Grid Stoppers for KT66s

[p2pAmps]

Ok fellas,

I am getting a few little squeals on my KT66 DR.  At first I thought I had a bad preamp tube but I changed em out and tonight I notice some high pitch sounds on certain notes coming out.

Now, I used 1.5K grid stopper on those KT66s and also wired the sockets to use EL34s too.  The screens are 1K.

Do I need to bump the value of those grid stoppers up and how far? 

Mike

[SILVERGUN]

Mike,
You may be better served to try the common application of adding a small value (47pf) cap from plate to plate on the phase inverter as a means to quell parasitic oscillation.

[p2pAmps]

Mike,
You may be better served to try the common application of adding a small value (47pf) cap from plate to plate on the phase inverter as a means to quell parasitic oscillation.

Yeah I thought of that but was worried it might kill the sparkle...  In your experience does adding the 47pF cap change the tone much?

[SILVERGUN]

Mike,
You may be better served to try the common application of adding a small value (47pf) cap from plate to plate on the phase inverter as a means to quell parasitic oscillation.

Yeah I thought of that but was worried it might kill the sparkle...  In your experience does adding the 47pF cap change the tone much?

No, as long as the value is kept low it doesn't seem to me to be a game changer as it is acting on ultra-high frequencies.


Another possible solution could be to increase NFB slightly, if you are using the values for DR 6V6s.
Comparing AB763 schematics, you'll notice the Super Reverb (6L6) uses 100ohm off the tail of the PI to ground, whereas DR uses 47ohm

A combination of those 2 moves should get you a favorable result.

[tubeswell]

FWIW, even running 100pF from grid to ground on those KT66s won't kill sparkle - but it sure will slug that HF oscillation at the dominant NFB loop pole.

[tubenit]

I am not a fan of the traditional plate to plate "snubbing" cap. When I've tried it a few times, I felt like it killed some of the tone and left it dull.
Just a personal preference to not use one.

I'd be inclined to get some insulated alligator clipped wires and experiment with 10p and go up in value from there if you're going the snubbing cap route.

In contrast,  I really like the "enhance" cap  across the plate resistor going into the LTPI.  I can't hear any loss in high frequencies but the higher frequencies sound MUCH smoother to me.  Again you could clip a 220p safely across the plate resistor and see if that works for you and you like it.  Any LTPI amp I build, ends up with the enhance cap.

http://el34world.com/Forum/index.php?topic=12723.msg153636#msg153636

And you can go to 5.6k to 10k range on grid resistors on the KT66's. I think you will find that still sounds fine to you.  On the SoLow Watt amp I designed using ECL84's (not EL84's),  I used 56k grid resistors.

with respect, Tubenit

[HotBluePlates]

... I am getting a few little squeals on my KT66 DR.  ... Do I need to bump the value of those grid stoppers up and how far?  ...

My experience with KT66's in a tweed style amp were the same; KT66's (at least modern production) tend to ring. I'd raise G1 & possibly G2 stoppers, but you should do it experiementally to see what it will take in your amp.

[p2pAmps]

Well so far I replaced my grid stoppers with some 5.6K but it was a little late for me to start testing last night.  I'm going to get after it this evening when I get home from Job#1...

I appreciate all the great feedback guys.

Another side note is I noticed the PT was pretty warm when I got home and tore that chassis out.  It's sized properly for sure @5A on the heater current.  Plate current on those KT66s are 430vDC, Bias set to around 39...

~Mike

[SILVERGUN]

Another side note is I noticed the PT was pretty warm when I got home and tore that chassis out.  It's sized properly for sure @5A on the heater current.  Plate current on those KT66s are 430vDC, Bias set to around 39...

Specs are solid for that PT and your application...
Probably just heat being transferred into the PT by the location of the output and rect. tubes being so close.

[Ugly Distortion]

You know how 5F6As/Marshalls can squeal with everything up and you twiddle the presence btwn 8-10? I patiently tried everything then lastly dug out my dusty box of long ignored tube shields and there it was. Just something to keep in mind. Ken Fisher thought the shields hurt tone, apparently they do have an effect on high frequencies so he might be right.

