sparking OT

[se7en]

I'm building a custom amp from scratch, a process which is turning out to have quite the learning curve. 

After a bit of muddling around with the preamp, the wiring and replacing shot tubes and blown fuses, I've finally got some decent amplified guitar sound coming out of the speakers.  The biggest problem I think I have right now (still a fair bit of tuning to do yet) is that my OT sparks when I hit heavy power chords.  I think it's just by luck that I happened to be standing in the right place to see through a tiny hole in the housing to see the sparks.  They only happen (that I can see) on power chords struck with a heavy hand.  Those chords are also very weak in the speaker.  Single notes are for the most part decent.

I'm wondering if I have a bad OT, or if I will kill it if I keep going, or if it's my circuit that needs some obvious work.  Or something else.

The OT is a Hammond 1650K.  Preamp is 12AX7 amp with cathode follower feeding a pretty vanilla tone stack.  Phase inverter is another 12AX7, and I've got a pair of 6L6GC running in a PP config for the power amp.

Thanks for any pointers.

[tubesornothing]

So my "red neck" side says:  "get a can of gas, place it by the amp, set up the video and play some AC/DC".  Send video to Darwin Awards.

My "conservative engineer" side says:  "sparks - bad - if its really coming from the OT, replace it"

[Maschinenmann]

My "conservative engineer" side says:  "sparks - bad - if its really coming from the OT, replace it"

Indeed.  

[Leevi]

I don't believe your problem is the OT. Do you have grid resistors on your power tubes?
Check the biasing of the power tubes. Are you using cathode or fixed biasing
Of course you can try to localize the fault e.g. by doing the following test:
Disconnect the preamp from the PI and connect you guitar directly to the PI e.g. by soldering
a jack to it or if you have an effect or another preamp out of another amp use it instead.
/Leevi

[stingray_65]

First off Welcome to the best darned Tube Amp Site on the net.

What OT are you using?

Is it new? or recycled?

Where did you purchase it?

The Hi Pot test that most trannys go through is pretty brutal, for any type of spark or corona to be generated inside a transformer I would think that the insulation has been comprimised somehow.

you say this is an ongoing scratch build/design. have you over heated your tranny during troubleshooting?

It will be important to track down the cause before you go much further. it is my fear that the transformer may never agin be useful, but the guys here will help you sort through all that I'm sure.

Ray

[DummyLoad]

The OT is a Hammond 1650K

So my "red neck" side says:  "get a can of gas, place it by the amp, set up the video and play some AC/DC".  Send video to Darwin Awards.

wheeeee!!!  

call me crazy, but if you see sparks coming from something that's supposed to be fully insulated, you definitely have a problem. you should seriously consider replacing it, but first try to determine what is/was the cause. 

[jojokeo]

OT flyback voltage condition???

[se7en]

Hey guys - thanks for the welcome and the comments.  To answer some questions:

It's a brand new Hammond 1650K.  I just noticed that the spec sheet suggests 4 tubes, and I'm only using 2.  I wouldn't think that would be my issue right now though.  How many 50W amps have 4 power tubes?

I don't think I've overheated it, I'm pretty sure everything is running pretty cool right now.  

I'm using fixed biasing, and I have yet to tweak it.  I just started to look at it, but neither of my DVMs will measure any current from the cathode to ground (voltage across a 1ohm is 0V at idle).  The bias right now is at -53.5, and I think I want to be up closer to -50, but I want to run through a few things first.  Plates are at 390.

I've not heard of the OT flyback voltage condition, so I'll have to look it up.

I really don't want to drop another $90+ on an OT, so I'm hoping it's that I'm doing something else stupid, I'm just not sure what.

If my drivers weren't connected well (physically) might that cause the sparks?  I tried another test, and I could get decent volume when I really hit some power chords.  I couldn't really get a good look at the OT since it's buried inside the amp, but it sounded better.  I'm taking it apart again now to get a better look at things and adjust the bias a little hotter.

Thanks for the help so far.

[phsyconoodler]

Your speaker connections could cause the issue for sure.The loose connections are like a car battery terminals.Loose connections make for lots of resistance so the flyback voltage is a reasonable assumption.
  There is no problem using the Hammond with two power tubes.Keep in mind that the impedance changes.You need to double the speaker load when using two tubes vs 4.
  One other thing:with 390v on the plates,you'll likely be looking for closer to -35v for a bias setting.

[se7en]

OK - I think I like the loose connection theory.  I haven't seen any more sparks yet today.

I don't want to get too far off topic, but I just want to be sure - when you say "double the speaker load", I'm assuming that means connect my 8 ohm load to the 4 ohm taps rather than the 8 ohm taps?

I think I'm going to have to go back and read up on bias voltage.  -35 sounds really high based on what I've read, and looking at the datasheet on the 6L6GC, it specs -22.5 V for 400V plates, and the -50V curve is *way* low.  So now I've got three conflicting sources....  :)

Thanks for the response.

[phsyconoodler]

Your spec sheet isn't your amplifier.And yes,that means double the load on each speaker tap.
A Super Reverb with two 6L6's at 450v likes to see about -43 to -45v to get about 40ma per tube.
   At 390v it needs to be at about -35v.Just about every other amp I've worked on is similar,with the tube brand being the only variable.And it doesn't vary by 10v.