News:

30% off all Honey Amp kits, check it out at https://store.ssguitar.com !

Main Menu
Menu

Show posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Show posts Menu

Messages - cin

#1
I've rerouted the yellow common wire to one of the power cap's ground leg and the buzz is gone!

And that does look like an elegant enough solution to my problem. I'll just have to do a good job at wiring the new ground path, so it looks all pro and such.

Splendid!

This is great news for the amp, but above all else it's what I've learned along the way... That sounds cheezy.
But you get what I'm saying. Thank you for showing me the way there. I now have a much bigger bag of tricks to attack my next amp project with, thanks largely to you.

This is certainly very addictive. Pretty sure I got some endorphins when I flipped the switch and the noise was... not there! very good.

Cheers!
#2
Well, my primary intention is to learn, and this little amp has been great for that so far... with your coaching that is.

But once I'm done it's got to go. I only have so much space, and I have another amp that I prefer (for now). So the plan was: repair it if I can, make it as close to original as possible, resell it.

But I hear you, it'll be hard to bring it back to the original state without the schematics or another one to compare to. There's another forum where I've found some hard-to-find schematics before.... music-electronics-forum.com I'll check there before giving up on the original design.

Will test the yellow wire re-route in a bit. I'm assuming I'm leaving the green earth-ground wire coming out of the power cable tied to the chassis here, for safety?
#3
Amazing! I'll review all of this carefully tomorrow.

Moving away from looking for faults, and into changing the circuit, pre-supposes that the amp is basically buzzy by design. It is a fairly low-cost unit, and it shows in the simplicity of the circuit compared to other amps I've seen, but I'd be unimpressed if it came out of the factory sounding like that. The buzz is about as loud as an electric toothbrush at any volume (a very scientific alternative unit to the decibel), it pretty much makes the amp unusable at anything lower than "quite loud" (like 500 or so electric toothbrushes).

I was also hoping to bring back the amp to a working state while removing all the modifications to make it as close to the original as possible. I won't be keeping it, and it's a lot easier to sell an amp that sounds good and has just been re-capped, than one with mods.

That being said, the primary objective is to get rid of that buzz. And bettering a flawed design is appealing. But I'm wondering if there isn't still a fault somewhere, other than a bad design. I can test re-routing the yellow wire to one of the cap's ground terminal without having to drill anything, so I'll try that to see if that does it, and report back.

Again, many thanks for the patient guidance and knowledge.
#4
I've gone back over what I thought was GND with my multimeter, and made some corrections to the image (attached). I've also made some corrections to the live circuit in red. And I've added, in dark blue, all the places I've checked to be <.4 Ohm when compared to the yellow common wire at the chassis (ie properly grounded?). Again, I'm assuming this means those spots can't have unexpected voltage potential, unless the chassis isn't grounded.

Where else should I check?

#5
I apologize for the many posts. I imagine I am as confusing as I am confused.  :dbtu:

I've been reading about ground loops, and if I understand correctly having many paths to ground is bad because there can be potential differences between the many points, which would cause current flow between the two. But how can there a be differences in potential between two points that are connected? Which is why I concluded that I was looking for a disconnect in the ground.

I thought I'd trace it out on my photo of the circuit, to make it easier to test everything. I also traced out some parts of the circuit that are not obvious in the backlit shots.

Ground in blue, missing circuits in red, yellow dots are the pins of the power amp. It's very likely wrong in some places, I can see spots where it may not make sense, like that ceramic cap in the middle that's not connected to anything??

#6
Alright, here's another backlit photo. Hopefully it's a little clearer. I labeled the pot connections and such.

I redid a couple of soldering points on the pots, like on the treble pot, and it removed the sensitivity but not the buzz.

I figured signal tracing is pointless if the noise can be coming from anywhere downstream of the reverb tank. The noise would likely show up everywhere past that.

So then, apart from the input jack, which may be part of the problem, I'm looking for places that should be grounded but aren't, correct? Spots that have voltage potential on the ground trace, where it is not supposed to be? So I can go around testing all the ground points against a known good one, looking for potential?

