Hammond Tone Cabinet Reverb Drive Transformers

[2Tracker]

...Hammond's original 1935 design involves the preamp in the organ sending a balanced-line signal to the power amplifier that's considerably hotter than what is now standard line-level, so the input to the reverb driver is fairly high.

Interesting....by comparison, I believe the “standard” Fender approach is to use a 12AX7 with linked plates to drive the reverb. Just on paper, I’m thinking this is a lot more gain than a 6SN7 in similar configuration, but if the Hammond preamp is already providing a higher gain….


I’m trying to puzzle out what would be best in a +4 dBm studio environment, and whether or not the 6SN7 will provide sufficient gain to drive the reverb. I could add a gain control, but I’m thinking it’s not best to place this in front of the first gain stage (in this case, the 6SN7 or 12AX7) for noise considerations….

[David Anderson]

Where did you source the mineral oil, David? Do you know the viscosity? Did it seem thicker or thinner than Hammond generator oil? Typical drug store mineral oil comes in three viscosities, but usually they only have "medium" on the shelf.

I guess I got "medium" from the drug store.

[MihevicB3]

There are a total of 5 springs in the Hammond fluid reverb system.

The 3 springs in the long oil filled tubes are just suspension springs that support the whole system. Oil in these tubes isolates the suspension system from the other two springs that are in free air. For these long tubes, Most of the damping fluid is required by these long springs. I would not be too concerned about the type oil used in these tubes. I have seen Type A transmission fluid or mineral oil as recommended substitutes for Hammond damping fluid.

The two free air springs are the reverberation signal springs. One spring goes to a crystal pickup (via a short ribbon strip) and the other spring goes into a short tube that has damping fluid . In operation, a first direct signal goes to the voice driver at the top. Immediately, the signal is transferred to a “lever” where it travels down both springs. The “lever’s” primary function is a signal direction changer. The first signal to hit the crystal pickup is delayed 1/15 second. That signal also hits the fluid in the small damping tube at the same time and is reflected back up the spring with a little less energy, through the “lever” and down to the crystal pick up. These reflections occur 4/15 second apart and continue until all the energy is dissipated by the fluid in the short damping tube. Changing the level in the short tube will change the true delay time of the repeated reflections. Less fluid means produces a longer delay time. Note that no spring unit (to my knowledge) has this feature. The spring necklace and tank units have fixed delay times determined by the spring construction. Another vintage unit, built in the ’60’s, was the Schober “Reverbatape”. It consisted of a continuous tape loop, a record head and 3 playback heads. A 3 gang adjustable pot that attenuated the signal from the playback heads adjusted the delay time.

In my opinion, it is the fluid in this short damping tube that is critical. Hammond Tone Generator Oil can be used as a substitute for the damping fluid, however, its viscosity is higher.

I did a crude viscosity test. See my pic below of my “lab” setup. I still have the damping fluid I ordered from Hammond years ago. For the test, I used the 4” medicine dropper you see in the pic. It holds 1/8 fluid oz. I filled it first with Hammond Damping Fluid (the “real stuff”) and then with tone generator oil. I timed each until the dropper was empty. The Damping Fluid took 3.1 sec to empty and the TG oil took 4.8 sec. Just by pouring, and the feel between my fingers, I could tell the Damping Fluid was thinner.

John M.

[Wes]

I googled "how to measure viscosity" and found this experiment:

http://www.wikihow.com/Measure-Viscosity

Then I had a flashback to 1992, PHYS 105 - we measured the viscosity of glycerine with this exact method. Did not expect that memory. I miss physics lab.

Wes

[2Tracker]

I can test one of the reverb drive transformers for you, but you are correct that there's nothing particularly unusual about the circuit. That the output of the crystal pickup was intended for a balanced recovery stage may or may not end up being significant.

Hi again,

Finally digging into this circuit and as it turns out, I'm wondering if you CAN shed a little more light on the reverb driver transformer. I'm wondering if I should keep the basic topology the same, and try to find an appropriate transformer (one set up for push-pull), or unbalance the incoming signal with a transformer, and then drive a single-ended reverb transformer with a "standard" (for a guitar amplifier) twin triode with linked plates.

[Tim_at_Jonas]

Interesting....by comparison, I believe the “standard” Fender approach is to use a 12AX7 with linked plates to drive the reverb. Just on paper, I’m thinking this is a lot more gain than a 6SN7 in similar configuration, but if the Hammond preamp is already providing a higher gain….


I’m trying to puzzle out what would be best in a +4 dBm studio environment, and whether or not the 6SN7 will provide sufficient gain to drive the reverb. I could add a gain control, but I’m thinking it’s not best to place this in front of the first gain stage (in this case, the 6SN7 or 12AX7) for noise considerations….

Fender uses a 12AT7 with the triodes in parallel (plates, grids, and cathodes), driving a transformer. The Fender transformer drives a tank with an input impedance of "8 to 10 Ohms" (DC resistance of .8 to 1.3 Ohms).

[2Tracker]

Hey Folks!

Wanted to thank you--especially David Anderson--for the input on this project. I wanted to share a photo of the completed unit; going to drop it off at its intended home soon and then fill the oil tubes.

Circuitwise, the best route was, not surprisingly, to follow Hammond's schematic closely, just using parts on hand and making small changes as needed.