Can I use my PSU as a 12v battery charger?

[TheNiceGuy]

Hmmm, 2nd go display became distorted at startup. Can't read it now. Related to psu?

 

[PottedMeat]

Hmmm, 2nd go display became distorted at startup. Can't read it now. Related to psu?

Maybe the voltage on the PSU dipped below 11V. Try disconnecting the charger from everything and letting it sit for ~5 minutes.

Also try attaching a +11 to +18V wall wart ( doesn't have to be 6 Amp rated ) without any batteries attached. If that clears it up, then the PSU isn't giving clean power to the charger.

 

[TheNiceGuy]

Thanks for the help. Tried a few things and didn't help. Sent it back yesterday.

I have a totally different, but related question. I made what's called a foam hotwire cutter from the 12v line. Basically, you make a bow strung with steal wire, connect the 12v line with alligator clips and the wire gets hot enough to cut foam. You can adjust the temp by moving the clips closer or further apart along the steal wire.
Mine works great, but is a bit too cool to use at 3+ feet apart.
Is it amps or volts that are affecting the temperature? Is there a way to increase the power going into the wire? The PSU has "dual 12v lines". Does that mean I could hook up 2 and double the power?
Any one know about this stuff?

Thanks again.

 

[bobdole369]

Ohms law uber alles.

 

[TheNiceGuy]

??

 

[PottedMeat]

??

V = I R

P = I V

Since your voltage is fixed you will have to vary the resistance to get the power output you want.

You said the wire is too cool to use at 3' so try using a little thicker wire. If it's too thick you'll probably set off the overcurrent protection. I don't really know what they use for foam cutters but I'd bet it's something like nichrome wire which should be a consistent fixed resistance per unit length so the power can be easily calculated.

 

[TheNiceGuy]

Hello all
Long time no see

As of yesterday I have both a new charger:
http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=7028
and VC97 multitester.
I still have the original 12v/14a power supply from an old 400w Seasonic PSU.
http://www.seasonicusa.com/PDF/Spec400HSAPFC.pdf
I have had 2 cheap chargers fail using it, but it could be a coincidence. On my 1st charger, IMAX, the display blew out 2nd time I used it. I'm really worried the LCD got bad power from the PSU and shorted out. The other, a small cheap 2/3s charger, lasted a few months before wildly unbalanceing during recharge. Balanceing was off quite a bit from the start though anyway.
I want to make sure the PSU is not killing chargers, and won't kill my new charger.
I tested the Seasonic with the VC97:
I get 11.87 v on the 12 v line. It holds steady.
The amps start at 14, and after about 10 seconds drop steadily to 13, beeping every second or so. Not sure if I tested correctly, as I am new to multitesters. The tester wires got hot, rated 20A 10sec max, so I pulled before 10 sec.
When I hit the power switch on, the volts jump to 20 for a split second. Although the VC97 shows the exact same reading when I test a 3s lipo, touching the contacts to the plug. It too jumps to 20v, then down to 12v, so not sure if its a readout error or a real voltage jump.

Any ideas?

Thanks

PS- if needed I have a spare, unused, OCZ 600 evostream still in the box:
http://www.techpowerup.com/reviews/OCZ/EvoStream600W/
Shame to cut up the pretty cables, so would prefer to use the SeaSonic, if its safe that is.

 

[uclabachelor]

I thought he was trying to charge his car battery? And most PSU's dont do close to 70A 12V

Who says the PSU needs to do 70A 12V?

 

[Gillbot]

I've done this before, but I use an old AT power supply that has an on/off switch and seems to provide a more stable output at 12v.

That said, the issue could be the low voltage. Many PC power supplies are rated at 12v, but as you load them up they tend to trail off on voltage, some dip pretty low in the 11v range. The lower voltage, the more current it's going to pull to make up for it.

 

[edro]

Do you have the manual for the charger?
What are the voltage limitations on the 12v input?
Most inputs are at least ±10%. (10.8-13.2v)

I think it would work fine.

 

[pm]

For what it's worth, I've been doing this same thing - using a computer power supply to power the 12V input for a RC battery charger (I have a Thunder Power TP820CD). I also use the same power supply to run my backup server, and I spliced out 5V and 12V (and 10V (using +5V & -5V), and 7V (-5V and 12V)) to replace all of the wall warts on my network... saved me about $30/year in electricity replacing a half-dozen wall power supplies with one computer power supply that was always-on anyway. I can post a photo if anyone is curious... it's pretty amusing.

I've never had any of the problems that you are describing though. But I switch to a much larger power supply when I want to charge any of the large lipo packs like the one in my electric car (which is a 50V 80Ah (16S/1P) Thundersky pack). My guess would run along the same lines as Gillbots.

 

[TheNiceGuy]

Thanks
Regarding my new charger, Ya, no manual included, just the info on the HK site, and perhaps on the internet somewhere. The site says input voltage between 11v-17v.
But my old charger:
http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=5548
Said simular, and the display blew out when I hit the power switch on. I had first turned on teh PSU, let it warm up, then hit the charger switch, and boom, the display lit up for a sec then all teh characters were distorted and unreadable. I didn't even have a battery connected to charge, so no load on it.

I am reluctant to fry my brand new charger in a simular way. Is it best to just buy an old suitable wall outlet from a junk shop? Just to bad to let this challenge go unsolved...

I can use my OCZ PSU, which looks much more robust?
EDIT: Tried the OCZ and it won't let me do it. I guess its too smart. Shorting the green/black, it will power up for a few seconds, then powers off. I tried puting a fan (load) on the 12v, 5v, then 3v lines - with the same result. Trying to read amps with the meter on 12v also same result.

