[Paul Running]

When the tubes are pulled, the grids are at -37V.

At 300VDC on the plates, bias should be hotter than -25VDC for the 7027A; -37VDC spec'd for 500VDC+ on the plates.
Those figures should be adjusted when the screen voltages are different from the spec.

 

[ook ook]

I removed the current limiter from the chain and voila, the amp comes on and makes sound. The tubes are sitting at around 430V on the plates, drawing ~41mA.

I played it at super low volumes for a few minutes. After raising the treble to about half and the volume to about a quarter, I played through it for a few seconds and smelled smoke so I quickly shut it off.

Don't know what exactly is unhappy, but at least I now know it at least partially functions

 

[ook ook]

Plugged it in through the current limiter again today and it makes sound with the limiter now. When I initially plugged it in, same tubes, same speakers, same guitar, same cables, same light bulb, I was getting no sound from it. Played it through the light bulb for a long while and had no catastrophes. Only issue was, apart from low voltage and current, that sometimes on low notes there would be a background buzzing sound.

 

[Paul Running]

apart from low voltage and current,

Current limiting, is also limiting voltage (Ohm's law); if the filament voltages are too low: operation will not be typical so, expect some odd occurrences.

 

[ook ook]

Current limiting, is also limiting voltage (Ohm's law); if the filament voltages are too low: operation will not be typical so, expect some odd occurrences.

Yeah, the low current and voltage is to be expected in this case. The filament voltages would certainly have been too low.

Everything seems fine now when running through the current limiter. I was getting no sound before but it seems to have just corrected itself?

I'm just wary since when I removed the current limiter, I smelled smoke after several minutes. Not sure where it originated.

A bunch of the carbon comp resistors measure high in circuit (e.g., 82k plate resistor on the PI measures 113k in circuit), so it might be time to go through and replace anything suspect (also clean up some of the mess).

I don't trust the filter caps. I don't really have any particular reason to not trust them, I just don't know how old they are and I'd rather have higher quality caps in there. The 40uf caps that would have been in there originally have been replaced with JJ 32uf cans with a 70° temp rating. They aren't visibly failing, but I don't have an esr meter and I haven't taken any measurements out of circuit yet.

 

[Paul Running]

I don't trust the filter caps. I don't really have any particular reason to not trust them

Do they get warm after the system stabilizes?

 

[nonreverb]

Yeah, the low current and voltage is to be expected in this case. The filament voltages would certainly have been too low.

Everything seems fine now when running through the current limiter. I was getting no sound before but it seems to have just corrected itself?

I'm just wary since when I removed the current limiter, I smelled smoke after several minutes. Not sure where it originated.

A bunch of the carbon comp resistors measure high in circuit (e.g., 82k plate resistor on the PI measures 113k in circuit), so it might be time to go through and replace anything suspect (also clean up some of the mess).

I don't trust the filter caps. I don't really have any particular reason to not trust them, I just don't know how old they are and I'd rather have higher quality caps in there. The 40uf caps that would have been in there originally have been replaced with JJ 32uf cans with a 70° temp rating. They aren't visibly failing, but I don't have an esr meter and I haven't taken any measurements out of circuit yet.

You had mentioned earlier that while testing the limiter was on but not bright? The fact that it was glowing at all probably meant that the amount of current left for the amp to operate was insufficient. Particularly where higher wattage amps are concerned.
You can narrow smoke down to a few select places. I'd check the screens for one to see if they show any kind of drift or browning, B+ caps as stated also B+ drop resistors. These areas are where things usually go wrong first.

 

[ook ook]

Do they get warm after the system stabilizes?

Left the amp running for a while through the current limiter and nothing was getting hot. Apart from the tubes, of course.

You had mentioned earlier that while testing the limiter was on but not bright? The fact that it was glowing at all probably meant that the amount of current left for the amp to operate was insufficient. Particularly where higher wattage amps are concerned.
You can narrow smoke down to a few select places. I'd check the screens for one to see if they show any kind of drift or browning, B+ caps as stated also B+ drop resistors. These areas are where things usually go wrong first.

All of these measurements I took in-circuit.

The 470 ohm screens are showing 122 ohms. I don't think I've ever measured screen resistors in-circuit so I don't actually know if this is accurate.

