Magnatone Vibrato/Reverb build Project

[wsscott]

There is no wire from the cap to the foot switch, just from the cap to the pot and it is shielded and grounded.  And there no longer is any connection to the Reverb at the foot switch.  Before these changes the wiring went:  Rev foot switch, to cap, to pot, to foot switch with no shielding in effect.

[sluckey]

There is no wire from the cap to the foot switch, just from the cap to the pot and it is shielded and grounded.  And there no longer is any connection to the Reverb at the foot switch.  Before these changes the wiring went:  Rev foot switch, to cap, to pot, to foot switch with no shielding in effect.

And that's what I wanted. We'll use this as a starting point. Now just remove all wires from the footswitch jack, including ground wires and the vib wire to the 820K. I'm trying to clean up that tangle of wires around the FS jack and reverb jack. Try the amp like this. Then use3 a gator clip test lead to connect the 820K resistor to chassis. Report back.

Then tomorrow am when I'm rested we'll begin trying to track down the source of the oscillation.

[wsscott]

Still hum and oscillation.  Before doing anything, I removed the foot switch pedal from the jack, and had the hum and oscillation.  So I then un-soldered the 3 wires connected to the foot switch jack, one wire each at the tip, ring, and ground, and left them safely suspended from the board.  Turned it on, before jumpering, and still the same hum and oscillation.  So then I jumpered the 820K between the chassis and the ground side of the resistor.  Same thing.  I also jumpered it on the other side of the resistor where it connects with the two .01 caps.  Made the loud static sound.  So it seems that its not related to the wiring at the foot switch jack.  I'll defer to you on the next step, but was wondering about the V-7 Pin 2 grid connection to the Reverb jack coming from the Output of the tank and into the board?  I just unplugged the RCA jack going into the board from the Ouput jack of the reverb tank and NO HUM or OSCILLATION. Of course there's no signal going to the board, so maybe that doesn't mean anything.  But it clearly means its the signal from the tank to the board that causes the problem.  I have a 4FB3A1B reverb tank and both of its jacks are GROUNDED.  Could that be the issue?

[wsscott]

I saw this thread and thought it might be relevant:


    Posts: 8208
        View Profile Email Personal Message (Offline)

Hoffman Amps Forum image
Re: need some help with nasty parasitic oscillation
« Reply #13 on: January 25, 2017, 10:40:57 pm »

    Like
    Quote

Quote from: markmalin on January 25, 2017, 07:04:06 pm

    The tank I bought has input isolated and output grounded.  The jacks I bought are panel mount RCA's Fender style, so they are not isolated from the chassis.  Without the jacks isolated, could that possibly cause a ground loop?


No, that's why 1 of the tanks jacks is isolated. (1of the tanks jacks has to be grounded so the tanks chassis acts as a shield.)

If both of the tanks jacks were grounded and both of the chassis jacks were grounded then there would be a ground loop.
« Last Edit: January 25, 2017, 10:45:59 pm by Willabe »

[shooter]

I just unplugged the RCA jack going into the board from the Ouput jack of the reverb tank

for fun;
put the output jack back
remove the INPUT to the tank and ground the***tank*** center pin, effectively grounding everything to the right - still hum osc?

[sluckey]

The oscillation you are experiencing is not parasitic oscillations. It's the vibrato oscillator coupling into the reverb circuit. That's why when you shorted out the 820k to kill the oscillator, the oscillations went away. This is not related to the reverb hum.

I still can't get past the idea that you can put a jumper across the 820k resistor (thus grounding the top side of the 820k)and kill the oscillation, but you can't connect a gator clip test lead to the top side of the 820k and chassis ground without causing all kind of nasty sounds. Makes no sense at all unless something is amiss in your grounding scheme, which BTW, is a mess. In fact, the entire board layout on the left end is a bit crazy. I suspect layout and grounding are big contributors to the issues you have.

If you can get to the point of being able to kill the vib oscillator by grounding the top end of that 820k, then we can move the FS to that point and eliminate the vib oscillation from getting into the reverb. Unfortunately, I don't think I can do that without putting my hands on the unit, and I'm not in that line of business. So, it's up to you.

[wsscott]

Shooter-I have the output of the tank connected to the board.  I removed the input to the tank from the board and jumpered the center pin on the cable going to the tank input to the chassis, and it has the same Hum and Oscillation.  So unless I misunderstood your instructions, no solution--same problem. 

