Plate stop resistors?

[R.G.]

Actually, even if the oscillation is at RF, it can sometimes cause audible artifacts... Also, repetitive peaks of noise that occur at a low frequency, like 50/60Hz rectifier noise, can induce audible hash because of the repetition rate, even though the oscillations themselves are above the audible range.

Another symptom can be an "angry" hiss.

[Enzo]

I usually recognize the sound of an amp oscillating at RF. As described above, the sound will be blanketed. The signal will pass, but will have a certain weakness about it, a certain lack of definition, and usually a little more background hum. But it is a particular pattern of all those things that says, "scope me please."

[alchemy]

Wow! This turned into an interesting thread while I was away!!

So from what I can gather from all of this info that is way over my head, I should be able to remove the 100 ohm resistors with no ill affect besides it possibly beginning to oscillate? And if it does oscillate (or if I decide I want to put resistors on the plates for other reasons), I should lower the value to somewhere between 22 and 50 ohms and bump them up to 5 watters. Does this sound like a reasonable solution without presenting any new problems?

Seems simple enough. I guess it's just watching the 3 watters go up in smoke that psychologically makes me think the plates are drawing too much current!! Of course 3 watts is too small, but still, watching them burn makes me paranoid as I'm still somewhat new at this!!

One last thing: Oddly, when I was playing the amp the other day at a somewhat low volume, everything was fine, until I unplugged a patchcord quickly to remove a pedal. In literally seconds, one of the 100 ohm resistors starting pouring smoke. Is this because the signal from the unhooked guitar cord acted as an antenna was inputing a much larger (hum) signal, and this could have pushed it over the edge as far as max dissipation?

Thanks a lot for the help, as always!!! Cheers!

[Steve Conner]

That's a pretty solid sign that the amp is oscillating when the input cord is unplugged. That is quite possible through just plain bad design, bad layout or whatever. My last homebuilt effort did it, but I recognized the symptoms and fixed it.

Nobody but a boutique builder could get away with actually shipping an amp that did it

[mooreamps]

Tubes like the 807 and 6146 are notorious for parasitics: they were designed for radio transmitters, so they have lots of gain at radio frequencies, and they just want to oscillate.

Steve is absolutely correct on these points.. We use power pentodes and beam power pentodes for HF and VHF transmitter power amplifiers because of the nature of the upper frequency response of the pentode vacuum tube... Thus, if using these tubes in the audio frequency range, additional circuitry "should" be employed to null out the tubes ability to operate in these frequency bands ; and/or break out into parasitic oscillations.... Yes, it was I who mentioned the whole tube does ring when this happens...

Also Steve brings out a good point about variations in screen current.. Screen current in a tetrode/pentode is a function of the mechanical screen wire alignment with respect to the control grid... For example, if the wires on the screen grid are in perfect alignment with the control grid ; or in the shadow of the control grid ; the electrons from the cathode should flow around the wires of the control grid, flow around the wires of the screen grid, and strike the plate. It was later on noted in one of my posts, when you put vacuum tubes in a guitar amp and lug it around from gig to gig ; with the tubes hot, there is the tendency to knock the wires on the screens out of alignment... Thus, using a screen grid resistor in such an amplifier helps keep screen current in check.. Using a SGR in an RF transmitter would also be helpful as well....

-g

[imaradiostar]

Interesting thread guys, lots of good stuff!

Jamie

[redelephant]

One last thing: Oddly, when I was playing the amp the other day at a somewhat low volume, everything was fine, until I unplugged a patchcord quickly to remove a pedal. In literally seconds, one of the 100 ohm resistors starting pouring smoke. Is this because the signal from the unhooked guitar cord acted as an antenna was inputing a much larger (hum) signal, and this could have pushed it over the edge as far as max dissipation?

If u had a very loud hum coming from the speakers, yes. A cable will act as an antenna if not connected into a source (guitar or effectpedal), so either unplug from the amp or turn the gain down.

I find it odd resistors that can get overheated like this are even in there.

Perhaps u should get a scope?

