Unplanned TSE mod, letting out the smoke

[Jpeg]

Hello All,
So I finally got around to finishing that Tubelab SE I started around Christmas time. I was finishing up all the connections at about 1 o'clock this morning when I apperently decided to swap T6 and T7 (the center tap goes in the center, right?) when connecting my filament transformer. I went through the checkout process and amazingly got 2.48v on my output tube filament. Everything else checked out and I popped in a pair of 45s, and hooked up some speakers. By the time I got the bias set on the tubes, my filaments had gone out on my 45s. My laser-guided thermometer told me that U1 was getting very hot. I assumed that I needed a bigger heat sink, so I found a much larger one in my motherboard graveyard. It seemed to be stable at that point and I listened to about half-a-dozen songs (sounded very nice BTW) I then swapped out the 45s for a pair of 2a3's and rebiased they sounded nice, for about half a song...

Then came the buzzing...
Then came the smoke...

A visual inspection of the board reveals a burned trace between D1 and D5. The is also a nice black spot in my filament transformer (thankfully I was using a seperate filament transformer, so that I would have the full 6 amps to play around with 2a3's, So I only burned up a $20 transformer.)

I soldered a wire in place of the burnt trace, and I will test it in the morning, but I am pretty sure I will be ordering a new x-former. My question is, while I am at it, should I order a new D1, D5, U1, or anything else I that I may have fried? I don't expect anyone to really be able to diagnose this from the information I have given, but if someone could point me in the direction of things to check, I would appriceate it.

Thanks,
Josh

 

[Jpeg]

Hmmm, seems to be 5d is no longer a diode...Off to Radio Shack, see if they have anything suitable...

 

[Jpeg]

D5, replaced D4 looked good, but it's traces were all burned off. Replaced them with wires, tested filiment x-former got 3.2 volts off each leg, wired it up correctly. Fired it up, more smoke. D1 also appears to be toasted, not sure how to test it, looks like it's time to fire off an order to Digikey.

 

[rknize]

I would get a few extras of all of those diodes and U1. Check to make sure C1 isn't shorted, though from your story it seems unlikely. Double-check your jumper settings as well. That area of the schematic can be a little confusing.

 

[Jpeg]

That just what I did, hopefully I will be listening again by mid-week. It is ironic you mentioned the jumpers. I agonised over the jumpers... But paid no attention to where the tranformer wires went. I built my board upside down, so the labels are not readable, and nothing is in quite the right spot. I checked it at least twice before powering up and again after the mishap.

Thanks for the help!

 

[rknize]

Yes, this board is a bit tricky when built upside-down.

 

[dagerman]

I am almost envious about your smoke, at least some components told you they aren't happy.
Mine just sounds awful and I can't figure out why. All voltages and other operating parameters are fine.
Once I figured out how to use my scope in a way that doesn't drive up bias current beyond 1A, I'll hopefully get an idea of what's going on.
Good luck with yours once you get the new parts!!

 

[Jpeg]

Smoke 'em if you got em.

You do have a point, as diagnostics go a big smoking black spot on the board is about as straight forward as it gets. Your signature cracks me up by the way.
Aside from the fact that the board now looks like a 2nd grader's art project, I am pretty happy. There was a digikey box waiting for me when I got home. I got a new D1, and mounted it on top of D4 and D5, bent the leads to fit on the transformer side of D4 and 5, soldered it on. Soldered on a wire and ran it to the one remaining trace where it is actully suposed to go. I tested D4 and 5 again, put a new fuse in and I was good to go.
I hooked up my best pair of thrift store speakers and cringed as I turned it on. (point away from face) No smoke, no buzzing, no burnt smell. When sound started coming through one of the speakers I jumped a little.
I finally got some time to listen to it, and I am impressed so far. I still didn't get to crank it much, and this is my first DHT amp, but not my first SE. I'll post an update after I get to play with it more.
What sounds so bad about your's Dagerman?

 

[dagerman]

Oh sweet moment of success! It's the best if you turn it on and are enveloped by what makes these lil' amps so special. Even if there were some 'victims' on the path to glory... . Congrats!

I decided to build mine for 300B tubes and followed George's directions after reading A LOT of the posts here with others' experiences building the TSE. Got everything together, double-checked, cross-checked and then re-checked all components. All initial setup and adjustments went fine after getting my B+ under control, but it just doesn't sound good. Kind of like a cheap transistor radio (well, maybe a better one) at low volume, but scratchy and 'overdriven' (I know, I suck at describing sound) at higher volume.

I am starting to think that my eBay driver tubes may not be in 'good working condition'.

Anyway, that's what you get for starting an 'intermediate' project when you are really just a bloody (tube) beginner. Humbling, but educational!

Stefan

 

[Jpeg]

Well, it looks like you have a scope, you are ahead of me on that one Did you wind up with a pair of "5842 compatable" russian tubes? I have seen posts about those not being very compatable... Have you tried just hooking up one channel on your RCAs and swapping both driver and output tubes back and forth to see if you can detect a bad tube? Just a thought, but it's easy to check.

 

[dagerman]

Well, it looks like you have a scope, you are ahead of me on that one Did you wind up with a pair of "5842 compatable" russian tubes? I have seen posts about those not being very compatable... Have you tried just hooking up one channel on your RCAs and swapping both driver and output tubes back and forth to see if you can detect a bad tube? Just a thought, but it's easy to check.

Well, not sure about the 'ahead' part... You know how they say "a fool's a fool independent of tool"... Right now I wold trade a thorough understanding of how the circuit works for my scope. I haven't figured out how to scope the driver stage. As soon as I hook up the scope probe to either end of the coupling cap, I get real bad static and the bias current goes through the roof.

Anyway, my 5842's are Raytheon and RCA (I have two pairs) and swapping them hasn't have any effect. The 45s however do seem to be different. On one channel I can't get the bias current under 45mA and if I swap the tubes, the issue follows.
Initially I had set the board up for my pair of new 300Bs and I was able to adjust all operating points as expected, but it still sounded off and distorted at higher volume.
I decided to 'park' the board and revisit after
A) I learned how to use my scope properly (I thought I did, but...) and
B) I read up on the theory, so I know where and how to measure

One day, there will be light (and sound)!
For the next couple of days I'll focus on Salmon fishing in the San Juans anyway.

Happy Labor Day,
Stefan