Antec Tru430 Fan Control Circuit

[jbbenson]

I had the fun and joy of recapping a True Power 430 this evening. All went well, with the exception of the fan control circuit. Voltages for all lines are well within spec, so that much was a success. However, on powerup something in the fan control circuit let out its magic smoke. Both the header on the board for the internal PS fans, and the two external molex connectors read highly erratic on my DMM. I can't really get a reading for any sort of reference, and I don't have a scope to look at the waveform. Does anyone have experience with that particular part of the circuit and a rough idea what might be wrong? I suspect a transistor or diode has bit the dust based on the smell, though nothing really looks cooked. Its tough to tell with all the wires in the way.

By the way, thanks to Topcat for the Rubycons!

Thanks in advance!

[999999999]

Re: Antec Tru430 Fan Control Circuit

Is this circuit sophisticated enough that a schematic is needed or it is a pretty basic thermistor sensed, biased transistors with a few resistors for current limiting? Well then again I think some do throw an opamp in or it's an option on some PCBs but left off depending on the features of a particular PSU.

I suppose it could be coincidence that the fan controller fried right after you'd recapped the PSU, but I wonder if there's more to it, maybe a short? Were you powering multiple particularly power hungry (high current and RPM rating) fans?

First confirm that the power (voltage) from the mainboard (PSU PCB) is ok.
Next, look at the last transistor before the output to the fans, trace that back visually on the controller circuit board. Transistors are cheap, I would just replace it with something that can handle more current. Since I don't know what the original is, the replacement isn't so easy either. Get the part # off the present one and look for a replacement with similar hFE but higher current rating.

It's not likely to be an exotic part, any generic transistor will probably work to some extent give enough current handling capability but the result will differ with hFE value differences of replacement part vs original part (if my guess about what this circuit is like is true, but it may not be). The limitation for reasonable implementation will probably be how much clearance there is below the controller PCB and the lead spacing on the PCB (so you're not hacking out something instead of it looking close to original design mechanically). Or you could just replace with exact same or try to cross reference to a suitable equivalent part that's easy or cheaper to source.

The other transistor(s) are probably generic 6 cents each parts. You could replace all of them or test each. Presuming they're BJTs, try this.
http://www.allaboutcircuits.com/vol_3/chpt_4/3.html

You mentioned diode(s) but I doubt it's a diode that went out. Regardless, practically any DMM has a diode check function.

If you get stuck, you might put the PCB on a scanner or take a good quality camera picture of both sides for someone to look over.

[jbbenson]

Re: Antec Tru430 Fan Control Circuit

Some more relative information:

The fan circuit was lightly loaded with only the PSU's two cooling fans connected. I was doing initial testing after the recap. I had plenty of time to kill, so I've pretty much done all the troubleshooting tips you listed. To make a long story short, the failure was (most likely) caused by a stray blob of solder. It had shorted the leads of several components together. It was likely from my desoldering, or tinning the tip on my iron The only reason I even noticed it was because it was more shiny than the solder surrounding it, and it was in an area that I hadn't replaced capacitors.

Anyways, I started tracing the fan circuit back from the output. I'm no EE, but the first description you gave is probably accurate. I had to remove an inductor temporarily though to actually check part numbers and be able to find and test components.

The transitor closest to the output is a 2SD880, and upstream of that is a 2SA684. I'm pretty sure the 2SA684 is the culprit. I don't have my test results with me (at work) but I remember a very low voltage drop (.094V or thereabouts) and some other weird results.

I'm going to go ahead and order the 2SA684 along with some caps for other repairs. I'd like to replace the 2SD880 as well for good measure, but Mouser and Digikey don't stock it. Mouser's suggestion for a suitable replacement is an NTE152. Would this be suitable, or could you recommend something else?

Thanks!

[999999999]

Re: Antec Tru430 Fan Control Circuit

At Mouser, 2SD880 -> KSD880
http://www.mouser.com/search/Product...y512-KSD880Y_Q

(or RoHS version if you like,)
http://www.mouser.com/search/Product...key512-KSD880Y

KSD880 Datasheet,

[jbbenson]

Re: Antec Tru430 Fan Control Circuit

Just thought I would post an update. While preparing to replace the questionable transistors, I found what was left of a resistor that burned completely in half. I just flat could not see it with the components packed in as tightly as they were. Live and learn I guess. Now to try and identify it for replacement. I'd take pictures, but my digital camera died several months ago.

[Sparky]

Re: Antec Tru430 Fan Control Circuit

I used to have a true430 I sold to a friend, he still has it. Maybe next time I'm over at his place I can pop the unit open and see if I can get you the value of that resistor if you haven't found it by then. Can you give a description of where the resistor is? Or maybe go out and replace your camera so you can snap a pic? That would help me find it.

[jbbenson]

Re: Antec Tru430 Fan Control Circuit

I've found the value. Just had to desolder what was left and put it under a magnifying glass to see it. 100 ohm, and I'm guessing 1/8 watt by how danged small it is. Verified against the True380 awaiting new caps too. PCB layout is identical. Off to Frys to try and find a replacement.

[Sparky]

Re: Antec Tru430 Fan Control Circuit

If it burned though maybe getting a higher rated resistor might help, like 1/4 W. If it'll fit that is.

[jbbenson]

Re: Antec Tru430 Fan Control Circuit

Yeah, that's what I was thinking too. Of course, it didn't help that I had a solder booger shorting something out which caused the initial problems to begin with.

[jbbenson]

Re: Antec Tru430 Fan Control Circuit

I finished the repairs and put the PSU back together yesterday. No magic smoke escaped this time!

Voltages at the ATX connector:
12V @ 12.11V
5V @ 5.14V
3.3V @ 3.37

Measurements were done with only a sacrificial half height 3.5" drive as a load.

The fan circuit is now functional with an output voltage of 11.07 when the PSU lightly loaded and at room temperature. Although this is higher than expected, I don't have a working point of reference to see exactly where it should be. The difference is probably in the transistors I substituted. I used a KSD880U in place of a 2SD880U, which is an identical sub. I matched the beta on both. I also subbed an NTE294 for the 2SA684, but I could not determine the beta on the 2SA684. Its a close enough match that I'm not going to bother with it further. Given the fact that these power supplies tend to run a little warm, I'm willing to trade off the extra fan noise for increased longevity.

If anyone is curious about where all these parts are located (for modding or whatever), look for the three big chokes on the output side of the PSU. Desolder the middle choke, and and bust out the magnifiying glass. The 2SD880 is the large heat shrink covered transistor that butts up against the 12V output caps. The 2SA684 is in the same general area, but closer to the 5V and 3.3V output caps.

The only other thing I'll probably do to this power supply is replace the cooling fans with some nice NMB-MAT Panaflos.