Debugging an integrated amp that blows a fuse

ctviggen · « **on:** 18 Oct 2005, 01:02 pm »

How does one debug something that blows the power fuse shortly after power up? I have a Creek 4140 integrated amp that's about 11 years old. It recently started blowing its power fuse. There are two other fuses (one for each channel) on the motherboard, and neither of these are blown. I cannot see anything that appears burnt, although I have not examined the underside of the board. I could desolder the connections from the transformer to the motherboard, but assuming the transformer is good, how then do I go about determining what is bad? All I currently have is a multimeter, so I could check resistances, but I can't check capacitances.

Dan Banquer · « **Reply #1 on:** 18 Oct 2005, 01:18 pm »

You should probably check the voltage on the main power supply capacitors as the unit comes up. You may have a defective capacitor or diode bridge. Transformers are real tough to blow. Do you have a variac to bring the AC up manually?
Hope this helps;
d.b.

ctviggen · « **Reply #2 on:** 19 Oct 2005, 12:43 pm »

Thanks, Dan. Sadly, I have nothing but a multimeter. (What I would give to have all the stuff we had at the engineering companies I used to work for -- we had tons of testing equipment.) I heard back from Creek, who said the likely culprit was the transformer, interestingly enough. I may search the Net and see if I can find a suitable transformer. I have two more fuses, so I may desolder the transformer, put the ends of the wires away from stuff and turn on the power to see what happens. If the fuse doesn't blow, I'll check the voltages on the two outputs. If the voltages are bad and iff the fuse does blow, then the transformer is bad. If the voltages are good and the fuse doesn't blow, I'll start looking at the caps.

ctviggen · « **Reply #3 on:** 19 Oct 2005, 01:11 pm »

You know, since I'm going to replace the transformer, I may also order nicer caps too, and just do my own rebuild. I do like Creek integrateds, though, as they're reasonably priced and sound great. Mine is a little on the low power side, though, as it can't hack 4 ohm speakers. The new integrateds are better at low ohm speakers.

ctviggen · « **Reply #4 on:** 20 Oct 2005, 01:16 pm »

You know, Dan, I think you're right -- upon closer inspection, there does appear to be a cap that looks bad. So, I'm going to order the caps for the power supply. I've also ordered the transformer. I'm going to replace the transformer and the caps for the power suppy and see if it powers up. If so, I'm going to order and replace caps in the audio chain. I might as well do it all while I have it dissected

Dan Banquer · « **Reply #5 on:** 20 Oct 2005, 02:17 pm »

If the cap is bad check the rectifier diodes to see if that might be the root cause of the cap going bad.
d.b.

ctviggen · « **Reply #6 on:** 20 Oct 2005, 05:08 pm »

It's funny you say that -- I was thinking of getting better diodes anyway. I'm not sure the cap is bad, but it does have "stuff" under it, which means it likely is bad. I'm going to replace all caps anyway, and I'll check/replace the diodes.

ctviggen · « **Reply #7 on:** 25 Oct 2005, 12:18 am »

Ok, I think the guy from Creek might be right -- it's either the two 6,800 uf caps or the transformer -- after this circuit, there appears to be fuses for each channel (though I'm not sure where the diode rectifier circuit ties into this -- I'm too tired to figure it out right now). Anyway, I'm going to replace the diodes, all caps in the power supply part, and the op amps. I may also replace the audio output caps.

peranders · « **Reply #8 on:** 27 Oct 2005, 05:11 pm »

A simple test is to use the DMM in diode mode and check the output transistors. If you get 0.10 or less it may be a sign of broken transistor, note: _may_.

Put RED cable on the base then put BLACK on the collector and the emitter. OK part is 0.7 V or less but how much less is a little dependent. This is valid for NPN transistors. Change the polarity for PNP.

This test is done when the amp is switched off and completely discharged!

ctviggen · « **Reply #9 on:** 28 Oct 2005, 12:58 pm »

Nice point! I'll try that, although my crappy VM appears not to have a diode test mode. From what I see, the power supply uses 25V rails (+25V to -25V), then uses a cross-coupled resistor/cap network that includes large caps to ground. From there, there are two fuses that run the rest of the equipment. I assume the diode circuit that converts to DC is in there somewhere, although it's hard to figure out without a schematic and it takes a long time to create a schematic by just following circuit runs. As the two internal fuses are not blown and the external fuse to the power supply is blown, I'm hopeful that I can replace the transformer and the PS caps and then call it a day for now. I have ordered a bunch of other parts to see what affect -- if any -- they have, but those will have to wait until later; I have a house to insulate before the coming winter.

ctviggen · « **Reply #10 on:** 28 Oct 2005, 03:28 pm »

Well, I'm an idiot! When I mapped out the PS circuit, the "resistors" I put on the diagrams were diodes for the diode bridge. Duh! Anyway, I have suitable diodes ordered, so this weekend, I'm up for soldering!