These forums are read only, please use our new forums here.

Main :: POD HD


Support forums for all POD HD products


HD500 FX Loop ... level drop or not?
by jimsreynolds on 2012-10-27 14:02:33

OK, there is some discussion in Meamboobo's high gain guide thread as to whether the FX loop is losing signal or not.    I want to get to the bottom of this and understand what drops/boosts exist (if any) hence the thread here.  This is key to some setups, not least of which is the 4CM configuration that I have been documenting.  Some sample previous posts are this, this and this.

To start the ball rolling, I have done some analysis of the input stages.  I use a setup like the one depicted below to test with (click on the pic to zoom in).  I use a true bypass loop pedal flipping between

1)  guitar to amp, via pod input and FXloop send

2)  guitar direct to amp. 

This allows me to do an instantaneous A/B test by flipping the pedal on and off and listening.  In practice, using a clean channel on the amp, I can hear a difference down to a single dB if I listen hard.to a sustained note decay and flip the pedal on and off.

bypass.gif

With this setup there is just the Pod circuitry between the  Guitar input and Loop Send to account for.  This entails:

  • The Pad switch ('normal'),
  • Input setup (1 = Guitar, 2 = Variax),
  • FX loop level switch('Stomp')
  • FX Loop Send level (0dB) to account for.  

With this setup, my tests have shown that I need a whopping +12dB of boost to get unity gain between guitar only and 'via pod' (courtesy of a Studio EQ).  This drops to +6dB with an input setup of Input 1= Guitar, 2=Same.

That is my observation but I think that others are reporting  unity gain with no boosts required and just a straight cable in the loop.  To me this suggests  that either:

1)  the signal is being boosted again after the loop at some point.

2)  there is an inconsistency across individual Pod's within the loop circuitry.  

I am going to do some more testing but am inviting others to give their views and observations on this.



Re: HD500 FX Loop ... level drop or not?
by meambobbo on 2012-10-27 15:42:31

I just wanted to point out my only real observation was the loop itself didn't change the level with just a patch cable in it.  So with 0 send and 0 recieve, toggling the fx loop on/off made virtually no change in tone other than a very mild noise increase (which is worse when you have a lower signal)



Re: HD500 FX Loop ... level drop or not?
by jimsreynolds on 2012-10-27 16:29:53

Cool Bobbo, gotcha.   I need to confirm the patch cable test myself.  If I get the same result as you then the implication is that there is a boost on the FX return or further downstream.  I need to work out some creative cabling with my bypass pedal to check this out ... should be doable but its late so mañana.now.

I am going to check:

  • Pod Guitar Input to Pod left 1/4" Output (no Amp, FX or loop, mixer path A at 0dB, mixer path B at Mute, Mixer Pans at centre, other settings as before).  My expectation is that I should see unity gain with the Master at 100%.  If not, I will find a reference level).

  • Pod FX Return to Pod left 1/4" Output.   In theory this should be unity but presently the evidence is suggesting the contrary.


Re: HD500 FX Loop ... level drop or not?
by jimsreynolds on 2012-10-28 13:31:15

Hokay, here is a first set of test results based on simple case of Guitar In --> 1/4" Left output (i.e. Output forced to Mono).  

I am not going to try and interpret anything right now cause there is more testing to do.  The general setup was as follows.

ParameterSetting
Master100%
Input PadOff
1/4" Output level

Amp

InputGuitar In
Output

1/4" Left out

AmpNone
FXNone excepting an FX Loop.  The FX Loop Block is in situ in all cases but is turned off when not being tested. 
FX Loop configuration

Patch from FX Send to FX Return Left.  Send and Return levels at 0dB, Mix at 100%.  FX Loop Level = 'Stomp'

Mixer ConfigurationPositioned last in the signal chain.  Modified as per test results below.

In each case I ran four tests based on the mixer configuration without an FX Loop, then with an FX Loop.  The Mixer Pan options have a significant effect on the levels and have been tested.  All results are expressed as the correction to the mixer required to get unity gain at the output.

The first set of tests used the standard Guitar/Same input configuration

Input 1Input 2Mix A PanMix B /PanFX LoopResult
GuitarSameCCNoneAbove Unity.   -3dB per channel to correct
GuitarSameLRNoneBelow Unity.   + 4 dB per channel to correct
GuitarSameCMuteNoneBelow Unity.   + 4 dB on channel A to correct.  Channel B not responsive
GuitarSameLMuteNoneBelow Unity.   + 10 dB on channel A to correct. 
GuitarSameCCOnUnity Gain !!!!
GuitarSameLROnBelow Unity.   + 4 dB per channel on the mixer to correct
GuitarSameCMuteOnBelow Unity.   + 4 dB on channel A to correct. 
GuitarSameLMuteOnBelow Unity.   + 10 dB on channel A to correct. 

