SE EL34 Plexi-verb

[Phil_S]

I believe I'm getting close to launching my next amp project. Many thanks to the early contributors for ideas and counsel, of which there was quite a bit.

Special thanks for the hardware contributions: an orphaned EL34 when I asked if anyone wanted to barter for used but not used up USA 6V6's or 6L6's, and the UTC R-8 PT that's good for very high voltage but only 100mA that's slated to go into this build.

I decided to attempt to go whole hog on this. I bought the Belton BTDR-3 reverb module. This latest release provides for both depth and level controls. For an extra buck (literally), I figured, why not? For me, this is breaking new ground as it is my introduction to working with sand. BTW, this brick is tiny...IIRC 28m x 38mm x 3mm. I bought a package of 10 pin headers to plug it into. I'm amazed at how inexpensive all the solid state gizmos are.

I've drafted a schematic. It's attached along with the Belton document on the BTDR-3. I haven't yet made the calculations for the B+ ladder, so the dropping resistors are likely to change. I imagine there could be more than one error. I'm not asking for a proof reader here, however, if something pops out, please feel free to let me know.

I found there are several variations on the Plexi preamp and, honestly, didn't know which one would be best. So I chose one and moved on. This is one area where opinions on tweaks would be welcomed.

On the reverb circuit, I wasn't totally sure where to tap the dry signal and where to insert the wet signal. The single ended topology and the cathode follower flumoxed me a bit. Opinions on the best point to take the dry and insert the wet signal would be appreciated. I realize as I'm writing, I didn't provide an "off" switch for the reverb. What's a good way to lift the reverb out of circuit?

The reverb circuit was challenging. Someone shared his circuit for the BTDR-1, but the BTDR-3 is a bit different. In particular, it has stereo outputs that need to be summed somehow for a mono application. Belton provides a sample circuit, but I was leaning more towards what someone here suggested. For a mono application Belton suggests dual pots for the depth and level controls and then mixes them. This seemed like a waste to me. Why not mix first and then feed to a single pot for each function. So, with abandon, I did that. I lack the expertise to know if it will work as I've drawn it. I did not post the Belton schematic I was given because it is not mine to post, though I did copy a significant piece of it.

I have a feeling it will be a challenge to stuff this into a 17" x 8" chassis, but I'm going to try. I think there are 8 pots (no P7 on the schematic), one input jack, pilot, and one switch (no standby.) That's 11 holes. This will impact layout, which isn't done yet. I feel it's best to get the kinks out of the schematic first.

One last thing on the reverb circuit. I'm intending to power it from the 5V 3A filament winding. Unloaded it is 6.03VAC. I am figuring this little brick needs no more than 100mA or roughly 3% of the winding's capacity. I'm gambling a doubler will generate 6V * 2.8 = 16.8VDC. Given the low current draw, I'm hoping it will give me the 16.5V input I need for for 7815 voltage regulator to put out 15V for the TL072 op amp. The 5V supply for the Belton should be fine. If anyone thinks there is only a snowball's chance in hell for this to work, I'd appreciate hearing your thoughts. There's plenty of current on tap to move up to a tripler, but I'm not all that pleased about 25V into the 7805. I suppose I can tap the voltage multiplier at first diode and get something like 8.4V, which is actually quite perfect. I imagine that would require modification to the regulated supply circuit, though.

Well, once again, many thanks in advance for your thoughts and good counsel.

Phil

[Phil_S]

Fun with sand. I received a couple of strip boards that I intend to use for the reverb. Still waiting for parts. Spacing is probably off as I don't know the size of everything 'til I see it. This is gonna be interesting.

[Littlewyan]

You might want to consider going single channel with this amp and using the spare triode as a fake PI. Either that or put a master volume between the tone stack and EL34 and use the JCM800 Pre amp.

With a 403VAC secondary your HT will probably be in the region of 500VDC+ won't it?

Can't comment on the reverb as I've never implemented it before. I'd imagine that if you wanted it for both channels you'd want it after the mixer resistors.

Also I think you'd need a Grid Leak resistor between the tone stack and EL34.

I've attached a schematic for the SL5 which has a fake PI and runs an EL34 on 250VDC so it only puts out 5W.

Edit: Just noticed the reverb in that schematic. They put it just before the fake PI.

[martin manning]

Re the need for a voltage gain stage (fake PI) after the tone stack, it will take ~30V p-p to fully drive the EL34 at the chosen operating point. Since the tone stack reduces the signal level 20-30 dB, I think that's a good idea.

