Signal Tracer Questions

[martin manning]

Measure the voltage to ground at each tube pin, and the supply node voltage for that stage (on the other side of the plate resistor). Calculate the current flowing through the triode as (Vsupply-Vplate)/Rplate and/or Vk/Rk. Depending upon the circuit, you should be seeing about 1-1.5 mA (0.001-0.0015) for a 12AX7. Typical cathode voltage will be 1 to 2V.

[husky]

A Signal Generator and scope would make it easy as pie

Actually, in this case, I doubt a signal generator and a scope would add any value at all. You'd simply see signal at the grid, and no signal at the plate. He already knows that is the case.

Doesn't seem like he is so sure his tracer is working. A scope and signal generator are invaluable tools.

I borrowed an old function generator, but it is really hard to get it set for 4.0 mV like the schematic calls for.

4 is very low
I usually use around 50mV 1K
Try and disconnect the stage after the coupling cap to isolate where you are getting a short. Or pull the tube and with power off see what kind of DC resistance you measure to ground after the coupling cap.

[Gibsonman63]

The schematic calls for 4.0 mV at 1kHz, but I am finding it difficult to generate a signal that low consistantly. 50 mV would be a more reasonable value.

http://www.webphix.com/schematic%20heav ... deluxe.pdf

I picked up a scope yesterday. I'll try to isolate the first stage a little better tomorrow.

I have a new trick. Painter's tape and wooden door shims work well for keeping the circuit board from shorting out while you power up and troubleshoot without having to spend 20 minutes putting everything completely back together.

[katopan]

Don't worry about matching that 4mV test signal. You're looking for signal / no signal, not correct stage gains and amplitudes. At this stage put in something of a decent amplitude just to test and get that first stage working.

Also did you measure those voltages as Martin asked? They can tell you a lot.

[Gibsonman63]

Measure the voltage to ground at each tube pin, and the supply node voltage for that stage (on the other side of the plate resistor). Calculate the current flowing through the triode as (Vsupply-Vplate)/Rplate and/or Vk/Rk. Depending upon the circuit, you should be seeing about 1-1.5 mA (0.001-0.0015) for a 12AX7. Typical cathode voltage will be 1 to 2V.

Something is definitely off with V1A

For V1A V-Supply = 366Vdc
For V1A V-Plate = 363Vdc
For V1A - R-Plate = 100K ohms

This works out to .03mA

For V1B V-Supply = 351Vdc
For V1B V-Plate = 230Vdc
For V1B - R-Plate = 100K ohms

This works out to 1.21mA, which sounds reasonable.

I pulled that cathode bypass cap and read 0Vdc at the cathode of V1A

V1B reads 1.8Vdc at the cathode with the cathode bypass capacitor in the circuit.

My cathode resistor checks out at very close to 1.5K and the grounded side of it reads 0.1 ohms to ground.

I know this is probably much easier than I am making it, but I am just not seeing it.

[martin manning]

So V1a is not conducting, and it appears that its cathode is not connected to the cathode resistor or the cathode resistor is not connected to ground. What's the voltage on the grid?

[katopan]

Also check you've got conductivity from the 100k resistor leg to the actual pin of the socket it's supposed to be connected to. Same for the cathode resistor to the pin it's connected to.

[vibratoking]

As mentioned, there ain't no bias current flowing in V1A so there ain't no signal.

I would turn it off, drain the voltage and check for continuity from the actual tube PINS to ground, supply, resistors, etc... Make sure that you don't let the probes complete the circuit. Maybe your probe is completing a connection during measurement and the connection is lost when you remove the probe. I have seen this quite a few times in the past. There must be something fairly obvious that you are missing.

[Gibsonman63]

I am still scratching my head.

I am injecting around 250mV AC to the grid. I don't measure any DC there. I have continuity from the grid resistor to pin 2.

I temporarily disconnected the plate from the coupling capacitor to disconnect downstream. Same voltages. The plate resistor reads very close to 100k Ohms. I am only dropping about three volts across the plate resistor.

I have continuity from pin 3 (cathode) to the cathode resistor. There is no DC on the cathode. The cathode resistor and bypass capacitor both measure continuity with ground on the other end.

Heater voltage is 6.3 vAC.

I pulled the tube so I could double check continuity at the socket.

I feel like I am missing something incredibly simple, but just have a mental block where I cannot see it.

[surfsup]

When you dmm'd the plate and b+ supply to get those voltages, did you have your scope probe connected to the grid still? If so, your plate voltage will be about the same as the b+ so that seems right.

[markr14850]

Post a picture of the wiring of the socket.

[martin manning]

You said you have filament voltage, but are you sure that the filament on the V1A triode is heating up?

[Gibsonman63]

You said you have filament voltage, but are you sure that the filament on the V1A triode is heating up?

Good question. It does seem a little dim. I will pull the tube and check at the socket tonight. I have tried a different preamp tube that I know works.

I'll post some pics tonight when I get home.

[Gibsonman63]

Gents, first a sincere thanks for the suggestions. This was very frustrating, but proved to be a good review of the basics.

The culprit was a loose tube socket pin. It really didn't look or feel loose, but a little light-duty prying with a jeweler's screwdriver solved the probem.

This amplifier came with a little foam thing that fits over the two preamp tubes to dampen vibrations. Probably good for it's intention, but it may be putting a little extra stress on the tube sockets as things get bumped around.

[martin manning]

Glad you got it solved. There had to be an "open" somewhere... which pin was it?