The second set of tests used the common Guitar/Variax input configuration

Input 1Input 2Mix A PanMix B PanFX LoopResult
GuitarVariaxCCNoneBelow Unity.   + 4 dB on channel A to correct.  Channel B not responsive
GuitarVariaxLRNoneBelow Unity.   + 10 dB on channel A to correct.  Channel B not responsive
GuitarVariaxCMuteNoneBelow Unity.   + 4 dB on channel A to correct
GuitarVariaxLMuteNoneBelow Unity.   + 10 dB on channel A to correct. 
GuitarVariaxCCOnBelow Unity.   + 4 dB per channel on the mixer to correct
GuitarVariaxLROnBelow Unity.   + 10 dB per channel on the mixer to correct
GuitarVariaxCMuteOnBelow Unity.   + 10 dB on channel A to correct. 
GuitarVariaxLMuteOnBelow Unity.   > + 12 dB on channel A to correct


Re: HD500 FX Loop ... level drop or not?
by CairnsFella on 2012-11-01 00:26:48

Thanks for putting time into this (I love these sorts of stats).

From my initial glance, there are enough consistencies to begin to draw some conclusions, but there does seem to be a bit of an anomoly. I am not at my most alert at the moment though, so I will have another look when I have more time to take it all in.



Re: HD500 FX Loop ... level drop or not?
by jimsreynolds on 2012-11-01 03:53:55

Still need to work up the numbers around the

     Input -->  FX Loop Send

     FX Loop Return --> Output stages

have been busy patch building for a cover set so distracted at the mo.  

I guess when I look back through the chart, the numbers look pretty consistent with my experience but I had not appreciated how much difference a combination of inputs, loop and Pan could make in attenuating a signal on the way through - over 12dB in the worst case, i.e. more that the mixer alone can cope with.



Re: HD500 FX Loop ... level drop or not?
by Astaroth_CY on 2012-11-01 04:01:37

Your move, L6.



Re: HD500 FX Loop ... level drop or not?
by jimsreynolds on 2012-11-01 04:18:55

I think that with the various options available there are bound to be variations introduced into the gain structure throughout the system.  I just hadn't previously registered that certain combinations of  settings  could result in such extreme gain drops.   I would not expect Line 6 to 'fix this'. I guess they could try some kind of expert logic to rebalance gain within the system based on combinations of  settings but that could potentially create a new can of worms with people trying to understand why levels have changed.   I would rather just understand what is happening and move on.

I see this as a documentation question.   There is suggestion that this kind of detail should be included in the Advanced Guide.  I am ambiguous about this because information deluge can be counter productive to people who are relatively new to the pod. 

Instead, I guess I would rather see Line 6 include this level of detail as 'Documents' on this site and referenced in the forums.  That way they are more dynamic, can be commented on and are available to those who seek the deeper knowledge.  The info does not clutter up the advanced guide .... which I really view as the advanced guide cause it tells you the basic operations rather open up the design ethic and advanced considerations

My 10 pence worth.



Re: HD500 FX Loop ... level drop or not?
by Akeron on 2012-11-01 04:36:08

If the documentation was more clear you wouldn't have 90% of the posts on this forum. Maybe it's done on purpose to keep the forum alive?



Re: HD500 FX Loop ... level drop or not?
by jimsreynolds on 2012-11-01 04:47:13

(Facepalm !)  



Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-01 05:36:30

Where did you have the FX Loop in the signal path?  If I read your one post correctly, it says that you have the mixer last in the chain, which would imply that your FX Loops is somewhere before it.  So why would doing anything to the mixer levels affect the signal going out of the FX Send?  Unless I have you wrong.

I'm going to do some testing this morning by sending a sine wave into the unit and measuring the signal using a meter.  Will let you know what I come up with in this thread.



Re: HD500 FX Loop ... level drop or not?
by TheRealZap on 2012-11-01 05:40:04

Akeron wrote:

Maybe it's done on purpose to keep the forum alive?

nice conspiracy theory.... but considering that L6 pretty much abandoned ship as far as the forums go...

i think they'd prefer it simply died.



Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-01 07:52:03

Edit:  Sorry, this test was flawed.  Be right back with a real one.



Re: HD500 FX Loop ... level drop or not?
by daferalo on 2012-11-01 08:24:45

Yes Karl, but in Jimsreynolds' test, there is a patch cable placed between FX send and FX Left (only) return, so that the output is the 1/4" left out.

It is quite interesting your results in comparisson to Jim's ones, as one can conclude that the sound drop only takes place when the signal is returning again to the POD.

Another interesting thing to me: According to Jim's results (post #3) at first set of test, and I guess that Jim has the FX loop placed before the path split (which is not clear), one should expect a sound drop in all cases, when comparing FX loop on Vs FX loop off, but it is not the case. It only applies when Pan A and Pan B are at center position.

