Parallel FX loops query

Voxman

I understand a little about how effects loops work, and broadly the differences between serial & parallel loops but am not an expert. I'm aware that each has advantages and disadvantages for certain types of effects/MFX units. Parallel loops can be good for noisy vintage stomp boxes that don't have mix facilities, and serial loops are better for certain digital time based effects and MFX units.  Serial loops 'insert' in-between pre-amp and power-amp stages, with 100% of the preamp signal going through the effect before its then shunted through the power-amp stage.  With parallel loops I think the signal is split into two; one path is a direct connection from the preamp to the power amp (i.e. as if there was no loop) and the other sends the preamp signal to the effect and then sends it back to the power amp, mixing it with the direct (dry) signal (have I got that right? - I did say I only have a little understanding of this!). 

My understanding is also that a parallel loop that allows a 100% of the wet signal in theory should act the same as a serial effects loop - but in practice this isn't always the case.

My Laney Cub 12R has a serial FX loop; my Laney VC30-210 and Marshall JCM2000 DSL401 have parallel FX loops. My Marshall Valvestate 8080 has an FX loop mix control and a -20dB and 0dB button - so I think that's parallel too but as it's a SS amp, I'm not sure.  

I'd like to try & find out, for each parallel FX loop amp I have, whether the wet/dry mix goes up to 100% or if not what the maximum percentage mix is.  Also, if not 100%, can it be altered to allow 100% and will that replicate a serial loop?  If so one (of the many!) things I don't understand is why do all parallel FX loops not then allow a 100% mix option so that it's more flexible? 

Hope that makes sense - all wisdom appreciated!

Voxman

Any thoughts on this guys?

ICBM

You understand more than you think you do . Basically that's all correct.

Yes, a parallel loop which goes to truly 100% wet is effectively series. The reason why some parallel loops (Mesa in particular) don't go to 100% wet is because the mixing is too crude and uses a single pot bridging the dry and wet signals, rather than a dual-gang which would turn either properly to zero, or better still a separate return gain stage. It should actually be fairly easy to fix it, although I've never seen it done - most people just disconnect the mix pot and make it into a simple true series loop.

There is no difference between valve and solid-state, the same applies to both.

jpfamps

You're correct.

A series fx "loop" (oxymoron alert), which would be called in insert on a mixing desk, interrupts the signal, so that all the signal will be going to the fx unit.

A parallel fx loop (which actually is a loop), and works like an aux send on a mixing desk, sends a signal to the fx unit and then mixes the effected signal back with the dry signal.

The series "loop" is the easier to implement, and avoids any problems with phase shifts in the parallel signal paths. This can be a problem with effects that cannot provide a 100% wet signal, where you could be mixing 2 slightly phase shifted dry signals, or worse some effects are phase inverting!

In the series "loop" arrangement the signal will be passing through effects loop buffers and effects, so the signal could be corrupted if the "loop" and or effects are poorly designed.

Of course the series effects loop could be part of an amp's sound.

Generally in pro audio you would use gain control fx in an insert eg compression, tremolo, volume pedals. 

A parallel loop requires more components to implement, but has the advantage that the dry signal will not be effected by anything in the loop. Ideally using a parallel loop you would run the fx return 100% wet, e.g. using a delay you would only have the delays coming into the effects return.

Some amps do indeed have a mix control for wet / dry balance from the fx return, which you are correct would render a parallel loop a series loop when set to 100% wet. Other amps also have the capacity to switch between series and parallel.

Both loops can work very well, however there are plenty of examples of amps with poorly designed loops.

One of the big problems in valve amps is that the signal level can be way too high for an effects unit to deal with, and if this isn't paid attention too then results will be less than optimum. Worse the effects loop is usually after at least one volume control, so the level to the effects loop will vary with use.

For example one of the Cornford amps (I can't remember which) output too much signal for a BOSS DD3 to handle without distorting excessively; a poor result given the popularity of the DD3.

ICBM

jpfamps said:

For example one of the Cornford amps (I can't remember which) output too much signal for a BOSS DD3 to handle without distorting excessively; a poor result given the popularity of the DD3.

