Please ensure Javascript is enabled for purposes of website accessibility Jump to content

4CM - using audio analysis software to find "correct levels", if there is such a thing.


zappazapper
 Share

Recommended Posts

I've seen quite a few questions on this forum about levels when using 4-cable method. Most of the answers that I've seen have been anecdotal at best, so I'm creating this thread in the interest of providing some objective information on the subject.

 

First off, I'm not an engineer. I am an enthusiast, a hobbyist, who knows what he knows from reading articles, forum posts, and watching YouTube videos, like many of you. As such, many of the techniques that I use and conclusions that I come to might be absolutely wrong. I'm not doing this to prove what I think I know, or to prove to anybody that I know something - I'm doing this because I WANT TO LEARN. If you see erroneous information or an incorrect way of thinking about something, please let me know. But also keep in mind that I don't really value the opinions of other enthusiasts/hobbyists any higher than my own - I would love for a REAL ENGINEER to come on here and set me straight (ie. not a "recording engineer"...  I'm talking about someone who has a university degree in Electrical Engineering or something...  "recording engineers" are actually just technicians - they don't use science to solve problems - they operate devices designed by people who use science to solve problems... rant over).

 

Second, my equipment is not "measurement-grade". I'm using a MOTU 896HD, a "prosumer"-grade sound interface. I will attempt, to the best of my ability, to demonstrate the limitations of this device as a measurement tool. Also, as per @DunedinDragon's advice, I downloaded RoomEQWizard (REW) as an audio testing platform. In all honesty, there's too much going on there for me to really understand completely, but I think I have enough of a handle on it to use it for my purposes. Again, if anybody has any advice on how to better use this program, I'm all ears.

 

So basically what I'm attempting to do is figure out what the correct settings and levels should be when using 4CM. Before I go any further, I'll fully concede that the actual data and conclusions that I come up with are really only ever going to apply to MY AMP - a Mesa Boogie .50 Caliber+ 1x12" combo - for various reasons. First, it's a tube amp, and as we all know, no two tube amps, even within the same model, are really the same. Second, not every amp model is designed the same (duh). For instance, the MASTER VOLUME on my amp is post-EFFECTS RETURN. On some amps, it is pre-EFFECTS RETURN. The major difference between these types of amps as it relates to 4CM is which control is used to set the actual LOUDNESS of the amp. In my case, I would use the MASTER VOLUME on my amp, and the VOLUME on the multi-fx sets the level going into the EFFECTS RETURN of the amp. In the other case, the MASTER VOLUME on the amp (I think?) would set the level of the signal being fed to the EFFECTS RETURN of the multi-fx, while the VOLUME of the multi-fx is what one would use to set the loudness. Anyway, the point is that what I'm really trying to do here is start a conversation about ways that a typical user can use his or her own equipment to do a little investigation into how their equipment actually works, not necessarily provide hard and fast answers, because there's too many variables.

 

Ok, so let's talk about my interface.

  • MOTU 896HD
  • 24 bit...  up to 192 kHz sample rate (I've never gone higher than 48kHz)
  • 8 Neutrik XLR/1/4" combo jacks for inputs, each with a sensitivity selector switch (LINE/+4/MIC) and a trim pot
  • 8 XLR jacks for outputs, each with a sensitivity selector (-10/+4)

Nothing special.

 

Anyway, I first wanted to test the interface itself, just to see if there was anything crazy bad going on there. I connected an XLR cable between an output and an input and set the sensitivity switches on both to +4. A quick summary of what REW does - it outputs a frequency sweep to the desired output, then analyzes the signal coming back into the desired input, and displays this analysis on a number of graphs. The first step to doing a test is to set levels - REW sends a few seconds of pink noise to the desired output and then measures the level coming back. Anyway, I had assumed that since both the output and input of my interface were set to +4, that the level at the output (-12 dB) would be the same at the input, but at the input I get -14.8 dB. I've tried different outputs, different inputs, different cables, but I always get -14.8 dB. But here's the thing - ultimately I'm not going to use balanced connections in my test, because my amp only has unbalanced jacks. This is only to test the interface in as ideal a situation as possible. And also, even if the levels at the output and the input don't match, it's not important that they do, only that I know what the level coming back to the input should be.

