hammond organ verb

[makepeace]

Hi guys,

Ok so I've pretty much 'finished' my standalone guitar reverb unit scavenged from an old hammond l-122 (just found a l-102, which i'm going to repair and leave as is, leslie plans are also in the near future).

Anyway - here's what I've come up with.



Its pretty much the standard circuit, with some controls added in.

All of the controls work as intended - the 'gain' control is a bit of a misnomer. It sounds pretty good, but there is still loads that I want to mess with. If anyone can see anything that stands out as being irrelevant, inefficient or that you would have done another (better) way, I would be incredibly grateful if you could let me know!

Regards

EDIT: to view picture properly, open in a new tab and go ctrl +, or save as and zoom

[Chuck H]

Well... You DID ask...

I don't see the 1M resistor on the input as doing anything useful. It could be a direct ground connection.

The way this is designed inverts the phase of the instrument. A peek at the typical "Fender" type unit would show a cathode follower circuit that does lose a tiny bit of gain but doesn't invert the phase. Not that it's a big deal.

It looks to me as if the dry signal is actually amplified over the input signal. Not that this should sound bad, but it's less predictable as a straight up reverb because this would require compensation.

Why is the recovery stage biased so hot? I would expect saturation of that stage long before cutoff. A more center bias condition would seem correct.

Is that a simple choke feed for the reverb driver plate load? Or is it a primary/secondary transformer?

The "gain" and "mix" controls are mixed up as indicated in the schem.

The "gain" control will turn down the dry signal as well as the reverb.

What are the loads on either end of the pan for?

I only looked over it quickly. I may have missed some things.

[makepeace]

Thanks for the reply and suggestions + questions.

Ok here we go:



Have done a bit of lypo.

Yep noticed the phase inversion happening. Not really anything to worry about? I might scrap the input stage - which will get rid of that problem right?

I don't know that, and why the recovery stage is biased so hot. Care to elaborate a little? Still don't really know enough about the maths and also I haven't seen that topology before, so don't really know what is going on there.

The reverb driver plate load is choke fed - simple 14H choke. Fixed the gain/mix, was an oversight. Also fiddled with that section a little bit.

I have no idea what the loads on either end of the pan are for, was hoping you could tell me . The tank impedances are 1475ohms into 2250ohms if that helps at all.

Would there be any sense in fiddling with the 12BH7 grid stoppers (like making them less, or taking them out - this a good idea?).

Also, going to have a complete relook at that recovery stage. It looks very funny.

Thanks again!

edit: Just realised I forgot the 12BH7 cathode resistor of 390ohms.

[Chuck H]

Was this the circuit that was actually in the organ? There may be some reason for the odd circuits that has to do with peripheral circuity no longer in use outside the organ. Perhaps there are drawing errors or some confusion in tracing the circuit as well. This definitely wouldn't function in the same way as other stand alone reverbs. What is your intended use for this and how do you want it to function? I've added some notes to the diagram as well.

[makepeace]

Thanks for the notes and comment!

At the moment, I have the circuit connected up as the first schematic I posted - and everything there works (reasonably) as its supposed to. Believe it or not, the dwell control really does make quite a large difference as its set up in that first schematic. As it stands, the whole setup is pretty much at unity when compared with the dry signal of the strat - so in reality: not much extra gain at all caused by that first stage..

In the second schematic:

The lack of ground reference resistor on the mix was an oversight, realize now that its obvious that it won't work at all like that. Will change that. Those spring tank loads, as well as the pre-drive and the recovery stages were from the original circuit, will see what happens if I take them out.

The stereo jack is there because I couldn't find a mono jack in KiCad, but its fairly irrelevant anyway.

Any comment as to the 12BH7 grid stoppers?

The intended use is as a stand-alone reverb box, in between the guitar (+whatever other effects in the chain) and the amp. I want to be able to vary the depth/thickness/intensity/(insert subjective tone adjective here) of the reverb as well as the wet/dry mix. It would also be cool to have some kind of equalization (for the verb part of the signal), but I'll work on that later.

The main aim at the moment is to thin the circuit out and tune it (wrt to cap values, amplification of the various stages, etc). There is a lot of junk in the first circuit as I kind of threw it together quite quickly. It works, and pretty well, but as always - yearning for more/better.

If I'm not going to use the 1st half of the 12AX7 at the input there, I'd also like to try and find another use for it.