Are there any tests I can do with my meter to make sure the old PSU is safe? or I just have to choose risking a blown screen or buying a new plug?

The more I consider it, teh more it loolks like its best to just buy a used wall plug. I saw a bunch between $15-$30. IIUC, any plug will do as long as its rated 12v and 7amps+? Lower amps will give a charge error, and higher than 7amps will be fine, just unused power, correct?

 

[Evadman]

EDIT: Tried the OCZ and it won't let me do it. I guess its too smart. Shorting the green/black, it will power up for a few seconds, then powers off. I tried puting a fan (load) on the 12v, 5v, then 3v lines - with the same result. Trying to read amps with the meter on 12v also same result.

Sounds like there isn't enough of a load on the 3.3V and 5v rails. That is what the CPU generally runs off of, so the supply is expecting a pull on those rails. (5v as source and 3.3 as drain, leaving a difference of 1.7v for the CPU). Look for a 1 ohm 5 watt resister and use it between the 5v (generally red IIRC) and 3.3v (generally orange IIRC) rails. That will pull a little less than 2 amps (~3 watts) which may be enough to get the PWS to stay on. Make sure to sue a resister rated for at least 3 watts.

 

[bobdole369]

The more I consider it, teh more it loolks like its best to just buy a used wall plug. I saw a bunch between $15-$30. IIUC, any plug will do as long as its rated 12v and 7amps+? Lower amps will give a charge error, and higher than 7amps will be fine, just unused power, correct?

You won't find a wall plug that'll do 7 amps at 13.8V.

In all likelihood as the load kicks in (10A is >100W) at the same time as the screen is turning on, the voltage drops on that PSU, which prevents the micro that runs the charger to power up.

Your fix is to get a higher voltage power supply as you suspected. 13.8V minimum for 12V battery packs. Even if the charger was a boost converter type (and I suspect it isn't, more likely a PWM based charger), I'd still have trouble thinking even a "massive" computer PSU is up to the task.

SInce you need at least 7A, look at Astron for your supply needs.

 

[bobdole369]

The other issue with a PC PSU - is that without a decent load - the filtering is ineffective. So when you turn on the charger, it might have a pulsed voltage with a lot of ripple. THis could cause the issue with hieroglyphics.

 

[pm]

I can use my OCZ PSU, which looks much more robust?
EDIT: Tried the OCZ and it won't let me do it. I guess its too smart. Shorting the green/black, it will power up for a few seconds, then powers off. I tried puting a fan (load) on the 12v, 5v, then 3v lines - with the same result. Trying to read amps with the meter on 12v also same result.

This sounds like a PWRGD/PWR_OK issue.

If I remember the timing diagram for a new ATX power supply, the system powers up the rails one by one 12V -> 5V -> 3.3V or something like that and then waits for the CPU to return "PWRGD"==1 back or else it turns itself off. This wasn't an issue with older AT power supplies - it used to be that you just needed a load on one of the rails to keep the supply on - but on the new ones, there's some error checking and the CPU communicates back to the supply that life is good. If the CPU doesn't indicate PWRGD, then the supply turns itself off after roughly 0.5s or so. It might seems like "seconds" because it takes a while for the rails to discharge.

See:
http://en.wikipedia.org/wiki/ATX and then search for "PWR_OK"

To emulate the CPU being happy, you need to jumper PWR_OK to a "1". I'm not sure if a "1" is 3.3V or 5V. So use a paper clip to jumper PWR_OK to 3.3V, and if that doesn't work, use 5V. I think 3.3V should work regardless though since it's high enough even for 5V. For the pin-out of where PWR_OK is, look at the Wikipedia link above.

 

[GammaLaser]

This sounds like a PWRGD/PWR_OK issue.

If I remember the timing diagram for a new ATX power supply, the system powers up the rails one by one 12V -> 5V -> 3.3V or something like that and then waits for the CPU to return "PWRGD"==1 back or else it turns itself off. This wasn't an issue with older AT power supplies - it used to be that you just needed a load on one of the rails to keep the supply on - but on the new ones, there's some error checking and the CPU communicates back to the supply that life is good. If the CPU doesn't indicate PWRGD, then the supply turns itself off after roughly 0.5s or so. It might seems like "seconds" because it takes a while for the rails to discharge.

See:
http://en.wikipedia.org/wiki/ATX and then search for "PWR_OK"

To emulate the CPU being happy, you need to jumper PWR_OK to a "1". I'm not sure if a "1" is 3.3V or 5V. So use a paper clip to jumper PWR_OK to 3.3V, and if that doesn't work, use 5V. I think 3.3V should work regardless though since it's high enough even for 5V. For the pin-out of where PWR_OK is, look at the Wikipedia link above.

PWR_OK is driven by the PSU, not the CPU.

 

[pm]

PWR_OK is driven by the PSU, not the CPU.

Yeah, you are right. It's not that way on the CPU that I work on (we have a 5+ stage handshake between the power supply and the CPU, ending in the CPU driving PWRGD high)... but the CPU that I work on doesn't have an ATX power supply. I forgot. Thanks.

So looking at this page:
http://www.wikihow.com/Convert-a-Computer-ATX-Power-Supply-to-a-Lab-Power-Supply

They talk about tying the "sense" wire to 3.3V.

 

[bryanl]

A 12V lead acid battery needs at least 13.0-13.5V to charge, and applying exactly 12.0V to it may actually discharge it since it's only at 30-50% capacity at that voltage. Also the charging current has to be limited through a resistor or active current regulator, the level depending on the battery and state of charge, anywhere form C/5 down to C/200 or C/500. For a typical 50 amp-hour car battery that would be 10 amps (C/5) until the battery is fully charged, or 250mA or 100mA to maintain charge.