Voltages idle and under load

[surfsup]

New report from bizzaroworld...and a recap:

Liverpool-esque build
Input to first stage (v1), gain, second stage (v1), TS, third stage (v2), PI, 2xel84

I think i have an issue with v2 not v1. Here's why...with v1 in and v2 out, i put the scope probe on v2 plate and i get signal with no tube. The dmm is then clipped to the supply and plate and i get 258 and 253 vdc so i am losing 5v across the 100k (not sure if that is normal). Then i test continuity between the grid and plate at the pins of the socket on v2. With amp off, i get an occassional minor beep. With amp on, the dmm beeps loud and clear (even though there is no tube). I assume this socket is bad? Or is that normal?

[katopan]

That could be it. Got a spare socket?

[surfsup]

Just to be clear, with no tube, losing 5v means i've got a slight amount of current going through that 100k plate resistor. With no tube this should not happen, right?

On the ax84 forum i posted this problem and someone responded continuity could be normal through the PS but i don't see how since there is a 100k off the plate pin and a 33k off the grid.

I have another socket but it is another ceramic...

[martin manning]

So disconnect the plate lead from the socket and see if the voltage drop across that resistor goes to zero. Then see if the signal still gets across to the PI.

[surfsup]

I disconnected it at the plate resistor since it's easier. Still have a 2.5V drop and guitar at the speaker. I went over the circuit several times today and last night. Nothing is wired "incorrectly". With the amp cranked the volume with V2 out is about speaking level so it is not minor, maybe 0.5W output?

One think I noticed was as it was powering up, the guitar was quite loud, then suddenly it dropped off in volume and stays at speaking level.

Looking at the circuit with V2 out, the signal must travel from grid to plate or cathode to plate, where the PI cap is connected? What other way would signal get to the PI? Or could it be getting to the Power Tubes?

[surfsup]

okay, so I had a bit of time again tonight. Still V2 tube not in socket, I disconnected the CC from the gain pot. No guitar sound in speaker. This tells me the problem is after this area of the circuit.

With V2 out I desolder both PI plate leads and the 1M from the primary side of the PI. I get no guitar. This tells me it is before or at the PI area in the circuit, not the Poweramp.

I put V2 in and connect the slave side plate of the PI. slight guitar (1/10th-ish). I connect the 1M again and the guitar gets louder. I disconnect it and for about 3 seconds there is no guitar, then it starts to gradually build over 10 seconds to once again about 1/10th the volume of before with all connected but V2 out. If anyone has any suggestions I am all ears or I'll report when I hit my next milestone.

[surfsup]

Okay no one replied so I give myself a bump and more info:

Here are a couple shots of the V2 socket and PI socket looking for arcing. Keep in mind I have re-soldered these leads many times so its a bit uglier than original but I am still trying to keep it clean and think it is okay. I don't see any charring or signs of arcing, anyone? :

[img1500]http://chicagocadcam.com/ChrisHahn/xen10/V2.JPG[/img]
[img1500]http://chicagocadcam.com/ChrisHahn/xen10/PI.JPG[/img]

With all tubes in except V2 (third stage), if I disconnect the grid to stage two, the speaker doesn't produce guitar. So I just want to make sure I am thinking about this right. Since the signal reaches the grid lead off the gain pot, but not the grid pin, and there is no guitar sound, that means the signal didn't reach the point where it could "couple" to the PI, so therefore the problem lies AFTER the gain pot. Correct? and this is weird since V2 is immediatly after...

[Reeltarded]

Not ignoring you here. Perplexed and need 10 more hours sleep and coffee.

This thing is still passing signal with V2 pulled? I see no arcing evidence, but those sockets confuse me in a profound way. Not used to those is all heh..

[surfsup]

This thing is still passing signal with V2 pulled?

Yea. Strange. If i pull the two PI plate leads, i don't get guitar, so i am hoping my thinking is right that it is between the gain pot and the PI...?

those sockets confuse me in a profound way. Not used to those is all heh..

AES was out of the NOS 13-pin triangular sockets so I settled for the round 9-pins...!