[wsscott]

Thanks Sluckey-I'll work on the 820K issue and see what I can figure out.  As long as the reverb output is not connected to its RCA jack on the chassis, the Vibrato is perfect, and no hum and no oscillation.  Its perfectly quiet.  Do you think the transducer for the output should be de-soldered so it's open and not grounded? 

[wsscott]

The schematic shows that the vibrato signal is being injected into the circuit at jumpered grid pins 2 and 7 on the V-6 reverb tube, and the signal with the vibrato is passed along from there through the tank, and out, unless that part of the circuit is bypassed with a dry signal avoiding the tank.  There is no background oscillation when the reverb is off, ie. "dry", so that part of the circuit must be okay. So the background oscillation must be "magnified" so that it becomes audible somewhere between the 1uf cap coming off of the V6 plates and the Reverb Intensity pot because manipulating that pot affects the loudness of the background oscillation.  I notice in my layout that I ran the wire from the V-7 Pin 3 to the 2.7K, as on the schematic, but then instead of taking the other end of the 2.7K directly to ground, I daisy chained it with the .047uf, 330K's etc. in the V-4 circuit to get the 2.7K from V-7 to Ground.  Could that be the problem?  Should I just go directly to Ground from the 2.7K?  Maybe that's what's picking up the background oscillation?

[sluckey]

There is no background oscillation when the reverb is off, ie. "dry", so that part of the circuit must be okay.

That could be a big clue. So, the vib may be coupled into V6, or the tank, or V7. But is it coupled through the shared B+ node, or ground connections, or layout?

Try moving the reverb tank around to see if you can eliminate the vib oscillations.

[wsscott]

I physically moved the tank around at all angles and the oscillation remained the same.

I looked at the layout and Node B is being shared by V-4 and V-6, and the way I connect the 270K and the 47K on pins 1 and 6 on V-4 is by jumpering the 47K with the 5K 10W which goes to V-6 and also to the Reverb Input!  This looks like an issue.  My wiring goes from Node B, to 270K (vib. circuit) then a wire from there to the 5K (reverb circuit) then by jumper to 47K (vib. circuit).  What if I removed the jumper between the 5K and the 47k, and then went straight from Node B to 5K, and ran another wire straight from Node B to 270K then from it to 47K?  So there would be separate wires supplying B+. Thoughts.

[wsscott]

Well, I tried that and no difference. I'll keep looking.

[wsscott]

I just changed the grounding I was thinking of in #408 for the 2.7K resistor so it now goes directly to the Ground Buss, but no change.  So that's not it.

[wsscott]

As for the 820K resistor, when I jumper it with test cables, the oscillation is gone but it has a loud static hum.  Even though its going to ground, there are still other components in the circuit that are still connected.  In fact the B+ from the B Node is still feeding that component via the 270K, the .02 and the .01 caps.

[sluckey]

the 5K 10W which goes to V-6 and also to the Reverb Input!

There better be a 1uF cap between the 5K/10W and the reverb input!

Try this... Completely remove the 1uf from the board. This will totally kill the reverb. Does this kill the vib osc?

[wsscott]

Nope, oscillation is still there.  If I disconnect the tank, the oscillation stops.  However, if I connect the input and output reverb jacks together with one of the cables, the oscillation is back and a bit louder.

[wsscott]

If there's nothing going into the tank's input jack (and the only connection to the input jack is the cable at the turret where the 1uf was located), then the signal has to be getting into the tank from the output jack.

[sluckey]

Leave the 1uF disconnected for now. Reconnect the tank cables as normal. Now use a gator clip test lead to ground the reverb output jack. Does this kill the vib osc?

[wsscott]

Clip 1 end of the lead to the chassis, and the other end of the lead to the ground lug on the output jack?

[wsscott]

Doing it as I describe above, the oscillation is still there and now it has some static in the hum.

[wsscott]

So I disconnected the Reverb Intensity pot from the lead coming off of Pin 7 on V-7 and the oscillation stopped. Still a hum.  However, the volume through the speaker was much lower, and now there was no vibrato at all.