Same with the Mesa Rectoverb, among its other faults.

Voxman

Thanks guys, I understood some of that, I think. But what about the amps I have...how much wet/ dry mix do they have?  

ICBM

Laney Cub - 100% (Series)
Laney VC - 0-50%. (Parallel loop with fixed dry level equal to max wet level)
Marshall DSL - 0 to somewhere high but not 100%. (Even though the loop return can be turned up louder than the dry, some dry will always be present)
Marshall 8080 - 0-100%.

Note that only one of these amps has a properly designed FX loop . Although the DSL's can be useful as a boost which also seems to improve the tone, as you know.

jpfamps

The Laney and Marshall DSL401 have parallel loops where the mix between the dry and wet signals depends entirely on the amount of "wet" signal coming back to the amp from the effect units.

The Marshall 8080 has variable mix between 100% wet (ie a series effects loop) and 100% dry (ie no effects loop).

The -20db and 0db settings refer to the signal level to the effects send. Pro gear runs at "line" level which actually +4db Effects pedal may not be capable of coping with line level as they are often designed with guitar level in mind and run off 9 VDC, so -20db may be more appropriate for effects pedals.

For best noise performance you want to run the sned level to outboard effects as high as possible with causing distortion.

Voxman

Thanks guys, very helpful. Jpf, I think you meant without causing distortion - but not sure what you meant by sned?

jpfamps

Voxman said:

Thanks guys, very helpful. Jpf, I think you meant without causing distortion - but not sure what you meant by sned?

Sned = send. Indeed I meant without causing distortion.

ecc83

Merlin Blencowe's book "Tube Pre amps for Guitar and Bass" has some excellent information on FX loops, several circuits and much else. You can ignore the maths!

Dave.

Voxman

Thanks everyone, much appreciated. 

clarkefan

ICBM said:

jpfamps said:

For example one of the Cornford amps (I can't remember which) output too much signal for a BOSS DD3 to handle without distorting excessively; a poor result given the popularity of the DD3.

Same with the Mesa Rectoverb, among its other faults.

Fryette Power Station with a cranked DSL 50 going into it, same thing.  It really burns me, I am *very* partial to the DD-3 and can't use it.

Even the DD-20 which *accepts* line level struggles, not as bad as the DD-3 but I still need to use a volume-pot-in-a-box to drop the send level to cope...

Do you learned chaps have a more sensible solution by any chance? Can something like a DD-3 be modified to cope with high fx loop send levels?

ICBM

clarkefan said:

Do you learned chaps have a more sensible solution by any chance? Can something like a DD-3 be modified to cope with high fx loop send levels?

No - or certainly not at all easily.

The best solution is a level-matching box which takes the high level from the FX send, reduces it to pedal level, then takes the output from the pedal and reamplifies it to the original level to go back the the FX return. The problem with is is that it tends to add noise because the noise floor of typical pedals is rarely as low as studio-quality FX, and the noise is amplified too.

Best is to find a studio-type FX unit which sounds similar to the DD-3 - I used a Boss DE-200, which has the same delay chip as the DD-2 and early DD-3 - but then you're into a rack case and more cable faff...

clarkefan

Thanks @ICBM  

I'll soldier on with the DD-20 and the volume pot box thing, but it's mental that amp makers do this line level in fx loops nonsense, we generally want to use a ton of instrument level pedals in fx loops ffs

The DD-20 is the only Boss delay I've tried that comes close enough to manageable, sad state of affairs...

ecc83

The level situation is going to be an issue so long as peeps are wedded to the PP3!

The best you can get from a 9V supply is around 3volts rms*, around +9dBV and so for a decent headroom, average signals have to be at -10dBV =316mV.

Since modern ICs can easily cope with supplies of 30V and more it WOULD be possible to use 4 (say) AAs and a DC-DC converter to give at least  + and - 10V . Trouble is the punters would moan, drainlike!

*You could have a "balanced" output stage giving another 6dB but I know of no one that does this?

Dave.