Adding Fender-style reverb (without distortion, please!)

[mr_hankey]

I built a 50W ODS inspired in an old Bassman 10 chassis a few years back. I always had the intention to use that extra room to add reverb, and I've finally gotten around to it.

I lifted the circuit straight from the Super Reverb. It's currently hooked up in my amp, but it certainly isn't working correctly. Before I get into all the different symptoms, can you please take a look at my circuit to see that I've actually adapted it correctly? I removed some resistors because there's only one channel feeding into it.

[img:800:544]http://i.imgur.com/jWpEPBb.jpg[/img]

The input comes from the clean channel's 0.047uF coupling cap; and the output goes into the 1M master volume which is just before the PI.

Now, the reverb cabling is a bit different than the usual Fender method. I'm using an isolated stereo 6.3mm jack which splits into two RCA plugs for the reverb pan. This means that the jack's sleeve is connected to the shields on both the RCA cables. Does that create a ground loop? It might be worth mentioning that all grounds in the circuit above are connected to my preamp grounding bus.


And while I'm posting this topic, I might as well ask about my relay supply.

[img:800:423]http://i.imgur.com/4xWoz6a.jpg[/img]

The inputs come from the PT (both yellow) and labelled as 5V and 0V on the spec sheet. The two outputs measure a tidy 4.98V between them, and about 300mV to ground. The relay switches as it should when I hold it against the contacts. I'd like to just tie the negative output to ground, so that I only have to run one wire to the relay and footswitch. Will that work?

P.S. Does anyone know of a good free OSX program for drawing schematics?

[Milkmansound]

it might create a ground loop if you are not using the proper reverb tank

I use tanks with the input isolated from ground

my 5W amp design uses a single TRS 1/4" jack for the reverb send and return - as long as you wire it up properly it should work great

make sure the tank is not wired up backwards - thats a common and easy mistake to make

maybe eliminate the relay for now and just get the very working properly and then go back and add it in once it does

[jjman]

The 2.3meg and 10pf cap may or may not be needed. It is attenuating the dry signal. You could use a 1meg pot on the reverb's input tube grids instead of the resistor to have adjustable dwell. You could reduce the 470k to increase the output strength range from the reverb. Fender was trying to match the 2 channels so you can take liberties with the values w/o that matching as a constraint.

[mr_hankey]

Right, let's get into the symptoms:

The reverb is switchable by switching the input and output with a DPDT. It's on the other side of the overdrive switch, which gives me the following options: clean, clean + reverb, overdrive. The clean and the overdrive settings work perfectly.

Switching the reverb in creates a lot of noise. It's not a low hum or a high hiss, but somewhere inbetween. The reverb control works as a volume control for the noise. When it's on 0, the dry signal sounds like it should. By 1-2 on the control it's already drowned out by the noise, and it gets loud.

At the same time, the wet signal isn't working either. I hear no reverb at all unless I poke the springs myself (so no signal). I tried reversing the tank cables, which doesn't appear to change anything.

Connecting headphones to the reverb send jack sounds alright, which leads me to believe V3 isn't causing any problems. As I mentioned before, the dry sounds good when the mix is off, so I think V4b is alright as well.

Further issue: when the reverb is on and I switch to overdrive (which disconnects the reverb section completely), I get an oscillation. I can't tell you the frequency (no scope here, unfortunately), but it's pretty high up in my hearing range. I can change the pitch slightly by moving the V4a grid wire and the input wire (the one running from switch to 500pF). The oscillation goes away when I turn the reverb control below 4-5. The two overdrive controls don't have an effect, and removing V1 and V2 doesn't change anything either. Removing V3 or V4 stops the squeal.

To fix this, I've tried moving the (V4a) 220k resistor from the board to the reverb jack, which eliminated 7cm of grid wire. I replaced the wire running from the reverb jack to grid of V4a with a shielded wire, and I put a 300pF snubber on V4a's plate-cathode. The noise may have decreased marginally, but the squeal is still there. And yes, I've tried other tubes.

I just checked the reverb transformer, and it appears to put out 0V. Bad?

EDIT: My reverb jack is wired incorrectly. I imagine the situation might improve when I'm no longer sending the transformer output straight to ground.

