CIssue Solved)Is this a good guess about a noise issue in a amp

[Planobilly]

Hi guys,

I have a small amount of random noise in a tube amp I am working on. The sound of the noise changes frequency from low to higher and last for apx 500ms then stops and repeats at random intervals between 1 to 4 seconds apx. All new tubes which I tested and all voltages are correct at test points and no distortion of the audio signal. I will do some signal tracing tomorrow to isolate the issue.

I am guessing this is caused by a cap.

Do you think this is a good guess?

Thanks,

Billy

[tautech]

Wow that sounds real weird. 

You'd expect oscillations from a cap to be repetitive not random.
Could it be some outside EMI?
Heat related?
Shift your test to elsewhere in your environment to eliminate noise from an outside source.

[alsetalokin4017]

Yeah, outside EMI was my guess too. Maybe it's picking up your soldering station?

[poot36]

Some tubes can become microphonic so that could be your problem.  I assume that the sound is coming out of the speaker but if it is not that could be a bad connection somewhere that has enough power going through it but enough resistance to cause it to expand and contract from the heating and cooling to make sound.  It could be one of the tubes making the sound but that is highly unlikely.  Maybe an inductor or transformer oscillation?

[Gyro]

It might always be a minor stability issue, maybe excited by external RFI. It's worth checking that the signal tubes have grid-stoppers and taking a look at the output with a scope. Dirty tube socket contacts have been known to produce some very interesting sound effects too.

[Andy Watson]

Have you excluded all possible external influences - like mobile phones, etc..

I am guessing this is caused by a cap.

I've known reverse-biased electrolitics to behave weirdly.

[Planobilly]

I have been busy on other projects today. At the very end of the day I connected the ground clip of the scope to the speaker input and the noise stop. So grounding it stopped the noise but then I get no gain. I will post the schematic. There are only a few things between the plates and the speaker. Two or three resistors and a cap or two and the out put transformer itself. Typical feedback circuit to the pre amp tube. So not much to check out tomorrow to solve this issue...R39, R40, C39....ect

The strange thing is the schematic shows the speaker jack connected to ground at the jack and that is the normal way things are done on a older twin. This made in China reissue POC ( I say that for several reasons) is not wired that way. Problem is if I ground the jack per the schematic I lose all the gain. Not sure why yet....

http://ampwares.com/schematics/65_twin_reverb_@20manual.pdf

Cheers,

Billy

[Shock]

Check it's not your mobile phone, they are a good source of intermittent noise.

For microphonic tubes plug something into the input, turn up the volume up and give them a slight tap with a wooden chopstick, cleaning the sockets is a good idea as well.

[Planobilly]

Hi Shock,

Checked the tube and moved the amp to another location, checked all the normal stuff. This is some other issue in the feedback circuit or output transformer.

Cheers,

Billy

[GAD]

I've built a few tube amps and one of them had a sort of rattling noise that happened intermittently and only showed up as notes decayed. The "rattling" wasn't physical like you'd expect with a resonant piece of metal, but rather a raspy kind of sound in the speaker. It was crazy trying to find the problem which turned out to be a bad solder joint on one of the tube pins.

[Falcon69]

I was having a similar problem with an amp for my guitar.  Turns out, the neighbor had a base station CB, and when he kicked on his linear, my amp picked it up. :/

This, of course, was long before Cell Phones...when everyone stopped using CB's.

[David_AVD]

Does the noise pattern only appear at certain times of day for about 10-20 seconds at a time?

[gadget73]

Grounding one side of the speaker connections is typical.  When you ground it, you close the NFB loop, which reduces the gain by however much the feedback circuit is.   Having an open feedback loop might contribute to instability. 

[David_AVD]

The OP hasn't been back to look at this thread for a while.  Must have found something else to do. 

[Planobilly]

After replacing the per amp tubes for the third time the issue was solved.

I never would have though that two sets of new tubes would have been bad!!

Live and learn...lol


Billy

[Shock]

Were they microphonic? It's unlikely that a whole set of tubes are bad unless it was a bad batch or brand.

[gadget73]

You cannot switch signal tubes with a rectifier tube to test anything, other than how pretty the purple flash inside the tube bottle is when you power it up.  Rectifiers have 2 elements, everything else has more.  The pinouts are rarely the same.  Quite simply put, power pentodes are not rectifiers and they don't make particularly good ones even if you do strap the elements together to operate it in that manner.

Oscillation troubles don't sound like tube issues to me, though this one being a push-pull parallel does require that the output tubes be well matched in order to behave themselves.  Get that too far out of whack and it can do odd things, though usually its confined to overloading a tube and causing it to die more rapidly than it should.

If you don't have the negative side of the speaker lead grounded, you still have no operable feedback loop.