[woodyc]

Interesting that it works. The dwell control must be wired differently than it is in your schematic? Also you must have a cathode resistor (390 ohms?) on the 12BH7 that you're not showing?

[makepeace]

i'm pretty sure i did the dwell part right, but i will check again when i update the schematic. oh and yep, i omitted the 390ohms cathode resistor by mistake (as stated). but yeah, as i say, the first schematic (bar the missing cathode resistor and swapped mix and gain controls) works.

[makepeace]

Ok, I have a couple of new ideas for this.

1st option: What if I drop the input stage, use the 12BH7 parallel stage to drive the tank with a 1Meg dwell control, followed with a 1/2 12AX7 cathode follower driving a passive baxandall tonestack, recovered by the other 1/2 12AX7 setup as a conventional gain stage. and then mix the wet and dry - maybe with a reverb level pot as well.

2nd option: Send the dry signal through a 1/2 12AX7 cathode follower, drive the tank with the parallel 12BH7 and recover with 1/2 12AX7 conventional setup w/passive baxandall stack on the end.

Any thoughts?

[Chuck H]

Option one is flawed because the signal coming from the pan is only about 6mV. That's really tiny. Running it through a cathode follower and then a tone stack would pad it down more. Down to maybe 2-3mV. At that point even noise is in competition with the signal. Most reverbs are noisy enough without further padding the tank output. I'd say send the tank to a high gain amp and then to the cathode follower and tonestack.

Option two sounds more like the normal Fender type.

[makepeace]

Ok cool. What would be the benefit of using the cathode follower on the dry signal (like fender does)? As far as I understand there won't be any gain per se, so whats the point?

[Chuck H]

My best guess is that it sets up the stage impedance for the mix control to reduce loading from that control and any connecting cables.

[makepeace]

Cool, thought it was probably something along those lines..

Anyway, had a bit of time to rework the physical circuit. It looks like this now (hope I got it right..):



I've also upped the voltage on the drive stage to about 350V, with the 12AX7 at about 240V

Sounds very good. The lack of gain stage on the front and the cathode follower on the dry circuit has done wonders for the quality and mixing of the dry signal. The reverb also sounds good, but it still want more!

Any ideas as to what I could do? I still feel like I could drive the tank a bit harder.. Any way to do this with the 12BH7 stage? I could try driving it with an el84, but thats probably a little bit overboard?

Suggestions, ideas, comments?

wooops! forgot about the gridleak of 470k on the recovery stage..

[Chuck H]

The reason you don't have more reverb is that the grid of the 12bh7 isn't being driven hard enough. Since you only have have two nine pin tube sockets, and one of them is strictly for driving the pan, one of the remaining triodes needs to serve double duty for both the dry and tank drive signals. This means that you need to do away with the cathode follower. Not enough amplifiers to squander one on a unity gain circuit.

The schem below should work. I left your mix control as is, though I probably would do it differently, you say it works well. The output impedances are less than ideal. If you don't run long cables you should be fine. Better would be to add another dual triode to the circuit so that a cathode follower output could be employed without sacrificing other needed functions.

[makepeace]

So basically, back where I started .

Jokes, thanks! I'm sure it will be better as you've drawn it - will hook it up this afternoon. I think I'll probably use another dual triode in the finished design and put a cathode follower on the end of each the wet and dry lines.

Out of curiosity -as I see you didn't correct that- do I need a grid leak resistor on the front of the recovery stage? Will the ground reference through the tank be good enough?

Also, with the stuff that I've got, what would be the best way to get rid of the bass filtering before the tank? Increase that 50uF cap?

Thanks!

[Chuck H]

Yes, basically back where you started. But a lot less confounded. The divider for the dry signal may require adjustment for unity gain. You may find there is more gain than you need for the 12bh7 grid. Good thing there's a Dwell control. I don't think you'll need less bass filtering once you start driving that 12bh7 grid. But if you do then the 50uf cap value is the limiting component. And if the unit works the way you use it as designed then there is no need to add other buffer circuits. But DO add a 1M resistor to ground from the grid of the recovery stage. The circuit doesn't need it to function, but should the tank ever become unplugged, accidentally or otherwise, that grid does need a 0V reference.

Since you want want the dry signal to alter as well as the wet, Why not include a gain control for each? Then you can jostle them any way you like. Then your getting even closer to where you started. But your refining the circuit.