5F1 Champ w/ Capacitor Coupled One-Tube 6BM8 Reverb (Eric Barbour/Ampeg-style)

[pietro_moog]

yeah, i think so. myself, i'm slowly going to build an external unit version.
i mainly miss a new enclosure and a power trafo.
i'll try both 250k and 500k as dwell pots, 470k and 1M as triode leak res.
i just need time. maybe a couple of months and it'll be done

[dchang0]

If you can, try putting in tone controls on the driver side. It seems that frequency shaping of the signal waveform makes a HUGE difference in the overall response of the reverb. I'm not sure which will be more effective: tone controls before the pentode or between the pentode and the tank. My guess is that before the pentode will make a wider range of tones but that between the pentode and the tank will result in less signal loss.

I am finding the 0.001uF cap to be far too muddy and annoying; it sounds great for overdriven playing but robs the clean sound of its glassy clarity. I'll be switching back to the 500pF fairly quickly because I play clean so often.

(BTW, Merlin touched upon this in his article on spring reverb drivers. http://www.freewebs.com/valvewizard2/reverbdriver.html It appears that I am sending far too much waveform into the tank, causing distortion and overworking the springs.)

[dchang0]

Hey, pietro--

I put the 500pF back in, and the clarity came right back. It also feels like the overall power has gone up. I have no way of measuring this, but it wouldn't surprise me if the 0.001uF cap was sending too much current into the tank and detracting from the overall output of the amp. My guess is that you were sorely disappointed in the loss in power by using the even larger input cap (was it 0.002uF?). I had forgotten that the reverb tank does not have linear frequency response in terms of impedance--my apologies.

If that is the case, then what you may be looking for (in order to get warmer reverb) is to keep using a larger input cap, but reduce the current gain of the pentode driver to take it easy on the springs, and then increase the gain of the recovery side to bring the volume back up.

That would require redoing the load curve calculations on both halves of the 6BM8--it's not enough to simply modify the grid leaks. I'll talk with my friend John to see if he wants to build a copy of my design, but with his build, we'll calculate a new, warmer reverb and test our theory with his build.

Or, if you're willing to experiment, here's what I suggest:

1) Switch to a 0.001uF input cap.

2) Keep the 220K grid leak on the pentode.

3) Increase the grid leak on the triode to 1M.

If that's not enough, then we'll work on increasing the gain of the triode side by changing the anode resistor, and that will take a little load curve calculation.

[pietro_moog]

i could try a 5f2a-ish tone stack around the dwell pot, i don't know..
wow, 2 tone stacks. it's gonna be freaky

[dchang0]

Well, it might actually be better to use a switch to select from different signal caps, like Tubenit does.

The reason I suggest this is that our experiment with different caps indicates that there's a huge change in tone and power dependent on the size of that "window." I suspect that there's a sweet spot between 500pF and 0.001nF that would pass through the maximum warmth without sending too much current out of the driver into the tank. Perhaps a 750pF cap?

Clearly, others have used 0.001uF caps before, such as in Tubenit's single-12AX7 reverb. So it's not that the reverb tank can't handle the larger frequency range. It's that the 6BM8 ends up shoving too much current through at the lower frequencies and overdriving the tank. What's the sweet spot without having to reduce the power of the 6BM8's pentode and thereby roll off the high-end response?

Perhaps reducing the 0.47uF cap to a smaller one would help. I haven't tried that yet.

[6stringer]

Can you point me to "TheTinMan posted a link to some of tubenit's schematics showing Psychonoodler's Champ with reverb" that is mentioned in the thread.
I can't find this schematic or info on the 12AX7 reverb circuit. Specifically, what reverb tank is used?
Thanks

[dchang0]

That circuit (like most other reverbs) uses a low-impedance tank like the 4AB series. It is transformer-coupled.

Strange--I can't find the schematic or thread either. I guess maybe it went offline or something. IIRC, it was over at The Amp Garage.

[6stringer]

Did it look like this one:

[dchang0]

Yeah, I'm pretty sure that's it.

[dchang0]

"Blessed" final schematic, v.1.0l

Okay, that's all folks. I've done all the experimentation I want to do on this design and this is the best possible wet/dry mix available within the original design goal (clean, clear, "big empty room" reverb, selected by the 500pF capacitor). This is as far as the reverb can be pushed up in the mix before it begins to distort.

The changes are in the grid leak resistors: 470K on the driver (triode-strapped-pentode) side (more than the 330K Eric Barbour originally used) and 1M on the recovery (triode) side.