Best Regards,

Daf



Re: HD500 FX Loop ... level drop or not?
by Akeron on 2012-11-01 08:26:51

In fact mine was an attempt at sarcasm I'm not that optmistic on the future of the product and the company itself...



Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-01 09:43:33

daferalo wrote:

Yes Karl, but in Jimsreynolds' test, there is a patch cable placed between FX send and FX Left (only) return, so that the output is the 1/4" left out.

You're right!  I must apologize, I read it wrong.  Okay, let me redo the test the correct way and measure the output.



Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-01 10:45:00

Okay, my first test was flawed.  I wasn't measuring what I thought I was measuring.  Reboot!   I had to calibrate the ouput of my UX-2 with an external meter.  Properly calibrated, here are the results:

Patch has FX Loop first in chain, Mixer last, nothing in between.

HD500 config: Guitar/Same

Input signal: -12dB

FX Send signal: -16dB

Left mono out signal (loop bridged): -10.3dB

Loop disengaged output signal: -10.15dB

So, going straight through you have to reduce mixer volume by 2dB on both the left AND right channel (because they are summed) to get unity gain.  With the FX Loop engaged and jumped, this still holds true.

The strange part here is that for 4CM purposes, it would appear that you are hit with a 4dB cut in signal going out to the amp from the FX Send.  That's not cool at all.

HD500 config: Guitar/Variax

Input signal: -12dB

FX Send signal: -22dB

Left mono out signal (loop bridged): -16.3dB

Loop disengaged output signal: -16.15dB

Yuk...Guitar/Varlax is just way worse in this scenario.

Now, let's change the FX Loop position to AFTER the output mixer:

HD500 config: Guitar/Same

Input signal: -12dB

FX Send signal: -10dB

Left mono out signal (loop bridged): -10.3dB

Loop disengaged output signal: -10.15dB

Now the FX Send signal is 2dB hotter than the input signal.  To get unity gain, you have to decrease the output mixer setting by 2dB on both R and L.  With the loop disengaged this also gives unity gain at the output.  If you only decrease the FX send level by 2, when you disengage or remove the loop, you will get a 2dB boost at the output.  Not desirable if you're running right into the front of the amp from the output.

For giggles:

HD500 config: Guitar/Variax

Input signal: -12dB

FX Send signal: -16dB

Left mono out signal (loop bridged): -16.3dB

Loop disengaged output signal: -16.15dB

Ugh...why even bother with that?

---------------------------------------------------------------------------

So what does all this tell us?  Well, for starters, I'm going to have to set up my patches for 4CM or as a pedalboard straight into the amp to use very differently from now on to ensure that I have the proper signal hitting the amp:

Unity gain at all points configuration:

Input Config: Guitar/Same

Mixer: Pre-FX Loop, -2dB R&L

FX Loop: Post-Mixer, if used

I don't pretend to understand why this is so, but I do know that it was a helpful exercise because getting unity gain at all points in the signal path with this unit doesn't happen by default.  You have to work for it.  There is no standard, default configuration that gets you there, but the one I just mentioned will get you unity at all points in the path and would appear to be the way to go.

A word of caution about my results:  I'm using an HD500 beta unit that was given to me well before these things were mass produced.  I kinda doubt this, but it is possible that maybe the hardware on production units is different and is producing different output levels.  Again, probably not a concern, but I want to cover my bases here.

--------------------------------------------------------------------------

For those that also might own a Boss GT-10, using this same methodology, I've found the following to get unity gain:

Output level: -10dB

Output level knob: 12:00

FX Send: 120

FX Return: 80

This surprises me because I've always thought the default settings of 100 and 100 on the send and return levels worked well.



Re: HD500 FX Loop ... level drop or not?
by daferalo on 2012-11-01 11:43:48

Very interesting Karl, thank you and Jim for the time and effort at doing these tests. So, it seems that the sound drop is located at the FX loop send jack by 4 db, even if the send level is at 0 db?



Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-01 11:46:59

Well, that depends on the configuration.  Note that the one that gets you closest actually has a 2dB BOOST at the effects send.  Not a drop. 



Re: HD500 FX Loop ... level drop or not?
by Akeron on 2012-11-01 11:48:36

Maybe this could be interesting for the discussion

http://www.grumpyrocker.co.uk/signal-level-4cm-pod-hd




Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-01 12:28:08

Okay...this is very interesting.  If I'm reading that article right, and set it up the way he did with everything in one branch before the mixer, that side centered, and the other side muted, I read a 4dB drop at the effects send and the main output.  That's bad enough.  The other problem is that there is no way, by compensating with the output mixer, to get the proper level at the effects send.  So I'm concluding that the way he's configured his POD is not optimal.  He can't compensate for the level drop without adding a gain block before the send.  Not a great way to do this.