 

So anyway, here are the frequency response, phase, and distortion plots for this first test -

444001097_MOTU896HDXLRouttoXLRintest-frequencyandphase.png.060ff82c2f35de445c2e5abb0bc46a6f.png

1568567479_MOTU896HDXLRouttoXLRintest-distortion.png.0f5bc3c24e224301f138ef85b8d1bb3d.png

The top red line in the first graph is frequency response. I'd say that's pretty flat, especially in the audible range. No issues there.

The second red line is the phase. Again, pretty good in the audible range.

The light brown line in the second graph is the same as the top red line in the first graph - the frequency response.

The dark brown line is the noise floor and the grey line is the distortion (THD). For whatever reason this program doesn't put the legend that I see in the program, which lets you see the value at a chosen point on the graph - the point is, the distortion never gets above 0.005% THD, and at any rate, it's barely more than the noise floor, which is inaudible.

 

OK so the second test I did is with an XLR > 1/4" TS cable connected between the output and input of my interface, because like I said, my amp doesn't have any balanced connections, so whatever is going on with a balanced signal doesn't really apply to testing my guitar rig; it was only useful to test the interface itself in as ideal conditions as possible. Basically I'm doing this test to get a target level at the input, and to make sure the frequency response, phase, and distortion are still in pretty good shape. 

 

The level I got at the input was -14.6 (I thought it would be significantly lower, but hey), and here are the graphs -

1023483134_MOTU896HDXLRoutto1-4intest-frequencyandphase.png.d72708a49a031202ff47bcc2805434cc.png

1122520589_MOTU896HDXLRoutto1-4intest-distortion.png.c17fecb76f3c201c25f2e8fd9b785592.png

No major changes between balanced or unbalanced connections, and no major problems in either than would affect the results of testing another piece of gear. Good. And now I know my target level at the input should be around -14.6 dB.

 

Ok so now what I want to do is test the power amp of my amp. The theory here is that if I have the output of a multi-fx plugged into the EFFECTS RETURN of my amp, with all the effects bypassed, I want the level to be the same as if I plugged my guitar into the EFFECTS RETURN jack on my amp, using a DI box to make the signal line-level. To do this test, I'm connecting the output of my interface to the EFFECTS RETURN of my amp with an XLR > 1/4" TS cable, and the DIRECT OUTPUT on my amp to the input of my interface with a 1/4" TS guitar cable. The DIRECT OUT is a line-level tap off of the output transformer, and it is affected in level by both the MASTER VOLUME and DIRECT controls on my amp. What I'm looking to do is find a MASTER VOLUME/DIRECT setting that gives me -14.6 dB at the input of my interface, so that for the next test where I connect the Helix output to the EFFECTS RETURN of the amp, I'm looking for a Helix VOLUME setting that gives me -14.6 dB. The MASTER VOLUME/DIRECT settings that I find in this test are not applicable to real-world use - only for the purpose of this test. In the interest of keeping the loudness in my apartment to a minimum, and to reduce the chance of power amp distortion affecting the results, I've set the DIRECT at 10 (full) and I'm using the MASTER VOLUME to get as close as I can to -14.6 dB at the input of my interface. I ended up having to set the MASTER VOLUME on the amp at just above 1. Also, I should mention that the PRESENCE control is set to 10, which is the setting for the least amount of negative feedback (I wouldn't be able to back up the suggestion that that setting provides ZERO negative feedback, only the least).

 

The results are interesting - 

2127638615_50caliberpoweramptest-frequencyandphase.png.277c2459c14d4875bf3ae4203d8d2ced.png

1408463096_50caliberpoweramptest-distortion.png.cfb09390d33abd6ba6b6463240c48807.png

Honestly, the frequency response and distortion characteristics don't bother me (it's a tube guitar amp, it's not supposed to be fidelitous) - BUT LOOK AT THAT PHASE!!  This would explain why I had to rewire a speaker cable to use 7CM with my JCM900 - apparently the power amp of my amp amplifies things out of phase and my JCM900 doesn't. Weird. Anyway, for the purpose of testing, I decided I had to remedy this. REW has an option to invert the polarity. It doesn't really explain whether it inverts the signal being outputted or the signal being captured at the input. Anyway, here's the result - 

280712724_50caliberpoweramptestinvertedpolarity-frequencyandphase.png.3dfdd74d4975709005431e6d8250ce70.png

58516245_50caliberpoweramptestinvertedpolarity-distortion.png.838e5d57499496beb53df59913917dda.png

So, not perfect, but much better.