[mooreamps]

One last thing: Oddly, when I was playing the amp the other day at a somewhat low volume, everything was fine, until I unplugged a patch cord quickly to remove a pedal. In literally seconds, one of the 100 ohm resistors starting pouring smoke. Is this because the signal from the unhooked guitar cord acted as an antenna was inputing a much larger (hum) signal, and this could have pushed it over the edge as far as max dissipation?

Thanks a lot for the help, as always!!! Cheers!

No, not from the signal from an unhooked guitar cord.. But, a 3 watt resistor at 100 ohms is only rated for 30 mills. You need to figure out what your average plate current is, and up the wattage accordingly... Well, I suppose 50 ohms is better than nothing at all ; or at a minimum, two 50's in series....

-g

[Albert Kreuzer]

But, a 3 watt resistor at 100 ohms is only rated for 30 mills.
-g

173.2 mA
P=I^2*R

[mooreamps]

173.2 mA
P=I^2*R

You're right. I was thinking something else...

-g

[redelephant]

No, not from the signal from an unhooked guitar cord..
-g

...if everything was working normal, and as soon as he unplugged the end of the guitar cable the hum came and the resistor smoked...you still don't think it can be the cable? If the gain and volume knobs are up, such low frequency hum on the input can produce enough signal swing over those under rated resistors to make them overheat. My guess anyways.

Platestoppers need to be rated according to usage. All parts have a derating curve related to temp, and u should asume the resistor temp is above ambient during use. So even if a x-wattage resistor should do, it's always good to go one step up. For stopping parasitics, which these are for, 50ohms is more than enough. Half the resistance and u only need half the power rating, obviously.

I had oscillations in my first build with parallel tubes and ever since then I've either used platestoppers or cathodestoppers (10-50ohm resistors) on fixed biased amps. On cathode biased amps I prefer seperate cathode resistors and caps for each tube, which serves the same function.

[mooreamps]

...if everything was working normal, and as soon as he unplugged the end of the guitar cable the hum came and the resistor smoked...you still don't think it can be the cable? If the gain and volume knobs are up, such low frequency hum on the input can produce enough signal swing over those under rated resistors to make them overheat.

Oh, you mean unplugged from the guitar, not from the input of the amplifier... Well, that could make the cable an antenna, and feed 60 cycle hum into the amplifier... Ya, I could probably see that....

-g

[alchemy]

Oh, you mean unplugged from the guitar, not from the input of the amplifier... Well, that could make the cable an antenna, and feed 60 cycle hum into the amplifier... Ya, I could probably see that....

-g

Yes the instrument unplugged, obviously. If the cable was unplgged from the amp, it would not be an antenna.

I added another 100ohm 3 watt in parallel on each tube, and so far everything seems fine. I haven't got to really push it yet, as the neighbors are probably already annoyed as hell at 4 kt88's at 1/3 power! I tried unplugging the bass again while monitoring the drop across the plate resistors and it did not go up so maybe it was something else.

Another thing I thought odd about this amp is that the screen resistors are only 250 ohms. Is it normal for them to be this low in an ultralinear connected design, or could it perhaps be that the voltage drop was too much on anything larger?

Also, I seem to remember reading that using boost/distortion pedals in front of big ultralinear amps (Marshall Major) was a bad thing as it would fry the OT. Is this something that is inherent in UL, or some other design aspect of the Marshall Major? I only ask because when I feel like playing bass (i'm not a bass player) I generally like to use a TON of fuzz, and I sure as hell do not want to buy any more kt88's or an OT!

All of this things issues are, I'm sure, from poor design. The faceplate says "prototype 1", and I am farely certain this amp never went into production (though the builder did go to prison for taking deposits and spending them on cocaine and hookers and never building the amps!!). A shame as he was building some very nice amps. I had originally got it to tear apart and use as a platform for another build (it was only $40) but then decided I would try to work out the bugs and use it for a while. I guess I really need to get around to tracing the circuit and drawing a schematic so I can get a better idea of what he was trying to do.

Thank you all for your input. I have learned a lot from this thread...