Is my OT fried?

[Gainzilla]

Hi guys,

Took some updated readings last night (havent yet replaced the 470k PI bias resistor with a 470r, so the voltages there still look weird). Any idea why they dropped so much from the previous readings? Things I did right before the reading were: Lowered the B+ dropping resistors. They're now 100r, 100r, 1k. I also re-attached the presence circuit, and the mosfet effects send stage. When looking under the board I noticed I missed the B+ connection to the Mosfet. I wouldn't think that would make the B+ drop though, would it? I'll connect that tonight and see if that changes anything.

B+: 249v
B2: 248v
B3: 247v
B4: 245v

EL84 2
Pin 2: 0v
Pin 3: 8.63v
Pins 4/5: 6.4vac
Pin 7: 245v
Pin 9: 247v

EL84 1
Pin 2: 0v
Pin 3: 8.63v
Pins 4/5: 6.4vac
Pin 7: 245v
Pin 9: 249v

12ax7 3:
Pin 1: 224v
Pin 2: 0v
Pin 3: 3.92v
Pins 4/5/9: 6.4vac
Pin 6: 227v
Pin 7: 0v
Pin 8: 3.92v

12ax7 2:
Pin 1: 137v
Pin 2: 0v
Pin 3: 1.79v
Pins 4/5/9: 6.4vac
Pin 6: 126v
Pin 7: 0v
Pin 8: .81v

12ax7 1:
Pin 1: 147v
Pin 2: 0v
Pin 3: 2v
Pins 4/5/9: 6.4vac
Pin 6: 160v
Pin 7: 0v
Pin 8: 2v

[nickb]

That PI is still wrong. Time to check it again. Measure on each of the 10K resistors. There no NO WAY the grids would be zero unless you have them wired wrongly.

[Gainzilla]

Thanks Nick, I discovered after I took the voltage readings that the PI bias resistor is actually 470k, so those values still reflect that. Ill pick up a 470r on the way home tonight. I assume those readings would be in keeping with a wrong value bias resistor?

[g1]

The lower supply voltages are probably because the first set of readings was before the power tube cathode resistor was grounded. The power tubes weren't running, so the supply voltages were reading high, the same as they would if the power tubes were removed.

[nickb]

Thanks Nick, I discovered after I took the voltage readings that the PI bias resistor is actually 470k, so those values still reflect that. Ill pick up a 470r on the way home tonight. I assume those readings would be in keeping with a wrong value bias resistor?

OK. I see, gotcha Yes the resistor would explain it.

Anyway, I find it very interesting that you liked the sound in that condition. It would not have been very loud, but hey, whatever.

Edit: No, that resistor does not explain it all. The plate voltages show that significant current is flowing in the tubes. It's almost as though you had the 10k's and 470 mixed up. Measure the volts both sides of each of the 470, and two 10K resistors and / or recheck.

[Gainzilla]

Aaaah. Makes perfect sense. Do you think those voltages are ok. My concern is that they're too low. I mean, it's a small amp, but if I had to I would consider looking into a PT with a bit more heft. Rather not, if I can avoid it though. This was supposed to be a scrap build.

[Gainzilla]

Sounded fantastic! But you're right, it was really quiet. Also very late at night, running through an attenuator (baby sleeping in the next room). I actually thought the low volume was because I hooked up the mosfet loop driver and didn't have the B+ connected to that stage. Hopefully once I get the new resistor in there and the B+ to the Mosfet it'll be loud and proud!

I'll report back with my results. Cheers!

[Gainzilla]

Ok, I didn't see your edit until after I left the house. I replaced the 470k with the correct 470r, and also changed the 10k to a higher quality one that I bought last night.

Results:
- Plenty loud
- New cathode voltage: 38v

Thoughts? Does that seem within ideal tolerances?

[Mick Bailey]

Cathode voltage seems fine to me. Big difference to what you had before. Out of interest, what are your other V3 voltages?

[Gainzilla]

Nice!

I didn't think to check the other voltages, assuming those wouldn't be impacted. Wish I would've grabbed those...

Here are the most recent readings, though, with the updated cathode added. How does that look (assuming for the moment that only the cathode changed)?

12ax7 3:
Pin 1: 224v
Pin 2: 0v
Pin 3: 38v
Pins 4/5/9: 6.4vac
Pin 6: 227v
Pin 7: 0v
Pin 8: 38v

I'll get fresh readings tonight.

Thanks!

[nickb]

Nice!

I didn't think to check the other voltages, assuming those wouldn't be impacted. Wish I would've grabbed those...

Here are the most recent readings, though, with the updated cathode added. How does that look (assuming for the moment that only the cathode changed)?

12ax7 3:
Pin 1: 224v
Pin 2: 0v
Pin 3: 38v
Pins 4/5/9: 6.4vac
Pin 6: 227v
Pin 7: 0v
Pin 8: 38v

I'll get fresh readings tonight.

Thanks!

I agree with Mick about the cathode voltage being OK

But note the plate voltages are unchanged from before and the grids measure zero? No way! I wonder, are you measuring the grid voltage by going across the 560K resistors? BTW, that is the right way to do it on a PI except you have to remember to add the voltage at the cathode end of the grid resistors to make to ground referenced as that is what is expected.

So now to the plate voltages. B2 is 248V and the plates are 224V and 227V. But, we know the current is 38/(20.47k)/2 = 0.93mA per side. So the plate voltage should be 248 -100K*0.93 = 155V yet you have around 225V. You'd need a 24k plate resistor to do that. Check the value of the plate resistors. It'll work but the gain will be a little less as is.

PS sorry about slow reply - sometimes I don't get email updates and have to manually check.

[Gainzilla]

No worries! You're helping me with my issues. I'm happy to wait (course, that doesn't mean I can contain myself from creating more mischief in the meantime. lol).

I think I understand what you're saying for the most part. The way I've been measuring everything is from chassis ground to tube socket pin (except for the filaments). So to clarify, I'll measure from the top of 560k to 560k, then add the measurement from ground to the junction where all three grid/cathode resistors meet? Then that will tell me the PI grid voltage (where currently I have 0v)?

Thanks, and apologies for the basic questions.

Cheers!

[nickb]

No worries! You're helping me with my issues. I'm happy to wait (course, that doesn't mean I can contain myself from creating more mischief in the meantime. lol).

I think I understand what you're saying for the most part. The way I've been measuring everything is from chassis ground to tube socket pin (except for the filaments). So to clarify, I'll measure from the top of 560k to 560k, then add the measurement from ground to the junction where all three grid/cathode resistors meet? Then that will tell me the PI grid voltage (where currently I have 0v)?

Thanks, and apologies for the basic questions.

Cheers!

See if this helps:



Have you looked at those plate resistors yet?

[Gainzilla]

THANKS, yes that will definitely help!

I'm at work right now but I have a pic... Funny, but I bought mostly mil-spec resistors. Some of them came with bands, and others had just the value printed on them (which I liked). I tried to install those so that the value was visible from the top. Wouldn't you know it, one of those was the only one in my build where the value is not visible. :|

They should be as per the schematic, 82k and 100k, but at this point I'm not taking anything for granted. I'll check when I get home tonight, and get the other voltages as well. It seems more likely that the values did change when I swapped out the cathode resistor. Hopefully we'll see that and the values will look normal.

[Enzo]

That is why the color code stripes go all the way around.