Peavey Windsor - Hum and White Noise Hissing

[TomCarlos]

Ok... retesting ....

The PreGain is at 0, the Texture is full power, Master Volume is at 12 (full warp speed), all other controls are at halfway.

Power Up and let amp warm up.

Adjusting the PreGain (slowly) until the Signal remains clean (or starting to distort) at V2, Pin 6. The results at V2 Pin 6 are approx. 2.5 Vrms. (Photo 1)

The Output looks “relatively” clean and I am measuring approx. 7.92 Vrms. (Photo 2)

Just for the heck of it, I am also including a screen shot of V3 Pin 6. You will see 11.7 Vrms. V3 Pin 8 looks almost identical.

I checked the output ac voltage using my HP 400e – it read close to 8.0 Vrms. (Photo 3)

[TomCarlos]

It is the FLIPPING Boost switch. Once you engage that, voila.

So same test... clean signal up to V2, Pin 6.

Output = 17.59 Vrms @ 4 ohms yields 77 watts RMS.

Oh lord... :-)

[Helmholtz]

What are tone controls settings?

[The Dude]

I thought the same as you are thinking (I believe). In an earlier post, he said tone controls were at 50%.

Edit: From post #53 - "all other controls at 50%"

[TomCarlos]

Helmholtz...

Perhaps I was reply as you were typing. The issue is the Boost Switch. My lord, it acts as a HUGE attenuation to the signal.

I tested the amp with tone controls at 50%.

It seems like it doesn't take much to drive this amp into a distorted signal. I guess that is how it was designed.

[g1]

Yes, they are supposed to be a British voiced amp (thus 'Windsor'). With tons of gain on tap and not much accommodation for clean.
The cold clipper stage dates back to the JCM800 series for mainstream production (2203 etc.). Those amps aren't much good for clean either.

[TomCarlos]

Tons of gain.... and I guess that is why I am seeing lots of threads on other sites where guys are performing Mods, using 12AU7 instead of 12AX7 tubes for less gain, etc.

The Pre Gain control is VERY sensitive up to the #3 level.

Anyway, I will let my friend give the amp a test drive. Maybe he will like it the way it sits. No sense in dickering with the Q1 transistor or anything in the FX loop if it not being used.

Thanks again for the help !!

[Helmholtz]

The Marshall 2203 uses a 10k cathode resistor in the "cold clipper" (second stage here) making its effect more moderate while preserving more clean gain. But there's another distortion cell, namely the DCCF which provides rather soft and symmetrical clipping.
Also as the second triode doesn't use a cathode bypass cap, grid conduction doesn't happen that early.

If you want to increase clean headroom and make the pre gain control less sensitive, remove C5 and lower the value of R21 (e.g. by paralleling a 15k resistor).
That would lower available preamp distortion while essentially preserving overall gain.

[TomCarlos]

A Post-Mortem Report ….

Immediately after releasing the amp, the owner took it home, used the amp for a day, and all was well. He did not use the amp again until two weeks later.

Two weeks after I released the amp to the owner, he called me to report a major issue. His report was that the amp had a loud hum, crackle, excessive noise, followed by Red Plating of tubes V7 and V5. He immediately turned the amp off. During this session, the owner told me he was using outboard gear in the Effects Loop.

From my notes, when I installed the matched El34s, the plate voltage was close to 440v. The bias point T29 was set to -41.22 volts. Each of the tubes were close to 36ma, which would have them running close to 60%. These were a Quad set. But I wonder if using the tubes (and a break in period) caused them to behave differently.

Upon receiving the amp, I pulled the power tubes to test for shorts – no problem. Each of the tubes passed an Emissions test. But I did notice that the Emissions reading for two of the tubes was off a bit. I don't recall if they came that way. I also have my test jig to look for similar current flow – all tubes conducted but I noticed the tubes no longer seem to match like they did prior to the incident.

I installed the tubes back in the amp, checked the negative bias, and it was set as before. I did make a slight adjustment this time to try and find the happy point as I noticed the tube current no longer matched as it did before.

While I was reinstalling and resetting the bias, I noticed the Phase Inverter tube was dark. Upon testing, the Filaments burned out. So V3 fried when the amp went haywire.

Back to the biasing and current levels -

Plate Voltage – 440v
V7 – 33.8ma
V5 – 31.1ma
V4 – 34.9ma
V6 – 29.8ma

So for a matched set, when I see something outside 10%, I would have some concern. But I guess these tubes are still close enough for this type of amp which is intended to be a heavy overdrive type output.

Using my Infrared Thermometer, I am aiming at the center of each tube (after a warm up period). Here are the readings.

V7 – 359 degrees F
V5 – 347 degrees F
V4 – 345 degrees F
V6 – 310 degrees F

I am wondering if something happened when using the Effects Loops. I know G1 mentioned (Post #12) that this amp “does not use the typical series FX loop” so I am wondering if something happened here and it put stress on V3 and the Phase Inverter. Perhaps pulling a cable and reinserting with the power on could have caused a problem?

So far today, with V3 replaced, the amp is behaving as expected and I will keep it on the bench for testing. But I am wondering, do I return it with a warning on using the FX loop and perhaps to avoid it?

Thanks!

[Helmholtz]

There's no operational stress that could blow the PI heaters - except excessive heater voltage.
But heaters filaments (just like incandescent bulbs) have a limited lifetime and if aged may burn out without identifiable reason.

[TomCarlos]

V3 was a relatively new tube... so could be coincidence that it failed. As for the hum, crackle, supposed red plating, I don't know how that would have taken out the V3 heaters.

[Helmholtz]

As for the hum, crackle, supposed red plating, I don't know how that would have taken out the V3 heaters.

As said: No way.

[g1]

Hum, crackle, red-plating of power tubes could all be due to oscillation. Could also be caused by something entirely different, but oscillation is a possibility.
Another slim possibility is that the PI tube failed in some way that caused oscillation. But some kind of feedback/oscillation could also be introduced through the FX loop.
In any case, I think the main worry would be that red-plating etc. was caused by loss of bias. Any kind of bad solder or iffy connector in the bias line could be a culprit.
A bad new tube that fails under use but tests good on a tube tester could also be responsible.

[TomCarlos]

Oh G1, I was afraid you would say that. For me, "oscillation" is like "porn" ... hard to describe (what it might sound like) but I know it when I see it (on a scope). Maybe it sounds like a high pitched squeal. I'll remember that.

I tested the amp using a single pedal (Boss Flanger) in the FX loop. I didn't hear anything odd. I have asked my friend to bring his pedal board tomorrow and we'll give the amp a thorough testing using the setup he has.

As for the bias circuitry, it looks pretty solid. I traced this out from the Power Transformer to points P23 and P24, through the connectors, to the circuitry, all the traces, up to the Output tubes. I think I can rule out a fault in the bias circuit. And today, no issues at all.

So let's hope it was a faulty V3 that somehow acted up which caused this problem.

I'll report back - if not tomorrow, next week once we have a few more days of real life testing.

[35L6]

While we're waiting , does C2 do anything ? Shouldn't be any DC and corner frequency is below guitar signal . Is there anything special going on ? Rumble filter ? Just because I don't know what they are doing doesn't mean they don't !