[wsscott]

Well I re-installed the Reverb Intensity pot going to the grid of V-7, and its back with its hum and proper volume levels.  Seems to me that the hum is created somewhere between the tank's output jack and the reverb intensity pot.  There's not much to check there and I've done it and don't find any issues. So the only thing I can think of is to change the transducer bridge on the output of the tank so that it is no longer grounded by removing the solder bridge.  If there's no change then I can always solder it back.  But that's for tomorrow. 

[wsscott]

Hope everyone has a great July 4th!

[wsscott]

I have eliminated the transducer in the tank theory without de-soldering the joint.  I just unplugged the cable at the output jack on the tank and left it plugged in on the chassis and the background oscillation continued. I also have determined that the volume of the oscillation is controlled by the Reverb Intensity pot.  Neither the Main Volume pot nor the Vib. Intensity pot have any effect on the volume.  The Vib. Speed pot does affect the speed of the oscillation.  The Output Level pot will increase the volume of the oscillation but it is limited by the Reverb Intensity pot setting. So that satisfies me that the cause lies between the Reverb Output jack and the Reverb Intensity pot.

[wsscott]

Could it be possible that the background oscillation signal is being "syphoned" back through the V-6 tube and along the Dry Signal path into the V-7 C Node part of the tube, and the setting on the Rev. Intensity pot on Pin-7 is controlling the amount of the "syphon" effect leaking it to the Main Output jack going to the amp?

[sluckey]

So that satisfies me that the cause lies between the Reverb Output jack and the Reverb Intensity pot.

I agree. I want you to connect another large filter cap in parallel with that 1µF cap at node E. Use a 10uf, 20uf, 30uf, 40uf, whatever you have. Any better?

edit... If you don't have any filter caps then put a jumper across the 10K that's between node D and E. This will short those two nodes together. Any better?

[wsscott]

I didn't have any caps with enough V protection so I jumped the 10K.  Its very much quieter, but still there.  Sounds like progress.

[wsscott]

Sluckey-any thoughts on how I should proceed given this positive result?  Any ideas on the reason for what's going on?  Do you think its related to the fact that I don't have an Phase Inverter and output transformer in this design that might otherwise act to prevent this?  Thanks. 

[wsscott]

Would there be any benefit to putting a resistor on the V-7  tube at Grid P-7 to ground?

[sluckey]

Any ideas on the reason for what's going on?  Do you think its related to the fact that I don't have an Phase Inverter and output transformer in this design that might otherwise act to prevent this?

I think it's either layour or a poor solder joint somewhere. You don't need a PI and OT to prevent this.

Would there be any benefit to putting a resistor on the V-7  tube at Grid P-7 to ground?

It's OK to just go ahead and try something when you have an idea. However, there is already a resistor from V7-7 to ground. Adding another one is not likely to do anything.

Try this... Ground V7-2 with a clip lead connected directly at the socket and chassis. Any better?

[wsscott]

I un-jumped the 10K and jumped from V-7 P2 to chassis and its about the same.  Still quieter than without being jumpered.  I have found though that if I turn the Vibrato Intensity all the way up that the oscillation stops!  Its there until that final minute turn of the pot.  I'm getting the linear pot today, so I may try that.  Should I still go ahead and replace the 1uf Node E cap with a bigger one?  I'll also be getting the smaller R for the Node A-B connection and will remove the existing 2K 10W one which is currently just being jumped.

[sluckey]

Should I still go ahead and replace the 1uf Node E cap with a bigger one?

yes. I had to do the same with my M2 to eliminate some different bad behavior. The original used a 0.1µF at node D. I replaced it with a 47µF. Look at the schematic, lower right corner.

     http://sluckeyamps.com/lil_maggie/Magnatone_M2.pdf

Meanwhile, unplug your footswitch, set it aside, and don't plug it back in until I say. Does this kill the oscillation?

Then we will revisit reply #382 and stay there until we can successfully kill the oscillator by grounding the top side of that 820K resistor.

[wsscott]

You've probably forgotten that you had me totally un-wire the foot switch earlier on.  So nothing is connected to it and nothing is plugged into the jack. Those 3 wires are just free in the air. Also, you had me remove the 1uf cap totally at the V-6 P1-6 connection to totally take the reverb out of the circuit at that point.  So its still oscillating except when the Vib. Intensity pot is turned up alllllll the way.  Back off a hair and it returns.  I ohmed the 820K and its reading 817K (actually my meter is in auto mode and its reading 0.817Mohms).  I just jumpered the top side of the R to the chassis with a clips and still makes really loud static noise.  So nothing has changed from when we started, ie. the 820 is still acting the same way.