#7
Good morning!

The 2 red wires coming out of the power tx do go to the PCB. And the yellow wire coming out of the power tx does indeed go to the chassis.

The ground trace on the PCB leads to the mounting screws that go into the chassis. The PCB definitely has to be mounted for the amp to work. So I've been screwing the PCB and the pots back in every time I need to test it.

The input socket is grounded to the chassis, but then so are most of the pots? I unscrewed the input jack and left it floating (everything else grounded) and the buzz is half as loud as when the jack is screwed in, but not gone. So this could be part of the problem, but not the only source of buzz.

I tried unscrewing the headphones jack and reverb pot from the chassis ground too, and the buzz is still there. I unscrewed the treble and volume pots, buzz is still there.

Poking around the pots, the treble pot, which is a 50K linear pot that's bigger than all the other pots except the volume pot (so maybe a replacement?), is especially sensitive to being touched, I can hear changes in the buzz when I touch the metal parts. The solder joints look fine at bother ends, but I might just redo them to be sure.

So a ground loop can be caused by many paths to ground? I thought it was mostly about some section of the ground trace actually having unexpected voltage potential. There are many paths to the chassis ground on this PCB. The 3 mounting screws, the input jack, most of the pots seem to be grounded via the chassis, etc.

Still on first coffee, might make more sense after the second cup.
#8
I also disconnected the Zorbel bits, and it made no difference. So I reconnected them for now.

I'll do some more signal tracing past the reverb tomorrow, and take better back lit photos of that area.

Cheers!
#9
LOL so much space for confusion. So, on the amp board there's no label of Input/Output, but you said earlier in the this thread:

"The output cable will be the one with the ground wire on the PCB (2 spades), while the drive cable only has one spade conn. (that is the usual setup)"

And I believe I have it right, now 8) The cable with 1 spade is connected to the In RCA connector on the Reverb tank, and the Out RCA connector is connected to the cable with 2 spades.

And I plucked the RC4558P opamp furthest away from the input out, and the buzz remains...

I was shopping for a reverb tank an hour ago. I owe you a beer.

#10
Hmmm, well the buzz is there wether the Out cable is connected to reverb tank or not, and is present wether or not I short the pin of the out cable.

So the noise may not be coming from the reverb tank after all! What a roller coaster hahaha.

But how can that be if the noise is there on the out, but not on the in??? Could the noise be coming up current? like against the flow of the signal? I'm guessing that could make sense with a ground loop, where there's unexpected potential somewhere.

So, at least I know that the ground problem is downstream from the reverb tank, right? Or is that also not necessarily the case?

Many thanks for all the help!

#11
Quote from: phatt on January 15, 2022, 08:38:23 PM
You said the tank was wired in reverse,,,,,,,,,Well that may well have overheated the coil. 8|

If you suspect the tank is the issue then disconnect the pickup cable and short the pickup end.

You mean short the middle pin to the sleeve on the male RCA connector that's connected to the IN of the reverb tank? Just making sure before I blow up something else -_-
#12
OR, or, I will repair this tiny coil, because apparently 8FB2A1B reverbs are hard to find and expensive....
#13
Alright, I checked the ground continuity from the earth pin of the power plug to the chassis, and to the sleeve of the input jack, and it's fine. Then I tried another power point, but the buzz is still there. Good things to check though. You wizards are teaching me to troubleshoot and that's awesome.

Instead of removing the reverb opamp, I did a little signal tracing with my oscope. But I quickly got lost after the first opamp, where the signal goes through the channel 2 overdrive pots, which are switched off to ground by the channel switch set to channel 1... So I randomly checked a few points to see if the signal was clean or dirty, and then I checked the reverb in and out...

The IN is clean, the OUT is dirty.

So I pulled out the tank and took some measurements inside, the signal wires going into the IN coil is clean, the signal wire coming out of the OUT coil is dirty. The OUT coil does look like it may have overheated, the tape is discoloured in one spot. Is that a thing? Can a small coil overheat if it's overstimulated? Would that indicate the amp was shaken vigorously while On?

