Failed Experiment

[Plexi]

Maybe failed in the outcome, but you also learned in the process.
Its great to explore and try things and see if they work or not, at least i like doing it...sometimes you never know till you give it a try.. I had a older 2203 in here.. it had been abused for most of its life. It had some oscilations in it,and i tried many of the tricks i knew to cure it. It was like i wasn't going to let the thing beat me..lol Sometimes you just have to tell the amp to give up, cause your going to be fixed.. Finally after pulling the board and working many hours,on what should have been an easy fix.. I fixed it. it turned out to be one of the quietest 2203 i've ever heard. Even the one i own,and babied since the 70s.. What i was getting at was.. even things you know should work..sometimes don't. But in the process,i learned a few more tricks to cure the problem for good..
Richie

[markh]

You know, Paul, I think it's really great how you spent all that time chasing that problem down, and finding an elegant engineering solution.

However, even though in my mind I understand how it works, and why it makes it sound better, something makes me really not want to put SS diodes (zeners or not) in my signal path.

I guess superstition is still alive in well in amp land.

--mark

[zaphod_phil]

I wonder if Jason's "PI Mods" help deal with that distortion in another way, by the use of different value resistors at the EL84 grids, as well as in the PI itself. Alright it also changes the sound somewhat, although Jason and others will argue that it's actually improved

Also, Mark, it looks l ike your Lite II layout shows those diodes added, even if you hate the idea - right??

ZP

However, even though in my mind I understand how it works, and why it makes it sound better, something makes me really not want to put SS diodes (zeners or not) in my signal path.

I guess superstition is still alive in well in amp land.

--mark

[zaphod_phil]

Hmmm, yet another case perhaps, where guitar amp design defies conventional EE wisdom! I remember from way back that we ere advised to bypass electrolytics with smaller caps to handle the higher frequencies which the electrolytics weren't good at. Thinking about it I wonder if the small caps were setting up some kind of a resonance effect with the series parasitic inductances in your eloctrolytics, which would causing ringining at some of the higher frequencies. Just thinking aloud here....

On my latest 18W build, I tried an experiment. I added 0.1uf mallory 150 caps in parrallel to all the electrolytic caps. Every last one. And, I added an individual 0.1uf cap from the B+ to ground at each plate resistor node to decouple each stage independently. I figured it couldn't hurt and maybe would help. Well... NOT!! The amp was harsh and fizzy when pushed hard. Clean tone was not noticably different but break up was terrible. Disconnecting all those caps brought the amp back to it's normal butt-kicking self.

[Plexi]

I would like to say.. the board pic you posted the other day,i liked that layout. With the added turets for the grounds along the back. Does away with using the buss bar thing.. makes it look neat and tidy.. and easy hookups. Similar to some hiwatt,sound city and some of hoffman boards use that .. layout in some amps can make or break having problems,nice work Paul.
Richie

[Moose]

The exception to this is that I always use shielded lines from the input jacks to the first tube sockets. That's a bit more work but is always a net benefit for noise immunity.

I find this an interesting point. I have NEVER had a problem with noise that was cured by using shielded wiring from the input jack to the socket, yet every amp I build anymore ends up with a shielded run from the jack to the socket. I guess it's just belt and suspenders -- the stopper resistor AND the shield both work to combat possible problems.

As for the final build, it's very nicely done. You're good at getting that Si wire to stay in place with the zip ties. I need to learn how to do that better. I switched back to cloth covered, solid core because it stays in place better, but it's kind of a pain in other ways.

[Dai]

I agree it's a belt and suspenders thing in many amps, but isn't much cost for insurance. This became a habit for me in building higher gain amps, especially those with solid state recto spitting out hash noise all over the place. Of course, I've also learned how to get rid of the SS recto noise at teh source to. So, belt, suspenders and safety pins.

Paul, would you care to share any tricks on reducing hash noise (I guess that would be the switching noise radiating out from solid state diodes?)? I'm thinking things like twisting AC wires for the B+, putting the run out of the diodes flat against ground or side by side with the B+ center tap wire, maybe caps across the diodes--things like that. I was wondering how meaningful it would be to do things like add an AC line filter, or electrostatically shield the AC primary and secondary (high voltage for B+) wiring with flexible copper braided tubing. Grounding seems to effect the background buzzing some. Have you found anything particularly effective?

(Nice work on the board and grounding/layout idea BTW!)

Dai

[markh]

I've found using the newer fast switching diodes (e.g., HF4007) helps a bit too with the switching noise.

--mark

[Plexi]

I think i posted a reply ,but it never showed up..

""""OK, one more bit of blue mojo""""

Still say to top it off you need the new blue multi can cap.. blue tie wires,or use the blue tape wrap or heat shrink... If they only made those EL84s in the blue glass,like some of the EL34s were..you used to be able to get them in blue or red glass.. And GT used to use blue labeling on there EL84s.. I guess you'll have to build a brown amp for the brown sound...
Don't even want to think how you'd build one for the "woman tone" I don't think RS sells those parts off the wall..
Richie

[Dai]

No magic to it... Add between 10 and 100 ohms in series with the diodes and a .001 or .01uf 500V ceramic cap in parallel with each diode.

Thanks Paul and Mark. I've already got 2200pf ceramics across the diodes (in a 50W). I will go ahead and try the resistors.

Dai

[Shea]

My goal is to minimize wiring (the hardest part) and do away with "flying" componets. I really don't like components jumping from the board to pots or tubes . . .

That's exactly what I was thinking when I did my latest board layouts. I even put my bright cap on the board. Unfortunately, I couldn't do a practical layout for a bax tonestack without using a couple of flying components and components mounted on the pots.

Originally, I even had the 1-meg resistor that is normally on the input jack mounted on the board, and the V1 grid stopper mounted on the board too. Since then, I've begrudgingly changed back to the old fashioned way to reduce noise. The 1-meg went back to the jack, I use shielded cable for the run from the jack to the tube, and the grid stopper is mounted right at this little terminal that they put on the center of my tube sockets (I still don't know what it's there for).

Shea