Re: HD500 FX Loop ... level drop or not?
by Akeron on 2012-11-01 12:56:01

He's a member of this forum, so maybe he could answer that, like in

http://line6.com/support/message/388854#388854">http://line6.com/support/message/388854#388854">http://line6.com/support/message/388854#388854



Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-01 13:44:24

Yep, that's the way I did it.  Doesn't work as well as the method I outlined above.  Using my method, you don't have to waste a block with a boost for the FX send.



Re: HD500 FX Loop ... level drop or not?
by jimsreynolds on 2012-11-01 17:52:09

Nice bit of testing Karl.  I Need to flag up a couple of considerations though.

You approach of mixer before FX Loop definitely saves you an effects block (i.e. Studio EQ to boost the FX send level).  The main downside though is that it does not allow you to use the mixer later to trim off the patch level.  Why is that important?  Well a couple of things, both related to 4CM.

  • The levels put out by many amps/preamps on the HD500 are significantly lower than those provided by some 'real' preamps.  This is a big old pain in the butt when you are trying to use a combination of real preamps and modelled pre-amps.   A fix for this is to set the 1/4" output on the HD to Line level, which pulls up the level of the modelled amps to the same gain levels as the real ones.  Of course by doing this you have also now boosted the output level of the real preamps which you now need to knock back down again .... using the Mixer.  With the mixer just before the loop you cannot do this effectively as all it does is cut down the input level to your amp.

  • If you have a multi channel amplifier with several modes individual modes/channels have quite a big volume difference.  The mixer allows you to balance volumes across channels/modes more effectively than the amp controls themselves.  For example the clean channel on the JVM410 has three gain levels, the last of which is percieved as way louder and more dynamic than the other two.  It is easy then to use the mixer at the end of the patch to balance the levels across all channels. This will not affect everyone with 4CM but for some of us it is very useful indeed.

There is a further trick relating to the pad switch which I use (but have not previously documented).   I tend to setup all my 4CM patches with the pad switch turned on and then boost the Studio EQ before the loop by 3dB to bring the levels back up to the point where they would be if the pad switch was turned off.  The reason for this is that I use mainly guitars with fairly powerful humbuckers but I also have some guitars with weaker single coils.   With the humbuckers I run with pad on but with single coils I run with Pad off.  This compensates quite nicely for the level drop between the two pickup types and is an easy change to make between songs/guitars.

None of these points are show-stoppers to the approach that you describe with Mixer before loop but these are the reasons why I have described the mixer at the end of the setup in my 4CM documents and a Studio EQ handling the boost. 

Finally on the Input 2 = Variax thing .... yeah, I know it is a point of contention on the board even now but many folks (including me) prefer the tone of the pod with that configuration, despite some of the knock on effects.  Each to their own preference on that.

These are all just considerations though and, if you do not object, I will tack on your approach as a variant on the setup to 4CM document that I have up on the forum (when I get time) as it may be useful to some people who can live with the limitations and need that extra effect block available.



Re: HD500 FX Loop ... level drop or not?
by daferalo on 2012-11-01 21:38:59

Karl_Houseknecht escribió:

Well, that depends on the configuration.  Note that the one that gets you closest actually has a 2dB BOOST at the effects send.  Not a drop. 

No Karl, from your results you would always have a 4 db drop in the signal at the FX send jack (i. e. you should have a 6 db boost, not 2 db as the signal is supposed to be doubled by the mixer). The interesting thing is that the FX send level is at 0 db. Would you believe it is due to a physical impedance mismatch at the jack? Have we discovered a bug?

Best Regards,

Daf



Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-02 04:48:39

daferalo wrote:

Karl_Houseknecht escribió:

Well, that depends on the configuration.  Note that the one that gets you closest actually has a 2dB BOOST at the effects send.  Not a drop. 

No Karl, from your results you would always have a 4 db drop in the signal at the FX send jack

But my results show that there's a 2 dB boost at the FX Send jack if the Loop is after the mixer and the mixer is at 0dB.  Are we saying that the mixer somehow adds 6dB of of boost even when it's at 0?  That doesn't really make any sense.

I'm not sure what we've found here, bug or whatever else, but it isn't very consistent, is it?



Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-02 04:57:45

jimsreynolds wrote:

Nice bit of testing Karl.  I Need to flag up a couple of considerations though.

You approach of mixer before FX Loop definitely saves you an effects block (i.e. Studio EQ to boost the FX send level).  The main downside though is that it does not allow you to use the mixer later to trim off the patch level.  Why is that important?  Well a couple of things, both related to 4CM.

  • The levels put out by many amps/preamps on the HD500 are significantly lower than those provided by some 'real' preamps.  This is a big old pain in the butt when you are trying to use a combination of real preamps and modelled pre-amps.   A fix for this is to set the 1/4" output on the HD to Line level, which pulls up the level of the modelled amps to the same gain levels as the real ones.  Of course by doing this you have also now boosted the output level of the real preamps which you now need to knock back down again .... using the Mixer.  With the mixer just before the loop you cannot do this effectively as all it does is cut down the input level to your amp.

