Problem with this fizzy sound

[Ross Grey]

Silvio,

Yeah, I was making so many changes, I bought Hakko desolderng tool.

If simply removing R38 did the trick, I would have stopped there. I remember some improvement, but obviously I felt a need to keep trying something because I was still having the issue too.

For my recordings, the changes were with the 'gain' control. The master volume was quite low because the mic was on the floor directly in front of the amp and the recorder would distort if i didn't keep the master volume low.

[Silvio55]

Ok, so you say I should try to remove R38 and increasing the Pi resistors (R32 and R31 in my schematics, both are 56k, R28 is a 2K7 resitor from the 3rd gain stage that goes between ground and the cathode resistor and cathode bypass cap, wich I donīt really undestand why is it there!) would fix it?
EDIT: looking at the ampīs board I think the resistor are actually R32 and R33, I know my schematic, the powerpress schem has some errors)

[Ross Grey]

R28 and R31 should be the right resistors to change. V4A is not a third gain stage but half of the phase inverter, although it may have some gain. R32 is the plate resistor for V4B, but I did not change this. My intent was to drop the gain by cutting the cathode current in both PI stages. This should be R28 and R31. Check the schematic at the link below.

I'm using the 'blinker' schematics which seem to match my amp. Original values for R28 is 2k7 and R31 is 56k.

Does this help?

Vox_Night_Train pictures by Blinker_2009 - Photobucket

[Chuck H]

Thing is, that is a concertina PI. The plate and cathode resistors should be of the same value to have a balanced output. So your amp is running quite an imbalanced PI output right now. If your ears say it sounds good to you then it's fine. I don't think it's what most players would like.

If it is a greatly imbalanced PI that killed the lingering crackle then it seems the problem is in the PI and that funny feedback loop. There should be another way to fix it other than imbalancing though.

[Ross Grey]

Thank you Chuck. This helps me a great deal.

Yes, I have 'experimented' my way into a solution that is actually quite pleasing to me and a few of my guitar buds.

In an off-line discussion with Silvio, it became more clear to me that the cathode current of V4B is the main driver for V2 -- I believe that this then is the nature of the 'concertina' PI -- the plate of V4B is driving V1 and the cathode current of V4B is driving V2.

Perhaps only the change to R28 was sufficient to reduce frizzy decay and an unbalancing the PI is not required.

In the darkest hours, I have considered ripping out the entire PI and implementing a more familiar design.

[Silvio55]

I have a question: Does the 3rd gain stage in this amp works like a cold clipping stage? Cause it has two resistor in series (one bypassed and the other donīt, donīt know why), giving a resistance of 4k2 (R25=1k5, R28=2k7) at the cathode, wich I think its pretty cold biased although not like the JCM800 or the TW express with 10k.
Could that asymmetrical clipping be the cause of the infamous fizz?

[Chuck H]

Assymetrical clipping in a preamp stage doesn't usually sound like your clip. It's more fuzzy and smooth. Not fizzy or crackly. And, FWIW and IMHE, it takes about 4.7k cathode resistance to get the cold clipping, assymetry effect audible. Higher resistances, of course, make the effect more obvious and also continue to reduce output correspondingly.

[Ross Grey]

Ok. I got my amp back from another friend who was checking it out. His unsolicited comment was "I couldn't get this thing to sound bad" -- he really liked it, and claimed that it is much louder than his TT.

I like my current configuration so well, I am reluctant to change it. On the other hand I don't want my friends and family to know that I am living with an unbalanced PI.

Here is my next attack: run a pot parallel with R31, incrementally bring the PI back into balance, and see where frizzy decay comes back.

When it does, I'll add a 1Meg grid stopper on V4B (per SickMan), and see if this has the kind of benefit identified by Merlin. If so, I will further balance the PI as long as frizzy decay does not appear.

Actually, I have no problem living with an unbalanced PI as long as it sounds good

[Chuck H]

An imbalanced PI isn't necessarily a bad thing. That's why I said that if you like it, it's right for you. I think there are some harp amps have adjustable assymetry. There's also been some speculation here that assymetry in the output usually improves the clipped tone of an amp. On my personal amps I go through many tubes in the PI socket and choose the one that sounds best. I don't even check for balance. In fact I suspect that the ones that sound best are the most imbalanced. But I've never checked that. Concertina's can do funny things when overdriven or when the power tubes are driven into grid current. It could be that the imbalance is creating some cancellation of the bad sounding atrifacts. I would experiment, but don't discount a happy accident.

[Ross Grey]

Ok, here are some useful results.

When I start bringing down the value of R31 back toward the original 56k spec from my 100k mod, the clean tone gets big and round as the PI gets more balanced (great!) and frizzy decay does not get distinct in my amp until I get to about 64K for R31.

So at least the PI can be much more balanced without frizzy decay.

At the same point, the OD tones start to get brighter, but it is still a pretty smooth transition from clean to OD. The over all effect brings this amp from a versatile practice amp back into small venue territory -- much bolder now.

This weekend I should have time to cut the trace and put the grid stopper on V4B.

Ross

[Silvio55]

That's great Ross! I told you before that I put a 10k gridstopper on the PI, I was wrong (bad memory, my bad!), the 10k was on the 3rd gain stage, I put a 1M gridstopper on the and didn't make any difference.
Did you restored the value of R28 for the testing?

[Chuck H]

An actual discussion with working models. Going where every man (at least the Night Train owners and people wondering about PI balance for concertina's) WANT's to go. This is great for the Night Train. And holds implications for other concertina amps, like the 5e3's and such that some people struggle with. Loving this.

[Silvio55]

Yes Chuck, I think we are really getting something here (finally!!).

[kleuck]

According to Merlin, the cathodyne PI fizziness can be canceled with an unusually big grid stopper.
Explanation : "The much greater problem with the cathodyne occurs when it is itself overdriven. Because it has such a large cathode resistor only a little grid current is required to 'jack up' the cathode voltage. When driven very hard this can cause an inverted copy of the cathode signal to appear at the anode, effectively creating a sort of full-wave rectified or frequency doubled signal at the anode."..."Fortunately, the cure is simple. We add a large grid stopper to the cathodyne, to keep this grid current in check. A value of 100k to 1Meg is usually necessary. Before you worry, this will not affect the treble response though, because the cathodyne only has unity gain! Therefore it's input capacitance is extremely low,"
The Valve Wizard
I've read often that diyers use most of the time 500K to 1M, anything under 500K seems to have little effect.

[Silvio55]

Kleuck, like I said, I have tried that already, a 1M gridstopper on the PI and 100k gridstoppers on the power tubes, and in this case it didnīt killed the fuzzy sound at all.