These are rated at 12v60a, but I've only done 8a max.
Oh wow, 22 AWG charge wires? Even 4A is pushing that wire.
I wasn't speaking electrical, more physical. Hitting stuff thats broken sometimes fixes it in my experience. Sometimes it just makes it more broken.
A 60A supply should be happy supplying 4A or so, the only issue I could see is how the setup is being current limited. You sure the meanwell isn't doing funny stuff dialing down the current that low? You were running it for a while exactly as it sits I presume?
I wasn't speaking electrical, more physical. Hitting stuff thats broken sometimes fixes it in my experience. Sometimes it just makes it more broken.
A 60A supply should be happy supplying 4A or so, the only issue I could see is how the setup is being current limited. You sure the meanwell isn't doing funny stuff dialing down the current that low? You were running it for a while exactly as it sits I presume?
icecube57
10 MW
Some of those supplies have LVCs also...how low did you tune it....auraslip said:Ran it for like 6 months until I adjusted the power down and started having problems. Will try to knock it with a mallet
auraslip said:Ran it for like 6 months until I adjusted the power down and started having problems. Will try to knock it with a mallet
Do you think it might have more to do with how the supplies are current limited, rather than the supplies themselves?
Limiting a bunch of series supplies with a single psuedo CC supply that makes up a minimal amount of the total series voltage is not like a lab power supply. It's limiting the current by limiting the voltage, which may or may not be possible for the voltage range of the supply and the range your pack can be in.
I'm suspecting the low current is the issue.
pguk
100 W
Been reading up on this topic today, came across this instructable to convert a PC smps to make an adjustable supply:
http://www.instructables.com/id/Adjustable-High-Current-Lab-SMPS-Out-of-a-Standard/?ALLSTEPS
Apparently this hack is very common as they have a dedicated page listing over 70 instructs.
http://www.instructables.com/id/Encyclopedia-of-ATX-to-Bench-Power-Supply-Conversi/
I haven't yet looked thro to see if one of these has adjustable current, needed for this application as described above.
Edit: actually these ATX PSUs are not robust enough for reliable use. Perhaps the same steps can be taken on a server PSU as for conversion of an ATX ?
Edit ii: the meanwell and clones are the best solution. Fechter has designed a circuit to easily adapt to use as a charger.
https://endless-sphere.com/forums/viewtopic.php?f=31&t=36958
http://www.instructables.com/id/Adjustable-High-Current-Lab-SMPS-Out-of-a-Standard/?ALLSTEPS
Apparently this hack is very common as they have a dedicated page listing over 70 instructs.
http://www.instructables.com/id/Encyclopedia-of-ATX-to-Bench-Power-Supply-Conversi/
I haven't yet looked thro to see if one of these has adjustable current, needed for this application as described above.
Edit: actually these ATX PSUs are not robust enough for reliable use. Perhaps the same steps can be taken on a server PSU as for conversion of an ATX ?
Edit ii: the meanwell and clones are the best solution. Fechter has designed a circuit to easily adapt to use as a charger.
https://endless-sphere.com/forums/viewtopic.php?f=31&t=36958
Similar threads
