Oahu Tonemaster-Hum and 21khz distortion

[shooter]

use 2 6.8K in ||  if you have 4 of 'em

[tubeswell]

At your suggestion, I put in another tube and the readings are the same, …

And the noise?

[wsscott]

I replaced each 1.5k with one 6.8k resistor.  The values look good.  It now has a voltage drop of 120VDC, and Plate Voltage of 90 now vs. 89VDC before these changes.  Remember the schematic showed 75VDC.

The hum seems to be much lower. 

But, that 21.k oscillation/distortion still exists in the same positions on the Vol and Tone pot settings as before.  When you play with the pots in those positions it's a very distorted sound. But turn the Volume all the way up to 9 and in cleans up.

So where we left off yesterday, the 21.k problem goes away with jumping the .006 cap to ground.  So I still need to resolve that.

I think I asked whether that 1Meg resistor was the issue, and maybe it should be smaller.  You guys know way more than I ever will.

[shooter]

When you play with the pots in those positions it's a very distorted sound. But turn the Volume all the way up to 9 and in cleans up.

 

the 21.k problem goes away with jumping the .006 cap to ground.

the cap should have been "jumped out" not shorted to ground!  so a wire from left side to right side, leaving ONLY the pot/input jack In-circuit


by jumping it to ground you have NO input signal, effectively "shorting the grid to ground"  look at the schematic.

[wsscott]

Sorry, I jumped it , not to ground. My mistake in describing it.

[shooter]

what's playing through this amp??


 as long as the input device has a pot for gain/vol , the amp pot/cap can just be trashed

[wsscott]

Shooter, I acknowledge that you present a practical solution to the problem, however, it's not a real solution to the problem from the owner's perspective.  Since this problem doesn't exist on the Instrument channel, it seems like there should be a solution.  Since the problem disappears when the circuit is grounded at various points, I hope that a solution exists.

I appreciate everyone's help on this frustrating problem.

[wsscott]

Why does the MIC channel not have any resistor coming off the Input Jack, when the Instrument Input jacks have a 100K ohm resistor?

The rest of the preamp circuit is the same for both channels.  However, only the MIC channel experiences the oscillation.

Is that the "canary in the coal mine"?

[scstill]

The absence of 100k in MIC is likely for impedance matching, as MICs are different than guitars.

In case I missed it have you tried another 6SN7? in case the internal capacitance has changed in that triode causing a 21khz distortion only in MIC channel. Also since the distortion goes away when you jump the MIC .006 why don't you suspect that cap as suggested earlier?

BTW - can you post some internal pics of this amp

[shooter]

why don't you suspect that cap as suggested earlier?

ya can lead them to water but.......


I would replace both the cap n pot since the pot "appears" to have a bad spot, but that could be the freq-range of the cap/pot filter causing the "muddy til 9" distortion.

[wsscott]

Yes I tried another 6SN7 with the same symptoms.

The distortion goes away when I jump that cap or when I jump the 1M resistor.  In fact the hum seems less when I jump the resistor vs. the cap.  Both the resistor and cap are new and their values are in spec.

I'm attaching a couple of photos.

[wsscott]

Shooter-The pot has been replaced twice, thinking it was a bad spot on the pot.  I even checked the "new" replacement that was replaced with another new pot, and there are no dead or weak spots on the pot.  So I've ruled out the pot.

The cap is new and reads to spec.

[wsscott]

When I jump the 1M resistor there's a slight hum, and absolutely no 21khz signal or anything close to it.

Of course I can't play a guitar and get a sound because there's a jumper across the resistor.

If "removing" that resistor from the circuit by jumping it eliminates the oscillation, would increasing or decreasing the value of that resistor do the same thing?  If so, which way do I go?

[wsscott]

Since the MIC channel isn’t going to be used for a mic but for a guitar with this sounding as a Bright channel, would adding a resistor at the input jack work, and would it be brighter than the other channel, or is some other mod needed?

[shooter]

with the plates wired together there isn't really any "voicing" room in front of V1-lower


if the cathodes are split you can leave off the bypass cap on the mic channel, that will "attenuate" bass to some extent, otherwise an eq pedal before the mic-in might be the simplest

[wsscott]

Shooter- sorry but I don’t understand your first comment.  Do you mean that because the plates of the preamp 6SN6 tube are jumped together, that the MIC channel’s tone can’t be changed?

If so, why is that?

