champ clone squeal

[tubby]

Hi All,

i recently finished a slightly modified champ clone and i'm very happy with it.

the one problem that remains is squealing.
if i turn the volume past 3 o'clock, a common problem, i guess. so i changed the lead dress, poked around with a wooden stick, replaced the pre-tube, etc. but still....
if i play guitar through the amp, there's ugly distortion at the point where the squealing starts, and with the pot maxed out the amp cuts out totally. also the voltages on the pre tube (ecc83) jump up at that point from around 180 to 240 and 200 volts on the respective triodes. to me it seems like oscillation but i don't have much experience (yet)...

is there any sense in changing the ot's polarity, since i've incorporated switchable nfb? (it squeals even without nfb)

otherwise i'll just use a 5751 and see if that helps or reduce the gain


and while we're at it, a few other questions that came to my mind while i was building the amp:

(1) i've read that an artificial center tap on the heaters works better than the grounded CT on the PT? what are your experiences?

(2) what's the advantage of taking B+ off the CT instead of pin 8 of the recto tube?

(3) what's a good value for a SE output stage grid stopper that is designed to work with el34, 6l6 and KT88 & 90 in maxed out overdrive conditions? yes, i've read the aiken article, but i'm hoping on responses based on your experiences..

thanks!

[tubby]

a little update:

today i reversed the OT's polarity. i slowly turned up the amp, then all of a sudden a LOUD squeal.
quickly turned off the amp. after that the amp was dead. it turned out i had killed the preamp tube somehow, cause after i had replaced it, it was back to normal.

i wonder how that happened, since the feedback loop was switched off. i'm also surprised that you can actually smoke a preamp tube that way....

after that i tried the following:
- reducing the gain of the first triode by raising the cathode R to 2,7k
- adding a grid stopper to the 2nd triode
- reducing the gain of the second triode by raising the cathode R to 2,7k (this helped a bit, but on last few millimeters of the pot
there's still that fuzzy, terrible sounding distortion). i reversed everything to the original state (1,5k and no grid stopper).

voltages on the 12ax7 are:
pin1 = 187 V
pin 3 = 1,45 V
pin 6 = 187 V
pin 8 = 1,43 V

this seems ok to me.

both 68k grid stoppers are soldered directly to pin 2 of the input tube instead of the input jacks as it is seen in many amps.
could this be the culprit?

i kind of ran out of ideas, so i'd be grateful for any suggestions before i rewire the whole thing.

[tubeswell]

Hi tubby

Can you post some clear pics of the guts of it?

[tubby]

ok, here's a guts pic...

i guess after the guts pics i've looked at now, it's rather for your amusement.
i probably can't expect the amp not to oscillate, since the wire dress is kinda messy indeed. however, it's my first build and it will only get better from here on....

some explanations:
- the right switch switches between the 22µF and .68µF cathode bypass cap on the first triode
- the left switch is NFB from the 8 ohm tap on / off
- the massive pot near the output tube is a 1K / 30W pot for adjustable bias; to the immediate left is the bias test point
- there's nothing special to the colors of the wires, except that yellow/green is always ground
- right vol. pot is pre, left is master

[MWJB]

That's not a champ "clone", not by a long chalk...compare your layout with that of a champ. You've somewhat reinvented the wheel there and it's come out kind of square.

You don't need an output stage grid stopper, nor do you need to rewire the heaters with a virtual CT if you are already using the PTs heater CT. What you DO need to do is basically rewire the amp's internals...might sound harsh but it IS what you need to do. Grid stoppers & other band-aid solutions are not curing your problem, just masking the syptoms. Tackle them at the root cause.

Observe the physical layout of the 5F1 circuit, ground PT B+ & heater CTs, main & screen filters & power tube cathode resistor & cap to the same PT bolt ground (grind away any paint for a good connection).

Ground preamp filter, preamp grounds pots & input jacks to a poiint near to #1 input jack.

Don't tie the grounds together & run them to the ground point, use seperate wires from each component to the prescribed ground point.

Remove the 30W pot, it's a waste of time, with 310-0-310VAC at the B+ secondaries just use a 390ohm cathode resistor with 5Y3, 470ohm if you want to play with GZ34 & SS rectifiers (5W rating is plenty for a 6V6).

Move the speaker jack from its current location and locate it where the 30W pot currently lives. Ideally your OT should not be located at the input jack end of the amp...you might get away with it?

What are you actually measuring at the bias test point?

How does the master work?

You should easily be able to lay out the circuit on just one of those 18-way tag strips...filter caps & 6V6 cathode resistor & cap could ground straight to the chassis, no need to mount them on boards. By cramming everything on to the 2 boards you have screwed the layout.

NFB switch should be located by speaker jack, between preamp tube & power tube.

That OT is twice the rating that you would need for a 6L6 tube, let alone a 6V6. How are you ruinning it? 4.5K into 8ohms is a bit low for 6V6...better match for 6L6, or a compromise between the two. 250mA rating on the PT B+ secondary is overkill (more than most 50W amps)...100mA would be more than enough.

[tubby]

Hey MWJB, thanks for the input!

Ok, champ 'clone' is putting it wrong, you're absolutely right.
Let's put it this way: a simple, yet high powered SE circuit that can be fitted with a variety of power tubes, e.g. EL34, 6L6 and KTs 77-90.