 

So the next test is to connect the output of my interface to the FX RETURN of my Helix using an XLR > 1/4" TS cable, the 1/4" OUT of the Helix to the EFFECTS RETURN of the amp using a 1/4" TS cable, and the DIRECT OUTPUT of my amp to the input of my interface using a 1/4" TS cable. The Input Block for the Helix is set to Return 1 (Input Gate "Off"), output to 1/4" (Pan "Center" and Level "0.0 dB"). There are no effects blocks. In Global Settings, 1/4" Outputs are set to "Line", and Volume Knob Controls is set to "1/4"". I can't think of anything that would affect this test (at this point I'm not using the Guitar In or the FX Send so those settings are irrelevant right now).

 

A couple thoughts about "line level" - there really is no agreed upon voltage that is defined as "line level", as far as I can tell. There's a range. I doubt that L6's line level is the same as what Mesa Boogie thought line level was in 1989. But in this case it's irrelevant because I'm looking for a Helix VOLUME setting that will give me as close to -14.6 dB at the input of my interface as I can get. But more importantly, "line level" is also a definition of impedance. I don't know what impedance the EFFECTS RETURN on my amp expects, what the Helix 1/4" Out sends, nor whether the LINE/INST setting has any affect on the impedance of the signal coming from the Helix at all. I suppose if there is a difference, it will show up in the frequency response graph?

 

So first thing's first - this is the VOLUME setting on the Helix that gave me 14.6 dB at the input of my interface -

20210324_163857.thumb.jpg.06d32dac09da457bf19c0799d010fa75.jpg

I don't know if I have an opinion on this, or if I need to have an opinion on this. Having the required VOLUME setting be at full or even 12 o'clock would make me feel like I've been doing everything right and actually got a volume setting that I could use and maybe EVERYBODY ELSE could use reliably. But until I see a compelling reason why my entire thinking is wrong on this, I'll accept that this is the VOLUME setting that is going to give me the same level as if I was plugged into the EFFECTS RETURN of the amp with a DI box. And so I put a little green triangular piece of tape on my Helix to mark that setting for future reference.

 

In case anybody's interested, here's the test results - 

1765044780_HelixReturntoOutinto50caliberpoweramptestinvertedpolarity-frequencyandphase.png.b1770b9c6fb8a83eef3b65dd0cdce4fc.png

669557947_HelixReturntoOutinto50caliberpoweramptestinvertedpolarity-distortion.png.468c1b23071cef31cbdd1acb416c7093.png

No major changes. I'm thinking that if the impedance were wrong, there would be a big difference in the high end content, which there isn't, so I'm going to assume that there's no point in investigating the INST setting for the connection between the Helix 1/4" Out and the EFFECTS RETURN of my amp.

 

