Fresh power tubes.......then smoke......

[ampgeek]

Here ya go mates!

Stethoscope background info about 1/2 of the way down here:

http://el34world.com/Hoffman/tools.htm

A picture of my humble little set-up attached.

I highly recommend the use of a 1M pot. However, do note that this will put a 1M load in parallel with the circuit that you are "probing". This could alter the results depending on where in the circuit you are listening in on.

Someone here made the excellent suggestion of including a pot bypass switch which I would do if I were building another one.

Also note that the amp being tested needs to have a load resistor installed in place of the speaker.

Yes indeed Joe...May Day must still be a welcomed event "back home"! My Dad is coming out west in a couple of weeks for Daughter #2 college graduation. Only one more daughter to go!!

I am playing the piss (well...as much as time will permit anyway) out of the "30 knob" Marshall to break it in before doing any final tweaks and shipping. I am astonished at how it's tone/behavior has changed after ~30 hrs of playing.

My buddy promised to replace the Explorer pick guard as it was TOTALLY busted to pieces (embrittlement). He is bogged down on that one so there isn't a great rush at my end yet. Truth be told...I rather enjoy playing something different from time to time. It is the first Marshall type circuit I have ever played.

Don't hesitate to shout-out if there are any questions on the stethoscope!

Rock on,
Dave O.

[Paultergeist]

Wow......that is pretty cool, Dave. Thanks for sharing it! (And I am certain that I will have some questions if i decide to build one of those myslef!).

As to my current challenging amp repair:

I turned the dang thing on last night and.....there again some of the intermittent noise. (Doh!) It did seem to be a bit less in intensity....but still unacceptable.

Last question I can think to ask: the amp has a 12AU7 at the PI -- I don't have a replacement 12AU7 -- I have been using a 12AT7 as a swap-in to test whether the issue was tube itself. I still get the intermittent noise with the 12AT7, so my assumption has been that the issue is not caused by tube. Is there any reasonable possibility -- due to the higher gain of the 12AT7 -- that using a 12AT7 to test in place of the (original) 12AU7 might be yielding an erroneous result?

Thanks for any ideas.

[JoeCon]

Dave

Thanks for the info on the 'scope"

Sounds like you might be keeping that Marshall or at least scale down to a "20 knob" version for yourself. My next build will be a 6V6 Plexi, I hope.
Currently playing the TwinTube Rocket 20 and Rocket 30 Octal at home.

See ya!

[ampgeek]

My pleasure as always mates!

Arghhh....so sorry to hear that the problem isn't solved Paul.

I must admit that I am not at all practically familiar with the cathodyne PI unfortunately and wouldn't know off the top of my head what the voltage toplogy should look like in this plan.

However, I did do some poking around and note that the venerable Fender Princeton uses one and it voltages are cited here:

http://el34world.com/charts/Schematics/ ... _schem.pdf

Note that a 12AX7 is in use here. The voltages imply a 1.1 ma quiescent current through the triode.

I thought that I saw some of your voltages that implied a ~2.2 ma current for your stand-in 12AU7. That might make sense given that the AU7 can source more current than an AX7..but again...I am not positive about that. In lieu of a known good 12AT7, throwing in a known good 12AX7 at the PI couldn't hurt just to get another "data point".

Is it possible that the previous gain stage is the source of the noise? I am not sure what components in that area have been checked out/swapped out. Old carbon composition plate load resistors (both before and at the PI) can produce similar, ugly artifacts.

Also, the 500K PI bias resistor connected directly to the PI grid would be a prime culprit. Although there is very little current going through it, its high resistance makes it prone to producing a significant signal "swing". A nice "fat" 1W metal film resistor there would eliminate that from my suspects list.

Good luck,
Dave O.

[Paultergeist]

Thanks Dave!

The PI tube voltages are indeed in the ballpark.

You insight about the resistors as suspects raises an entirely new question in my mind -- can a reisitor check as "right" with respect to measured resistance with a DMM and still be a contributor to a noise issue?

My ignorance is no doubt shining through, but I "checked" the resistance of the virtually all the resistors in this amp with my DMM -- thinking that I was thus eliminating suspects. It never dawned on me that perhaps resistors which measured correctly could be a source of noise?

Maybe I should do some more reading......

Thank you again. I am putting a second beer on your ledger.....

[ampgeek]

Thanks for the good thoughts Paul..but....I am not sure that I have earned the first one yet!!

