GK 800RB with AC ripple on output

[Andy S.]

Folks,

I have an GK 800RB that was given to me, dead and already partially disassembled. I noticed a burnt resistor on the PS and was getting nothing out of the 60v and the 15v on the PS.

Replaced a couple of Xsitors, a burnt resistor and a zener and now I have voltages, but I have a annoying ac ripple , very small , on the 60 v dc. The 15 DC is clean,and it comes off the same rectifier. The 85v comes from a different section so it is not affected.

I replaced the rectifier, the main filter caps and STILL have this nasty ac signal. I've looked for some stray connection possibly left by the person that had it before me, nothing that looks suspicious.

I've attached a drawing with where I'm seeing the ac. I can see getting some right off the rectifier but am wondering why the filter caps aren't smoothing it out more.

Any insight on this would be great. It's probably something right in front of me, but as usual, I'm just not seeing it.

Andy S.

[Enzo]

Well the arrows point to two ends of the same wires, so all we have is some ripple on the HV rails. These are not regulated rails, so there will always be some ripple. And it is a single stage filter. There may be more than one cap in parallel, but they are all connected together. You mention 60VDC, but how much ripple? Just flip your meter to ACV and give that reading. It would take a lot of brute force filtering to make that real clean DC.

Both pos and neg do this, though the ripple waveform would be inverted, right? All the more reason to think this is normal.

Is the amp humming? Or are you just expecting flat like a straight line DC there?

Another reason an amp might have excess ripple is excess idle current. Put an ammeter in the mains lead, how much current does this thing draw at idle? No load. If this amp is drawing 100 watts or something just sitting there, it is biased too hot.

[Andy S.]

................... There may be more than one cap in parallel, but they are all connected together. You mention 60VDC, but how much ripple? Just flip your meter to ACV and give that reading.

It's approx. .7vac. While on the 85vdc it is .017.

Both pos and neg do this, though the ripple waveform would be inverted, right?

Yes, both pos & neg.

Is the amp humming? Or are you just expecting flat like a straight line DC there?

Both. Since the 85v supply is smooth, I was expecting the 60v to be similar.

There is NO buzz on the 4 ohm,300w output, which is the 85 vdc supply. As well, there is no ripple on the 85v rails. Nice & quiet.

But on the 8 ohm, 100 w output, that is powered by the 60vdc, it is buzzing.

Another reason an amp might have excess ripple is excess idle current. Put an ammeter in the mains lead, how much current does this thing draw at idle? No load. If this amp is drawing 100 watts or something just sitting there, it is biased too hot.

That's a high possibility. As I mentioned before, the last couple of projects I've been given had already been torn apart and some attempt had been made at fixing them. One amp had two resistors literally tied together to add up to get close to the prescribed resistance. So it's real possible the previous person tweaked pots just to see what would happen.

(on a side note, I've told this shop owner that I don't want anymore "projects" that have been tampered with or attempted to be fixed. This is number 3! All were given to me at one time, so i think I've learned my lesson in that department! )


I'll check the current draw and the output section tonight and let you know what I find.

Thanks!!

Andy S.

[Enzo]

Well 7/10 of a volt is not a lot of ripple. Those dropping circuits for the LV rails are drawing more current through the 60 than the 85 which stands alone. So I expect more.

I also note the two channels come from separate inputs, so we need to determine if the hum is coming in through there.

[cjlectronics]

I gave up on repairing the GK 800RB I have. I replaced a shorted MC15023, its .33 ohm power resistor, and the MPSA06 in its circuit as well. The amp powered up with an 8 ohm load for 10 minutes and then took out two 15022's and not to mention other support circuitry.

I had enough....you can buy another one for $200 + shipping on the net.

Is latency failures common on high powered amplifiers?

CJL

[Enzo]

What happens is that it was never fully repaired. One replaces an output or two and an emitter resistor, and fires the amp up, and it seems to work. But that is not a job completed. Resistors in the driver circuits can be damaged, as can the drivers themselves. Limiter xstrs can be leaky, etc. Even simple things like bias matter. When I fire up an amp after repair, I monitor the mains current for sure. It may be working just fine as fa as we can tell, but if hte thing idles at 300 watts from the wall, something is wrong. High idle bias means the trransistors will be fighting each other.

Once the amp is truly repaired, then it will be reliable. In other words, when an amp I repaired blows up again after a short time, it was My fault for doing a half asssed job, NOT the amp's fault.

[cjlectronics]

Point well taken and you are exactly right. I let my frustration get the best of me sometimes.

Thanks for the kick in the ass Enzo!

CJL

[drewl]

I've seen pretty many of these blown up, and GK just replaces the board with a different one that use TO3 type devices.
Even though Enzo the master has probably fixed a miilion of these, I think there's a design flaw somewhere which is why the replacement boards are different.
GK has a flat rate for the board swap repair, like $200-$250.....
My friend with a music store is a dealer, so he just sends these back to GK.
So luckily I don't have to deal with them anymore!

[Enzo]

Well, OK, but he has MJ15023 outputs, which already ARE TO3.

I would say that switching output xstr types and making a new board for it is more likely a matter of reliable supply that circuit problems. Circuit problems are fairly simple to correct on the board.