RGs secret transformer tester - question

[Brymus]

OK I finally got my high gain SE amp sounding the way I wanted.
3 gain stages a CF - TMB tonestack - MV - FX loop with NFB circuit,switchable cathode caps for boost.
All stuffed in a Valve Jr.
Anyway I was so stoked that when I took the FX loop out into another amp I forgot the dummy load.....
After it started smelling I started sniffing and noticed I forgot the dummy load so I shut it down and pulled the chassis.
The first dropping resistor was fried....
So I pulled the OT and used RGs Super Secret Transformer tester to check it.http://www.premierguitar.com/Magazine/Issue/2007/Apr/The_Super_Secret_Transformer_Tester.aspx
I used a 9V battery and a 120V neon lamp (be careful 9V stepped up through the OT primary bites pretty hard )
It flashed on all the windings didnt matter if the DC was applied to the primary or secondaries the bulb flashed on the winding it was attached to.
I also have another Hammond 125ESE in another amp so I checked it too and got the same flash responses.
I then measured the windings on both the suspect OT and the known good one,both OTs measured the same.
So I replaced the burnt resistor and put the OT back in and using a different set of tubes and fired it up.
It sounded good for a couple of minutes then the smell...Sure enough the dropping resistor is burnt again (R10 from the EVJ schematic http://www.muzique.com/news/images/Epi1.gif mine is much different using a 1k2 for R10 otherwise the power rail is basically the same)
I guess the OT has to be bad,but why did it pass all the tests and measure the same as a new one across all the windings?
Is there something else I am overlooking?
This has me stumped.
My thinking is without a load on the OT the OT was pulling way to much current through the first filter node frying the dropping resistor.
But why did the OT check OK flashing on every winding ?
And could the problem be something other than a fried OT?

[Brymus]

Here are some pics of the build- testing out many mods one by one adding to the circuit as I went,they arent in order yet.
http://s419.photobucket.com/albums/pp276/Bry928/High%20Gain%20EVJ/

[R.G.]

I guess the OT has to be bad,but why did it pass all the tests and measure the same as a new one across all the windings?
Is there something else I am overlooking?
This has me stumped.
My thinking is without a load on the OT the OT was pulling way to much current through the first filter node frying the dropping resistor.
But why did the OT check OK flashing on every winding ?
And could the problem be something other than a fried OT?

At least one problem is that you're not believing what your test instruments tell you. 

If you decide ahead of time what the problem is, then measure, and the instruments tell you that's not the problem, it's very human to keep trying to measure some way that what you thought was the problem really is. That obscures what the real problem may be. Conan Doyle had it right, speaking through Sherlock Holmes - "... eliminate the impossible, Watson, and whatever remains, no matter how improbable, must be - the truth."

You haven't totally eliminated the OT, but it's sure looking improbable. Here's something you may have not thought of: OTs don't stop DC from coming through. The DC resistance of that winding is small. If the rest of the amp was working properly - IF, mind you! - you could replace the OT primary with a shorting wire and the circuit would not know from a DC perspective that anything was wrong.

So it's time to think: what else could even possibly be pulling too much current through the dropping resistor?

In this case, you've decided the OT is bad. But the tests say, no, it's good. I promise you - if the windings are not open, and not shorted, it's most likely good. There is still some possibility that it only arcs over internally with 300+ volts on it instead of the 100V or so the neon fires at, but that's a very small window.

[Ripthorn]

You should also check to make sure that none of your power tube pins got shorted or something (possibly due to an internal arc).  I know this has happened before when it was thought that the OT was bad.

[Brymus]

OK thank you for the reply RG
I too was under the impression the OT was probably good after it passed all the winding tests.
Which is why I replaced the resistor and power tube and tried again.
I cant find any shorts to ground,which is the only other thing I can think of to cause that much current to pass through that node.
Would you be so kind as to advise what else to look for or other possible causes ?
I am thinking of putting the stock OT back in to see if it still pulls enough to burn yet another resistor.


Ripthorn,thanx for the tip - no signs of arcing no shorts found with the DMM

[Brymus]

OK another new dropping resistor and installed the stock OT.
No smell no blackening of the dropping resistor.
The only other change was I had added another coupling cap after the MV before the FX loop and EL84s screen.
This removed the remaining hiss I had at high gains,but being unsure I removed this cap as well before firing it up.

Does the EL84s screen have to have a reference to ground ?
Could this have contributed to the meltdown of the dropping resistor ?
I guess I could put the cap back in and see if it burns again,but I would rather understand whats happening than just solder and observe/
trial and error it.

The stock OT just doesnt come close to the sound of the Hammond ESE in tone or volume...

[R.G.]

OK another new dropping resistor and installed the stock OT.
No smell no blackening of the dropping resistor.
The only other change was I had added another coupling cap after the MV before the FX loop and EL84s screen.
This removed the remaining hiss I had at high gains,but being unsure I removed this cap as well before firing it up.