As far as the question of level settings for the preamp of a guitar amp in the loop of a multi-fx, I don't really see how this kind of testing is relevant to this application. In the case of trying to find an output volume to feed your amp's effects return, it's a question of trying to maximize fidelity and volume (even if, in the case of my amp, the power stage of your amp isn't particularly fidelitous at all). When it comes to a guitar preamp, fidelity isn't even a remote concern. It's a tool for shaping the tone of your dry electric guitar signal (which nobody likes anyway), so even if there is a slight difference in gain or output levels between plugging your guitar straight into the front panel of your amp and putting the preamp of your amp into the FX loop of a multi-fx, you're setting that GAIN control to taste anyway, so that kind of accuracy in settings is less important than if the user can find a setting that "feels" the same as plugging straight into the front panel. And that is a question of impedance, since a low-impedance line-level signal will result (in my experience, at least) in an unbearably "screamy" sounding preamp, I'm assuming because a high-impedance instrument-level signal comes with an inherent high-frequency roll off that a guitar preamp is designed to account for. And so the Helix has a setting that seems to be intended to deal with that - the Line/Instrument settings in Global Settings > Ins/Outs > Send/Return. I've asked the question a few times but never really got a definitive answer as to whether or not this setting actually does anything to the impedance of the signal coming from the FX Send, or if it just dials down the voltage. Back in the X3 days, I started using a reamp box to convert the line-level signal coming from the FX Send of my X3L to an instrument-level signal and that made my preamp "feel" like I was plugged into the front panel when it was actually in the loop of the X3L, and so I continue to use the reamp box with the Line setting with my Helix because it just feels better than the Instrument setting. Whether or not I got the same amount of gain for a given GAIN knob setting is irrelevant, or at the very least the difference was too small for me to notice. So I don't know what there is to gain by running pink noise or a frequency sweep through a guitar preamp for the purposes of finding the "correct" settings, because there isn't a correct setting on a guitar preamp. Right? Maybe someone can give me a compelling reason why it would be worth the time and effort to do such a test. I think that if the question is whether the Line/Instrument settings actually do anything in terms of impedance, it's probably better to answer that question with a multi-meter or something, and I'm not knowledgeable enough to do that test.

 

I might not even be knowledgeable enough to do the tests I just did. Haha.

 

Anyway, I hope this is interesting to SOMEONE and I'm happy to see and respond to any CONSTRUCTIVE comments or questions or criticisms.

Link to comment
Share on other sites

Sorry, I passed out about half way through that...;)

 

There's nothing wrong with knowledge for it's own sake, but by the time you figure all of that crap out, some orderly will be dumping you in a wheelchair and rolling you into the day room...

 

If it sounds good and nothing catches fire, then all is well... rock on!

 

  • Haha 1
Link to comment
Share on other sites

3 hours ago, cruisinon2 said:

Sorry, I passed out about half way through that...;)

Ya, a bit of a long read I guess.

 

3 hours ago, cruisinon2 said:

There's nothing wrong with knowledge for it's own sake, but by the time you figure all of that out, some orderly will be dumping you into a wheelchair and rolling you into the day room...

It really only took a couple hours.

 

3 hours ago, cruisinon2 said:

If it sounds good and nothing catches fire, then it's all good... rock on!

Well, that's kind of the whole point. I don't know what your experience with 4CM is/was, but for me it sounded better than it...  uh...  sounded. I mean, when I first read about it, it looked like the perfect marriage of old school, real gear and next generation technology. But when I hooked it up for the first time, the reality didn't live up to the promise. And there really isn't a ton of really in-depth information on 4CM, so I had to look into myself, with my bargain basement gear, free software, and hobbyist-level technical knowledge. And since whatever I know is what I've learned from reading this forum and others like it, I feel an obligation to pass on what I've learned, even if I have serious doubts as to the veracity of my own conclusions. I really wish someone would come on here and tell me where I'm wrong and why and how to get the right answer.

Link to comment
Share on other sites

1 hour ago, zappazapper said:

Well, that's kind of the whole point. I don't know what your experience with 4CM is/was, but for me it sounded better than it...  uh...  sounded. I mean, when I first read about it, it looked like the perfect marriage of old school, real gear and next generation technology. But when I hooked it up for the first time, the reality didn't live up to the promise. And there really isn't a ton of really in-depth information on 4CM, so I had to look into myself, with my bargain basement gear, free software, and hobbyist-level technical knowledge. And since whatever I know is what I've learned from reading this forum and others like it, I feel an obligation to pass on what I've learned, even if I have serious doubts as to the veracity of my own conclusions. I really wish someone would come on here and tell me where I'm wrong and why and how to get the right answer.

 

Though mostly tongue in cheek, my point was this:

 

There is no "right" or "wrong" with this stuff... and if I've learned nothing else from years of playing with modelers, it's that what works for you, may or may not for me, and that doesn't make either of us deaf, stupid, or crazy. And fixating on some theoretically "correct" or "optimal" combination of settings is the quickest path to utter failure... which is why commercially sold patches that claim to make you sound like Eddie, Metallica, or SRV are a fantasy, and Customtone is a wasteland of sounds you wouldn't use with a gun to your head... but I digress.

