Serial/Parallel loop...... once and for all

[Joeytpg]

I don't know much about loops shamefully. When is it better to run Serial and when is it better to run parallel? what FX sound better through each one?

 

[King Crimson]

"Which is better for you?

There are 2 ways to handle effects signals. If you use the serial return, then the signal path of your amp is interrupted, the signal is sent to the processor, gets more or less processed, then sent back to the serial return into the power amp. Digital effects units often digitize this signal, then process it, then convert it back to analog, then send it to the amp. This is called ADA conversion. It is necessary for digital effects units to do this to your guitar signal, so that it becomes a digital code, which the processor can read and understand. Your tubes, however, need an old fashioned analog signal, so the processor needs to convert the signal back to analog before it goes back to the amp. Generally, even in highest quality effects processors, this causes a change in the original signal, typically a loss of tonality and warmth, also noticeable as a "harder" sound.

When you use the serial loop for an effects unit like this, then your signal will have been ADA converted at least once. Tone junkies and vintage freaks alike will more than likely have hives developing by now. But, as always, there is a better way. Using the Parallel loop and the mix (labeled "Volume") control on the back of the amp determines how much effect signal is being added to the original signal, which now still flows through the amplifier. There is always an analog connection between the send and return jacks: a parallel loop!

Important: You must set the mix control on the effects unit to 100% wet when using the parallel loop. Otherwise there will be nasty phasing problems resulting in unsatisfactory tone. The signal portion that is unaffected by the mix control in the effects unit would reach the amplifier at a different time due to the cabling, and cause phasing cancellations."

As far as to which effects are more suitable for the serial or parallel it's simply a matter of testing.

 

[Joeytpg]

thank you King, another great opinion

 

[moltenmetalburn]

thank you King, another great opinion

His post was factual not opinion.

 

[dass101]

You can also think about this this way: Effects that "add" something to the original signal, like reverb, delay, harmonizers, can and should be used, if possible, as parallel effects. Obviously, this depends on the fact that your FX processor has kill dry, or 100% wet settings.
Effects that modify your signal, like chorus, phaser, distortion, etc, must be used in series, because adding a dry signal to the processed one would defy the very role of those effects.

 

[King Crimson]

His post was factual not opinion.

Actually both; note the quotation marks.

Focus on the subject matter.

I’m just a connoisseur for factual opinion - envy.

 

[moltenmetalburn]

His post was factual not opinion.

Actually both; note the quotation marks.

Focus on the subject matter.

I’m just a connoisseur for factual opinion - envy.

I should have elaborated but was busy at the time. Whwt i really meant to express what that the considerations taken into account about what loop to use in wht situation doesn't change. Poorly expressed im my haste I agree.

 

[King Crimson]

No worries MM. We're all family here in Diezel land.

BTW - Love Futurama, I watch it all the time.

 

[Skrln]

I have a follow up question on this matter.

They say to avoid the loss of tone because of the ADA conversion in the Serial loop, you should go with the parallel and set your fx unit to 100% Wet or KillDry.

I understand the issue with the serial loop. But how would the parallel loop fix this? The signal is still being processed through the fx unit. And what does the "100% Wet" function do exactly on the unit?
The way I see it, the parallel loop combines both the pure amplifier signal and the processed fx unit signal. Either way with serial or parallel loop, at least some of your signal is ADA converted.

Also, as stated, some kind of effects go better through serial, and other through parallel, would this mean you need 2 fx units through each of the loops? One that handles reverbs and delays, and the other that handles modulation effects?

Thanks for clearing this up!

 

[i.ak]

I have a follow up question on this matter.

They say to avoid the loss of tone because of the ADA conversion in the Serial loop, you should go with the parallel and set your fx unit to 100% Wet or KillDry.

I understand the issue with the serial loop. But how would the parallel loop fix this? The signal is still being processed through the fx unit. And what does the "100% Wet" function do exactly on the unit?
The way I see it, the parallel loop combines both the pure amplifier signal and the processed fx unit signal. Either way with serial or parallel loop, at least some of your signal is ADA converted.

Also, as stated, some kind of effects go better through serial, and other through parallel, would this mean you need 2 fx units through each of the loops? One that handles reverbs and delays, and the other that handles modulation effects?

Thanks for clearing this up!

As you said yourself "at least some of your signal is ADA converted". Some of your signal is processed, but in the serial loop all of your signal will be processed.

100% wet means, that the signal going through your unit in the parallel loop will be converted and processed 100% (aka completely wet) and thus you are able to avoid the very annoying phasing issues caused by combining the dry signal and the wet signal in the return of your parallel loop.
See this: viewtopic.php?f=14&t=88779

You do not neccesarily need two separate units. Also a two channel unit that is able to operate in "dual-mono" -setup is OK.

 

[durbodill]

100% wet means, that the signal going through your unit in the parallel loop will be converted and processed 100% (aka completely wet) and thus you are able to avoid the very annoying phasing issues caused by combining the dry signal and the wet signal in the return of your parallel loop.

Somewhat correct. The signal will always be converted and processed 100% (A/D conversion), the question is what percentage of the outgoing signal towards the amplifier's effects return will contain effects (wet) as opposed to the original signal (aka dry).

This is easily understandable in delays: If you have your effects unit set to 100% dry you will only hear you original unaffected signal (still A/D/A converted). No delay whatsoever.

Now on the other hand, if you set the unit to 100% wet, you will only hear the delays, no original signal whatsoever. For example, delay time 1000 ms: You play a note and the first thing you hear out of your unit is the first delay after one second. You don't hear the actual note you were playing, only the delays generated by the effects unit.

