Problem with my 18W TMB

[tboulette]

Hi,

I built an 18Watt TMB about 7 years ago. It worked great for a bit, but then developed a problem where it would lose power, then recover, lose it again, etc.

For various reasons, including a gig that dried up, I haven't used the amp in several years.  I mentioned on this forum that I had a problem (in an unrelated post about purchasing an OTS FM) and someone encouraged me to post pictures and a description of the problem here.  I'm hoping y'all can help me figure out what's wrong.

Pictures are below.  This was my second amp build (after a Gilmore Jr. kit from Gary Gerhart), and it shows in my soldering skills.... 

I tried the amp again today and here's what I found:

After playing about 10 minutes at moderate volume on the normal channel (i.e., clean), during which it sounded great, there was a sudden, somewhat soft "phfftftf" sound, and the power dipped way down.  The sound distorted a bit in an unpleasant way.  After maybe 5 seconds of fits and starts, the power came back, and it was fine for another 15 minutes (before I had to put it away for now).  In the past, the problem has cycled on and off every 5-15 minutes or so.  Throughout, I was getting occasional interference -- what sounded like RF interference to me.  That may or may not be coincidence.

That's all the data I have for now, but I'll try to play it some more in the coming days and weeks and perhaps post more observations.

Any help much appreciated!!  Now for the pics:

http://photobucket.com/18WattTMB

[wyatt]

Well, you've already had the advice about checking for cold colder joints in the other thread.  And it's the best place to start.

As the amp warms up or cools off, parts expand or contract and then an pull a cold solder joint around from it's eyelet/lug. turret, etc.  You should probably chop stick you way through the circuit and see if you can locate it.  I would pay particular close attention to the filter caps and other electrolytic.  These must have breathing room, they expand and grow during use...if they don't have enough space or lead length, they'll pull free of solder joints.

[tboulette]

I've done the "chopstick check" before without finding anything, but I'll repeat it with a particular look at the filter caps. Thx.

[T Wilcox]

Rather than chopstick I would suggest doing the Texas shotgun approach and just re-flow every joint. I do see a few points where the wire is just inserted and then soldered rather than crimping a loop before soldering. On those I would desolder, re-strip and reconnect. All connections should have a good mechanical connection before applying solder. I would also desolder the caps that fly from the board to the pots and put heatshrink over the exposed leads as well.
Good news is it is just an intermittent problem so we dont really have to chase down any miswiring.
A small set of needle nose with 45 degree bend makes doing these connections so much easier.

Todd

[tboulette]

Thanks Todd.  As I said, my soldering skills are pretty basic. Do you have (or can you point me to) any good pictures of how you do what you describe?  I'm a visual learner!!

[T Wilcox]

Hey
I dont really have any pictures of a step by step.
But heres what I do.

a. Strip the wire about 1/4 to 3/8 ''

b.  If stranded tin wire with very small amount of solder to keep wire from fraying. ( Not too much though or else it might not fit through hole you are soldering to.)

c. With tip of needle nose pliers bend the stripped section in 1/2 making a hook with the stripped part of wire.

D. route hook through terminal hole that it is to be soldered to, then with needle nose squeeze the wire together rather tight.

At this point you should have a good mechanical connection, meaning you do not have to hold it in place and it will not move while soldering.

when soldering hold the iron on one side of the connection and apply the solder to the other side once it is hot enough. The solder should never even touch the iron but just melt when applied to the heated wire and terminal connection.

[T Wilcox]

Another suggestion.
I have found crimp connectors ( the ones you have used on the star ground ) to be unreliable in past projects. This may not be the cause of the problem you are having  and would probably make sure re-flowing the solder doesn't fix the problem first but I do recommend using solder type ground lugs.

Also noticed the chassis is powder coated/painted. Make sure it is bare metal at any ground points. I use a dremmel to grind down to bare metal but a file or even sand paper would work.

Todd

[tboulette]

Well, good news I think.  I reflowed all connections at the power tubes (and for good measure) at the output socket.  I haven't had time to put it through a real stress test, but I played for about 15 minutes with no loss in power or weird noises.

I took this route because (a) the turret board was constructed by Nik, so I was assuming at least for the time being that the issue wasn't there, and (b) I simply couldn't face the idea of redoing EVERYTHING.

(One minor point for Todd -- while those are crimp connectors on the grounds, I actually did, in the initial build, flow some solder on them, so I think they're okay.  Your point about the powder coat is a good one though, and maybe one of these days I'll clean it off at the ground locations to be sure it's good.)


Thanks everyone for their suggestions.  Hopefully I'll have some time to play through this amp some more soon and will let everyone know if the problem persists.

[T Wilcox]

Thats great, let us know once youve put a few hours on it.
If you have a meter and want to check that ground you can place one lead on the ground prong of the IEC connector and then check continuity between it and the grounded wires where they land at the board. Should read about 1-2 ohms at most. i woiuldn't worry about it though unless it starts acting up again

Todd