Reverb Tank Selection for a Univox U-1221

[astronomicum]

I am starting restoration of a pair of Univox U-1221s. One unit came with a 4F reverb tank, and the other a 4A tank. Both tanks are Gibbs, both appear to be of proper vintage. Interestingly, I have another pair of U-320s that I am nearly finished with that have the exact same issue. I will limit this discussion to the 1221 due to its schematic availability.

To determine the appropriate tank input impedance, I have calculated the 6AN8A (Triode) output impedance based on the schematic at 390 ohms. I would like to think that the tank selection would be as simple as selecting a tank with the closest corresponding impedance, either a 4E (600 ohms), or 4D (250 ohms) but this is foreign territory for me. If I where to pick one, I would be conservative and go with the 4E. An online search and the information I have on selection is limited. Any advice would be appreciated. Thanks.

[sluckey]

To determine the appropriate tank input impedance, I have calculated the 6AN8A (Triode) output impedance based on the schematic at 390 ohms.

How did you do that?

[astronomicum]

 Followed the instructions in this lecture on cathode followers.

Worksheets and load lines attached.

[PRR]

Tank is a coil. Tank impedance varies a LOT with frequency. Can't "match". The spec impedance is taken at 800Hz or 1kHz. You want to drive below 500Hz and maybe as high as 5kHz, a 10:1 range of frequency and impedance.

In most audio, we want load impedance higher than source impedance (not matched).

The tank has multiple frequency effects. One common technique is to pick load -lower- than source, so the fit improves in treble.

> 4E (600 ohms), or 4D (250 ohms)

I suspect even Univox was unsure. Try both, can you hear the difference?

[astronomicum]

I suspect even Univox was unsure. Try both, can you hear the difference?

Thanks PRR. I suspect the same or they used whatever they had in inventory . Going to try what you suggest. I calculated the U-320 output impedance at 689 ohms so going to compare the 4F and 4E on them as well.

[astronomicum]

I have conducted some testing on a completed U320 (689 ohm calculated reverb output impedance) on different tanks (Revisit 4F, 4E, 4D, all new). I threw the 4A tank on for good measure (vintage Gibbs). I measured the mA and mV to the tank input coils under the same conditions. Although the impedance varies with frequency as PRR mentioned, I did testing at just one frequency. Here are my results:

Test Conditions: 600Hz input, 10V(pp) output (1.70A-6.84V rms, 11.63W, about 1/2 full output)

Tank 4F (1475 ohm input impedance): 6.83mA-13.3mV (0.0882mW)
Tank 4E (600 ohm input impedance): 7.32mA-14.1mV (0.1032mW) +17%
Tank 4D (250 ohm input impedance): 0.02mA-2.8mV (NADA)
Tank 4A (8 ohm input impedance): 0.01mA-2.7mV (NADA)

I thought there was a difference in sound between the 4F and 4E tanks, although it was not easy to discern. An A-B switch would have helped  I thought the 4E tank sounded more "full" sounding than 4F tank which sounded more "tin canish". There did not seem to be an appreciable difference in level but I did notice the decay from the 4F (decay 2) tank was longer than the 4E (decay 3). Interesting... The 4E will be my choice for the U320.

As the testing results indicate, I had no sound at all out of either the 4D or 4A tanks. I expected to get something out of the 4D tank though Looks like for this circuit (very similar to the U1221), once the tank input impedance drops too low, the reverb driver output drops off a cliff  It would now seem apparent that reverb circuits in the units I received with 4A tanks did not work, although I have not tested the U1221 yet. I will report back when I have completed that testing.