If you mix both situations you will hear both the wet and dry signal according to the percentage usually specified in something called a MIX-parameter on an effects unit. So you'd hear your playing _and_ the delays, which in most situations is what you'd want.

Now why would you want something like a "kill dry" option? It means that what will come out of your effects unit will always be _only_ the effects (e.g. only the delay - none of the original playing). This makes perfect sense of you use a unit in a _parallel_ effects loop. So the MIX-parameter talked about earlier would be a knob on your amp (like the "Volume parallel Loop" on the VH 4). You can now decide how much of that 100% wet signal from the effects unit is going to be added to the original really UNconverted and UNprocessed signal from your preamp. Both of the signals are then mixed together and fed to the power amplifier.

So what happens if you can't set your unit to kill dry or 100% wet and still use the parallel loop? Well, it's not the cabling (like someone mentioned before) but the inevitable latency from the unit's A/D/A-conversion that causes the well-known phasing issues. A regular effects unit will probably need about 5-10 thousands of a second (ms) to convert the incoming analog signal to digital and then back to the outgoing analog. You typically won't notice that by itself. But if you don't have a kill dry or 100% wet setting, some of the dry (original) signal will spill through to the output of you effects device - and later be mixed together with the output of your preamp (as stated before). Now if you mix two same waveforms (sounds) together, one being the original preamp signal and the other being the (not killed) dry part from your effects unit's output, which lags behind about 5-10 ms, you will get wave cancellations. That's very basic physics (wave trough + wave peak = no wave at all, etc.) and results in what you hear as a weird change in sound, maybe best described as "hollow".

Man, sorry I didn't mean to write that much, just had a little time and thought I could share my opinion ;-)

Greetings, Till.

 

[i.ak]

Thanks for the specifications Till. I was in hurry and did not dare to write the long detailed answer here at work.

In some units you have analog hard-bypass as well so there is also unconverted signal present. Usually this is only done by relays on thus is either full on or off, but basically it could be done with mixing options as well.

 

[Skrln]

Thanks guys. It's starting to come together (no pun intended while talking about parallel loops and such ).

One more question. How do you take care of the fact that you'd want some of the effects in serial, such as modulations? Only a 2 channel fx unit could handle the split between serial and parallel?


//EDIT:

However, as the other effects (e.g. chorus, flanger, phaser, compressor, etc.) require serial loop it is often the decision wether to use only the serial loop for everything or not to use other than time-based effects. I would say that one needs to experiment and know his/her own equipment so well that knows which works the best.

This was in the other thread. And it seems to acknowledge my current idea about it. So the only solution to use both time- and modulation- based effects is to use a very good processor which won't colour the sound too much with ADA conversion and go with the serial loop?

 

[durbodill]

Thanks for the specifications Till. I was in hurry and did not dare to write the long detailed answer here at work.

In some units you have analog hard-bypass as well so there is also unconverted signal present. Usually this is only done by relays on thus is either full on or off, but basically it could be done with mixing options as well.

Hey i.ak, good for me that it's my week off

Thanks guys. It's starting to come together (no pun intended while talking about parallel loops and such ).

So the only solution to use both time- and modulation- based effects is to use a very good processor which won't colour the sound too much with ADA conversion and go with the serial loop?

I think using both the parallel and the serial effects loop for different kinds of effects really mostly makes sense if you have a pedal board with a dedicated stomp box for every single effect. This way you can easily line up the pedals you want to use either for the parallel or the serial loop, respectively. For example: serial fx send -> reverb/delay pedal -> serial fx return, parallel fx send -> chorus pedal -> parallel fx return.

Using one single guitar multi fx unit will not allow you to be as flexible - in most cases. So you'll have to choose whether you want to use it in the serial or the parallel loop. Just try it both ways and see what's best for you.

I'm using a tc electronic Nova System in my VH 4's serial loop - ended up being the best solution for me. This way I can take full advantage of the unit (including noise gate, different mix levels in different patches...), while the loss in tone - if at all - is barely audible. There might be some fx units out there that really color the tone, even when on bypass. I got an extra true bypass switch that I turn on if I want the unit out completely.

Again - there definitely is a difference in tone when the Nova System is not bypassed but you can only hear it in an A/B comparison and with no other instrument playing in the rehearsal room. No chance anyone in the audience would ever notice this in a live situation. That's good enough for me. You'll have to find out if you can live with such a tiny compromise. Hope I could help you a little bit, I have to go to sleep now.


Greetings, Till

 

[Skrln]

Thanks till!

Very clear now!

 

[durbodill]

Thanks till!

Very clear now!

My pleasure

 

[phil b]

I do need to add that when using different types of effects with either a serial or parallel loop there is no specific effect that will always sound better with one or the other. Lets take a standalone chorus unit for example. The "general" rule is a chorus effect will "usually" sound better in a serial connection, but thats not accurate all the time. You need to take into consideration the unit itself and how transparent it sounds with the amp itself. Again, there are basic guidelines to use but its only a guideline and not written in stone. I use a TC 1210 chorus with my Einstein and it sound great on cleans in serial, but I run it thru a mixer in parallel and it also sounds great. I have a newer TC unit and in serial it colors the tone too much so parallel is the only option for me. Now if a take that same 1210 and move to my Mesa for cleans it will ONLY sound good in serial and not parallel. I use a mixer with the Einstein in the serial loop as I like that much better than running all my effect units straight into the parallel loop. Bottom line is to let your ears be the judge and always remember the "general guidelines" and you shouldnt have a problem. There are just too many differences in multi effects, standalone units and amplifier technology {specifically loops} these days to say definitively what will work better every time.