 

You're missing the forest for the trees, if you will. Trust your ears... they're the best tool you have. The rest is tedious grunt work until you hit on a formula that allows you to squeeze the sound you've got in your head out of whatever gear you have. And sometimes gear has to go...I've forgotten half of the stuff I've played through over the years.

 

In the end it's all entirely subjective assessments of what something sounds like. You either like it or you don't... and the only reliable way to turn a sound you don't like into one that motivates you to play, is to turn knobs until you find your secret sauce. Just because something seems like it should work, doesn't mean that it will.

 

Personally, I've done the 4CM thing, and I'm not a huge fan either... because no matter what you do, the inherent characteristics of the power amp (or power section of the head, combo, whatever) and the cabinet will color anything you pump into it. Imho, this defeats the purpose of having a modeler in the first place. The whole allure for me is the realistic representation of lots of very different amps... if I want the sound of a Plexi with greenbacks, it wouldn't do me much good to amplify a Plexi model with a real-world Boogie or a Fender Twin... you can't "un-Boogie" a Mesa Mark V.

 

Ymmv...

  • Upvote 1
Link to comment
Share on other sites

I think it's a valid argument that objective, easily defined facts can be useful in trying to find some subjective quality that can't easily be defined. I mean, even L6 didn't boil this stuff in a cauldron. They used science and engineering to create sounds that are pleasing in an artistic way. I don't see why anybody has to be tied to one or the other. I don't think about phase response graphs when I'm playing, but stuff like that can tell you something when you're not hearing what you think you should. 

Link to comment
Share on other sites

Hoo boy!  I had a hard time with all that as well.  To toot my own horn, I do have a EE degree and have done a fair amount of audio and DSP design work.

 

Few comments.

a) You show about 350 dB range in your response graphs.  This is about 3 times the dynamic range of human hearing.  Probably more!  I'd make the full range more like 60 dB.

b) I have a Mesa Boogie Express + 5/25 and the Master volumes are BEFORE the effects send.  There's a "Solo" knob after the FX loop which adds level but it doesn't work unless you use a footswitch.

c) I wouldn't worry too much about phase, unless as you say you're trying to send the same signal to two amps at the same time in which case there may be cancellations.  But understand that every time a signal goes through a capacitor, there is going to be phase shift.  My opinion is that phase is the least of your worries.

d) Distortion measurements I am not quite sure how to interpret as they seem to mostly be tracking the noise measurement.

e) I'd be more likely to use the real-time (oscilloscope) display and look at actual signals going in and out of the effects loop to the HX box.  It's much easier to see what's going on when you have a constant, smooth sine wave as your test signal.  The challenge there is to pick a sine wave that more or less is at the same peak level as your guitar signal, which is easy enough to measure.  Just pick 1 or two volts peak to peak as a starting point.  Then look for clipping on the tops of the waveforms coming out of the HX Effects box, which you can probably hear as well.

 

Having your gain levels set too low will reduce the likelihood of clipping but since you might wind up boosting the levels afterwards, any noise will come up as well.  So again, speaking generally, you want to try to have the signals as hot as possible without clipping anything.  This way you can reduce the final gain as much as is possible, which pushes noise down with it.  I would take noise over unintentional clipping any day.  You'll want to do this with all effects blocks turned off, but not "bypassed" (whatever that means) as you want to make sure the signal is actually going through the A/D and D/A stages.

 

Now add to that the fact that some of the effects blocks distort the signal on purpose, and some do it in more subtle ways, like all the flanger blocks with "headroom" settings to emulate the  (I always thought undesirable) distortion you'd get sending too hot of a signal through a BBD type device. 

Link to comment
Share on other sites

On 3/26/2021 at 11:49 AM, Digital-Larry said:

Hoo boy!  I had a hard time with all that as well.  To toot my own horn, I do have a EE degree and have done a fair amount of audio and DSP design work.