Does the EL84s screen have to have a reference to ground ?
Could this have contributed to the meltdown of the dropping resistor ?
I guess I could put the cap back in and see if it burns again,but I would rather understand whats happening than just solder and observe/
trial and error it.

What you're describing is something that's turning on the output tube. Inserting/removing caps is changing the bias on the tube.

[PRR]

> thinking is without a load on the OT the OT was pulling way to much current

No. It will do odd things, and maybe spike-through the OT windings, but won't "pull way too much {plate} current". (It might pull excess screen current, but that would be R12 smoking.)

> a couple of minutes then the smell... dropping resistor is burnt again (R10 from the EVJ schematic http://www.muzique.com/news/images/Epi1.gif mine is much different using a 1k2 for R10

With R10 near 1.2K, it dissipates about 2 Watts (about 5X more than with the specified 220r). If my math or your EL84 is a little off, maybe 3 Watts, which would stink-out a 2 Watt resistor in a few minutes.

Get a 1K 10 Watt resistor; they used to be stocked at Radio Shed. That should give the drop you apparently want, and last roughly forever. If it does fry, something is real wrong with your EL84 or its bias. (What is R14 value and voltage?)

[Brymus]

Here is the schematic of what I did the coupling cap in question is in a dotted line box following the MV.
I dont see how adding/removing that cap could alter the bias,please explain.
And is it needed or not it definetly sounded better with it.
Please excuse my hand drawing skills its alot faster for me than using the DIYLC
I just noticed I drew in a plate resistor on V2B and there is not one there ,
Here is the corrected version everything else looks right by what I did
http://i419.photobucket.com/albums/pp276/Bry928/High%20Gain%20EVJ/schematic002.jpg

[Brymus]

And the power rail

[R.G.]

Here is the schematic of what I did the coupling cap in question is in a dotted line box following the MV.
I dont see how adding/removing that cap could alter the bias,please explain.
And is it needed or not it definetly sounded better with it.
Please excuse my hand drawing skills its alot faster for me than using the DIYLC
I just noticed I drew in a plate resistor on V2B and there is not one there ,
Here is the corrected version everything else looks right by what I did

OK, we have the problem in sight. The grid on your power tube needs to be tied to ground by a resistor all the time. Put in a 220K resistor between grid and the ground side of the cathode resistor right on the power tube socket. Whenever the cap is inserted, it lets the grid float, turning the power tube into a power diode - no current control. Add the pull-down resistor.

[Brymus]

Thank You RG ! at least I finally got on the right track.
SO is the cap in question needed ? Or are the caps in the tone stack enough to isolate the DC ? It definetly sounded better after I added it before it went thermal.
And I am guessing the other OT is still good since it passed all the tests ? 
Do you think the two power tubes I used may have been damaged ?
I didnt see them redplate but I know that doesnt mean they didnt .

If any one has any suggestions for this amp please ,dont hold back. I know very little about electrical design as witnessed today 
Talk about learning the hard way,at least I am smart enough to use RGs advice 

[R.G.]

Thank You RG ! at least I finally got on the right track.
SO is the cap in question needed ? Or are the caps in the tone stack enough to isolate the DC ? It definetly sounded better after I added it before it went thermal.

I'd leave it in there. What is critical is making sure that the grid of the output tube is pulled to ground DC voltage somehow. Inserting the cap breaks the DC path and lets the tube overheat.

And I am guessing the other OT is still good since it passed all the tests ? 

Most likely.

Do you think the two power tubes I used may have been damaged ?
I didnt see them redplate but I know that doesnt mean they didnt .

They're probably fine.

Whip in a grid leak resistor (that's the Golden Age of Tubes name for that grid resistor) and see how it sounds.

Notice that the extra-hot currents with the cap in there may have been affecting your tone too. Once you get the bias stable, you can adjust it by the value of the cathode resistor - higher for lower currents, lower for higher plate current.

[Brymus]

Thanks RG ! Once again your logic and method proved right,the transformer test was correct and it was human error on my part.

I put the 220K grid leak resistor in,as well as the Hammond ESE.
Everything works,OK  Listening to it today its sounds the same as it did before the thermal incident

Could I use a 470K grid leak , or whats the highest value I could get away with?

Voltages - As it sounded sweet with them ramped up,I am now willing to sacrifice tube life for tone and plan to tweak the voltages to try and achieve what I heard before.
What effect did (would) the lack of a grid leak have on the pre-amp voltages ?
I am guessing they would have dropped a little as well as the heater voltage too.

At least I learned first hand the importance of the grid leak , and again had I drawn the schematic before starting the thread I would have saved alot of time.

EDIT - I had said it sounded different now but after playing it again today,it sounds right.