I'm working on rewiring the board and incorporating the grounding tips you gave me. Thanks again.

There's a few things I'm not getting yet:

- Why is there no need for an output stage grid stopper at full blast?
The tech article on grid stoppers at aikenamps made me think. Please explain.

- I wired the virtual CT cause it supposedly has the advantage of providing a perfect center reference (if close tolerance resistors are used), whereas the center tap of the PT is off the middle point resistance-wise.
(see also: http://www.freewebs.co.uk/valvewizard/heater.html)

- About the 30W pot: it's probably overkill, but since I want to be able to run different tubes (see above), I guess I need some kind of pot. They only had 5W or 30W at the store, that's why I'm using this one.

- Quote: "Move the speaker jack from its current location and locate it where the 30W pot currently lives."
What's wrong with the jack where it is? I thought one design philosophy is keeping wires as short as possible. Could you elaborate pls?

- Quote: "Ideally your OT should not be located at the input jack end of the amp...you might get away with it?" Isn't it worse having the PT on the input jack end of the amp? Maybe a middle position would be best?

- The master works as a voltage divider to ground after the coupling cap (.02) on pin 6 of the second triode before the ground reference resistor (220K) and the grid stopper (2K2) on pin 5 of the output tube.

- At the bias test point I'm measuring the cathode current via the 1 ohm resistor method. I'm aware I'll need further test points for plate and cathode voltages respectively to calculate exact plate dissipation. At the moment it's just to make sure I don't run the tubes too hot.

- The OT's primary of 4.5K is a compromise between the above mentioned power tube types. The OT has 4, 8 and 16 ohms secondaries, at the moment only the 4 ohm tap is used for the speaker jack. I based the PT's mA rating on the max idle cathode current of a KT88 which is 235mA. The higher rating wouldn't affect tone or am I mistaken?

[MWJB]

"- Why is there no need for an output stage grid stopper at full blast?
The tech article on grid stoppers at aikenamps made me think. Please explain." - it's a simple circuit, good layout should negate the need for a grid stopper...there are other champ style amps that run the tubes you mention, at higher voltages, that do not need grid stoppers.

"About the 30W pot: it's probably overkill, but since I want to be able to run different tubes (see above), I guess I need some kind of pot. They only had 5W or 30W at the store, that's why I'm using this one." If you want to rebias every time you change tubes why not go adjustable fixed bias? You just need an additional 100K 2W resistor, 25uf 100v cap, 1N4007 diode and 50K trim pot. What most folks would do is bias to 90-100% plate dissipation for the lowest wattage tube that you intend to use (25W) and leave as is...alternatively you could switch between 2 different cathode resistors like 470ohm for KT66/6L6/EL34 and a smaller value for the big tubes? As you draw more current through the power tube, plate voltage will drop and gains in power output will be negligible...you might well enjoy the differences in tone & dynamic response however.

"Quote: "Move the speaker jack from its current location and locate it where the 30W pot currently lives."
What's wrong with the jack where it is? I thought one design philosophy is keeping wires as short as possible. Could you elaborate pls?" Which wires are as short as possible? Certainly not your NFB wire, nor your OT primaries & secondaries? You have the main B+ & NFB wire running right up to the preamp end of the amp. It would make more sense to have a short, direct run on these wires.

"Quote: "Ideally your OT should not be located at the input jack end of the amp...you might get away with it?" Isn't it worse having the PT on the input jack end of the amp? Maybe a middle position would be best?" - PT should be at the rectifier end of the amp, OT should be as far from the input jack as practicality allows, yes a central mounting should be good...redo the layout and see how things are after that if you still need to move the OT? There's no need to have ANY transformer at the input jack end of the amp.

"The master works as a voltage divider to ground after the coupling cap (.02) on pin 6 of the second triode before the ground reference resistor (220K) and the grid stopper (2K2) on pin 5 of the output tube." OK but it gives you an unnecessarily long wire from the end of the coupling cap...great for picking up oscillations.

"At the bias test point I'm measuring the cathode current via the 1 ohm resistor method. I'm aware I'll need further test points for plate and cathode voltages respectively to calculate exact plate dissipation. At the moment it's just to make sure I don't run the tubes too hot." You don't really need to calculate "exact" plate disipation with cathode bias. If you switch to fixed bias you can assertain plate voltage with the tube with the least current draw (6L6/5881 whilst at a reasonable level, 40mA min) then you know as you fit bigger tubes/more current that plate voltage will drop...you should still get into good tone territory before risking burning anything up.

"I based the PT's mA rating on the max idle cathode current of a KT88 which is 235mA. It wouldn't affect tone or am I mistaken?" - Let's see, your PT choice means that your max B+ voltage is unlikely to exceed 400vdc, less cathode voltage, with a reasonable currnt draw from the power tube...we'll guestimate 350-360vdc-ish assuming reasonable current draw. 40W for a KT88 * 90% plate dissipation, = 36W divided by 360v = 100mA. Even if you realised a full 400vdc at the plate in fixed bias...that would be 100mA @ 40W dissipation. Even with a low voltage rectifier and say 300v on the plate you're not going to idle at anything like 235mA (70W dissipation...keep a fire extinguisher handy!).

[tubby]

thanks for explaining MWJB!
i never thought about adjustable fixed bias, maybe I'll use it in my next build.