The 10k dropping resistor before the preamp filter cap is measuring over 11.2k. It seems to have drifted above it's 5% tolerance. The 4.7k is measuring 5.15k, pretty close to it's 10% rating. Nothing is visibly burned or anything like that.

Several other resistors throughout the circuit are measuring too high

 

[nonreverb]

Left the amp running for a while through the current limiter and nothing was getting hot. Apart from the tubes, of course.


All of these measurements I took in-circuit.

The 470 ohm screens are showing 122 ohms. I don't think I've ever measured screen resistors in-circuit so I don't actually know if this is accurate.

The 10k dropping resistor before the preamp filter cap is measuring over 11.2k. It seems to have drifted above it's 5% tolerance. The 4.7k is measuring 5.15k, pretty close to it's 10% rating. Nothing is visibly burned or anything like that.

Several other resistors throughout the circuit are measuring too high

122 ohm seems awfully low. You "should" get an accurate reading of the screen resistance in circuit as the screens are infinite resistance when the amp is off....or are supposed to be.

 

[ook ook]

122 ohm seems awfully low. You "should" get an accurate reading of the screen resistance in circuit as the screens are infinite resistance when the amp is off....or are supposed to be.

The 122R reading turned out to be an issue with my meter. It's a cheap meter which doesn't like high voltages or low resistance, giving wonky readings. I replaced the 470s anyways since I had some 3w metal oxide 470Rs on hand.

I have now replaced the phase inverter plate resistors, the power supply dropping resistors, the screen grid resistors, and I replaced the V1 cathode bypass cap.

I ran the amp without a current limiter last night for about 30-40 minutes and nothing seems to have gone wrong. The 6.3V heater winding reads 6.3V with an input of 110VAC and at that voltage I was getting 405-410V on the plates of the 7027As and 39-41mA plate current.

 

[ook ook]

I'm not so worried about the amp blowing up now, but my work is still not over. Anyone who's looked at the pictures of the inside I posted earlier can tell that there's a real mess in here. I haven't done any testing at high volumes so that's something I'll have to do at some point.

I still plan to make the swap to EL34s/6CA7s. I know that Traynor themselves recommended this change in the 60s, but is there any reason to ignore their advice? I know EL34s have a higher heater current rating. Can the PT handle it under all conditions?

Gotta add some grid stoppers on the power tubes as well. Might experiment with the nfb loop down the line. I don't feel bad about experimenting with this amp since it's already been heavily tampered with.

Here's a schematic of the amp as it currently stands as well. Mostly for my own future reference, but it's here for anyone interested. If anyone sees anything that concerns them, I'd love to hear about it.

 

[Paul Running]

The output tube is a personal choice that only you should make.
Try the 7027s and the EL34s. If you like the 7027, you could also try 6L6GC. the 7027 is modelled from the 6L6. Compare both internals, referencing the datasheets on the 6L6GC and the 7027a, noting their similarities.

 

[ook ook]

The output tube is a personal choice that only you should make.
Try the 7027s and the EL34s. If you like the 7027, you could also try 6L6GC. the 7027 is modelled from the 6L6. Compare both internals, referencing the datasheets on the 6L6GC and the 7027a, noting their similarities.

I have looked at the datasheets and noticed the similarities. I was more so wondering if there was any chance of overloading the power transformer or something when running the amp at max volume. I suppose not. I have some el34s which I'll try at some point, but for now, these sketchy looking 7027as seem to be doing the job.

 

[diyfabtone]

I don't know if Traynor changed the power transformer on the YBA-1's when they switched to 6CA7's or not? Swapping in 2xEL34's requires and additional 1.2A more current @ 6.3VAC for the secondary to provide compared to 7027A/6L6 types which is more than a 20% increase - that might be too much even for the conservatively rated hammond trannies?

I'd try it myself by checking how hot the pwr tranny feels after playing for 10 mins with the 7027A's then swap in the EL34's (check that PIN 1 is tied to 8 or grounded first) and repeat ... if the temp increase doesn't seem too excessive you might be ok.

 

[Paul Running]

To improve heat transfer from transformer...85W CPU thermal sinker.

 

[Mark Brown]

To improve heat transfer from transformer...85W CPU thermal sinker.
View attachment 465056

Waterblock or nothing man!

Thats actually a really smart idea. You could even lose the fan and just add passive sinks. Wouldn't have the same effect but something tells me finned aluminum will dissipate heat a lot faster than laminated iron sheets.