Phase issue with two amp blocks (one path each)

[JulienVD]

Hi.

 

I've ran into problems multiple times when creating a dual amp setup inside the Stomp.

ex.: Path A -Cartographer set clean

       Path B - Placater Clean

 

There is an audible conflict between those two in my setup.

In real life, you'd move a microphone until both signals are in phase.  In this instance, however, You only get a switch to reverse the phase.  (the sound is bad in both positions)

I even tried varying the "distance" function in the cab settings.  But it doesn't change the phase relationship as it would in real life.

 

Anyone use two amp setups with Helix?

Anyone ran into a similar problem?

What are your favorite amp combinations?

 

Thank you.

 

 

[OmniFace]

Steps to reproduce:

 

Are you using the full amp+cab for this?  or separate cab blocks?  (Don't know if the Stomp is different than the full Helix)

If you use the full amp+cab, you could try moving the mic position in one cab section to see if it doesn't anything.  No idea if L6 compensated for phase when adjusting that or not. 

What cab and mic settings are you using?

Could also try adding the Simple Delay to one side with a 100% mix and 0% feedback.  It'll let you set the Time as low as 0, and increment 0.1 ms increments.

[waymda]

1 hour ago, JulienVD said:

Anyone use two amp setups with Helix?

Anyone ran into a similar problem?

What are your favorite amp combinations?

 

1. Yes - and have run up to four amps for very specific blended sounds - at all times using separate amp and cab setups - typically streaming 2 amps into a dual cab block panned hard left and right before the block

2. No never - the biggest 'issue' I've had is balancing the output of individual amps which is just down to channel volume

3. It depends on what I'm doing but I use the litigator a lot - sometime 2 of them with different settings, JC120, Vox, Plexi, The boogies, Baddonk, the twin models, the princess (now its available) - I tend to try lots of amps in combo to find the thing I'm looking for and may blend them differently in snapshots and/or push them differently at different point - not lots of help I know, but I work across a very wide range of styles

[rd2rk]

5 hours ago, JulienVD said:

Hi.

 

I've ran into problems multiple times when creating a dual amp setup inside the Stomp.

ex.: Path A -Cartographer set clean

       Path B - Placater Clean

 

There is an audible conflict between those two in my setup.

In real life, you'd move a microphone until both signals are in phase.  In this instance, however, You only get a switch to reverse the phase.  (the sound is bad in both positions)

I even tried varying the "distance" function in the cab settings.  But it doesn't change the phase relationship as it would in real life.

 

Anyone use two amp setups with Helix?

Anyone ran into a similar problem?

What are your favorite amp combinations?

 

Thank you.

 

 

 

Attach a copy of the preset you're using and we can test it on our Stomps.

[JulienVD]

17 hours ago, OmniFace said:

Steps to reproduce:

 

Are you using the full amp+cab for this?  or separate cab blocks?  (Don't know if the Stomp is different than the full Helix)

If you use the full amp+cab, you could try moving the mic position in one cab section to see if it doesn't anything.  No idea if L6 compensated for phase when adjusting that or not. 

What cab and mic settings are you using?

Could also try adding the Simple Delay to one side with a 100% mix and 0% feedback.  It'll let you set the Time as low as 0, and increment 0.1 ms increments.

I'm using amp+cab.

I did try moving the mic position using the "distance" function in the cab settings.  (doesn't seem to be another way to do it)

Delay is an interesting idea.

[JulienVD]

17 hours ago, waymda said:

 

1. Yes - and have run up to four amps for very specific blended sounds - at all times using separate amp and cab setups - typically streaming 2 amps into a dual cab block panned hard left and right before the block

2. No never - the biggest 'issue' I've had is balancing the output of individual amps which is just down to channel volume

3. It depends on what I'm doing but I use the litigator a lot - sometime 2 of them with different settings, JC120, Vox, Plexi, The boogies, Baddonk, the twin models, the princess (now its available) - I tend to try lots of amps in combo to find the thing I'm looking for and may blend them differently in snapshots and/or push them differently at different point - not lots of help I know, but I work across a very wide range of styles

When hard panning the amps or cabs, the effect of a bad phase relation will be less noticeable.  It might add even add something that you like.  A certain "spread'.

