Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: musicfixerupper on November 19, 2010, 07:48:26 am
-
I was reading about a trouble shooting technique in on of Gerald Webers books where you can "shut off" a section of a triode, by shorting the plate to the grid? I lost the book so i can't remember how to do it exactly.
thanks
Joe
-
Don't do that!
I haven't read that from Gerald Weber, but if you short the gird to the plate, you turn the triode on as hard as it can possibly be; you turn it into a diode, with the plate current limited only by any series resistance connected to the tube. Damage is likely to follow.
Maybe the directions for a special case? I say that because if you connect the grid to the cathode, you effectively reduce the bias to zero, and plate current goes way up (though not quite as bad as grid-to-plate).
Lastly, connecting the grid to ground would make sense only if you were trying to prevent an a.c. signal from entering the grid. The triode is almost always connected to ground through a largish resistor (200k-1M), which holds the grid at ground potential. Connecting a jumper from the grid to ground should also short any incoming signal voltage to ground, and prevent a signal from entering.
-
Don't go shorting stuff ad lib in a tube amp, and probably even more in a solid state amp. Destruction is likely to be the result; and is a SS amp, it will be instant.
With tube amps, you can in 99%+ of cases pull a tube out of the chassis with no bad effect in an effort to isolate a problem as coming before or after that tube in the circuit chain. This doesn't often get you much unless you have the ability to proceed further, to isolate a problem down to particular component. Or in the rather rare case of a completely dead tube. That happens, of course, but in my experience, if it was working yesterday it's likely to be working today. Isolation requires voltmeters and at least minimal understanding of how the stuff works.
The classic method of troubleshooting in DIGITAL circuits is "half-splitting"; where you conceptually cut the circuit in half, then disable each half one at a time. You locate the bad half, then cut THAT half in half yet again. That has some measure of applicability in tube stuff, but more important is the experience you gather working on commercially-built amps that were once working and are now not working. If you are troubleshooting an amp you built, you really do not have the ability to assume that ANYTHING about the amp was properly built. Unless of course, you measure things, or see lit-up heaters or hear hum or can shake the reverb can or can see plates glow red or smell smoke or the peculiar smells that parts emit when they are being overworked....and again, it's a matter of experience either directly building tube amps, or other circuits.
For a time, I worked at Audiotronics in LA. My job was to troubleshoot stuff coming off the assembly line after someone else had given it the very first "smoke" test. Some things had exploding electrolytic caps, very obvious. Some days, Consuela put all her diodes in backwards and all day I just de-soldered two diodes and flipped them around. That form of troubleshooting was exceptionally visual: It was pretty rare that the first thing I did was to plug the thing in. (These were mostly little CRT monitors and indestructible record players for schools.) No, I was looking for yellow and red to be reversed, diode bands wrong-way oriented, components left out. So it was different than troubleshooting something that was definitely working at some point, and it was different than troubleshooting something I just built. Because these were built on PC boards, there was very little chance of a bad wiring connection. There was a slightly larger chance of a solder bridge, but the boards were visually inspected after being flow soldered, so that was fairly rare. By the end of the day, I had fixed 47 little monitors (they were OEM open-frame type affairs so I didn't have to do much disassembly) and the visual familiarity with literally each resistor built up pretty quickly.
-
A good visual inspection is often the most overlooked step in troubleshooting. No pun intended.
-
Glass magnifiers and bright lite do wonders when looking for bad solder joints.
I have yet to put together the listening amp Doug describes on the Library.
Also, the disturbance test works well too. just poke on plate and grid pins from power stage to input stage hearing for noise when you actually touch voltmeter probe to socket pin. Found many problems pretty quick doing so..
Hope this helps.
Best Regards,
Rzenc