Bugs in a scratch built JTM45

[mort]

Still around 760 with the recto removed and also with the 5U4G installed.

[big_teee]

Still around 760 with the recto removed and also with the 5U4G installed.

The 760AC won't change.
It is the Rectified B+ voltage on V4,& V5 pins 3.
If the rectifier is working put it back together and measure DC on Pin 3.
The 5u4 should put out less B+

[mort]

With the 5u4 I'm getting 530v on pin 3 of v4 and v5 with the standby switch in the on position. Standby off the voltage slowly just keeps dropping....

It's an old old tube so maybe it's bad, even though it's working??

[MWJB]

The voltage of the PT is not causing blown fuses.

You are getting more voltage in "play" mode, than with the amp in standby? Doesn't scan, you can only assess the idle voltage with power tubes in operation and drawing current.

380VAC through a 5U4 should give you around 460-480vdc on the plates, depending on current draw (plate & heater)?

Install a 1ohm 2W resistor from each power tube cathode (pins 1 & 8 tied) to ground. Then you can read the idle current with a DMM. Fuses are blowing because of the current demands, not the voltage.

The 5U4 hits the amp with a lot of voltage right off the bat, then it drops as the tubes draw current....leave the amp in standby for about minute before throwing into play mode, then the voltage should be settled when you take a reading.

[guitician]

Check your circuit to the schematic VERY carefully...assuming all the components are fine, there's nothing that could be wrong...right?

[mort]

Ok, all I could get my hands on for 1 ohm resistors was 10W, but that should work right? An ohm is an ohm is an ohm...

With the bias pot all the way counter clockwise:

V4
P3 - 490v
P5 - -67.4
across resistor - 10.7mv

V5
P3 - 490v
P5 - -67.4
across resistor - 11.3mv


With bias pot all the way open:

V4
P3 - 486v
P5 - -65.0
across resistor - 13.2mv

V5
P3 - 482v
P5 - -65.1
across resistor - 14.2mv



According to what I'm reading LINK, this means I'm getting between 5 and 6 watts per tube (??)

That doesn't seem right, the amp sounds like it's 30 or 40 watts at least, generally speaking.

[big_teee]

Ok, all I could get my hands on for 1 ohm resistors was 10W, but that should work right? An ohm is an ohm is an ohm...

With the bias pot all the way counter clockwise:

V4
P3 - 490v
P5 - -67.4
across resistor - 10.7mv

V5
P3 - 490v
P5 - -67.4
across resistor - 11.3mv


With bias pot all the way open:

V4
P3 - 486v
P5 - -65.0
across resistor - 13.2mv

V5
P3 - 482v
P5 - -65.1
across resistor - 14.2mv



According to what I'm reading LINK, this means I'm getting between 5 and 6 watts per tube (??)

That doesn't seem right, the amp sounds like it's 30 or 40 watts at least, generally speaking.

It appears now that you have your bias voltage too high.
With the bias set you should be reading in the -35 to -45V.
-65V is turning the output tubes way down, but you are getting it figured out.
You may have to readjust that bias ground resistor.
T

[mort]

You may have to readjust that bias ground resistor.
T

That's what I was thinking... back down to the 56k and read again

[big_teee]

That's what my 2204 uses.
a 50k pot in series with the 56k resistor.
It will adjust a very wide range.
T

[MWJB]

Idle dissipation of a few Watts per tube can still see your amp make full power...there is no linear correlation between idle disspation W and W audio output in a fixed bias amp.

Because you have fitted such large load resistor to your bias voltage divider, you have reduced the useful sweep of your bias trim pot. You are looking for a .79:1 ratio...you have a 10K drop resistor after the diode, so your load at the bias pot's mid setting wants to be around 39K - funnily enough, you might recall you started with a 50K pot and a 15K fixed resistor...50K/2 = 25K, 25K+15K = 40K, 10K dropper and a 40K load gives you a 0.8 divider...hey, what a coincidence, it's almost as if it was designed that way!? :-0 :-)

The negative bias voltage that you are looking for with 6L6/EL34 is 11% of the power tube plate voltage...more negative bias voltage (bigger number) with KT66. So at 470vdc you'd want to see -53vdc at the junction of the 220K resistors, at 500vdc you'd want to see -55vdc - at the mid pot/nominal setting. Set coldest (most negative -dc) when installing new/different tubes.

[mort]

Yep, you had mentioned losing useful sweep when upping that dropping resistor. Now I see what you mean.

With the 56K installed I'm getting a range of 13 to 20mv. Gonna drop it again to a 23K and see what I get.



The reason I was upping the dropper value was to get my negative voltage up. Now I'm just dropping it back down because I have too much... what changed??

Wait, it was adding that cap at the diode that brought the negative voltage way up... I guess that was the problem all along. Cool, I'm learning something here

[mort]

With 23K dropper


With the bias pot all the way counter clockwise:

V4
P3 - 476v
P5 - -62.0
across resistor - 16.6mv

V5
P3 - 476v
P5 - -62.0
across resistor - 17.9mv


With bias pot all the way clockwise:

V4
P3 - 443v
P5 - -48.4v
across resistor - 44.5mv

V5
P3 - 440v
P5 - -48.4v
across resistor - 47.4mv


That puts me at ~20W per tube with the bias pot at its lowest resistance setting.

Let me stop right here before lowering the dropper back to the 15K that I started with in the first place. ***If I want to be able to swap EL34's into this amp with only a bias pot adjustment, should I stay where I am with the dropper, or should I put back the 15K?

[guitician]

Where did all the hum go? It seems you were running the EL34's cold at -67v dc so the bias wasn't causing it.

[mort]

Where did all the hum go? It seems you were running the EL34's cold at -67v dc so the bias wasn't causing it.

The loud hum was cured by swapping the OT secondary wires around.

Also, I had gotten ahead of myself and switched to KT66's. They have been in for all of yesterday and today's readings and I will keep them in until all the wrinkles are ironed out AND when I understand what needs to be taken into account when switching them back to EL34's.

[MWJB]

I'd keep the plate current under 40mA...13-17W dissipation should be fine.

Depends on the brand of tube, different 6L6s will bias up differently, as will different brands of EL34, but there is not a specific setting for EL34 vs. 6L6.

What changed with regards to the bias voltage was adding that 10uf cap after the diode.

This formula is useful for bias, NFB & line out voltage dividers in particular:

We'll call R1 the dropper in series with the voltage, R2 is the load connected to ground.

R2/(R1+R2)*voltage in = voltage out.

So 23K/(10K+23K)= 0.7
0.7*-70vdc= -49v.

Also notice you now have a plate voltage of 441.5vdc average, 441.5*11% = -48.6vdc...pretty close to what you are reading.

You shouldn't need to go to 15K with the tubes you have at hand, reducing the 23K to ground will just make everything hotter.

You now know pretty much everything there is to know about setting fixed bias in a 6L6/EL34 amp running 430-530vdc. :-)