  • If you have a multi channel amplifier with several modes individual modes/channels have quite a big volume difference.  The mixer allows you to balance volumes across channels/modes more effectively than the amp controls themselves.  For example the clean channel on the JVM410 has three gain levels, the last of which is percieved as way louder and more dynamic than the other two.  It is easy then to use the mixer at the end of the patch to balance the levels across all channels. This will not affect everyone with 4CM but for some of us it is very useful indeed.

Both interesting and very valid use cases that I probably wouldn't have thought of, especially the second one.  I've always used the amp channel volumes on my multichannel amps to balance the volume.  Never touched the HD's mixer, but it certainly makes sense the way you explain it.  Regarding the differing preamp volumes, I guess the ones I typically use are more consistent volume-wise across preamps, so I never found the need to level them with anything other than channel volumes.

I think the bottom line here is that the HD has a level "problem".  Whether that problem is by design or chance remains to be seen.  In the meantime, maybe to get the best all around use the HD might require external help in the form of a line level shifter like the Boss LS-2 or even a clean EQ pedal that has a level control.  Put that in the FX Loop chain right after the send so you can boost the signal up a few dB and free up the internal mixer for the uses above.



Re: HD500 FX Loop ... level drop or not?
by jimsreynolds on 2012-11-02 10:51:44

Karl_Houseknecht wrote:

maybe to get the best all around use the HD might require external help in the form of a line level shifter like the Boss LS-2 or even a clean EQ pedal that has a level control.  Put that in the FX Loop chain right after the send so you can boost the signal up a few dB and free up the internal mixer for the uses above.

Yep, that idea is a keeper.   Another device, PSU and patch cord  to manage I guess but it covers all the bases and falls into the 'set it and forget it' category of gear,

A lifetime ago I offered up a feature request to allow the loop send to boost as well as cut but I am not overly optimistic that this will find its way into firmware.  Que sera.



Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-02 10:53:31

jimsreynolds wrote:

A lifetime ago I offered up a feature request to allow the loop send to boost as well as cut but I am not overly optimistic that this will find its way into firmware.  Que sera.

Something I will certainly lobby for myself.  It really should also have a boost like the GT series loop does.  Don't know if that's physically possible or not but worth a shot asking for.



Re: HD500 FX Loop ... level drop or not?
by MartinDorr on 2012-11-03 10:58:31

Thanks to both of you for checking deeper into this topic. This is really interresting. I have a few questions (forgive me if I missed it while skimming through the thread):

- Did you use a stereo or mono effect loop patch cable?

- How was the line/stomp switch set in Karl's experiment?

Two things I noticed that did not seem to be mentioned:

- The difference of 6dB between Guitar/Same and Guitar/Variax seems to indicate 2 summing operations, i.e., adding the input signal 4 times. A single summing of 2 signals with the same level should only get you 3dB. I fear that the HD internal digital signal is being intentionally lowered to assure we're not running out of bits, then the HD effects assume some usual summing via mono effects and amps, and then an effects loop which would normally be betyween the pre and power amp stages. I guess what I am trying to say is that the experiment may be an unusual use mode that makes it look different (hopefully worse) than it normally is. Obviously it would help a lot if we'd know what's really going on so one can plan for it.

- On the surface it looks like big signal level losses but on the other hand we deal with preamp output signal levels that can easily exceed a 100 dB range if you use the full range of channel volume and drive. And even if you change only one of them (e.g., Channel Volume=50%, Drive=0-100%) we still get a range 50-60 dB in the output level for clean Amps. So if you want to run your amp in the clean / uncompressed range (low Drive and low Channel Volume) you need to dial in large gain values after the preamp and usually before any effects loop that can't be done with the mixer alone (it only gets you +12dB). Again, I think the key missing thing is Line 6 documentation to get out of the guessing and measuring game and back into dialing in tones (while actually seeing signal levels and clipping indicators in the HD editor or HD edit menues) and playing.

Thanks again for your work and detailed results. I guess we slowly build up what's not document and live with lots of remaining uncertainty.

Martin



Re: HD500 FX Loop ... level drop or not?
by jimsreynolds on 2012-11-03 13:16:00

The FX Loop now detects if a mono link cable is used and mixes/sums/whatevers to mono.  In my case I used a single mono patch cable from Send to Left Return. Karl can confess his own sins.

MartinDorr wrote:

Obviously it would help a lot if we'd know what's really going on so one can plan for it.

I think that is the bottom line.  We can test, document and draw conclusions (right or wrong) but, from what I see, Line 6 are reluctant to make changes to some of the fundamental operations on the POD unless there is a clear bug in evidence that affects a significant number of users.   This makes sense because changes to things like how levels are mixed/summed wouldl cause merry hell with many peoples configurations.  I assume that issues such as these will not necessarily be fixed then (but will be happy to be proven wrong ).so plan B is to understand and document them well enough to be able to effectively compensate.



Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-05 09:00:28

jimsreynolds wrote:

The FX Loop now detects if a mono link cable is used and mixes/sums/whatevers to mono.  In my case I used a single mono patch cable from Send to Left Return. Karl can confess his own sins.

I used both a mono cable from the send to the return, and a stereo insert cable with stereo from the send into both right and left returns.  This does not affect the results.



Re: HD500 FX Loop ... level drop or not?
by MartinDorr on 2012-11-05 09:26:26

Couldn't agree more. Something does not seem to add up. Can't check anything right now. Maybe more ideas tonight. One thing I noticed in the past: If you center both L and R you get 3DB more output than when you only use L or R and mute the other on the Mixer. There may be some things going on in the Mixer in relation to how it treats its 2 stereo inputs that we do not fully understand. I am also not sure whether Karl's result comparing stereo verus mono patch cable on the effects loop really makes sense because I though the loop is a stereo effect, i.e., it would support 2 independent signals to be looped and thus if a mono cable is plugged in I would expect summing to occur and a raise of 3dB in the effects loop out voltage level. But obviously this is not what you found.



Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-05 13:07:40

Right, the summing makes no difference in overall level in the loop if you run it mono.



Re: HD500 FX Loop ... level drop or not?
by daferalo on 2012-11-05 20:47:29

MartinDorr escribió:

- The difference of 6dB between Guitar/Same and Guitar/Variax seems to indicate 2 summing operations, i.e., adding the input signal 4 times. A single summing of 2 signals with the same level should only get you 3dB.

Hi Martin. This is not necessarily true, as it depends on the "coherency" of the signals. Well, that is true only if you have two incoherent signals. But if you have two coherent signals (the same frequency), then you should consider also a "phase" term , which should be the case when you have input 2 as "same". And if there is no "delay" between those two signals (known as "constructive interference"), then the sum would bring a 6 db increase, due to the "phase" term. This increase has been proved before by another forum member (maybe MeamBobbo or Hurghanico?). Anyway, I think Meambobbo believes that the phase is other than 0°.

Another thing to consider. Summing two acoustic signals could be different than summing two electric signals. If you sum two signals based on its amplitudes, then you would have around a 6 dB increase, but if you sum two signals based on its power quantities, you would have around a 3 dB increase. What does the POD do?

Best Regards,

Daf



Re: HD500 FX Loop ... level drop or not?
by daferalo on 2012-11-05 21:23:15

Karl_Houseknecht escribió:

 

But my results show that there's a 2 dB boost at the FX Send jack if the Loop is after the mixer and the mixer is at 0dB.  Are we saying that the mixer somehow adds 6dB of of boost even when it's at 0?  That doesn't really make any sense.

I'm not sure what we've found here, bug or whatever else, but it isn't very consistent, is it?

This is something I would expect due to the way the experiment has been carried out. Anyway theory is one thing and practice could be another one.

I am not sure about what you have found here (I mean, if it is a bug or not). Today I suspect that the 4 dB decrease at the FX loop send jack is intentional, as the FX loop send level is additionally affected by the level switch (line Vs stomp). I guess your tests were done at stomp level maybe?

There is something very funny to me: In spite of the fact that there is a level difference at the FX loop send jack by placing the FX loop effect before or after the mixer (which can be explained according to the concepts of "monosizing a stereo signal" and "summing two signals"), the POD HD device seems to be smart enough to know that, so that you always got the same signal level at the 1/4" output jack, no matter if the FX loop effect was placed before or after the mixer!!!

Best Regards,

Daf



Re: HD500 FX Loop ... level drop or not?
by meambobbo on 2012-11-05 21:39:51

my tests show the FX Loop created a 40 sample or ~42 ms latency in whatever channel it was placed in.

it's important to note that two audio signals can be out-of-sync, whereas two specific frequencies can be out-of-phase.  When two identical, or near identical, signals are out-of-sync, some frequencies will be in phase, while most will be out-of-phase, to varying degrees.

and I have also noticed there is a difference in the way the loop works when before an amp model vs. behind one.  I find it adds much more noise in "pre" than "post" regardless of the levels fed into it.



Re: HD500 FX Loop ... level drop or not?
by MartinDorr on 2012-11-06 06:50:33

The simplest explanation I could come up with was that

  1. The Input 2 setting of Same causes a -4dB change in the HD internal signal level sent out of the effects loop
  2. The Input 2 setting of None (Variax, Mic, Aux not connected) case a change of -10dB for the above (at least when you come in via Guitar as Input 1).
  3. The Mixer adds +5.7dB even when set at 0dB (at least for the no Amp use case).
  4. The effects loop in or out of the path causes a level change of 0.15dB (I found the same effect with the Studio EQ)

This seems to be the simplest explanation, but that does not make it necessarily right.

