Short piece of wire attached to pin 9 of a 12Ax7. What is your role?

[mooreamps]

Seriously, though, from the picture it looks like the wire lays right across two coupling caps. I think the "induce some hum cancellation" is probably most accurate. Glued into that position, it radiates 3V of 60Hz AC. As close as it is to the caps, it has a good chance of coupling some of that into the signal path.
Just a theory. .

It is more than likely a part that was pre-wired for ground, and got kitted for an amp that did not require it. JFC, don't you people know anything on how an assembly line works ?

[R.G.]

It is more than likely a part that was pre-wired for ground, and got kitted for an amp that did not require it. JFC, don't you people know anything on how an assembly line works ?

Hi, Gary.

How far do you want to play this one out?

[bob p]

Radio amateurs were all familiar with the "gimmick" cap, two wires twisted together and clipped to length to tune the interwire capacitance. You get about 2pF per inch...

I wonder if Ken Fisher picked that up in the Navy.

[bob p]

Boogies are sensitive to the phase of the moon as regards oscillating or not. 8-)

My Mark IV is notorious for oscillating if you dime all of the controls on the lead (red) channel. But realistically speaking, the circuit has so much gain that if you do that then you're just asking for it to oscillate. No problems otherwise.

[Old Tele man]

I wonder if Ken Fisher picked that up in the Navy.

...can't speak for Mr. Fisher, but the Navy *did* teach it to me back in 1963-1964...and, although I'm not a radio "ham," they are the other people who typically use such "tricks."

...gimmick = "loosely" coupled, capacitive, AC-coupled NFB circuit.

[loudthud]

IMHO the wires are used to cancel hum picked up on the PCB. In the upper right photo, the wire is glued next to pin 2, the grid of one triode. In the left photo the wire is glued to some capacitors, you would assume coupling caps. In the lower right photo the wire is routed to pin 4 where it's next to a small ceramic cap. The wires are probably glued in place by techs that look at a scope and trim the wire to get the lowest hum. A procedure would be setup outlining control settings and expected results. If the wire is glued to capacitors, I bet they care about which end gets the outside foil.

[R.G.]

No response, so I'll reply just to the face of it.

It is more than likely a part that was pre-wired for ground, and got kitted for an amp that did not require it.

There is literally no way to tell why it was put in there. As I noted in my (semi) serious response:

... I think ... it has a good chance of ... remove it and see ... This looks like...maybe ... I suspect ... Just a theory. Could happent [sic] that way. Your mileage may vary. I'll respect you in the morning.

I think that about sums it up unless and until either (a) a veteran of the line that made those steps up an enlightens us or (b) someone experiments by removing them and listening/scoping one of the amps.

I personally doubt that it was put there just because the parts were kitted; the photos are all of sockets mounted on PCBs, no other hanging wires, and the wires were each dabbed down with a dot of glue. Not much reason to do that if you're a bored assembly line worker and you have an extra inch-long wire in the bag of parts. More to the point, PCBs are not generally stuffed by the bag-of-parts-kit method in a real assembly line. And PCB sockets are not usually pre-kitted with inch-long wires already attached before they're stuffed into the boards.

But as I said, I doubt we'll see the real explanation without more data.

JFC, don't you people know anything on how an assembly line works ?

John's Fried Chicken?

Please tell me about how this happened if you know.

IMHO the wires are used to cancel hum picked up on the PCB. ...
If the wire is glued to capacitors, I bet they care about which end gets the outside foil.

I think you're right.

[Enzo]

I instantly thought gimmick, but figured most of us here would not know what a gimmick was.

I have seen gimmick caps in schematics. It would say 2pf or whatever with "(gimmick)" next to it.

[J M Fahey]

They were used in "neutralizing", a trick to allow higher RF gain without getting into uncontrolled oscillation.
Being that cap values ranged into the 2 to 5 pF and had to be adjustable (If you put a new tube there you had to re-neutralize), the gimmick trick was a good practical solution, and much cheaper than an impossibly small trimmer.
Check any old ARRL handbook.

[mooreamps]

Hi, Gary.

How far do you want to play this one out?

I'm not. If you think it's some voodoo part for hum cancellation on the filament power, go for it. :{

-g

[R.G.]

I'm not. If you think it's some voodoo part for hum cancellation on the filament power, go for it.

Not wanting to play is an answer as well.

And if you want to think it's there because parts were kitted by people who didn't get the engineering change notice, and then the assemblers were too dim or bored to raise a fuss about unexpected parts, it's OK by me if you want to think that.

[mooreamps]

Not wanting to play is an answer as well.

And if you want to think it's there because parts were kitted by people who didn't get the engineering change notice, and then the assemblers were too dim or bored to raise a fuss about unexpected parts, it's OK by me if you want to think that.

I can see the photo's now. I see no reason for it to be there. I looked at the schematics. Appears to be A/C for filament power for all the preamp tubes, except for V1, which it's filaments are D/C ; but "no wire" drawn attached to pin 9.

-g

[Enzo]

I am sorry, I have a hard time wrapping my head around this proposal. Are you suggesting that on this printed circuit board, they anticipated some alternative wiring scheme involving pin 9 of the heaters and added an inch of wire as a connecting point? As if someone would splice something to it instead of simply soldering whatever it was to the board directly?