Flipping the phase on two amps blended in mono will let you hear what is being lost due to comb filtering.

Ideally, you'll hear a pleasant sound when in phase and a tiny, weird sound when out of phase.

If both are a bit problematic, then there is likely a phase issue usually remedied by moving a mic. (in the real world)

 

In my case, I had spread both amps 60% each side and found the result a bit "woofy"

When I flipped the phase on one of the amps, the sound was usable, but scooped. 

That's when I realized that I had comb filtering altering my sound in a way that I didn't like and couldn't control.

 

 

 

[JulienVD]

13 hours ago, rd2rk said:

 

Attach a copy of the preset you're using and we can test it on our Stomps.

I just did a factory reset after the 3.0 update. 

Figured i could do a better job starting over.

If I run into the same problem today, I'll send the patch.

[waymda]

Did some testing using:

• a split A/B path even split
• WhoWatt amp and Cab default settings
• Join block default

When the B Polarity is inverted 0 output gain - so phase aligned

 

When either mic distance from cab or different mic used - varies amounts of output gain.

 

So I can only conclude that phase alignment is affected by mic choice and mic distance - although more accurately it is probably time alignment resulting in phase misalignment.

 

I don't know if this has always been the case, but have never noticed before (on V3.0 now).

 

I also tested with separate amp and dual block approach in my DAW and got the same result when inverting one of the tracks (recorded each separately).

 

 

[donkelley]

Well, in REAL miking, phase is definitely affected by microphone position/distance.  That's why in a studio you'll see the tracking engineer come and move one mic just like 1/4" inch or so closer or further from the cab and then go check things while you play, over and over until it's just right.

 

Maybe line6 actually emulates mic position with time delay as well.... I dunno.

[waymda]

I'm pretty sure that's what is happening - I found a post from 2018 that mentions the need to realign phase alignment when using different cab/mic/distance combinations.

 

I don't know why,  but I thought/assumed (and you know what happens when you do that) that the L/R signal from the Helix would be phase aligned.

[waymda]

Having finally realised the need/desirability of ensuring time/phase alignment in dual amp/cab setups (so slow) - can any one advise how to determine the delay required?

 

The comment below the video about this is unanswered and any advice welcome.

 

"Wondering about the phase shift. The 10 ms delay: Did you just "trial and error" or is it just a good common solution (and value) when you run into this issue? Thanks for another great video full of good insights. And yes - I am a sucker for Sultans as well :-)"

 

 

 

[CraigGT]

On 11/28/2020 at 5:12 AM, waymda said:

Did some testing using:

• a split A/B path even split
• WhoWatt amp and Cab default settings
• Join block default

When the B Polarity is inverted 0 output gain - so phase aligned

 

When either mic distance from cab or different mic used - varies amounts of output gain.

 

So I can only conclude that phase alignment is affected by mic choice and mic distance - although more accurately it is probably time alignment resulting in phase misalignment.

 

I don't know if this has always been the case, but have never noticed before (on V3.0 now).

 

I also tested with separate amp and dual block approach in my DAW and got the same result when inverting one of the tracks (recorded each separately).

 

 

 

I'm not sure that test is entirely valid. 

 

With B polarity inverted you only hear the difference between the two signals so with the same amps, cabs and mics the difference should be zero and that's what you hear.

If you use different mics then you would expect to hear the difference between the two mics,  whether that's due to phase response or frequency response is irrelevant, it just proves that the mics are different. the same goes for anything else on the parallel paths.

[qwerty42]

What @CraigGT said ^^^

 

You're only going to get total cancellation (or something approaching it) if both signal chains are using the same amps, effects etc. And indeed if you do this (at least on my own Helix), things sum to zero or very close to it.

 

If you have two different sounds, effects, whatever and mix them back together, the most you're going you get is filtering of what they have in common, which might not even be much especially if they've been altered by other effects like modulation.

 

FWIW I have intentionally tried forcing phase offset problems in my own patches as a test in the past using the stock cab blocks, and I was never able to do it. If the Helix stock cabs are IR-based, they behave like MPT IRs where phase alignment isn't an issue. I've never been able to see any phase shift from varying mic selection or distance parameters on the stock Helix cabs.    edit: The Jason Sadites video that @waymda posted above seems to clearly show phase alignment issues coming from two Helix cabs in stereo, so I might be totally wrong about this! Will look into it further...