To tinkerers: changing the 500pF cap to a larger value will require that you reduce the gain on the driver side, because the midrange frequencies will distort in the tank.

The impromptu sound samples I provided give a very accurate idea of the tonal characteristics of this design, though because of the grid leak resistor changes, the wet/dry mix is different than previously specified.

v.1.0l:

[dchang0]

Just for sh--- and giggles, I dug out my electric violin from storage and hooked it up to illustrate the new and final final final 1.0l schematic.

Forgive my rustiness--it's been years since I actually played, but the sound clip definitely highlights the gorgeous reverb circuit better than a guitar would. This really, really sounds like the violin is being played in a concert hall, but it's just the teeny little Champ. Keep in mind that electric violins are not self-resonant like acoustic violins, so the "body" you hear is not from the violin itself, it's from the reverb. Also, the recording level was a bit low--it sounds much bigger in person.

The recording mic was again a Shure WH30 vocal condenser mic shoved right into the grillecloth, about 2" off axis. Violin was a Yamaha SV-200 with volume set at 5/10. Amp was set with 12/12 on volume and 6/10 on reverb. Any hiss is from the violin itself--all the noise went away when the violin was shut off.

violinreverb1_0l.mp3

[Storeinfinity]

Millions of engineers and technicians are using cheap testers less than $10 and scope

I did some research on popular cheap testers and scopes present in China.
Data at disposal are collected from the largest online shopping platform taobao.com by searching top sellers+the most popular deals.
1 Clamp meter: 11107 items online.
Top ten best deals are from $4 to $20.
Top3 are respectively $4, $20, and $5 each.
2 DMM Digital Multimeters : 84393 items.
Top ten deals are from $10 to $30.
Top3 cost $6 (nearly 1,000 piece sold every month), $10 (nearly 379p/month), $4. (311p/month).
And the Fluke 17B around $85 are also popular, sold 100 pieces per month.
3 Oscilloscope: 23969 items.
Popular ones are all around $300.
Best seller number 1 ADS1102cal with 100MHz and 1G sample rate sold nearly 60 pieces every single month by one seller
These data are only from one online shopping platform in China among numerous market places in this billion-population country.
Actually the bigger scale of sales are mostly off line in China.
According to one of the top online forums in China ourdev.com, a place where more than 170,927 topics with 2,130,682 technical-only replies shared by 41,344,468 visitors, (among them 335,088 regular registered and strictly screened by a standard of at least 5 technical replies after registration，and always around 5000 netters active online) lots of tester&scope users, most as technicians and engineers working for state-owned enterprises in China, use cheap testers and scopes off work and those first class ones are only purchased with reimbursement and fund from government.
And lots of them argued, in terms of usage, in most cases cheap testers work really well and sometimes even better. (esp. considering diode or transistor hFE etc.) I was thrilled to find these cheap scopes and testers on sites like storeinfinity.com as well as some other distributers of scopes like ADS1102cal or fluke 17b on ebay, which considerably saved my budget on these equipments. For most of us who are not rocket scientists, cheap testers and scopes are really helping for daily circuit debugs and appliance repairs so that we can save money on something more challenging in our lives.

[dchang0]

Awesome info--might help others if you start a new thread with the same info! Thanks very much for the specific model names.

[bazookajoe]

very stupid question

sorry, very stupid question, this circuit does NOT need a reverb transformer??? although this is very unusual, after reading this whole thread i think that fact is never mentioned and to be more confusing there are several other 6BM8 reverb schematics on the web (including the documents from Eric Barbour you posted from the tubetown website) that show a reverb transformer in the schematic.
And somewhere on the tdpri forum i found one of your schematics (silverlake and all, Rev. 1.b) that has a reverb transformer, unlike all the other schematics in this thread including the final final final version.
Confused. Dont know it you still look into this old thread.
thanks.
possibly i will try the one tube 12AX7 reverb posted on the Hoffmann Forums.

[dchang0]

No reverb transformer is required for this circuit. It is capacitor-coupled like the Ampeg reverbs. You could convert it to a transformer-based design with minimal changes on the driver side and a different model tank.

Be sure to use the final schematic version 1.0l, as that has the widest dynamic range possible with the intended bandwidth.

You are of course welcome to tweak the bandwidth to suit your tastes. If you want more mids, you'll have to reduce the power of the driver side, as it was distorting the reverb tank heavily when I tried it last. Perhaps switching to a reverb transformer may solve this easily because of the different impedances involved.