Help with new build. Loud squealing noise. High Gain SE amp

[Shitak1]

Hello! My name is Vitor and this is my fist time posting, although I've been a long time reader. Sorry if I make mistakes in formatting the post and, if this is in the wrong place, please delete the post.

I'm in dire need of help in my Single Ended with a high gain switch. Squealing appears in the high gain switch only when the Master Volume reaches halfway. Treble and middle controls affect the squeal and the capacitor bypass switch in the second stage gain affects it too.

Now some basic information: It uses 2 12ax7 preamp tubes (Plate Voltages > V1a = 210V; V1b = 198V; V2a = 280V; V2b = 192V) and 1 6L6WGC/5881 as a single ended output tube (Plate voltage = 400V; Screen Voltage = 397V).

Things that I tried:
1. Changing preamp tubes, output tubes, speakers, electrical sockets around the house and guitars - No affect on the squeal.
2. "Chopsticking" the amp - No obvious component or wire affected the squeal.
3. Reversing O.T. leads - It made squeal even worse, so I reversed right back.
4. Adding Grid Stopper in the later stages - Helped a little, but the squeal persists.
5. Shielded Cable in all input grids - Helped a little too, but the squeal persists.
6. Tried to re-arrange the lead dress - No change (moved the input grids to the top and the plate wires to the underside of the board)

I attached the layout and schematic below. Thanks for the attention and for the help! Cheers from Brazil.

[shooter]

disconnect the NFB, better?


overdriving an SE amp at the PA tube will get you squeal!  my guess is you're slamming the PA tube hard enough to output squarewaves which tend not to be very musical.


got a scope?

[Shitak1]

I forgot to add that, disconnecting the NFB improves a lot, but the overdrive becomes very harsh.

Sadly I don't have a oscilloscope.

PA tube you mean the 6l6wgc/5881 output tube? If it is, I will lowering the Master Volume pot to maybe 100k log, but I fear that the overall volume will drop a lot.

[Latole]

PA tube ???
NFB go to Phase Inverter tube. ( PI )

[SEL49]

Four gain stages is a lot! Lead dress and layout are critical. Your OT is mounted right over the sensitive preamp and tone control circuit. It would likely be beneficial to move the OT so it's near the power tube.

[bmccowan]

Latole, it's single ended.
Agree with SEL49 - that's a lot of gain.
How does it sound without the high gain switch engaged? How does it sound just short of the squeal?
When it squeals, is the initial gain pot maxed? In other words, what is the interplay between the two volume pots?
Unless you find a mistake, I'd consider cutting back on the gain boost until you reach a comfortable level.
Would be easy to try it with only one half of V2 in the circuit.

[shooter]

a quick n dirty;
change the bypass cap on V1a to something like 5-10uF
change V2 from an AX to AU7

[Willabe]

Four gain stages is a lot! Lead dress and layout are critical. Your OT is mounted right over the sensitive preamp and tone control circuit. It would likely be beneficial to move the OT so it's near the power tube.

He has the output jack right next to the 1st preamp tube.

[Willabe]

I forgot to add that, disconnecting the NFB improves a lot, but the overdrive becomes very harsh.

This is not clear.

Disconnecting the -FB only lessens the squeal but it still squeals?

Or disconnecting the -FB stops the squeal but now the distortion is harsh?

What happens with only 1 preamp tube and the -FB disconnected? Still squeal?

I would think that with 4 gain stages you turn it up, it's still going to squeal with or with out the -FB hooked up. 

[Willabe]

The way he has the extra preamp tube, when it's in the circuit, it's also in the -FB loop. So, the power tube, V2 and V1b are all in the loop.

That could be a problem.

I'd change that so only the last preamp triode was in the loop with the power tube.

[shooter]

That could be a problem.

that's why I ask him to remove it,
but yep for designers that are lurk'n NFB in an SE amp should ONLY be used IF you want a nice clean amp, if you want hair on an SE amp, don't bother designing it in

[Shitak1]

Sorry for the delay in the response and first of all I really appreciate all the feedback and help.

Now for clarifying some questions and what I plan to do based on the suggestions.

Four gain stages is a lot! Lead dress and layout are critical. Your OT is mounted right over the sensitive preamp and tone control circuit. It would likely be beneficial to move the OT so it's near the power tube.

This chassis is from my very first tube build, an ax84 Hi Octane and from a time I knew very little tube amplifiers. And I not gonna lie... I knew that is nice to make short runs on the primary and secondary leads of the Output Transformer but I was lazy when I reused the chassis and now I'm gonna pay for that with more work. The OT will be put in a better place.

Latole, it's single ended.
Agree with SEL49 - that's a lot of gain.
How does it sound without the high gain switch engaged? How does it sound just short of the squeal?
When it squeals, is the initial gain pot maxed? In other words, what is the interplay between the two volume pots?
Unless you find a mistake, I'd consider cutting back on the gain boost until you reach a comfortable level.
Would be easy to try it with only one half of V2 in the circuit.

It sounds very very nice without the high gain engaged, if you look closely at the schematic the first two stages is just an Fender AA764. About the relationship of the volume and gain, basically with the gain and treble pot maxed and high gain switch on, I get the squeal about halfway on the master volume pot. If I max the master volume and treble pot, I get the squeal in the gain pot about 75% turned up. I will try to reduce the gain and build up from a stable amplifiers first. About the sound before the squeal appears, if I back up about 10 to 20% of the gain or master volume pot, the overdrive that the preamp produces is actually a very nice sounding one (basically I made to sound like a JCM800). The nice cleans (with the growl when you max the gain e master volume without the gain switch) and the great overdrive sounds are the main reasons that I want this amp to work 100%.

a quick n dirty;
change the bypass cap on V1a to something like 5-10uF
change V2 from an AX to AU7

Will try that too, thanks!