I have no time to check anything for a while but thought of doing a little experiment using a Studio EQ in front of the Mixer with gain +4 or +10dB to make up for the suspected drop on the input. If the 0dB Mixer setting really causes +6dB gain on the mixer (or if the input did not get lowered and the Studio EQ gain will exceed 0dB)  we should see clipping on the output (or more than 0dB output). Now if the Studio EQ caused the clipping we should see the output to come down -6dB if we pull the mixer down -6dB. On the other hand if the mixer caused the clipping (because it really added 6dB gain on the 0 setting) we should see the output not come down (or not below 0) if we pull the mixer down -6dB because the Studio EQ will drive 0dB to the mixer.

I can't do Karl's experiment setup due to lack of equipment. But I can come in with a Variax on the digital input and check output via USB. From the past I already know that for a specific guitar model I will have -10.7dB out with a pre Mixer Studio EQ and a post Mixer Studio EQ, no amp and mixer set at 0dB. Hope that works out and we can determine whether the 0dB mixer setting is really something else ... or we have more puzzles .



Re: HD500 FX Loop ... level drop or not?
by daferalo on 2012-11-06 08:55:07

meambobbo escribió:

my tests show the FX Loop created a 40 sample or ~42 ms latency in whatever channel it was placed in.

it's important to note that two audio signals can be out-of-sync, whereas two specific frequencies can be out-of-phase.  When two identical, or near identical, signals are out-of-sync, some frequencies will be in phase, while most will be out-of-phase, to varying degrees.

and I have also noticed there is a difference in the way the loop works when before an amp model vs. behind one.  I find it adds much more noise in "pre" than "post" regardless of the levels fed into it.

Hi MeamBobbo, thank you very much for your clarification. As you have pointed out, the proper term in this case should be "sycn" rather than "phase", as we have a group of waves rather than a single one. In fact, what you have described is the explanation of the "comb filtering" effect. Anyway, what I said should still apply, because when input 2 is set to "same", you have exactly the same group of waves going for the two inputs in the performed experiment.

I have to say that I am not an audio engineer and I am not familiarized with the "sample rate" concept for example (but I would like to!!! ), everything that I said is based on physics, so I might be missing something. I would expect always some degree of latency unless we have unlimited power processing capabilities. Of course, my words are not God's words, and I could be wrong!!!!

Best Regards,

Daf



Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-06 09:09:28

meambobbo wrote:

my tests show the FX Loop created a 40 sample or ~42 ms latency in whatever channel it was placed in.

That sounds awfully high.

Thinking about latency in terms that we can quantify:  Latency via distance from an amplifier:

Speed of sound in air at roughly 72F = 1131.29 ft/s

If I'm 10 feet from my amp, that translates to 8.8 ms latency.  Negligible, probably not able to feel that.  Typical use case, right?

What you're saying is that the latency of the effects loop is the equivalent of me being 47.5 feet from my amp???  That's definitely audible, and annoying I might add.  You couldn't play that way.  And I definitely play using the effects loop 100% wet, so something seems a little off about that result.



Re: HD500 FX Loop ... level drop or not?
by phil_m on 2012-11-06 09:30:33

If the latency introduced by the effects loop block is 40 samples, that works out to be less than 1 ms according to my math. Something doesn't add up here. Another way to look at it - if the sample rate is 48kHz, that means there's 48 samples every millisecond. So 40 samples is less than a millisecond. Even if the sampling rate is 44.1kHz, it would be less than a ms.



Re: HD500 FX Loop ... level drop or not?
by Karl_Houseknecht on 2012-11-06 10:00:34

Thanks for that, Phil.  I never was good with the maths.



Re: HD500 FX Loop ... level drop or not?
by phil_m on 2012-11-06 10:05:57

I was told there would be none...



Re: HD500 FX Loop ... level drop or not?
by daferalo on 2012-11-06 11:58:23

MartinDorr escribió:

The simplest explanation I could come up with was that

  1. The Input 2 setting of Same causes a -4dB change in the HD internal signal level sent out of the effects loop
  2. The Input 2 setting of None (Variax, Mic, Aux not connected) case a change of -10dB for the above (at least when you come in via Guitar as Input 1).
  3. The Mixer adds +5.7dB even when set at 0dB (at least for the no Amp use case).
  4. The effects loop in or out of the path causes a level change of 0.15dB (I found the same effect with the Studio EQ)

This seems to be the simplest explanation, but that does not make it necessarily right.