Also, how did the circuit designer determine that no resistor is needed on the input jack of the MIC channel?  How is impedance for the input jacks determined?

[AlNewman]

Does the mic channel make noise only when something is plugged in?
Or all the time?

If when nothing is plugged in, when you jump the tip of the mic input to ground, there's still noise?

[SEL49]

The single tone control affects all three inputs.

The two 100ks on the instrument jacks are mixing resistors. They are used to isolate the two jacks from each other. If you had two guitars plugged into those jacks but did not have the mixing resistors, then turning the volume pot on one guitar would also affect the other guitar.

The mic channel only has one input. Therefore, a mixing resistor is not needed.

Make the two changes shown in the attached pic and I think your oscillation problem will be solved.

[shooter]

he's already done those




you really want V1 for one purpose, to give the guitar signal some gain so you have "room" to add, subtract, multiply..... to the original guitar signal. 
by trying to "add tone" to the input, you attenuate the guitar signal to the point where it's not much above the amps noise-floor
pretty much every E-guitar has a tone knob, use it, you might be surprised it actually works.


like I said, separate the cathodes, leave the bypass cap off, get rid of the cap .006 n 1 meg pot, verify the amp works making guitar noises, THEN we can tweak "tone"

[scstill]

Those pictures show why you might be getting a 60/120/180hz hum
The original heater wiring could be the cause as it is poorly dressed.
With wall at 120vac (vs 110 in 1940) it could be a cause to more hum compared to running this amp in the 1940's
If you have a variac try running the amp at 110 and see if the hum is reduced.
If so, you might consider to apply modern techniques of heater wire dress and signal line separation as discussed earlier. Elevating the heater as discussed earlier might also help.

[scstill]

Hey Shooter,
if the 21khz distortion is only in the MIC channel how would mods in the power tubes improve this?
Seems like it should be in the MIC preamp circuit.
Unless you are addressing something else that I missed...

Nevermind - sorry I see that the cathodes you are separating is the preamp

[AlNewman]

Obviously, Sluckey's right, you can place a resistor at the grid and limit upper frequencies/oscillations.  You could also wire in a capacitor to ground, to shunt those frequencies.

The circuit is the same at both sides, so you could also place a resistor between the tip of the input and the volume/tone pot to reduce RF frequencies.

What I don't understand is why in an amp which was designed with these features, there's an issue.  Perhaps it's because grounded jacks were added?  Maybe there's no shielding on the amp when testing, and your cell phone/router is in the area?  Maybe the capacitor is out of tolerance?  There's only 4 parts which could be an issue, compared to original design.

[scstill]

What I don't understand is why in an amp which was designed with these features, there's an issue.  Perhaps it's because grounded jacks were added?  Maybe there's no shielding on the amp when testing, and your cell phone/router is in the area?  Maybe the capacitor is out of tolerance?  There's only 4 parts which could be an issue, compared to original design.

I had a noise issue when restoring the Silvertone 1484. No grounded input jacks but a noise in one channel with no input. It turned out that the PT was super close to the channel. When I added a shunted jack (just one) the noise was gone (thinking sluckey suggestion). Maybe different situation, but a change for improvement to the original design. I also removed the death cap but left the unconnected switch so the next guy can feel good when switching it. :-) I'm a big fan of leaving original as much as possible...
https://stillampd.com/silvertone-1484-restoration

[wsscott]

Thanks for all the comments.  Here are my responses.

When I shunt the Mic input jack tip to ground, with nothing plugged in, the noise remains.  It sounds the same even when and instrument is plugged into the jack.  It only disappears when either the .006 cap or the 1M resistor are jumped across themselves.  The only time there isn't 21khz noise is when the Vol pot is set at 2 or below, and Tone is about the same level.  As Tone increases with Vol, it becomes more noticeable, and when Vol is 7-9 then it's distorted and very noticeable.  With Vol past 9, it immediately clears up.  But that's with the amp at full Vol.

I'm going to install a new Socket and see if that helps, and address the filament wire dress at that time.  The old socket has some wobbly lugs.

Separating the cathodes and adding the caps and resistors substantially reduced the hum.  However the 21khz distortion is still there.

It had this problem even with the original non-switching jacks.  Changing them to switching made no difference. No cellphone or router nearby.  Does this in different houses.  All of the components except the 5M resistor (whose value is about 6M and in a different part of the circuit) are new and test within specs, including the subject cap.  Yes I agree, those 4 parts are all there is, or so it seems, and their values are fine. 