So I'm guessing I'm stuck replacing the whole tank, it's all pretty much riveted together...

As for the headphone jack, it is wired with the tip and ring connected, I believe that provides mono to both left and right. @Phil That's probably what you meant. I do use the jack, I would only remove it for troubleshooting purposes, and then put it back. I'd like to bring back this amp to a functioning state that's as close to the original as possible. So I'm looking suspiciously at everything that looks janky, like that Zobel cap/resistor deal on the headphones jack. If it was a Zobel network from the factory, it probably would be done better, closer to the LM1875, would you say that's a pretty good reason to think those components were an additions?

I realize the schematics would be useful in finding out what is original and what isn't...

The 3W resistor is indeed a 0.27R.

I can take better backlit photos of the board if I've jumped to conclusions about the reverb tank being at fault, but that would surprise me at this point.

@g1 Yes, I'm learning with every rookie mistake I make, but I'm having a great time. 8) Thanks for the clarifications. I now understand that. I'm not sure how the empty TRS jack would help though. So as to get the speaker out of the equation? Since the buzz is present in the headphones too that would make sense. I'm still not really sure how to proceed with the "resistance reading" and wiring tracing, I'll google that.

THANKS GANG!
#14
That's a great trick, Will. I went through all the ElCaps with my good MM on AC V and there's no AC voltage across any of them. I checked the film capacitor next to the phones jack (or is it a paper cap?) and it leaks about 30mV, but then I'm not sure those caps are polarized, so that might be normal.

I may be wrong, again... about the phones jack. So all pins on the phones jack have continuity to ground (duh), but the signal pins are at -4mV compared to ground, and test with 7.1 Ohm of resistance to ground, so I don't think they are "grounded". I was expecting the ground pin of the phones jack to have very little resistance to ground (to be actually "grounded"), but it actually measures at 10 Ohm. And.. looking again the datasheet for that jack, the pairs of signal pins are connected, I believe, so then the second wire is probably not for switching purpose but actually the signal coming in to the jack, then leaving for the speaker.

Would bypassing the phones jack, and cap/resistor to ground assembly, be a worthwhile test?
Or, should I be following the buzz upstream on the signal path?
Or, should I be looking for something that's connected to ground that shouldn't be?

I'm probably thinking about this wrong, so yeah, your help is greatly appreciated hahah.

Cheers, and a happy new year to you all!

EDIT: Ok, so when I said the "ground pin" of the phones jack I meant the sleeve pin. Turns out that's probably not ground or it would be connected to the reverb knob chassis right beside it... I don't yet understand the use of negative voltage in DC circuits, outside AC for signal and such. Can you tell?  8)
#15
So there actually was some user error with the oscope... There's a tiny bit of noise (<10mV) on OUT1 of the IC closest to the power end of things, while there's no such noise on the other OUT pins. Still, I'm not sure that's the noise I'm looking for, at under 10mV on the out of a 12V fed opamp, it wouldn't be that loud? But again, I really know nothing. Making assumptions.

I put my oscope on the phone jack, knowing that it was all grounded, and it showed what I think is probably the buzz I'm looking for. 60Hz 157mV peak to peak. Which makes sense since I can hear it loud in the phone jack... captain obvious here.

I made sure the square capacitor and resistor between the phone jack and the reverb pot were good, they check out. And there's no continuity across the square cap, so that's not how that path gets grounded.

My guess right now is that neither of the 2 white wires on the phone jack should be grounded, the black wire should be the ground. I'm assuming one of those 2 wires carries the signal, the other one acts as a switch to turn off the speaker when headphones are connected, but not sure how. One of those 2 white wires goes to the speaker terminal, and that confuses me... I don't really understand the signal path here. I did check that there was nothing stuck in the phones jack connecting it all together.

I may be getting ahead of myself... getting late, I will investigate further tomorrow.

Cheers!