 

Few comments.

a) You show about 350 dB range in your response graphs.  This is about 3 times the dynamic range of human hearing.  Probably more!  I'd make the full range more like 60 dB.

b) I have a Mesa Boogie Express + 5/25 and the Master volumes are BEFORE the effects send.  There's a "Solo" knob after the FX loop which adds level but it doesn't work unless you use a footswitch.

c) I wouldn't worry too much about phase, unless as you say you're trying to send the same signal to two amps at the same time in which case there may be cancellations.  But understand that every time a signal goes through a capacitor, there is going to be phase shift.  My opinion is that phase is the least of your worries.

d) Distortion measurements I am not quite sure how to interpret as they seem to mostly be tracking the noise measurement.

e) I'd be more likely to use the real-time (oscilloscope) display and look at actual signals going in and out of the effects loop to the HX box.  It's much easier to see what's going on when you have a constant, smooth sine wave as your test signal.  The challenge there is to pick a sine wave that more or less is at the same peak level as your guitar signal, which is easy enough to measure.  Just pick 1 or two volts peak to peak as a starting point.  Then look for clipping on the tops of the waveforms coming out of the HX Effects box, which you can probably hear as well.

 

Having your gain levels set too low will reduce the likelihood of clipping but since you might wind up boosting the levels afterwards, any noise will come up as well.  So again, speaking generally, you want to try to have the signals as hot as possible without clipping anything.  This way you can reduce the final gain as much as is possible, which pushes noise down with it.  I would take noise over unintentional clipping any day.  You'll want to do this with all effects blocks turned off, but not "bypassed" (whatever that means) as you want to make sure the signal is actually going through the A/D and D/A stages.

 

Now add to that the fact that some of the effects blocks distort the signal on purpose, and some do it in more subtle ways, like all the flanger blocks with "headroom" settings to emulate the  (I always thought undesirable) distortion you'd get sending too hot of a signal through a BBD type device. 

I wasn't necessarily looking for the point at which the input of the power amp would clip. What I was looking for was a level setting that would be the same as not having the Helix connected at all, so that the behavior of the amp would be more or less the same as if I was just using the amp. My basic building block Preset has just the FX Loop 1 block and I wanted that to sound the exact same as just using the amp without the Helix, and then build off of that.

 

You have a point about the range of the response graphs, although I was really just using them to detect MAJOR problems that would show up even at this extreme range.

 

Wouldn't using a sine wave at a particular frequency only tell me about the behavior at that frequency? That's why I used a sweep, so that it would tell me about the behavior at all audible frequencies.

 

On 3/26/2021 at 11:49 AM, Digital-Larry said:

You'll want to do this with all effects blocks turned off, but not "bypassed" (whatever that means) as you want to make sure the signal is actually going through the A/D and D/A stages.

 

Can you expand on that a bit? I have yet to read anything that suggests that it is possible to completely bypass the A/D and/or D/A conversion stages on a Helix. Where does it say that?

Link to comment
Share on other sites

All I really meant by suggesting scope view is that (for me anyway) I relate more directly to the waveform than its spectrum.  And even that isn't particularly true as I relate pretty well to the spectrum.  If I were doing this I'd check it at different frequencies as well.  Nothing wrong with your approach.

 

I don't know if it's possible to bypass the A/D/A conversion on this device - more of a hypothetical concern.  Simply meant to say: run it through A/D/A processing with nothing else going on so that you know that any problems are being introduced by the I/O rather than the headroom setting of the flanger, for example.

Link to comment
Share on other sites

On 3/24/2021 at 5:57 PM, zappazapper said:

I've seen quite a few questions on this forum about levels when using 4-cable method. Most of the answers that I've seen have been anecdotal at best, so I'm creating this thread in the interest of providing some objective information on the subject.

 

First off, I'm not an engineer. I am an enthusiast, a hobbyist, who knows what he knows from reading articles, forum posts, and watching YouTube videos, like many of you. As such, many of the techniques that I use and conclusions that I come to might be absolutely wrong. I'm not doing this to prove what I think I know, or to prove to anybody that I know something - I'm doing this because I WANT TO LEARN. If you see erroneous information or an incorrect way of thinking about something, please let me know. But also keep in mind that I don't really value the opinions of other enthusiasts/hobbyists any higher than my own - I would love for a REAL ENGINEER to come on here and set me straight (ie. not a "recording engineer"...  I'm talking about someone who has a university degree in Electrical Engineering or something...  "recording engineers" are actually just technicians - they don't use science to solve problems - they operate devices designed by people who use science to solve problems... rant over).

