I dumped the Cathode Follower. Check out the new design.

[leadfootdriver]

I'm totally digging the Parallel Triode driven tone stack. It's so much more dynamic, with a strong frequency response. The Cathode Follower I was running turned out to be everything I hated about my amp.

http://home.comcast.net/~leadfootdri...rsion%20II.JPG

Here's a hacky blues clip. It would be nice if I used a Fender.

http://home.comcast.net/~leadfootdriver/ZZ3.mp3

[Chuck H]

Looks like a lot of gain (not a bad thing). Sounds like a fun amp to play through.

The only thing about the schem that looks odd is that the signal from V2 is decoupled and then sent to the TS (which serves to decouple already). I might try using 33k and 15k resistors in series for the plate load and use the junction to feed the TS directly (split plate load). Anything you can do to decrease series resistance will reduce hiss with high gain. I suppose smaller series resistance and loads on V1 could also be done. Like a 500k vol pot and 100k series off V1A and a 220k load with a 22k series off V1B.

The tone is really good. Interesting and different. A little skwanky on higher lead notes. Maybe a small shunt filter across the OT primary (like a 6.8k/5W in series with a 2200pf 1kV cap). Easy to install and easy to remove if you don't like it.

But it's really cool just like it is. Nothing at all "wrong". These are just places where "I" would experiment.

[leadfootdriver]

Hmm. The 3rd 22n coupling cap was there because I was experimenting with 4 gain stages, but bailed on that because I just didn't need all' that gain! I left it on the board not even realizing it. I just jumpered that cap and it made no change in sound. So I guess I can dump that one eh?

I went with small bypass values on the 1st 2 triodes because that's all' I had. I'd like to try some higher values in there like maybe a .22uf, and a .47uf on the 1st 2 triodes to maybe round out the upper midrange a bit. What do you think?

SO with splitting the plate load of v2 with a 33 and 15k, do I get the same amount of current drive thru the tonestack as I would with just a single 47k? I'm all for reducing hiss!

[Chuck H]

In all honesty I'm not sure how much reducing the series resistance on the V1 stages will reduce hiss, Since the series resistance is AFTER the load there's not much current through them, so probably not much difference in thermal noise reduction. Maybe worth a try though.

To find the split load value for V2 you could temporarily use a 50k pot for the plate load and feed the TS from the wiper. Dial in the amount of gain you want and then remove the pot (without turning it) and measure each end to the wiper for your values. This way you know you'll have the same drive level. My values above were just a guess. You should actually increase available current a tad due to the lower impedance of the split load.

The knee frequency of cathode bypass caps if relative to the cathode resistor value. For your values I think you could try 1uf on V1A and 2.2uf on V1B for more mids. You could also try a 10uf for the bypass on V2 to open up the frequencies prior to the TS, but you may end up back at 1uf if the amp gets muddy.

[leadfootdriver]

I'll need to stock up on parts to try this! I'm way late to the party on this stuff.

I tried a 22uf on the v2 cathodes, and it oscillated a lot; even with a 10uf! When I went into it, I had the idea of running a 47uf in that spot, since I read that you need to double the bypass values for a parallel triode to achieve the same gain as a normal triode. But it didn't work out that way. Maybe it's my board layout.

But all is not lost as I've achieved a better quality and level of bass with much smaller bypass values since I tried the PT driven TS.

This pic is my board right now. I thought it was a good idea when I set out to do this first build, was to connect plate resistors and coupling caps together in series, and then run wires underneath to save space. But I really don't know anything about how components interact together in proximity. You can see the orange drop 22n's in series with the plate resistors. What do you think about this?

I originally had the board setup to do a bootstrapped cathode follower, but that oscillated too! That's when I tried the regular CF.

BTW, thanks for your input!

[pdf64]

'I tried a 22uf on the v2 cathodes, and it oscillated a lot'
I think that may be due to B+ supply de-coupling for V1 / the first stage - there doesn't seem to be any!
Another 10uF to 22uF cap to ground, after the 1k 2watt dropper resistor, should help.
Whatever you decide with the cathode bypass caps, supply decoupling is necessary, otherwise there will be unpredicatable (positive and/ or negative) feedback paths between the stages, which will cause weird things to happen.
Also, the 470k grid stopper on the el84 is a very high value; it may be rolling off high end (I'm not sure how the Miller effect works on pentodes) and more importantly may result in the 1M max grid leak resistance being reached or exceeded. Pete.

[Chuck H]

good catch on the decoupling cap and grid giant stopper. These are definitely two things to address. I missed it somehow.