AINewman, what values for those caps and resistors should I try?

Shooter, I'm leaving the cathodes separated.  I think I'll leave in the bypass cap since the change improved the hum even if it didn't affect the 21khz distortion.  I'll try jumping at the same time both the .006 cap and the related 1M resistor.  The result would then be a direct connection from the grid, through the new 50K grid stopper to ground.

Could there be anything going on internally with the PT that is creating this 21khz oscillation? 

Why does the 21khz oscillation stop when I jump the Tip of the Input Jack to the lead going into the 1 Meg resistor?

[shooter]

Why does the 21khz oscillation stop when I jump the Tip of the Input Jack to the lead going into the 1 Meg resistor?

1. the original wire is close to the 21Khz "source" and coupling it into the signal path
2. the ground for your new shorting jack is sketchy not shunting the 21k to ground, so it sneaks' in

[wsscott]

Shooter-the 1M resistor is wired directly to the Ground lug on the MIC Volume pot on one end of the resistor, and the other end of the resistor is wired directly to the Grid Pin 4. 

So I don't know what it could be.  It is close to the 100K resistor connecting the Tip on the Instr. Input Jack to its Volume Pot, and also close to the coupling wire connecting the Tip of the MIC Input Jack to its Volume Pot.  BUT I have previously disconnected both of those wires at their Volume Pot connections, and the 21khz is still there and distorts.

It only goes away when I jump the MIC Tip to the 1M resistor.

[passaloutre]

It sounds like the input volume pots are causing more problems than they solve (naturally, they don’t solve any problems since almost anything you plug into this will have its own volume pot), and one may be just plain broken. Everything I’m reading here points to that volume pot as the problem.

If this were my amp, I’d wire the input jacks straight to the grids with grid stop resistors on the tube sockets. Then I’d put the two volume pots after the first stage, ala 5E3.

[wsscott]

Good point-They are new pots and I've checked them for dead spots, but no issues.

I did think it was unusual to have the Volume pots before the Grids.  I've not seen that before, but that doesn't mean much.  But of course, this is the way it was designed and sold, so one would assume the circuit worked.

[passaloutre]

It was designed for lap steel guitars at a time when many lap steel guitars didn’t have volume knobs.

And just because it worked doesn’t mean it remains a good design.

[wsscott]

True-things were pretty basic back then.

I just did the grounding test on the 1M resistor again.  This resistor goes from the Grid Pin on the socket to the Input Jack Ground lug.  When I jump the resistor on its lead side connecting to the Grid Pin on the socket, and the other end of the jumper to ground, there is a very low background hum and the 21khz oscillation is gone.  When I connect the lead on the resistor to its lead to the MIC Input Jack Ground the oscillation is back.  Of course, the jumper is just another ground lead in this configuration.

If I remove the 1M entirely from the circuit with only the .006 cap in circuit, then I get the same oscillation/distortion.  Nothing is connecting the Grid to Ground.

If I connect one lead of the 47k grid stopper to the Grid Pin on the socket, and the other lead to the lead coming from the .006 cap that connects to the Wiper of the MIC volume pot, then same problem as above.

Finally, if I connect the Grid stopper directly to the Wiper, then the same problem remains.

It doesn't matter if a guitar is plugged in or not.  The hum and distortion are present in all configurations described.

I've ordered a new socket, and maybe that will be a solution.  I don't see what's left to be the cause.

[shooter]

what happens when you wire like this???

[wsscott]

I'll give it a try tomorrow.  The cocktail hour has started out here, and I stay away from electricity!!

[AlNewman]

It was designed for lap steel guitars at a time when many lap steel guitars didn’t have volume knobs.

And just because it worked doesn’t mean it remains a good design.

True, but this side is a microphone input, and most microphones don't have a dedicated pot for volume.

Again, I wonder if everything is shielded and/or electronics are out of reach when testing.

[wsscott]

The oscillation is present regardless if my cellphone is on the workbench, or upstairs in my bedroom to floors up.

Any suggestion on shielding? I don't think that's the source since the problem exists in different locations miles apart.

[AlNewman]

Maybe a cookie sheet on top of the chassis when testing?
The jack has been mentioned, but it's been tested.  Maybe if the routing of wires has been changed, shielded wire may be necessary, or chopsticking/rerouting while testing.

Again, there's only a handful of parts that could be faulty or installed improperly.  If you're sure everything is good, a modification like the grid resistor in series with the signal should solve your problem. 