The power supply is a choke input design worked out in another thread:
https://tubeamparchive.com/viewtopic.ph ... sc&start=0

I agree the EL34 needs a grid leak, and a grid stopper too.

Edit: Phil, your schematic above is showing a FWB rectifier. It should be a FW, I believe.

[Firestorm]

In my SEs, I put the tone stack in the middle, Fender style. Added gain in tone recovery.

[martin manning]

I think another R-C filter stage is needed to get the ripple down to an acceptable level, and I'd also increase the filtering after the choke to 100uF, making it Rectifier, Choke, 100uF, 150R, 100uF, 5k, 22uF.

[gui_tarzan]

I'm curious about putting the tone stack right before the EL34, what was your inspiration for doing that?

[Phil_S]

Martin: good catch (thanks), it should be FW, not FWB rectifier. I don't know what I was thinking! Grid leak and stopper and the EL34 are an oversight. Another good catch, thanks. I guess, I've have realized after I built it. Good thought on adding one more R/C stage -- where in the B+ ladder?

Littlewwyan: with choke input and FW rectifier, I'm expecting 340VDC. I've got this on a breadboard right now with the actual PT and choke, so I don't expect much variance.

Firestorm and Guitarzan, the CF tonestack follows the typical plexi, where the t/s immediately precedes the power stage. Since I don't know any better, I put it there. I suppose I could move it. Though I'm wondering aloud now, I could add a tube like a 6AV6 (half a 12AX7) for a recovery stage. I'm thinking this will also resolve the problem raised by Martin and Littlewyan.

Thanks everyone for input and for looking! I'll post a revised schematic after a while.

[martin manning]

Good thought on adding one more R/C stage -- where in the B+ ladder?

Right after the choke. See in the SE Marshall SL5 schematic posted by Littlewyan (thanks for that!). You could add it to your breadboard rig and measure actual DC and ripple voltages.

[gui_tarzan]

I wasn't suggesting you should move it, I just wondered.

[martin manning]

I'm gambling a doubler will generate 6V * 2.8 = 16.8VDC. Given the low current draw, I'm hoping it will give me the 16.5V input I need for for 7815 voltage regulator to put out 15V for the TL072 op amp... If anyone thinks there is only a snowball's chance in hell for this to work, I'd appreciate hearing your thoughts.

In a low voltage supply the diode junction drop becomes significant. Suppose your PT will hold 6 VAC. That means the peak voltage is 6*1.414=8.48V. But the diode will clip 0.7V off of that leaving 7.78V. Doubling that you only get 15.56.

There's plenty of current on tap to move up to a tripler, but I'm not all that pleased about 25V into the 7805. I suppose I can tap the voltage multiplier at first diode and get something like 8.4V, which is actually quite perfect.

I like this idea. You can get around 23V for the 15V regulator and 7.7 for the 5V regulator. I would use large reservoir caps, 2200uF for the two in the doubler, and a 4700uF for the HW one.

[Phil_S]

...the diode will clip 0.7V off of that leaving 7.78V. Doubling that you only get 15.56.

There's plenty of current on tap to move up to a tripler, but I'm not all that pleased about 25V into the 7805. I suppose I can tap the voltage multiplier at first diode and get something like 8.4V, which is actually quite perfect.

I like this idea. You can get around 23V for the 15V regulator and 7.7 for the 5V regulator. I would use large reservoir caps, 2200uF for the two in the doubler, and a 4700uF for the HW one.

I forgot to consider the 0.7V drop, so I'm grateful you pointed it out. This is simple lack of experience, not lack of knowledge. Funny how this works in real life. If you like the tripler solution, so do it. I'll bump up the caps and make it so! Thanks very much. There are a number of places on this amp where I'm getting experience with new stuff. It's nice to get out of the comfort zone. It's even better when there is reliable guidance.

On the c/f without a recovery stage (re: Jim's comments), as I understand it, it is too lossy to adequately drive the SE EL34, so it turns out that it's the right question. The answer, to be really frank, is that I didn't know better. Either I've got to move it back a stage, go with the typical Fender tone stack, or add a recovery stage. I thought I had a drawer full of 6AV6's from an organ harvest to provide the single triode I need, but it turns out they are 12AV6. Having looked through the tube bin and found I've got 2x 6AQ7 and expect one of them will be good and provide the single triode. I hate to waste a triode.

[martin manning]

Study the SL5 schematic above. It has several of the features you are considering.

[xtian]

Phil, I'm very excited about your build of the digital reverb. Keep good notes!

[Littlewyan]

Do you really want both plexi channels? If not then lose one and use the spare triode for the recovery stage.