On this Mesa as on most tube amps we see with parallel heaters, the thing is wired already for 6v, which means pins 4 and 5 are wired together. SO that eliminated pin 9 as any center tap access for the heater, unless part of this alternative wire connection was also to go under and cut traces between 4 and 5 and presumable then ADD wires there for alternative 12v power. Hence whether it is one of the AC heaters or the DC heater, it still represents a wire to one side of the heater string, not any center tap.

SO "pre-wired for ground"? They intended to ground one side of the heater string on some models? and to do that they added an inch of wire on a socket? Considering they already have the two 100 ohm resistors on the board for false CT, part of that regrounding then would be to snip out those resistors.


I can't buy that at all.

[R.G.]

I can see the photo's now. I see no reason for it to be there. I looked at the schematics. Appears to be A/C for filament power for all the preamp tubes, except for V1, which it's filaments are D/C ; but "no wire" drawn attached to pin 9.

I removed my original response to this. As I myself posted in other threads, the right thing to do is when one finds one is wrong, admit it and move on. You've done that, it's a good start. I commend you.

=====

I'll go through my reasoning just to be clear.

A wire all by itself is kind of a monode - no possible current flow, and hence electronically a "don't care" until one remembers that the wire on pin 9 carries the voltage of pin 9 and that all conductors are capacitively connected to every other conductor in the universe. What varies is just the capacitance.

The capacitance from one conductor to another is a somewhat complex function of the geometry of the conductors and other nearby conductors. Capacitance varies inversely with the square of distance for all the geometries if you can write down the electrostatic equations for the geometry, but still inversely with the square of separation. So you can vary the capacitive coupling between two conductors easily by changing the separation and geometry.

Attaching a wire to pin 9 and laying it near a high impedance point maximizes the capacitance from pin 9 to the high impedance point. The capacitance is small, but for something like a grid with a grid leak of 1M or so, even small capacitances let some signal get transferred.

For a back-of-the-envelope calculation, a grid being driven by a 1M source, we could roughly guess at maybe 2pF of capacitance from the wire to the grid. This is handwaving, as the actual capacitance is a complex function of the geometry, but "gimmick" caps have this range, as noted earlier in the thread. A 2pF cap at 60Hz is an impedance of Xc = 1/(2*pi*60Hz*2pF) = 1300M ohm. With a 1M resistive coupling to ground through some source impedance and grid leak, the 2pF cap would transfer
the voltage on the wire in the ratio of 1M/(1301M). So 3.15V on the wire would couple over as 3.15V*(1/1301) = 2.42mV of 60Hz transferred to the grid. a few mililvolts is all you have to work with so even if you hose it up, the hum doesn't get all that bad.

2.42mV of transferred 60Hz signal is reasonably in the range of hums from other sources. It's phase and magnitude will then add to the other power line hums that are unavoidably picked up in the amp. If the phase is such that it adds, hum will go up. If the phase is such that it cancels the other hums, hum will go down. In this way it's like a variable hum-cancelling pot on the heater winding itself.

Putting such a gimmick on several tubes where the signal is inverted after each tube will let you add more or less of one phase or another, and select the wire to tune that makes hum go down. The earlier in the amp that the hum is induced, the more it's amplified by later stages, so picking which tube you tune lets you dink with how big the induced hum is in the final output, so you have a step-change in tuning out (or in, if you're a glutton for hum!) hum.

Since the hum varies by geometry, you can experiment on test with which wire in your tuning set lets you tune hum in or out by bending the wire closer or father away from the target pin or capacitor with a non-conductive probe. This lets you pick a wire to tune, as bending the wire away changes the capacitance in the same direction (smaller) as does clipping wire lengths off. So one reasonable way to tune hum out would be by starting with all the wires bent away from the target pins or caps. This mimicks the "no wire at all" length reasonably well. You then bend the wire on the tube closest to the front end of the amp and see what happens to the hum by listening or watching a scope.

If the hum goes down as you bend the wire in, great. If it goes down, then rises again, you can get too much using the full wire, so you clip off a bit of the wire and try it again. When you get down to the length which only makes hum decrease when bent fully down, glue it into place. If you've clipped almost all the wire off, clip it all off and go to the next tube. With several to tune with, you can possibly do a good job of tuning out hum. You have amplitude of hum as adjusted by wire position and length, and phase as adjusted by selection of the triode, available to tune hum out.

In any back-of-envelope calculation, you need to validate the result by identifying possible errors and omissions. The grid impedance can easily be lower than 1M, perhaps down to 100K. The capacitance is adjustable by how you bend and then clip off the wire. So both the R and the C in the AC voltage divider are variables. And that's just the effect at power line frequencies. The capacitance this sets up also adds to the grid-to-ground capacitance of the tube, so it has an effect on the high frequency response of the triode section, appearing in parallel with the internal grid capacitance. Probably not a big deal a audio, but it's there.

And that's what ran through my head when I saw the pictures. And that's why I guessed it was a hum tuner. It could plausibly be. Might, could be. Maybe even worth having your final tester do this by ear. Even a not-too-bright final tester could be taught this pretty quickly.

[Enzo]

This all reminds me of some discussions we are having elsewhere, like the hum in the AC30. And the recurring "mystery" where someone reports that they have hum that is least not at zero on the volume control, but at 2 or 3. That is of course no mystery, just out of phase hum from different sources cancelling out in that stage. I have no trouble imagining a little 3VAC antenna introducing enough reverse phase AC to cancel some other hum source.