 

Second, my equipment is not "measurement-grade". I'm using a MOTU 896HD, a "prosumer"-grade sound interface. I will attempt, to the best of my ability, to demonstrate the limitations of this device as a measurement tool. Also, as per imageproxy.php?img=&key=183b093d1325c787imageproxy.php?img=&key=183b093d1325c787imageproxy.php?img=&key=183b093d1325c787@DunedinDragon's advice, I downloaded RoomEQWizard (REW) as an audio testing platform. In all honesty, there's too much going on there for me to really understand completely, but I think I have enough of a handle on it to use it for my purposes. Again, if anybody has any advice on how to better use this program, I'm all ears.

 

So basically what I'm attempting to do is figure out what the correct settings and levels should be when using 4CM. Before I go any further, I'll fully concede that the actual data and conclusions that I come up with are really only ever going to apply to MY AMP - a Mesa Boogie .50 Caliber+ 1x12" combo - for various reasons. First, it's a tube amp, and as we all know, no two tube amps, even within the same model, are really the same. Second, not every amp model is designed the same (duh). For instance, the MASTER VOLUME on my amp is post-EFFECTS RETURN. On some amps, it is pre-EFFECTS RETURN. The major difference between these types of amps as it relates to 4CM is which control is used to set the actual LOUDNESS of the amp. In my case, I would use the MASTER VOLUME on my amp, and the VOLUME on the multi-fx sets the level going into the EFFECTS RETURN of the amp. In the other case, the MASTER VOLUME on the amp (I think?) would set the level of the signal being fed to the EFFECTS RETURN of the multi-fx, while the VOLUME of the multi-fx is what one would use to set the loudness. Anyway, the point is that what I'm really trying to do here is start a conversation about ways that a typical user can use his or her own equipment to do a little investigation into how their equipment actually works, not necessarily provide hard and fast answers, because there's too many variables.

 

Ok, so let's talk about my interface.

  • MOTU 896HD
  • 24 bit...  up to 192 kHz sample rate (I've never gone higher than 48kHz)
  • 8 Neutrik XLR/1/4" combo jacks for inputs, each with a sensitivity selector switch (LINE/+4/MIC) and a trim pot
  • 8 XLR jacks for outputs, each with a sensitivity selector (-10/+4)

Nothing special.

 

Anyway, I first wanted to test the interface itself, just to see if there was anything crazy bad going on there. I connected an XLR cable between an output and an input and set the sensitivity switches on both to +4. A quick summary of what REW does - it outputs a frequency sweep to the desired output, then analyzes the signal coming back into the desired input, and displays this analysis on a number of graphs. The first step to doing a test is to set levels - REW sends a few seconds of pink noise to the desired output and then measures the level coming back. Anyway, I had assumed that since both the output and input of my interface were set to +4, that the level at the output (-12 dB) would be the same at the input, but at the input I get -14.8 dB. I've tried different outputs, different inputs, different cables, but I always get -14.8 dB. But here's the thing - ultimately I'm not going to use balanced connections in my test, because my amp only has unbalanced jacks. This is only to test the interface in as ideal a situation as possible. And also, even if the levels at the output and the input don't match, it's not important that they do, only that I know what the level coming back to the input should be.

 

So anyway, here are the frequency response, phase, and distortion plots for this first test -

444001097_MOTU896HDXLRouttoXLRintest-frequencyandphase.png.060ff82c2f35de445c2e5abb0bc46a6f.png

1568567479_MOTU896HDXLRouttoXLRintest-distortion.png.0f5bc3c24e224301f138ef85b8d1bb3d.png

The top red line in the first graph is frequency response. I'd say that's pretty flat, especially in the audible range. No issues there.

The second red line is the phase. Again, pretty good in the audible range.

The light brown line in the second graph is the same as the top red line in the first graph - the frequency response.

The dark brown line is the noise floor and the grey line is the distortion (THD). For whatever reason this program doesn't put the legend that I see in the program, which lets you see the value at a chosen point on the graph - the point is, the distortion never gets above 0.005% THD, and at any rate, it's barely more than the noise floor, which is inaudible.

