Powering up my amp

[Paul P]

I finally mustered the courage to turn my new build on today (the
FrankenstAmp one over in "Build Your Amp") and I may have a problem
but I'm not sure.

I plugged my amp into my Dimbulb with a 15w bulb in it. When I turn it
on with no tubes in the amp (which uses a tube rectifier) the bulb glows
at maybe 25% brightness. My PT is 320-0-320 and I get 222vac at
pins 4 and 6 of the rectifier socket. Heater taps are 2.3 and 2.2 vac
(I've got a balance pot). My 50vac bias tap is at 35vac. So far things
seem ok though I'm not sure if I should be getting any kind of glow from
the bulb.

Next I plugged in two different RCA 5U4GB rectifiers which both test off
the scale good on my somewhat basic Triplett tube tester. When I turn
the amp on the 15w glows at full on. I haven't dared go any further than
this.

Is it possible that the lamp is glowing brightly while my 2x20uf first stage
filter caps are charging up and that I don't necessarily have a short ?
I've gone over all the wiring a few times and nothing looks wrong.
When I saw the bulb go full on I immediately shut the amp off. How much
time does it take to charge the caps ? I did notice that B+ measured at
the filter caps stayed at zero but perhaps the multimeter didn't have enough
time to display a reading before I switched the amp off.

With the amp off if I measure the resistance across the filter caps it starts
off at about 320K then falls constantly down from there. I presume the
caps are charging with the resistance-measuring voltage but this causes
the resistance to go down ?

Any help would be much appreciated.

Paul P

[tubeswell]

Try plugging in an old rectumfrier that you know works and some old reliable tubes, and switch it on with the meter connected to the B+ and watch closely for smoke.

[Paul P]

Problem solved, which was the 15w bulb. I think there wasn't enough
power getting through to operate the rectifier tube, even though it was
enough to light the bulb brightly. I replaced it with a 100w bulb and was
overjoyed to see my B+ climb up to 412v using a new JJ GZ34. I then
put a 5U4GB back in and B+ went immediately to 399v.

Now I'll measure all the other voltages.

Paul P

[Paul P]

Hey, the thing actually works !

Well, sort of. I get a real nice clear tone from my signal generator but
there's still something not right at the power tube.

My circuit right now is close to a Champ. One Tung-sol 12ax7 and one
SE 6v6GT, also Tung-sol. I switched the rectifier to a NOS RCA 5Y3GT.

When I first took the amp off standby I measured about 70ma through the
power tube cathode both with fixed bias, at about -40vdc, and also when
switched to cathode bias. Needless to say when I went around back to
see what my power tube looked like there was this faint red glow coming
from inside. (The amp sounded great though and NO hum).

I shorted my bias circuit's newly installed dropping resistor and got the
current through the cathode down to 39ma with -56v worth of bias and
the power tube stopped glowing.

So something isn't right. Here are the numbers :PT secondary 329-0-330vac (spec says 320-0-320)
B+ first filter on standby 442 vdc
B+ first filter off standby 430-440
6v6 at plate 382vdc
6v6 at screen 385-387vdc
bias -56vdc
B+ to preamp 353vdc
VIa,b plates 240vdc
First of all, isn't my B+ kind of high for a 5Y3GT ? I also find it a bit strange
that the screen was higher than the plate, though this may just be due
to variations between measurements of my AC mains which seems to vary
quite a bit. I'm using a 470 ohm screen resistor and have a 100 ohm 3H choke
between the plate and the screen node. And a 470 ohm cathode resistor
bypassed by a 25uv 50v cap when switched to cathode bias.

Any ideas where to go from here ?

Paul P


,

[tubeswell]

Hey, the thing actually works !
First of all, isn't my B+ kind of high for a 5Y3GT ? I also find it a bit strange
that the screen was higher than the plate, though this may just be due
to variations between measurements of my AC mains which seems to vary
quite a bit. I'm using a 470 ohm screen resistor and have a 100 ohm 3H choke
between the plate and the screen node. And a 470 ohm cathode resistor
bypassed by a 25uv 50v cap when switched to cathode bias.
,

You might just have an efficient 5Y3GT.

As for the screen voltage, try chucking a (1k5? - 5k?) 5W grid resistor at the screen grid pin and see what that does

[Paul P]

I retook the measurements this morning. I may have gotten a few things
confused yesterday but the fact remains that the voltage at the screen is
higher than at the plate.

Here are some more accurate numbers, using a NOS RCA 5Y3GT :

On standby :AC mains.......... 120 vac
PT secondary ..... 336-0-335 vac (spec 320-0-320)
B+ first filter... 449 vdc, 150 mvac
Heater bias....... 48 vdc
Filaments ........ 3.4--3.3 vac
Off standby :B+ first filter... 391 vdc
6v6 at plate ..... 384 vdc
screen node ...... 389 vdc
6v6 at screen .... 388 vdc
preamp node ...... 355 vdc
V1a,b at plate.... 236 vdc

6v6 cathode current at idle :
cathode bias.......... 48 ma
fixed bias at -57 vdc, 39 ma
Paul P

,

[scole]

paul, i went through this with my champ. for some reason only fender can make a 6v6 sound good at about 400v on the plates, the rest of us have to get that plate voltage down to about 350, otherwise the amp sounds stiff as a board. refer to this thread:
http://music-electronics-forum.com/showthread.php?t=4879

[Paul P]

Thanks for the help tubeswell and scole. I changed my screen resistor to
1500 ohms (only had a 2w on hand but that should be ok for the moment)
and that brought the screen down to below the plate. As far as the amp
sounding stiff, I don't have enough experience with amps to really know
what that means. I've been strictly an acoustic guitar player though my
daughter has a Fender MIM Strat Jr and a small Vox Pathfinder 15r ss amp.