[mr_hankey]

It works! I feel relieved and annoyed.

Now I've also discovered why reverb generally is switched like this: it's way louder than the clean channel by itself.

[mr_hankey]

I was a bit premature yesterday. I was playing with my guitar's volume very low, and it turns out that when I hit the reverb circuit with more signal it distorts (preamp volume above 3 and guitar at 10). It's sort of a fizzy, fuzzy frying sound. It goes away when I put the reverb mix on 1 or remove the driver tube. Using other tubes makes no difference. I've tried reflowing the solder joints on and around that tube, but that also had no effect.

[Milkmansound]

That'll happen

Gotta get the gain right or it can sound bad - try pulling the bypass cap from the driver tube

[rdjones]

Are you using the shared cathode resistor/cap at "point A" as drawn ?
This might be a source of distortion when stages are in series.
It's usually used when stages are in parallel, like the typical dual preamp Fender arrangement.

reddog Steve

[mr_hankey]

Are you using the shared cathode resistor/cap at "point A" as drawn ?
This might be a source of distortion when stages are in series.
It's usually used when stages are in parallel, like the typical dual preamp Fender arrangement.

reddog Steve

Yes, as per the ab763 schematic. I'm finding it strange that it isn't working for me, as my voltages are quite similar to the Fender example (that is, if I didn't make any mistakes).

[jjman]

The reverb drive will probably cold-clip when driven hard. A scope is probably needed to see exactly where the distortion is coming from. On my AB763 DR setup I reduced the 2200R resistor after noticing the later SFs (e.g. Twin Reverb) had done this. They also removed the bypass cap. You have about 4ma going thru that tube and thus the reverb tranny. The 12at7 can handle more current but your tranny should be verified as capable before moving in such a direction.

[mr_hankey]

I pulled the 25uf cathode cap on the driver tube. Now I can turn the preamp volume up to 9-10 before it starts to distort slightly. It also made the reverb more subtle, which makes the control range more useful.

How do I get rid of that last distortion? How about a small grid stopper (10k or so) on the driver tube?

[soma_hero]

you are most likely overdriving the driver tube. voltage divide the signal going into the driver. After your 500pf put a 220k or 470k.

Sharing the cathodes between stages can get wonky when you're talking massive signals of the same or opposite phases, but it is fine here since the reverberated signal has little phase relationship to the dry signal.

[statorvane]

After your 500pf put a 220k or 470k

Yes, it appears there is an additional resistor that's needed there. I have a few reverb schematics I have collected over the years. The Fender style reverb w/o dwell control as implemented on different amplifiers typically has a 3.3M/1M voltage divider at the driver grid. So try fitting say a 1M or higher between the 500pF cap and the 1M resistor.

That would explain the apparent high reverb level and resulting distortion - the signal is unattenuated going to the driver tube, but the dry is choked going to the recovery tube.

If you need to bump up the reverb after this, the suggestion to fiddle with the 470K resistor after the reverb level is a good one. Around 220K you'll get massive surf (or tsunami) level reverb when wide open.

Hope this helps.

[pdf64]

Yes, grid stoppers on the 12AT7 reverb driver can be very beneficial; somewhere in the range 10k - 47k is likely to be appropriate.
There's a lot of gain and phase shift between the input jack, grid of second stage, and reverb driver plate.
Give some serious consideration to getting a scope; a whole lot of stuff will become apparent.
Pete

[mr_hankey]

I've added a 39k grid stopper to the 12AT7, and it doesn't distort now, even with the volume at 10. Success!

I also added a 68k grid stopper to V4b, as well the 220k(series)/220k(to ground) 'channel mixer' resistors (that are present in the orginal circuit) after the last coupling cap. I'm trying to get the signal down, because the volume is much too big when compared to the overdrive channel. I suppose it isn't such a problem, except for that fact that I'm having to run the OD volume higher than I would like: 3 o'clock instead of 12 o'clock, for example. Those resistors have helped a bit, but it's still too loud. How can I attenuate the signal a bit more without affecting the tone? Should I just decrease the 220k resistor to ground?

To clarify: I set up the channel switch to switch between the overdrive circuit and the reverb circuit.