 

I guess I should add though -- if you are using stereo IRs (by that I mean two mono IR blocks in stereo paths), then yes you can create comb filtering problems if the IRs aren't made correctly and aren't minimum phase transformed. And also if you have a delay somewhere in one of your paths which is forcing an offset, or other time-based modulations, those of course can create static or dynamic phase cancellation when L+R get mixed back together. In those cases, you might find it helpful to add the simple delay block mentioned by @CraigGT above, or you can also use the 'dual delay' on the stereo path with them hard-panned. Flip the final mix to inverse polarity, and adjust the delay until it has the 'most-cancelled' sound. Then when you flip the polarity back, that sound give you the least-cancelled result, but again you're not going to hear anything approaching total cancellation unless the stereo effects paths are very similar prior to combining them.

yet another edit: another thing I often do with stereo patches is collapse the path to mono at the very end of my chain to see how it sounds, partly for a phasing check and partly just to make sure things that sound glorious in stereo don't sound muddy/thin/awful on a mono playback device. You can do that by temporarily putting a mono gain block at the end, which barely uses any DSP. With it bypassed it won't do anything, but with it enabled it will immediately sum to mono so you can toggle back and forth to A/B them.

Edited December 4, 2020 by qwerty42
speakin outta my posterior maybe

[qwerty42]

On 11/24/2020 at 1:07 PM, OmniFace said:

Could also try adding the Simple Delay to one side with a 100% mix and 0% feedback.  It'll let you set the Time as low as 0, and increment 0.1 ms increments.

This is what I do too if I'm intentionally trying to force a shift.  If you run the stereo version of simple delay, you can even get sub-0.1 ms increments by using the 'scaling' parameter (it sets the relative length of the delay for the right channel compared to the left. So if the base value is 0.1 ms, and you set the scaling to 50%, the right channel will be delayed by 0.05ms while the left is 0.1ms).

[waymda]

Thanks @qwerty42 and @CraigGT - I really hadn't fully thought that through, and it possibly explains why I'd never perceived it to be a problem before reading the post

 

I wonder if the explanation "it sounds different because it is different" answers the OPs concerns?

[qwerty42]

2 hours ago, waymda said:

Thanks @qwerty42 and @CraigGT - I really hadn't fully thought that through, and it possibly explains why I'd never perceived it to be a problem before reading the post

 

I wonder if the explanation "it sounds different because it is different" answers the OPs concerns?

Yeah, I think OP has the right idea, it's just that the cabs and mics in Helix don't respond the way they would in real life, where distance would definitely affect phase alignment. (If they did, I can just imagine all the posts we'd get about 'when I use stereo cabs everything sounds thin and quiet!')

But if he's getting some undesirable phase behavior from other parts of his chain, then using the delay trick mentioned above is a way to accomplish the same thing as moving the mics would. The only time I've ever had to do this was when I was using two IRs that hadn't been trimmed correctly, but maybe there are other ways to create phase problems in Helix that I just haven't discovered yet... (using the effects loops in a parallel path would be one possibility!)

[qwerty42]

On 12/2/2020 at 3:35 AM, waymda said:

Having finally realised the need/desirability of ensuring time/phase alignment in dual amp/cab setups (so slow) - can any one advise how to determine the delay required?

 

The comment below the video about this is unanswered and any advice welcome.

 

"Wondering about the phase shift. The 10 ms delay: Did you just "trial and error" or is it just a good common solution (and value) when you run into this issue? Thanks for another great video full of good insights. And yes - I am a sucker for Sultans as well :-)"

 

 

 

Hmmm... I just watched this to see what you were referring to, and that is very interesting! I'm going to have to play with those two amps and see what's going on there. Looks like I could definitely be wrong about the stock cabs/mics always playing nice together without phasing issues. He doesn't have anything else in his chain prior to that which might be causing a phase shift, so it has to be either the amps or the cabs. Thanks for sharing this and pointing that out.

At any rate, the fix would still be the same, using a delay just like Jason did. I don't know of any other way to adjust phase offsets with Helix (but maybe I'm wrong there too!)