Rocket Build...noise issue.

[matt h]

(deleted)

[tubeswell]

I'd suspect a leaky coupling cap letting dc onto the grid and throwing out the bias. (I've found that the quality of those black mica caps can be a bit hit and miss).

[Richie]

that would be correct. any of those mica caps leaky? what happens if the input goes direct to the tube pin, and no connection to the board? maybe take a dental mirror and look under the board for a miswire or a piece of wire or solder.

[Firestorm]

Is that bypass cap in correctly? I'm not familiar with that type (what the band means) but it almost looks like the end you have going to the cathode is marked with a "-".

[MGW]

The bypass cap is correct. I reverified last night. The black band furthest from the tubes is ground. Not sure how a leaky mica cap on the pin 6 would impact pin 7 of the same triode. I see how it would leak voltage into the next stage, but it is clean.

[matt h]

(deleted)

[MGW]

Just to clarify, we ARE talking about the input triode for the amp, right?

Therefore, there is NO coupling cap preceding its grid?

That is correct. This is the input triode...no coupling cap preceding it's grid. Well...there is a cap and grid leak resistor now, but I still don't think the issue is solved as it shouldn't be there.

[tubeswell]

Ah I see - I thought you were talking about the grid on the 2nd stage

The grid of the 1st stage shouldn't have any DC on it (unless you don't have a grid leak (resistor) path to replenish electrons that otherwise get lost from the grid because of its close proximity to the cathode? You say you've put a cap in series with the grid? Is it between the grid and the grid leak resistor by any chance?)

[MGW]

The grid leak resistor is between the cap and the grid. I added that after the initial issue. I'd like to be able to remove all of that crap though and still not have DC on the input.

[Firestorm]

Grounding issue? It looks like you're using the chassis as ground for the input jack and the grid load resistor and then there's the grounding buss along the backs of the pots. Is there 0 ohms between the jack ground and the ground buss?

[Tillydog]

Input jack is hiding it.

OK - just double check that it's wired correctly - the circuit appears to be acting like this resistor isn't making contact with ground.

When you measured 1M between grid & cathode did you have anything plugged into the input jack?

How do you determine the appropriate value for a grid leak resistor?

For what I was talking about (providing a grid to ground path for DC[*] after you have put a capacitor in series with the grid stopper on the input) then 1M is fine. You could move the 1M from the input jack (may not be best practice, but it should work - certainly for troubleshooting).

Connection between chassis and B+ ground? the filter caps are grounded at a lug on the recitifier socket. Is that what you mean?

Yes - I didn't want to make any assumptions about how you had done this.


[*] I believe this is still a grid leak resistor, whether or not it is used to generate a bias voltage, but I'm not going to fight about it!

[MGW]

There is not a 1M on the input jack between tip and ring. Tip goes to a .022uF cap before the 33k grid stopper. There is a 1M from the junction of the cap and the 33K to ground. That junction is seeing about -120V. I can also hear a very low (volume) low frequency hum even with the volume off. It varies with movement on the bass pot. Don't know if these are related, but at this point I assume everything is relevant.

Measurements between the grid and cathode have been taken with a cable both plugged in and unplugged, with a tube and without. Any variance is negligible.

Input jack ground to buss bar reads 000.4 ohm. I don't think that is the issue.

Also, It's dropping about 100 volts across that plate resistor going to pin 6.

Here's the whole gut shot.

[matt h]

(deleted)

[MGW]

back the truck up...

I thought you had negative-point-something volts (hundred something millivolts).

HA...yes...back the truck up. That was a very significant typo. It is indeed floating around -120 mV. Sorry bout that. Too many moving parts at the moment.

[Tillydog]

There is not a 1M on the input jack between tip and ring. Tip goes to a .022uF cap before the 33k grid stopper. There is a 1M from the junction of the cap and the 33K to ground. That junction is seeing about -120V. {assume mV}

...

Measurements between the grid and cathode have been taken with a cable both plugged in and unplugged, with a tube and without. Any variance is negligible.

OK, you added the cap and moved the 1M resistor...

Before you moved it, you should have seen 33k ohms between grid and cathode pins with no cable plugged in (due to the shorting contact on the jack), and 1M with a cable (no guitar) - if you always saw 1M, it makes me suspect that something was amiss with wiring of the 1M resistor on the input jack, but things seem to have moved on now.

Butting out!