The Crystal SET

[mikebse2a3]

Thebes is fearless....!!!! A little smoke and fire only encourages him more....!!!!!!!![]

[White_Hawk]

VERY nice work!

[mike stehr]

Time to fire her up. hey what could possibly go wrong?

[Marvel]

Nice, Mike...[:|]

[DU73]

what everyone said. [Y]

i struggle to put batteries the right way in my tv remote let alone build a tube amp from scratch and make it light up.

you rock!!!!

[Video Guy]

Awesome amp! I'm going to build my first amp this summer.

[seti]

Time to fire her up. hey what could possibly go wrong?

Now that is funny!!!!!!!!!

[thebes]

Thanks for all the nice comments guys. It was a learning experience for me and very much helped me to understand how to sorta read a schematic, figure out the logic involved and learn about things about heaters, screens and such.

Mike Stehr, that's a wonderful idea you've given me. A thermonuclear tube amp, now we're talking! I

am just the man to do it.

By the way Mike take a gander at the schematic given in the link and see if you can think of where I might apply a little damping.

[Mike Lindsey]

As usual Marty, you never cease to impress! Nice work, and would love to hear it one of these days!

Mike

[Timmikid]

Beautiful! Now if you would put the electronics in transparant tubes and bags, we have a nice aquarium instead of a "come watch it live" mouse electrocution chamber. And if you put a small door in it you could hatch chicken eggs in there! $&%$ Where are my pills? [:#]

[colterphoto1]

painting is a bit slow, is it?

That is a beautiful piece of machinery- what are you going to have it hooked up to?

M

[thebes]

Painting sucks, and has all winter and seems to be carrying into the Spring.

Actual hours building this were not as many as you would think, but the time in trouble-shooting becasuse of a major omission in the shcematic, plus noise and gain issues ate up tons of time.

Hey Timmie, I've already done the Tubes-O-Amp-Acquarium:

[leok]

Hello Marty,

Nice job on the amp. Interesting design too. Use of the ECC83 in a no feedback design with a cathode follower driver is a differen approach (at least for what I can remember) . Makes me wonder if my tube amps are a bit under driven. A couple of thoughts

Hum will be tough without a ground plane right next to the wires. Some re-dressing of the wires around the ECC83, isolating signal from everything else might help. If your 3rd wire line power safety Gnd connects to the amp signal and power gnd you might cut that and insert a big (10MegOhm or so) bleeder resistor between safety and signal gnds and let the amp get its Gnd from the signal source. Keep the safety Gnd connected to the metal plate (for safety).

On the sound: maybe try disconnecting the 100uF 6L6 cathode caps (disconnect the cathode side not Gnd side). You can always reconnect them. Also, a 6922 instead of the ECC83 might help. Check the pins: they might not all be the same.

Can't think of anything else. Looks like a lot of fun and hours of tinkering. Enjoy!!

Leo

[Dmannnnn]

That is a very impressive project.

Great Job!

[Marvel]

a major omission in the schematic

I realize you added the tube rectification, but what was the major omission? You mean different than what is posted on the website?

Bruce

[thebes]

Thanks for looking it over Leo, I'll definitely give your suggestions a try. Ultimately the blaring may simply be a function of using cheap $40 OPTs which I had on hand and were all I could afford. It sounds better modifying the circuit by using the OPT's screen taps, but the screen voltage actually ends up higher than the plate voltage on the 6L6's which is probably not a good thing.

I've also just found out that they used 6L6GC which is quite a different animal from the 6L6GB's I'm using so maybe getting a pair of them will also help things out.

Marvel, what was omitted initially and has now been corrected on the schematic is R6 a 470K resistor to ground between the screen and the coupling cap. Without this resistor the tubes would unbalance with one running away while the other dropped it's signal. I literally spent days and days trying to figure out what was gong wrong assuming (naturally) that I had screwed up royally yet again. It wasn't until an email to Costas led him to revisit the schematic and discover the error that things got back on track.

I'm actually using 120K at this spot on the suggestion of Mike who notes that David Halfer did it that way.

[mike stehr]

By the way Mike take a gander at the schematic given in the link and see if you can think of where I might apply a little damping.

Hi Marty,

That screen tap may not be for UL purposes, unless the output transformers indicate for ultra-linear use. They may be just for humbucking. (a type of filtering.) The screen voltage shouldn't be higher than the plate voltage I would think.