Yes! Resistors can measure spot-on perfect in value but still cause the noises that you are experiencing. I couldn't begin to explain the electrical theory behind it. However, I know it to be a hard fact from dozens of experiences mostly in repairing aged audio gear of various flavors.

I fact, that is what has led me to be such an advocate of the stethoscope for us pure hobbyists. I shot gunned a really ugly distortion in one of my very first builds for weeks. Everything measured perfectly and I could tell that there was a beautiful sounding amp just "below the surface". Learned about the device at the Hoffman forum, whipped one up and within, honestly, 2 minutes I identified a noisy, brand spanking new carbon film plate load resistor in an area of the amp that I hadn't even considered as being problematic. Replaced it and found total joy!

Carbon composition (especially really old ones) are the most prone to being noisy. Carbon film are next followed by metal film. Not sure where metal oxide falls because they are not commonly used in the signal path of our standard fare. Hmmm...maybe that is the reason why?!?

Cheers,
Dave O.

[Paultergeist]

Dave, I don't want to jinx myself by celebrating just yet, but, I think the issue has been nailed.......(fingers crossed).

I was doing some more poking around, and I was puzzled as to why -- even with the plate, grid, and cathode of the PI all disconnected -- the amp still made a little bit of sound....AND I could still hear some of the noise issue manifest in that little bit of sound. At the same time, both the noise and ALL guitar signal vanished if I physically pulled the PI tube. That PI tube, however, also serves as a gain stage with the second triode. I had long ago checked all resistor values and replaced virtually every coupling cap in the amp. Your information, however, about resistor noise got me to thinking...... what if the other half of the PI tube had noisy resistors?

At any rate -- even though the values checked as good with the DMM -- I went and replaced several resistors servicing the other half of the valve used at the PI location (grid leak, cathode resistors on then other gain stage in that tube). I have powered the amp on and off quite a bit the last couple of hours and........sounds good with NO NOISE!

I really want to say "thank you" to you (Dave) -- and everyone on the forum -- for all their help with this. Noisy resistors.......wow......I never knew they could make THIS much noise.

Thanks again to everyone for all the help.

[ampgeek]

Right on Paul!! Congratulations. Glad to hear that you got to the heart of the problem and made it right.

Although I hesitate saying this...but...sometimes the satisfaction of simply fixing a problem(s) including the knowledge gained along the way exceeds that of having a working amp when done. Kind of like the "means justifying the end"!

Just out of curiosity (and to add to the data base): What types of resistors did you replace in the gain half of the triode? Carbon comp? Carbon film? Other? Presumably they were original issues with the amp.

Rock on,
Dave O.

[Paultergeist]

Hi Dave,

Honestly, I never would have thought to have considered the resistors if it had not been for your insights -- insights you generously shared during this thread. Yes, I had heard of "noisy" resistors before, but I thought the issue was subtle background *hiss* which only a dog or a sensitive scope could pick up -- this was NOISE which completely obliterated the guitar signal when said noise showed up. The intermittent nature of the problem made it particularly hard to locate.....

I do not know how to define the old resistors -- but they must be original. That would make them 50 years old. I would describe them as looking like little gray cylinders with slightly larger-diameter ends.

Here is the best photo I could take -- they are a bit small for my camera to catch in focus:

[ampgeek]

Always a pleasure to help whenever I can!!

I don't think that I have ever seen resistors that look like that. My initial sense was that they were wire wound construction but I can't be sure.

Regardless, congratuations on getting the amp fixed!

Cheers,
Dave O.

[Paultergeist]

Thanks Dave!

And don't forget that I owe you a beer!

Paul

[gktamps]

They look like wire wounds, used in old radios and TVs. Have many of those from my Dad's stash, never used them.

[ampgeek]

Roger that Paul!

I was always under the impression that wirewounds were taboo in the "sensitive" signal path amp sections (grid stoppers, plate loads, tone filters) because we seldom (ever?) see them in the "standard" guitar amp configurations.

Is it likely that we are seeing evidence of that in Paul's experience here?

Dave O.

[Paultergeist]

Interesting....all the original resistors in this amp are of the same look/design.....even the ones early on within the signal path.

From what images I have been able to dig up on the net, I would venture to say that the original resistors are, indeed, of the "wire-wound" type.

[Firestorm]

Really old carbon comps look like that, too - the "dog bone" resistors found in table top and console radios. But they're a 1930s thing. Hard to to believe Teisco would still be using them that late.