I forgot to add that, disconnecting the NFB improves a lot, but the overdrive becomes very harsh.

This is not clear.

Disconnecting the -FB only lessens the squeal but it still squeals?

Or disconnecting the -FB stops the squeal but now the distortion is harsh?

What happens with only 1 preamp tube and the -FB disconnected? Still squeal?

I would think that with 4 gain stages you turn it up, it's still going to squeal with or with out the -FB hooked up. 

Disconnecting the Negative Feedback lessens the squeal, but it still squeal when everything is maxed out and the distortion is very harsh. With only 1 preamp tube and gain switch off the amp works just fine, even maxed out and with the gain switch on there is no output since it cuts the circuit without a second preamp tube.

The way he has the extra preamp tube, when it's in the circuit, it's also in the -FB loop. So, the power tube, V2 and V1b are all in the loop.

That could be a problem.

I'd change that so only the last preamp triode was in the loop with the power tube.

Will try that too, thanks. Actually this was on my mind when I was making the schematic, where should I put the NFB circuit. Now I'm gonna try that too and without Negative Feedback also.

[bmccowan]

I will try to reduce the gain and build up from a stable amplifiers first. About the sound before the squeal appears, if I back up about 10 to 20% of the gain or master volume pot, the overdrive that the preamp produces is actually a very nice sounding one (basically I made to sound like a JCM800).

That's a great sign. The method(s) you choose to back off that gain will each have their own sonic character. Changing one or both preamp tubes, modding voltage dividers, using only one half of V2, wiring V2 parallel, etc. are all worth monkeying with IMO.
Have fun.
And...Shooter and Willabe have a lot more knowledge on high gain amps than I do. Their ideas on taming your beast will be well grounded (unintended pun.) You also may want to take a look at dwinstonwood's current and recent projects. He likes pushing the gain up while keeping things under control.

[Merlin]

2. "Chopsticking" the amp - No obvious component or wire affected the squeal.

Your power tube anode wire (blue) appears to be very close to the yellow grid wire of the 2nd preamp tube. Push those preamp wires away from the blue wire.

You could also try shileding the anode wire; a piece of kitchen foil taped down to the chassis would work, just as an experiment. I can't see where the blue wire goes, but a permanent solution may be to re-route it outside the chassis.

Is your AC socket wired correctly? The earth terminal is usually in the middle but yours is not, but maybe they do things differently in Brazil.

Where is your circuit ground connected to chassis?

[Shitak1]

Hi Merlin, thanks for the feedback!

Well, actually the hole besides the output tube socket is there because I tried re-routing the anode wire outside the chassis haha. Sadly it did not help at all. Right now I'm moving the output transformer and output jack to a better place and then I going to start testing and making changes one by one to see what effects what.

For the AC socket, it's wired correctly just that the photo make it hard to see, but the ground pin is indeed in the middle. And for the ground connection they are connected via the grounded tag strip where the heaters are connected.

[Shitak1]

So, a little update about the amp. The squeal is gone! and the sound is great too!

Now to the actual modifications. First of all, what removed the squeal was moving the output transformer to a better place (making the run of the primary and secondary wires very short) and removing the Negative Feedback from the amp. Shooter is absolutely right, the nfb has no place in a high gain single ended design for guitar it causes more trouble than it is worth (I tried the nfb in all preamp cathodes to see the effects with the gain switch on, with a lot of different values and even made little resistors dividers to test that too and in all cases it causes some kind of problem or the nfb is so low that might as well be removed since it makes no difference to my ears).

And now with a stable amp at least, I needed to tone shape the overdrive again. So, as bmccowan said I started changing the voltage dividers and treble peakers caps first and eventually the cathodes resistors and capacitors. In the end, I removed the capacitor from the second gain stage cathode and changed the fourth gain stage cathode resistor to 1.5k. Funny enough the voltage dividers and treble peaker caps stayed the same. In the end, sadly, after all this testing the overdrive was still kind of bad (to my ears at least) so I through that all this was actually just a waste of time.

Now the kinda funny part is, now that the amp has less gain I thought it could use a bright cap in the first pre gain potentiometer, even through I never liked the bright cap in any build that I made before. And, to my biggest surprise a 1nf bright cap completely changed the sound of the overdrive, from a kinda dull, flat and harsh overdrive it became a very lead like and articulated overdrive sound and I loved it. A 4.7nf capacitor made it sound more full sounding (obvious a bigger cap literally let pass more low frequencies, but holy hell it changes the overdrive so much). Sadly, with the gain switch off I have a little too much treble but I can live with that easily.

Never thought about using a bright cap to shape the overdrive of a amp. The most common thing that I think first is coupling caps, NFB, Local NFB, cathodes resistor and tone stack, so that was fun to learn.

Thanks you all for the help! Now I'm very happy with the amp.

[kagliostro]

If you want to experiment to achieve the best on your amp here there is an NFB you surely didn't gived a try

At DIYItalia we are planning an SE amp and we have already give a try to this NFB circuit, the plan of our amp is not yet ultimated but this NFB was tried on BreadBoar with very good results

Franco

[DummyLoad]

Your NFB insertion is wrong. It wants to squeal at around 1.5KHz. If you want to re-insert, lose the 100nF bypass cap wrapped around the 820R. Lead dress looks like it's probably an issue as well.


--Pete.