VTA ST-70 strange initial start-up

Built Bob's ST-70 2 years ago for a friend. One channel not working all of a sudden. Removed tubes and bottom cover and inspected wiring, etc. All look OK. Went through the start-up procedure. Left 12AU7A lit fine, center and right did not. . Measured the AC at the driver board and is 6.5VAC. worked my way back to pins 2 &7 on the other tube sockets towards the PT and all the same. changed out the two tubes not working and all of a sudden all OK. Then started on the left channel, installed EL34's and rectifier. Was able to adjust bias to 0.400 VDC ok. Noticed some static in the left speaker though. Rectifier tube getting really hot, then static again. Noticed the same 2 driver tubes not lighting up again. Turned it all off, let things cool and pulled EL34's and rectifier. Replaced the rectifier with a 5AR4, replaced EL34's and turned back on. Rectifier flashed, and fuse blew. Removed those tubes. this morning started over only with driver board tube and once again, left one lit, center and right did not. Kind of scratching my head here. Help appreciated. Going to order up new EL34's and a new rectifier now just in case.

First the obvious suspects, in no particular order:

Do you have any way to test the tubes?
As you have a VOM, could you test the tubes (Cold) for any internal shorts? Follow the pin-out to know what should, and should not conduct.
Any idea how many hours on these tubes over the two years?
Had it been sitting for a while, then re-introduced to service?

Now, the first diagnosis procedure:

Obtain a dental-pick or similar small hooked-end tool.
Tug on, and test *every* soldered connection you made when assembling the kit. Look for loose, broken or cold-solder joints throughout.
Look for pinched wires as well - it takes time for heat-and-age to cut through the insulation.

Please report back, and best of luck!

If you built the driver board I would also check all the solder points on the driver board. Yes it's a pain, but in most cases, a cold joint on a through hole connection is the culprit.

Thanks to both. The intermittent occurrence of the center and right driver tubes working / not working point to solder joints for sure. Visual inspection shows all are nice and shiny but will redo all of them. Will start that process this evening. Yes, have a dental pick just for that reason! Also the static heard in the left speaker when the tubes were warming up and bias was increasing could point to the tubes. Ordered up a new set of Mullard's yesterday. So strange, when I initially completed the build 2 years ago, burned it in for over 50 hours before handing it over.

Brap wrote:Thanks to both. The intermittent occurrence of the center and right driver tubes working / not working point to solder joints for sure. Visual inspection shows all are nice and shiny but will redo all of them.  Will start that process this evening. Yes, have a dental pick just for that reason!  Also the static heard in the left speaker when the tubes were warming up and bias was increasing could point to the tubes. Ordered up a new set of Mullard's yesterday.   So strange, when I initially completed the build 2 years ago, burned it in for over 50 hours before handing it over.

Just a thought:

Intermittent static that is more of a hiss than a crackle usually points to a leaky cap - and that may be thermal.

The intermittent speaker noise was crackling, not a hissing sound. It would come and go.

Reflow the joints on your input board and I suspect you'll be good to go. FYI: I probe with the amp on using a pointed 1/4" wood dowel (using disposable speakers, not your Quads). Bad solder joints tend to be... apparent when probed this way.

Wonder if that might be an issue with the CCS transistors?

(assUming the board you're using has those)

Great help here -- stay tuned for update tomorrow!

haven't had any LM334s fail in normal use in the past 10 years, but of course if you're probing around indiscriminately, who knows?
Easy to check by looking at pin 3 and 8 on the two outer tubes of the driver, should be around 10vdc, and/or the plates pins 1 & 8 should be around 210vdc,
both of which are noted on the schematic included with the boards.

I've seen this - reflow the bias pots.

Will do that next, thanks. Reflowed the driver tube sockets and all light up fine now. Cycled on off several times. Went to left channel setup, decent bias holding at 0.405vdc, but intermittent static in speaker. Rectifier really hot. Didn’t do pinout checks on tubes yet but getting mullards tomorrow to try. All other underside continuity ok. Nothing loose either, connections solid fortunately gave my friend my personal ST70 I built earlier so he has tunes!

For the tube rollers out there... when you re-tension your PCB mounted sockets do you also re-flow them? Probably a good idea.

Retensioning a noval socket isn't normally necessary as long as you don't try to rock the tubes out. Reflowing the joints more than necessary just puts undue wear on the board, increasing the chances of lifting a trace.

Now, octals are another story. I retension those every time I swap tubes. I also tend to crimp the tabs after string the wires and before soldering to make the connections as strong and tight as possible.

Well thanks for the suggestions and help. Went through the driver board, sockets, quad cap and connections and reflowed. New mullard EL34’s. No speaker static and bias all good with trim pots at 12 o’clock roughly. Will let it sit tonight, rebias tomorrow if necessary and start the burn in. Got a local JJ rectifier but mullard coming Monday. A spare not a bad idea anyway. Thanks again to all

Glad it's sorted but the fact that 2 tubes didn't light would make me first look at the fillament supply connections to those tubes.

On the driver board two are in parallel and the other on its own. If you followed the instructions then the fillament supply goes from the power transformer to each of the output tubes and then off to the driver board on each side. The fact that you didn't mention anything about the output tubes not being lit would say that the connection from the power transformer to those was ok and the fault could have been in the link from the front tube over to the driver board. This link goes directly to the fillament pins on the tube and does not use any PCB tracks. My guess at this stage would have been on a dry joint on one of the front tube fillament pins.

Thanks. Did continuity checks and re flowed that path fromdriver board back to power transformer first. Same problem. Reflowed the driver board tube sockets, and along with other tubes, so far problem solved. Also crackling/static gone, quiet as a mouse. Oh, also took the suggestion to reflow the bias pots