I also played with the decoupling resistors. Here are the updated voltages,
off standby :B+ first filter... 384 vdc (2 vac)
6v6 at plate ..... 378 vdc
screen node ...... 383 vdc
6v6 at screen .... 375 vdc
preamp node ...... 307 vdc
V1a,b at plate.... 205 vdc

6v6 cathode current at idle :
cathode bias.......... 48 ma
fixed bias at -51 vdc, 48 ma
Not sure if this is ok or not. Looking at the power tube with all lights off
I see a very faint glow in a band about 1/4" wide top to bottom. Is this
just the heater showing through ?

I'll talk about the amp's performance back in the main thread.

Paul P

,

[tubeswell]

Hi Paul P

That screen voltage still looks too close to the plate voltage. You might try going to a 2k2 or 4k7 5W screen grid resistor in there. If it destabilises the rest o fthe power supply too much, then decouple it with a 10uF 450V cap to ground with the +ve end at the screen grid pin

[Paul P]

I'll try a 4.7K screen resistor later on today.

My main problem is that my poor 6v6 is dissipating 17w. Too bad I
can't leave it like this, it sounds good. I can get this neat sort of "whump"
or "pwop" to each each note and chords are full of all sorts of harmonics
bouncing around all over the place (maybe a bit too much so, but I'm using
my daughter's strat with its 22-3/4" scale and the strings are a bit like
rubber bands).

Paul P

[tubeswell]

What sort of tube is it? I'm running one of those new-fangled New Sensor TungSol 6V6GT 'reissues' in that 5F2A I just made and it seems to happily handle a plate voltage of 364 at idle, and seems to do some of those neat things you describe

[Paul P]

What sort of tube is it?

I'm also using a Tung-Sol 6V6GT.

I'm going to move my choke to before my first filter (just have to switch
a few wires). Hopefully the choke won't radiate too much hum. This should
drop my B+ down to some reasonable level. Then I can easily drop the
screen to something lower.

Paul P

[Paul P]

Well, changing to "choke input to filter" dropped B+ alright. Right through
the floor ! All the way to 280vdc at the first filter caps using the 5Y3GT.
With a GZ34 it move up a bit to 296vdc.

Doesn't sound so great either. Kind of dead, no prettying up of the sound
of the guitar strings. Like an acoustic only louder. Not terribly loud and
there is some distortion at the upper end of the volume. Also, a funny
thing, one note sounds clean but two notes or more together causes a
mumbly background confusion. None of the neat sounds I was getting
before.

This change played havoc to the other voltages. Here they are for the record :

With 5Y3GT :

...On standby :AC mains.......... 120 vac
PT secondary ..... 334-0-334 vac
B+ first filter... 417 vdc, .85 mvac
...Off standby :B+ 1st filter cap. 280 vdc, .75 mvac
6v6 at plate ..... 275 vdc
screen node ...... 272 vdc
6v6 at screen .... 267 vdc (1500 screen resistor)
preamp node ...... 220 vdc
V1a,b at plate.... 144, 146 vdc

6v6 cathode current at idle :
cathode bias, 470 ohms .. 32 ma
With GZ34 :

...On standby :B+ first filter... 421 vdc, .80 mvac
...Off standby :B+ before choke... 296 vdc
B+ 1st filter cap. 292 vdc
6v6 at plate ..... 288 vdc
screen node ...... 285 vdc
6v6 at screen .... 281 vdc (1500 screen resistor)
preamp node ...... 230 vdc
V1a,b at plate.... 151, 153 vdc

6v6 cathode current at idle :
cathode bias, 470 ohms .. 34 ma

Back to the drawing board :).

Paul P

[tubeswell]

In that post Scole put up earlier, Bruce from Mission Amps suggests 9.1v to 15v, 5 watt zener diodes (reverse biased to ground off the center tap in the high voltage secondary of the PT) to lower the B+

[Paul P]

In that post Scole put up earlier, Bruce from Mission Amps suggests 9.1v to 15v, 5 watt zener diodes (reverse biased to ground off the center tap in the high voltage secondary of the PT) to lower the B+

Bruce also suggested using a choke but I think scole was using a PT with a
higher secondary than mine. I don't have any zeners lying around and adding
one or two in my amp could be a bit of a pain but if it's the only way...
Another thing is I'm going to be adding more circuits to my amp which
should lower the B+ some.

Tomorrow (today to you) I'm going to get the other voltages to their
proper levels and see what the amp sounds like. This will give me some sonic
experience at the "low B+ high current" end of things.

It's really nice to be able to hear the differences in tone I've only read about
till now.

Paul P

[Paul P]

I put everything back the way it was. If I bias the output tube with
-56v I get a current of 36ma and, with the plate at 390 or so, I get
a dissipation of 14w. So I'm ok. I could also achieve this with cathode
bias by changing the current 470 ohm cathode resistor to 600 ohms,
which I don't have right now. This with the 5Y3GT.

I also changed the screen resistor to 4.7K and that dropped the screen
voltage to 380v.

The amp sounds pretty good like this so I'm going to leave it for now and
go on to phase 2 in which I'll add the second channel preamp and effects.

The numbers as they now stand :

With 5Y3GT :

...On standby :AC mains.......... 120 vac
PT secondary ..... 331-0-331 vac
B+ first filter... 440 vdc
...Off standby :B+ 1st filter cap. 393 vdc, 1.3 mvac
6v6 at plate ..... 389 vdc
screen node ...... 393 vdc
6v6 at screen .... 380 vdc (4.7K screen resistor)
preamp node ...... 305 vdc
V1a,b at plate.... 202, 204 vdc

6v6 cathode current at idle :
fixed -56v bias .. 36 ma
6v6 dissipation at idle...... 14 w
Paul P