[surfsup]

With all tubes in except V2 (third stage), if I disconnect the grid to stage two, the speaker doesn't produce guitar. So I just want to make sure I am thinking about this right. Since the signal reaches the grid lead off the gain pot, but not the grid pin, and there is no guitar sound, that means the signal didn't reach the point where it could "couple" to the PI, so therefore the problem lies AFTER the gain pot.

Also, I disconnect the three leads coming off the caps to the tonestack pots. I still get guitar! So that tells me the problem is BEFORE this spot...YES???

[img758]http://chicagocadcam.com/ChrisHahn/xen10/preamp2.jpg[/img]

[John_P_WI]

Surf,

A guitar amp is a modulated power supply, in simplistic terms. I am guessing that you have inter stage modulation between the 1st and 2nd stage as they are not decoupled. I also question the way the following filter caps are grouped together at the power transformer.

Vacuum Voodoo had once posted a simplistic scheme of grounding, I believe that he called it "client server" or something like that. I can't seem to locate it now, it may have been on another forum.

Anyhow, I would seriously look at the power supply side which is the "other half" of the signal. Maybe look for Merlin's or Larry's ground info as well.

Good luck

[Gaz]

It was my understanding that since two sequential cascaded stages were out of phase it was okay to power them from the same node. It is definitely worth decoupling them to rule it out.

Again, I had a 'bleed' in the same area as you, and every stage was decoupled with adequate filtering.

I think it's more likely capacitive coupling than migrating from pin to pin on the socket but that's nothing more than a hunch. If it were me, I would replace the socket anyway to rule it out.

EDIT: Also, the 2nd and 3rd stages are out of phase and decoupled.

[Structo]

How is there a voltage drop between B+1 and B+2?

Don't you need to decouple those two nodes?

At least a 300-500R resistor or 1K is normal there.

[surfsup]

It was my understanding that since two sequential cascaded stages were out of phase it was okay to power them from the same node.

The original that the clone was based on was this way as well...

How is there a voltage drop between B+1 and B+2?

Sorry, in my haste I forgot to add the text to indicate there is a choke there.

Okay here are some scope shots and some comments:

[img:960:2574]http://chicagocadcam.com/ChrisHahn/xen1 ... eShots.jpg[/img]

All tubes in except V2 like before
First column is gain control at 10
Second column is gain at 5
Rows are labeled on the left (the stage 2 plate didn't save for some reason so it is blank)...

As I crank the gain from 0-10 the signal we see on B+5 increases to a maximum of 80mV peak/peak. So this is definitely affected by stage 2. But, isn't the filter cap supposed to shunt that to ground? How is there signal there with a 10uF to ground?

Secondly even if not, how does signal get to the PI through two decoupled PS nodes (one 10k and the other 22k)? B+4 and B+3 don't really show any signal so how does it get all the way to the PI with V2 not in even in the amp?

[VacuumVoodoo]

I think i have an issue with v2 not v1. Here's why...with v1 in and v2 out, i put the scope probe on v2 plate and i get signal with no tube. The dmm is then clipped to the supply and plate and i get 258 and 253 vdc so i am losing 5v across the 100k (not sure if that is normal).

Your DMM needs to pull some current to make the measurement.
Try measuring voltage across that resistor with one end of in free air. Bet it reads 0.
Rule No.1 in electronic test & measurement: TEST INSTRUMENT AFFECTS OBJECT UNDER TEST

Then i test continuity between the grid and plate at the pins of the socket on v2. With amp off, i get an occassional minor beep. With amp on, the dmm beeps loud and clear (even though there is no tube). I assume this socket is bad? Or is that normal?

Continuity test injects current into circuit therefore test object must not be powered ON. Occasional beep on NOT powered amp indicates a cap charging up by the injected current.
See rule No.1 above
Signal leaking through with V2 removed: I would suspect a grounding issue or a faulty cap at node 4 or 5 (a speculative guess, I'd need to have the amp to investigate).
Also, not unusual to see is leakage through component board material itself. Had seen it couple times on newly made boards from NOS sheet.
On the other hand I would not obsess with quirks with one tube removed. The important thing is if it works correctly with all tubes installed.