[sluckey]

I just jumpered the top side of the R to the chassis with a clips and still makes really loud static noise.  So nothing has changed from when we started, ie. the 820 is still acting the same way.

Use that same clip lead to just short out the 820K, One clip on each lead of the resistor. Does it kill the oscillator?

[wsscott]

No luck.  Still makes the loud static noise, same as if I connected the top end of the R to the chassis.  The static is so loud I can't tell if its oscillating.

[sluckey]

No luck.  Still makes the loud static noise, same as if I connected the top end of the R to the chassis.  The static is so loud I can't tell if its oscillating.

Remember this from reply #380?

Okay, so I just jumpered the 820K resistor, and as would be expected no oscillation--because its out of the circuit.  There's still hum when the Reverb is On, but no hum when its turned off--and of course no oscillation whether Vib  is on or off.

I don't think your test lead is any good or you are not getting a good connection. Check the end to end resistance.

Cut a fresh piece of wire. Solder one end to the top side of the 820K. Solder the other end to the negative lead of the Node B cap. Does this kill the oscillation?

[wsscott]

The test lead is reading 0.2 ohms across it, so its fine.  I went ahead and soldered a wire as you said and still the same result.  The result is that it creates a very loud screechy scratchy staticey noise.  I can't hear any oscillation but the noise may be covering it up.  If what you are expecting is a relatively quiet sound with maybe a little noise or hum, then that's not what's happening.

[wsscott]

I only see 1 820K R in the whole schematic and that's where the V-4 Tube is. So that's what I've been checking.

[sluckey]

I'm expecting to kill the oscillator without introducing any noise.

OK. Leave the wire connected to Node B cap negative lead and totally remove the 820K from the board. Connect the wire to the turret where the top side of the 820K used to be. Any luck?

[wsscott]

I'll give it a try.  Attached is a pdf image of just that section of my layout that we're working with.  This is how its wired.  Any problems with it?

[wsscott]

Same noise as described above.  Nothing different than when the resistor was still in the circuit.

[sluckey]

OK. Leave the wire connected to node B cap but remove from where the 820K was connected. We'll be using that ground wire again. So, the 820K is still off the board. Any noise? Oscillation gone?

BTW, the partial layout you just posted is a bit different form the one I'm using. Please re-post the layout that is exactly like your wiring.

[wsscott]

The layout is basically the same, I just separated the B Node into separate wires, but this part of the circuit hasn't changed.  Attached is the current full layout from which the last "focused" image was posted.  Okay, good news I guess.  I disconnected the ground at the turret, leaving it soldered to the Neg. B Node, and the sound is clear with NO OSCILLATION but some hum.  But not loud static etc. like before when I jumped the 820 K.

[wsscott]

Oh, and of course there's no vibrato anymore since the circuit has been broken.

[sluckey]

OK. Put the 820K back on the board. Cut that black jumper between the bottom of the 820K and the 3.3M. Move the green ground wire from the bottom of the 820K to the bottom of the 3.3M. This will leave nothing connected to the bottom of the 820K but we are about to change that.

Connect a new wire to the bottom of the 820K. Connect the other end of this wire to the VIB lug on the FS jack. Now use your test lead to short the FS VIB lug to chassis. Shorted will enable the VIB. Remove the short to kill the VIB. Test your Vib and reverb operation. Is all OK?

[wsscott]

Thanks.  Sorry for taking so much of your time. 

[wsscott]

Do you want me to re-install the 1uf cap at the V-6 P1-6 connection that you had me remove previously?  I also just received the replacement 100 ohm 2W Resistor for the 2K 10W at the Node A-B connection that is currently jumpered.  Should I install the replacement Resistor at this time?

[sluckey]

yes and yes

[wsscott]

Ok, got it re-wired as directed and replaced the R at A-B Nodes with the 100 Ohm one.  I jumpered the VIB footswitch and it turns on and off the Vibrato.  Still has background oscillation which stops when I take the jumper off of the chassis. Only issue is now I'm not getting any Reverb at all!

[sluckey]

Look over everything that's been done and/or undone while you tracked down where the osc was getting into the reverb.

Measure and post voltages on V6 and V7.