With regard to damping, the designer claims good damping, which the amp probably does have. Tetrode connection with no feedback, would probably result in good damping, or strong thumpy bass I guess.....

I did some breadboarding last summer with single-ended tetrode connected 6L6, using one 12AX7 for each tube. Just a simple two stage amp, with a gain stage feeding a 6L6. I used 20 watt James 6123HS universal OPT's, which has a 43% screen tap that can be used for UL. I experimented with pretty much all the primary taps, 2.5K, 3.5K, and 5K. I even tried UL. I also experimented with feedback from the OPT's secondary to the 12AX7 cathodes.

I dunno what the designer has against feedback, but in my experiments with SE 6L6, feedback helped with noise, which can be pretty bad. It also helped tame the sharpness of the upper midrange of 6L6, which can have quite a bite. I've played with 6BQ5 without feedback, and they always seemed shouty/sharp in the midrange. I've always prefered some sort of feedback, it made things sound more relaxed, less unglued. The damping always seemed ok...

Now directly heated single-ended triodes without feedback is a different matter......

You can maybe establish a feedback resistor value by ear to help tame some of the noise/sharpness. But if the OPT's have some sort of upper bandwidth ringing or overshoot, you will have to find out where in the frequency bands it is happening, and compensate with frequency comp capacitor bypassed across the feedback resistor. Of course it requires a signal generator with sqarewave fuction and a scope. You would probably want to apply the feedback to the cathode of the second half of of the driver tube. (cathode follower) But I'm not certain about that, maybe others can chime in.

You could try plate to plate or partial feedback from the plate of 6L6 to the second driver tube's plate, maybe start with 100K and work up to 300K.......I dunno......google search "RH84" or "RH807". They will show circuit examples to examine. A 807 is not much unlike a 6L6, you may be able use the value of the feedback resistor from that schematic for a starting point.

If you eventually get sick of the 6L6, convert it over to 6V6. You may have to lower B+ voltage a little, but a 5K primary is about right for SEP 6V6 I think. 6V6 can sound rather nice, they aren't as sharp as 6L6 in the upper midrange. Not as much power, however......pretty much a like a SE 6BQ5.

But some feedback would still apply.

Mike.

[thebes]

Thanks Mike, I really appreciate you taking the time to post an extended response to my question.

Your description of your experience with the 6L6 seems to eerily echo mine. Although I don't have a scope I will take a shot at it. It appears from this schematic for an RH84 SE that the plate-to-plate resistor is 100k. Saw the same thing on a schematic for an A807 so I'll start there.

http://homepage.mac.com/planet10/forum/rh84_schX.gif

Just to be perfectly clear so I don't blow something up, I would run this resitor to the plate of the 6FQ7 not the 12AX7 preamp tube.

[kaiser SET say]

I know you'll track down that hummer Marty, great looking piece[]

[mike stehr]

Thanks Mike, I really appreciate you taking the time to post an extended response to my question.

Your description of your experience with the 6L6 seems to eerily echo mine. Although I don't have a scope I will take a shot at it. It appears from this schematic for an RH84 SE that the plate-to-plate resistor is 100k. Saw the same thing on a schematic for an A807 so I'll start there.

http://homepage.mac.com/planet10/forum/rh84_schX.gif

Just to be perfectly clear so I don't blow something up, I would run this resistor to the plate of the 6FQ7 not the 12AX7 preamp tube.

I guess I should have paid more attention to circuit you posted. I didn't pay attention to the 12AX7 feeding a 6FQ7.

To be honest Marty, this is where you are starting to cobble two designs together.......who knows what you might come up with. You could try it real quick in operation with clip-leads attached to the feedback resistor value of choice.

It may work and sound better possibly, or it could be worse. It could get rather frustrating........

Maybe email the orginal designer of the circuit and explain to him the situation with regard to the sharpness, and whatever else you may find objectionable. Ask him what his opinion would be of using negative feedback, and how to apply it to his design.

I've breadboarded the RH84 circuit using my James OPT's. The circuit sounds just as good as my little Magnavox SEP console amplifiers on the mid and top end, with stronger bass. It's a foolproof design, easy to do, and sounds good.

Out of pure boredom I'm going to convert the RH84 circuit from EL-84 to 6V6. It should work.....I dunno, I just never could get into single-ended 6L6.