[leadfootdriver]

Okay fellas, here's the revision. All I can say is wow! The extra filtering did wonders for the low end. I did my best to keep it tight, but I was still able to up the bypassing values on v1a-b. It was better to go smaller on v1b to keep the attack sharp. I tried a 2-2.2uf on v1a, but it was to bloated for the little EL84. I need to get equivalent values now, because I have a bunch of candy in parallel on my board!

I shot down the helicopter by bypassing the anode of v1a with a 100p. I kept the 470k grid r on the EL84 because I'm getting plenty of high end.


This thing shreds now, but I'll prolly tear into it again in a day! Who knows?...

http://home.comcast.net/~leadfootdri...%20Revised.JPG

[riz]

470K grid stopper isn't high for EL84. IMO. That's the value that did the trick for the last EL84 amp i worked on, and if IRC, Merlin says something along that line as well. Absolutely right about the added filter. Pdf, maybe it was you who suggested that to me a while back, on a project of mine... It was a screen supply node, but it wasn't there. Big help!

[Chuck H]

I'd be surprised if Merlin said to float an el84 grid 1.5M above 0V !?! IIRC the max is 470k, so a 220k grid stopper and a 250k master volume pot would do it. The extra gain from removing the 220k series resistor feeding the tone stack should be plenty to compensate for the smaller load on the el84 grid.

Also, the extra filtering suggested wasn't bigger filters. It was another can added to the end of the rail at the top of V1A plate load.

But I digress. If it's working for you then that's the most important thing. Don't be surprised if you get a runnaway el84 at some point though.

[riz]

I'll have a look at my work, Chuck. I'm way open to criticism, so don't digress on my account! I want to say Merlin said even up to 1Meg, but I probably shouldn't post such things without double checking. As to the filtering, yeah, that's what I did, in a way. The screen supply was off on a branch of it's own. Dropper with no cap. Just had a look. Please tell me where I'm wrong...

The Valve Wizard

Wrong or right in theory, it worked out VERY right in my amp in question.

[leadfootdriver]

I went with upping the existing filtering after looking at a Trainwreck Express schematic. Ken had each triode filtered with a 22uf cap! I really like the way he laid out his amp if the Ceriatone amp is at all accurate. I unknowingly gave myself an extra challenge by choosing a tiny amp as a 1st rebuild to stuff all' these components in.

I thought about doing a negative feedback circuit, but I think I'll stick with what I have on this one. I really don't want to drill another hole in the back of the amp for a pot. I'd rather have all' the pots up front, and there's no more room. Although, I could probably get away with doing larger bypassing with a NFB circuit.

I wanna do a fixed bias 18, or 30 watt now!

http://www.webphix.com/schematic%20h...ck_express.pdf

[riz]

I'm by no means an expert, but in my tinkering I haven't been all that jazzed by adjusting the feedback, or making it variable at least... Seems to me amps *generally* sound good more or less as they were intended. There are so many different, great circuits out there, some with feedback, some without. I say tinker with it on the bench so you get to know what you like, then build in the value you like (or none at all!) if you don't want to drill.

[Chuck H]

The write up you posted is a general tutorial of the principals. It applies to preamp triodes without the need for additional consideration. But when a grid stopper is used on a power tube the max grid resistance has to be observed or the tube could fail under certain stresses. This is likely to become more critical as the tube ages. Not to mention that the max is rarely chosen for additional safety reasons. Also, considering that this is an amp that will likely spend most of it's life cranked up and pushing limits, well... I sited 470k above. I was wrong. For self bias the max is 1M (here is a copy of the max parameters on the data sheet)

Plate Voltage ................................. 300 V
Grid No. 2 Voltage ............................ 300 V
Plate Input ................................... 0 W
Plate Dissipation ............................. 12 W
Grid No. 2 Dissipation ........................ 2 W
Cathode Current ............................... 65 mA
Grid No. 1 Circuit Resistance Fixed Bias ........... 0.3M Ω
Grid No. 1 Circuit Resistance Self Bias ............... 1M Ω

The OP's design presents a max grid resistance of 1.5M. I can't say it won't "work" but time will tell if it presents any stability problems.

[riz]

Right on Chuck, I always appreciate your posts! Just to be clear, Merlin does say on that page that power tube stopper values up to 1meg should be fine. I happened to use that value cuz it worked and sounded good. I am going to be thinking about what you're saying here, to see if I get where you're coming from. The more I study, it's actually starting to happen...

Quote from Merlin, copied straight from the link:

"For a good, modern design, all valves should have grid stoppers, even if it a small value of around 10k say. Values up to 1Meg are quite reasonable though, and are a powerful tool in tweaking the overdrive and treble characteristics of the amp. Except for the input stage, the stopper does not need to be connected directly to the valve socket. Stages enclosed within a feedback loop (usually the output stage) are even more suseptible to blocking distortion so using large grid-stoppers will be even more necessary for these."