Look to Merlin's explanation on Miller capacitance.
https://www.valvewizard.co.uk/gridstopper.html

[shooter]

I'll give it a try tomorrow.

Any suggestion on shielding?

quit following all the rabbits at once, shoot one, move on to next!

[wsscott]

Shooter-When I wire it like your last suggestion with the MIC Vol Pot removed from the circuit by disconnecting the lead from the Mic Input Jack tip from the pot and connecting it in-line with the 50k resistor and the 1M resistor that goes to Ground, I get a clean sound.  No distortion. Only a very slight hum.  Of course the Mic Vol pot and the Tone pot do not function, and the Volume is all the way up from the amp, but controllable from the guitar.

[passaloutre]

Shooter-When I wire it like your last suggestion with the MIC Vol Pot removed from the circuit by disconnecting the lead from the Mic Input Jack tip from the pot and connecting it in-line with the 50k resistor and the 1M resistor that goes to Ground, I get a clean sound.  No distortion. Only a very slight hum.  Of course the Mic Vol pot and the Tone pot do not function, and the Volume is all the way up from the amp, but controllable from the guitar.

This is how it should be wired IMO, and how 99% of amps are designed for a reason. The grid stopper makes a low pass using the tubes internal capacitance and cuts out your HF noise. Just put the volume pot after the first stage and move on.

[wsscott]

So how would you wire the Volume Pot back into the circuit? 

Insert it between the 50K and the Grid, by connecting the 50K's grid side lead to the Input lug on the Volume pot, and the Wiper to the Grid?  What about the .006 cap?  Leave it out of the circuit?

[passaloutre]

Like I said above, wire it like a 5E3 with vol/vol/tone in between the first and second stages.

[wsscott]

Would the attached wiring work if I wanted to keep the Instrument channel as is?

[shooter]

the plates are TIED TOGETHER, what is the 500k labeled that has a cap to ground???  it should act both as tone and volume to some extent

[wsscott]

I got no response from either the Mic Vol or the Tone pots when I tested it today.  It was putting out volume at full level with no reaction to turning the Mic Vol pot, and also no reaction when I turned the Tone pot.

The oscillation/distortion is gone in this configuration, and minimal hum.

So I'm just trying to see how I can control the Mic Channel's volume.  Tone may be an entirely different issue that can be dealt with later.

[passaloutre]

You'd have to separate the plates. Give each one its own load resistor (double the shared value). Send the signal out from the mic channel plate through a coupling cap (say 0.05uf) to the top of its volume pot. Then send the signal from the wiper to the grid of the PI through a 220k mixing resistor.

On the instrument channel, go from the plate through a coupling cap (0.05uf) through a 220k mixing resistor to the PI grid.

[shooter]

What does the amp have THIS pot labeled AS??????

[wsscott]

Tone

[shooter]

so make it look like this then call it MV for testing since earlier you said it had little to no effect on signal.

[wsscott]

I'll give it a try.  Thanks.

I remembered just last night, that when the Mic Vol pot gets past 9, the distortion cleans up and of course the amp is at full Volume.  That's the way it is with the Vol Pot removed from the circuit.  So I don't think the underlying issue has been resolved.

Another point of mystery, is that the socket for the 6SN7 preamp tube appears to be original to the amp; however, it appears to have been removed from its riveted original orientation with the other tubes, and turned 90 degrees and then bolted with machine screws and lock nuts.

Originally all the tube sockets had their locating pin at 6 o'clock.  This one is now at 9 o'clock!  Maybe someone was trying to solve this same problem by changing its orientation?

[shooter]

I remembered just last night, that when the Mic Vol pot gets past 9, the distortion cleans up and of course the amp is at full Volume.  That's the way it is with the Vol Pot removed from the circuit.  So I don't think the underlying issue has been resolved.

can you clarify this??  there is no mic-vol now, what happens sound-wise as you roll the guitar volume from 1 to 10???
my sense is you are running into clipping/compression, which causes distortion, not a "bug in the system"

[wsscott]

Shooter, nothing happens now when I roll the Mic Vol or Tone pot.  It puts out full volume.

Before this change when the Vol pot was still in the circuit, it would start getting the distortion around a setting of 7 on the volume pot which continued through about 9.  Then from 9-10 full volume, it clears up and sounds fine, just like now with the Vol pot out of the circuit.

These numbers are dial number positions, and not referenced to the position on the dial of a clock.