I have no time to check anything for a while but thought of doing a little experiment using a Studio EQ in front of the Mixer with gain +4 or +10dB to make up for the suspected drop on the input. If the 0dB Mixer setting really causes +6dB gain on the mixer (or if the input did not get lowered and the Studio EQ gain will exceed 0dB)  we should see clipping on the output (or more than 0dB output). Now if the Studio EQ caused the clipping we should see the output to come down -6dB if we pull the mixer down -6dB. On the other hand if the mixer caused the clipping (because it really added 6dB gain on the 0 setting) we should see the output not come down (or not below 0) if we pull the mixer down -6dB because the Studio EQ will drive 0dB to the mixer.

I can't do Karl's experiment setup due to lack of equipment. But I can come in with a Variax on the digital input and check output via USB. From the past I already know that for a specific guitar model I will have -10.7dB out with a pre Mixer Studio EQ and a post Mixer Studio EQ, no amp and mixer set at 0dB. Hope that works out and we can determine whether the 0dB mixer setting is really something else ... or we have more puzzles .

The test you proposed is very interesting. Mainly, due to the fact that the output level through USB is not affected by the master volume knob, as opposed to the 1/4" outputs. That is why it should not be very important the "number" measured at the 1/4" output in Karl's experiments (as it should be affected by the master volume position). But what I have found to be very interesting in Karl and Jimsreynolds experiments is the "tendency".

In fact, I have a question: the stomp level is an instrument level (i. e. -10 dB) and line level is +4 dB for the FX loop send? If so, it will explain the FX loop behaviour we have found here.

Best Regards,

Daf



Re: HD500 FX Loop ... level drop or not?
by MartinDorr on 2012-11-13 15:16:15

A small update with some interresting measurements via USB (adding more to the puzzle):

  1. The experiment did not work. I could not make Studio EQ to really clip as far as I can tell. It causes gradually increasing distortion/compression if its output gets above -6dB or so. Mixer will reduce or up output level as expected and dialed in no matter what the Studio EQ was set at. The mixer's dialed in values seem to exactly match measurements.
  2. Mixer with channel A and B panned hard left and right will produce 6dB less gain than with a setup where both channels are set to center. I checked that this is the mixer's doing by recording a small sample with the looper in pre position to check output and then moved the looper to post position (by passing the mixer) and checked again. I had no effects and no amp configured, i.e., only the mixer is measured.

Started to measure the Studio EQ's real gain knob setting to output level function to find what's going on with the compression/distortion. It can easily get up to 2dB off what you'd expect despite having 0.1 dB steps on the dial. The only real good match is around 0 and +/-18dB. But watch on the +dB side. It seems that everying resulting in an output signal larger than -6dB gets compressed and sounds distorted as if clipping. But the latter may not really what happens because the output level is far from 0dB. Will keep checking till I have some solid data to post.

Would be still interested how Karl set the mixer and the effects loop line/effect switch in his external effects loop test.

Martin



Re: HD500 FX Loop ... level drop or not?
by anglepod on 2012-11-25 02:02:15

Done with this disaster of time and money, I am out so much money on this thing. $500 HD 500, $130 Mission expression pedal, $200 high ohm headphones. Oh how I wish I would have dropped all that on more pedals and just stayed with my board.

There is a level drop on the main out as well. If you boost to compensate the noise floor hiss gets terrible, unuseable for me, obviously they know this and obviously they hope no one tries to use external better effects and stays w the 4CM amp hook up. With the 4CM you just compensate the amp level, same with recording you do not notice anything. Try to drive  an external pedal from the loop or the main out and it does not work.

I gave up on the loop and redid a bunch of my board pedals. Since I have a great amp I do not need amp models with it. I wanted to try just using a wah matrix and a few filters front of my chain and run the POD into my line of drives and gains I use. The level at master on full and mixer at "unity" zero, panned to center, is at least 6db off and cannot drive a pedal correctly. I hooked up my chain straight to the amp and hit my Wampler Pinnacle and my god I think I heard it for the first time. The Pod was seriously killing my tone and I was compensating using clean boost. I had been using "guitar" and "aux" with no input on the aux as this is less noisy. Why is that, because using "same" increases the level and more level equals more hiss. It stares you in the face at what is going on here. More level more noise.

Both the effects loop and main outs are NOT unity, somewhere like 6db off. Line 6 is hoping most users do not progress past the 4Cm and internal effects useage. That is why they never respond to requested minor fixes expecially on the effects side of things. They do not care and no doubt believe most guitar players have a house plant IQ and will not notice the real problems. They are simply hiding the noise floor digital hiss of this unit by rolling off the levels and hoping no one ever actually tests it.

I'm done. Back to my pedal board. I just wanted to use some wahs I cannot be guessing at what is unity level, I need unity to be what the unit claims it is. It is NOT.



Re: HD500 FX Loop ... level drop or not?
by perapera on 2013-01-02 17:07:03

hi

I came to very similar results and found a solution for this problem

see my thread here (in particular point 2):

http://line6.com/support/thread/90441

bye

Lore




The information above may not be current, and you should direct questions to the current forum or review the manual.