Adding an effects loop

[Hamfam47]

I have managed to buy a set of Dagnall transformers so I can build a Marshall 1959/2203 amp. But I would like to insert an effects loop into the circuit. Any suggestions how and where in the circuit to do that? I would prefer to avoid using anything solid state, so perhaps do I need to include an additional 12AX7?
Thanks for any suggestions.
Glen

[txstrat]

I'd add an effects loop right before the coupling cap to the PI. You don't necessarily need an additional tube.
Have a look at this: http://www.schematicheaven.com/marsh...h_50w_2205.pdf. Just one of many ways to manage it.
You might also have a look at some Mesa/Boogie or Soldano SLO schematics.

[defaced]

There are ALOT of ways to do a loop. What do you want your loop to do will dictate how you implement it. Serial, parallel, full bypassable or in circuit, tube, solid state, send/return levels, mix for parallel loop, active, passive, before/after tone stack, before/after master volume... that's a good start at the options, but I think you see what I'm getting at. So, what do you want your loop to do?

[Wilder Amplification]

FX loops are pointless on non-master volume amps IMHO. The whole point of having a loop in the first place was to give people who get their overdrive from the preamp a place in the signal path to insert any "post overdrive" effects so that the clean power amp can reproduce post preamp/overdrive effects cleanly (i.e. delays, reverbs, etc etc). Think "slaved rig" in a single head.

However, without a master volume you're not going to overdrive the preamp before you overdrive the power amp and once the power amp is overdriven this renders a loop a moot point.

That being said, I prefer to install the loop on a Marshall after the tone stack but before the master. Installing it after the master allows the master to control the FX Send Drive level, which alters the signal being sent to your loop effects as you change volume, which makes you have to change your input/output levels on your loop effects to compensate, etc etc. However, having the loop BEFORE the master allows the master to just change volume only without altering the FX Send drive...basically just controlling the volume of the whole enchilada with the FX already mixed in.

[Chuck H]

The schem that txstrat referenced actually does have a buffered send and a re-amp stage. On a 1959 you actually DO need to add a tube to do this. And a buffered send is the only way to have a fully functional effects loop. If you can omit one channel from the standard Marshall front end it is possible to do it with only three preamp tubes instead of adding a fourth. You could revoice the one input channel to be brighter than the standard Marshall "normal" channel but fuller than the standard Marshall "bright" channel. Like a hybrid of the two. Then you could buffer after the tonestack/send to effects/return effects/re-amplify with the unused triode/send to the PI or a master volume/PI. I've done it and it worked out well.

Chuck

P.S. While I do agree that running your effected signal into distortion is not ideal compared to running your distortion into the effects I do think there is still some advantage to adding some effects after the preamp rather than into the guitar input even on non master volume amps. For example, try running a delay into the input of a non master volume amp that your cranking up for distortion. If the repeat drops in volume, as they almost always are run, the repeats get cleaner and cleaner as the signal from the delay unit drops. This effect is lessened a bit by adding the delay after the preamp. Consider an EQ. An EQ at the amps input will dramatically change the harmonic content of any preamp distortion and any final EQ benefits will be lost due to the clipping of the preamp. This effect is lessened by running the EQ in a loop even if you are driving the following stages into clipping. But IMHO effects loops are NEARLY useless on non master volume amps IF you run the amp cranked for distortion. The advantage is so small as to be more trouble than it's worth.

JM2C

[Hamfam47]

Effects loop

Thanks so much to those who replied - txstrat, defaced, jm2c and chuck. The effects loop is mainly going to be used for adding a mixture of straight signal with some distortion and reverb signal, and occasionally a phaser when I play keyboards. I should be able to add an extra 12AX7 without overloading the power supply, but I will have a look at some of the other amps suggested to see if I can adapt something, particularly if it looks like I need a buffer.
Glen

[Chuck H]

You really only need 1/2 a 12ax7 to do it. All the schmancy effect loop schems are great, but really over kill IMHO. Simply measure the preamp signal where you plan to put the loop, then pad the signal down to about one volt and send it to your effects loop. The effects return then goes to the added triode set up as a common gain stage to be re amplified to the original measured preamp output. The sonic degradation and any extra noise of this simple circuit is tuely mice nuts, you probably won't hear any difference at all. So why bother with the "dumblator" or whatever. Keep it simple.

Chuck

[Wombaticus]

You really only need 1/2 a 12ax7 to do it. All the schmancy effect loop schems are great, but really over kill IMHO. Simply measure the preamp signal where you plan to put the loop, then pad the signal down to about one volt and send it to your effects loop. The effects return then goes to the added triode set up as a common gain stage to be re amplified to the original measured preamp output. The sonic degradation and any extra noise of this simple circuit is tuely mice nuts, you probably won't hear any difference at all. So why bother with the "dumblator" or whatever. Keep it simple.

Chuck

I assume you're talking about driving the effects from the output from a conventional gain stage... which might work OK except when the input impedance of the effect happens to be on the low side.. and this is evidently the case surprisingly often. In that instance, driving the loop with a low impedance output from a cathode or anode follower really would seem the way to go.

[EFK]

Here's how Selmer did it. It works, requires 1/2 triode only, but one thing I can tell you (having a couple of these amps) is the "Echo" tube and entire circuit is crazy microphonic. If you so much as breath on the tube, the tube cover, or any of the wiring, you will hear it. It's like having a mini-microphone in there.

[Chuck H]

I assume you're talking about driving the effects from the output from a conventional gain stage... which might work OK except when the input impedance of the effect happens to be on the low side.. and this is evidently the case surprisingly often. In that instance, driving the loop with a low impedance output from a cathode or anode follower really would seem the way to go.

A cathode follower is a good way to drive effects. It's more ideal. And I once thought that if I wanted to include an effects loop in one of my amps I'd want to do it that way. But I had an extra triode on a build once and didn't want to waste it, so...

Yes I am talking about a conventional gain stage. But remember that we need to pad the signal down. The preamp signal voltage is as high as it's going to get at that point so it takes a lot of padding. Usually 22k or less. So without going into detail about other considerations that affect this circuits impedance we're going to have less than 20k. Not very high at all. I've NEVER had any problem, even digital multi processors that tend to be finiky with certain amps work fine in my loops. Just sayin'.

Chuck