There's been web-talk of Classictone PTs running hot. PTs in orig Fenders, at least the 6V6 ones, ran crazy hot - and for like 60 years - and still going strong. A lot of it is just the upside down chassis design. If the Classictones are built well they should be ok.

[SILVERGUN]

Mike, I'm curious...
How did you wind up wiring the 4/8 ohm OT outputs of that CT#40-18008?...I can't tell from the thread pics.
I'm assuming that because the green and black wires were twisted that you used those for a 4ohm speaker out and used the yellow (8ohm) tap for NFB?
 
Is my assumption correct?

[p2pAmps]

Mike, I'm curious...
How did you wind up wiring the 4/8 ohm OT outputs of that CT#40-18008?...I can't tell from the thread pics.
 
Which one did you use for the NFB loop?

I used the 8 ohm tap only and just shrink wrapped the 4 off.  Running 2 16 ohm speakers...

[SILVERGUN]

I'm stuck on the NFB 'ratio', but honestly don't know the math enough to help you fix it.
I do know that the voltage will be greater when using the 8ohm tap instead of the 4.
 
You might actually benefit from decreasing NFB voltage, but I'll need more time to wrap my remaining brain cells around it.
Maybe one of our cranially blessed friends will chime in and straighten me out.
 
Until then I personally would experiment with 820ohm series feedback resistor value to try to dial it in, and dial the noise out.
You could also switch the tap to the 4ohm wire and see if that helps. (if that's an easier experiment for you)
 
Here's some reading on the subject if you're interested:
http://www.aikenamps.com/index.php/designing-for-global-negative-feedback
 
 
 

[p2pAmps]

I'm assuming that because the green and black wires were twisted that you used those for a 4ohm speaker out and used the yellow (8ohm) tap for NFB?
 
Is my assumption correct?

The green wire on my tranny is 8Ω, the yellow is 4Ω.  I have an 8Ω load on the amp so impedance is not an issue at all.  The NFB is coming off the 820Ω resistor like most if not all AB763 circuits. 
I have built quite a few of these so I am pretty confident I am good on that respect.  However, I may mess with the 47Ω resistor in the NFB and stick a 100Ω to see the difference.  To my knowledge 100Ω is
normally used on a 4Ω load and 47Ω on 8Ω loads.  That is another thread...

My issue is a small high pitch sound on just a few notes that I play.  Almost sounds like a bad preamp tube but it's not.  I believe what I am dealing with here is Parasitic Oscillation.  My grid stoppers on the KT66s were 1.5K which would
be fine for 6v6 but since I'm running the KT66s I changed them to 5.6K as a starting place.  This whole amp is sort of an experiment for me in some respects.  I still need to test tonight.  My next move will be adding a snubber cab
to the PI plates starting with a 10pF and testing along the way.  I don't want to disturb the tone at all as it is amazing the way it sits, just need to lose that high pitch tone I hear sometimes.  Another thing is when I first fired the amp
I thought it was way too bright so I cut the 47pF off the Vol pot and that tamed her down quite a bit.  This is trial and error for me at this point for sure...


Mike

[SILVERGUN]

I'm assuming that because the green and black wires were twisted that you used those for a 4ohm speaker out and used the yellow (8ohm) tap for NFB?
 
Is my assumption correct?

The green wire on my tranny is 8Ω, the yellow is 4Ω. 

Mike, I was going off of this pdf from CT:
http://www.classictone.net/40-18008.pdf

Either way, it's not a bad idea to try to dial it in.
The noise you're hearing 'could be' a result of NFB but I'll be curious to hear the results of your snubbing cap experiments.

Keep up the trial and error...we've all done it and are better off for the lessons we learned.

[p2pAmps]

Well I fired the DR and cannot make it squeal now.  All I changed was the 1.5K grid stoppers on the KTs to 5.6K so far.  I played it a good 20 minutes and nothing no matter what.  I'll be keeping a close eye on this thing...

[Paul1453]

I'm always interested in the little tweaks sometimes needed to dial in the sound.   

Please let us know if this did it, or further tweaking is required. 

[tubenit]

I fired the DR and cannot make it squeal now.  All I changed was the 1.5K grid stoppers on the KTs to 5.6K so far.

   Thanks for letting us know!  Glad you have it resolved.

with respect, Tubenit