Silvertone / Danelectro Piezo reverb mod

[456Onno456]

Hi,

this short post is motivated by a repair of a Danelectro DM-25 amplifier. Nice little amp, the downside is the horrible Piezo Reverb (in this case driven by two 6CG7s in a push-pull configuration). It is trashy at best, but for most users it is simply unusable. This post is directed at the latter users, looking for a decent reverb in this amp.

The reverb driver is arranged as a poor mans constant current source (credit goes to Merlin Blencowes and the nice instruction(s) on his website). Instead of paralleling two 12AU7s, here a single 6CG7 is sufficient to drive a F-type Accutronics tank (the LT-Spice files are in the *.rar. Have a look at it if you don't believe me) from a 280V HV-node. The second 6CG7 can be used to drive the constant current driver itself (think Ampeg VT22). The reverb recovery can be implemented with a LND150.

Pros: No additional tubes or transformers. Same current draw as the push-pull driver of the original circuit.

Max

PS:

1.) Keep R25+R19 ca. 220k in total. The individual values can be adjusted (dwell pot).
2.) Adjust C7, C8, and C4 such that each highpass is to your taste (Fender uses 470p 1M, ca. 320Hz)
3.) R2 and C1 are purely optional (more info on Rod Elliots 'ESP' site)
4.) C3 and R6 have the typical Fender values. This really depends on the way you are going to inject the wet signal back into the main signal path. I don't see a reason why this shouldn't work with the DM-25. But if you want to keep the original 1M pot, than reduce C3 to 330p.
5.) R5 and R7 should be 150k in total. The individual values can be adjusted. Increase R7 if the reverb is too strong on max. setting of the reverb pot.

PPS:

the LT-Spice models for the two 6CG7-sides have been fitted from curves I measured with my (well calibrated) utracer (see Ronald Dekkers website for more info).

[tubenit]

One of my favorite amps was the Dano Centurian 275 that I owned.  It came with the piezo reverb (just awful).  This amp is somewhat similar to the Silvertone 1482 (which doesn't have any reverb)

I simply replacedthe 6CG7 with a 12 A_7 tube (rewiring the socket) and the "one tube reverb".  I was very happy with how that turned out.   It used a standard Fender reverb trannie and reverb tank such as Hoffman sells.

http://el34world.com/Forum/index.php?topic=7957.0

With respect, Tubenit

[456Onno456]

Hi Tubenit,

I agree, these old amps produce a charming tone. I recall reading your 1-tube reverb thread, although I never really implemented your circuit into a build. Do you know the primary DC-resistance of the standard Fender 125A20B? And out of curiosity, do you know who started using 1/2 12AX7 to drive a reverb tank (I can come up with early amps designed by Randall Smith, but he cannot be the inventor ;-))?


Now the not so nice part:

I have my problems with the electrical design behind most single 12AX7 driving an A-type tank through a 125A20B. The load is completely mismatched (Ra=68k vs 25k) and depending on which HV-node is used in an amp, the tube may be overdissipating (some Mesas certainly do, even with the added resistor). Please, I am not claiming, that it does not sound nice (which is nevertheless subjective). I am just making an technical statement.

also with resepct, Max

[tubenit]

priimary 25k/8k

The earliest example I know of a dual triode "one tube reverb" is the Gibson Scout GA-17RVT.  There may be other early examples.

http://el34world.com/charts/Schematics/files/Gibson/Gibson_ga_17rvt_scout.pdf

I know there are a number of guys on this forum and on The Amp Garage that implemented the "one tube reverb" into a significant number of different amp designs.  I was first exposed to it by the Dumble cloners using this.  (maybe BrownNote?)

IF you look, you'll note a lot of similarities between the Silvertone 1482 and the Dano Centurian.

http://el34world.com/charts/Schematics/files/Silvertone/Silvertone1482.pdf
http://el34world.com/charts/Schematics/files/Danelectro/Danelectro_centurian_275.pdf

The attached schematic is close to how I implemented this in my Dano Centurian.

With respect, Tubenit

[PRR]

> The load is completely mismatched (Ra=68k vs 25k)

We always "mis-match" into a reverb. The coil impedance is largely inductive, impedance varies with frequency. For other reasons, if you drive with low-Z it sounds dull. Drive with high-Z, the drive rises with frequency and the low-high tone balance is good. Driving nominal 25K with 68K is about right.

> the tube may be overdissipating

Looking at 12AX7 with 1K cathode resistor and 280V supply, counting on thumbs, I get 0.75W plate dissipation. That's under the 1W rating, but close enough that I would want to get actual operating values, and not deviate too far. However tube abuse is an old-old custom, things don't always burn-up at 1.01W, or not before the warranty runs out. You are wise to be wary, but some care in set-up should give a happy-enough reverb.

[PRR]

Here's the plot. Rk=1K, 1 Watt line. RT primary DCR nearly zero. At 280V we have 0.5W dissipation. It looks like we could go past the 330V rating on 12AX7 before we hit 1W dissipation. We could go higher, but the increase of drive is small so why push it?

[munkeyboy]

Max,

Old post, I know, but I hope you're still around.  I'm working on a Danelectro DM-25 now.  I'm not sure I understand what you did here.  The schematic posted doesn't make a lot of sense to my noob brain.  I'm just now to the point where I prefer to rely on schematics instead of layouts for amp building/repair, but I'm having trouble with this one.  What are V1 - V2 - V3 and L2 in the schematic?   It really doesn't seem like it is based on the same DM-25 or even a silvertone circuit I've seen (i.e. cathode components are different).  Did I misunderstand this?

[PRR]

> What are V1 - V2 - V3 and L2 in the schematic?   It really doesn't seem like it is based on the same DM-25 or even a silvertone circuit ....  Did I misunderstand this?

I believe it is a clean-sheet redesign from Piezo tank to common Hammond tank. Since the Hammond wound-coil tank is less efficient it needs more gain-stages.

No V3. V2 is the audio signal from somewhere in the preamp. L2 is an approximation of the tank's coil. V1 is the simulated audio from the tank output. Both are set to nominal 1V for easy math; real levels will be different. V4 is a 320V DC source such as a tap on the B+ filter chain. Yes, all the "V" say "SINE", but this is really a multi-function source which can have a DC component.

[munkeyboy]

Ok that make a little more sense, thanks.   I think I might attempt a different route... if at all.