 

OK so the second test I did is with an XLR > 1/4" TS cable connected between the output and input of my interface, because like I said, my amp doesn't have any balanced connections, so whatever is going on with a balanced signal doesn't really apply to testing my guitar rig; it was only useful to test the interface itself in as ideal conditions as possible. Basically I'm doing this test to get a target level at the input, and to make sure the frequency response, phase, and distortion are still in pretty good shape. 

 

The level I got at the input was -14.6 (I thought it would be significantly lower, but hey), and here are the graphs -

1023483134_MOTU896HDXLRoutto1-4intest-frequencyandphase.png.d72708a49a031202ff47bcc2805434cc.png

1122520589_MOTU896HDXLRoutto1-4intest-distortion.png.c17fecb76f3c201c25f2e8fd9b785592.png

No major changes between balanced or unbalanced connections, and no major problems in either than would affect the results of testing another piece of gear. Good. And now I know my target level at the input should be around -14.6 dB.

 

Ok so now what I want to do is test the power amp of my amp. The theory here is that if I have the output of a multi-fx plugged into the EFFECTS RETURN of my amp, with all the effects bypassed, I want the level to be the same as if I plugged my guitar into the EFFECTS RETURN jack on my amp, using a DI box to make the signal line-level. To do this test, I'm connecting the output of my interface to the EFFECTS RETURN of my amp with an XLR > 1/4" TS cable, and the DIRECT OUTPUT on my amp to the input of my interface with a 1/4" TS guitar cable. The DIRECT OUT is a line-level tap off of the output transformer, and it is affected in level by both the MASTER VOLUME and DIRECT controls on my amp. What I'm looking to do is find a MASTER VOLUME/DIRECT setting that gives me -14.6 dB at the input of my interface, so that for the next test where I connect the Helix output to the EFFECTS RETURN of the amp, I'm looking for a Helix VOLUME setting that gives me -14.6 dB. The MASTER VOLUME/DIRECT settings that I find in this test are not applicable to real-world use - only for the purpose of this test. In the interest of keeping the loudness in my apartment to a minimum, and to reduce the chance of power amp distortion affecting the results, I've set the DIRECT at 10 (full) and I'm using the MASTER VOLUME to get as close as I can to -14.6 dB at the input of my interface. I ended up having to set the MASTER VOLUME on the amp at just above 1. Also, I should mention that the PRESENCE control is set to 10, which is the setting for the least amount of negative feedback (I wouldn't be able to back up the suggestion that that setting provides ZERO negative feedback, only the least).

 

The results are interesting - 

2127638615_50caliberpoweramptest-frequencyandphase.png.277c2459c14d4875bf3ae4203d8d2ced.png

1408463096_50caliberpoweramptest-distortion.png.cfb09390d33abd6ba6b6463240c48807.png

Honestly, the frequency response and distortion characteristics don't bother me (it's a tube guitar amp, it's not supposed to be fidelitous) - BUT LOOK AT THAT PHASE!!  This would explain why I had to rewire a speaker cable to use 7CM with my JCM900 - apparently the power amp of my amp amplifies things out of phase and my JCM900 doesn't. Weird. Anyway, for the purpose of testing, I decided I had to remedy this. REW has an option to invert the polarity. It doesn't really explain whether it inverts the signal being outputted or the signal being captured at the input. Anyway, here's the result - 

280712724_50caliberpoweramptestinvertedpolarity-frequencyandphase.png.3dfdd74d4975709005431e6d8250ce70.png

58516245_50caliberpoweramptestinvertedpolarity-distortion.png.838e5d57499496beb53df59913917dda.png

So, not perfect, but much better.

 

So the next test is to connect the output of my interface to the FX RETURN of my Helix using an XLR > 1/4" TS cable, the 1/4" OUT of the Helix to the EFFECTS RETURN of the amp using a 1/4" TS cable, and the DIRECT OUTPUT of my amp to the input of my interface using a 1/4" TS cable. The Input Block for the Helix is set to Return 1 (Input Gate "Off"), output to 1/4" (Pan "Center" and Level "0.0 dB"). There are no effects blocks. In Global Settings, 1/4" Outputs are set to "Line", and Volume Knob Controls is set to "1/4"". I can't think of anything that would affect this test (at this point I'm not using the Guitar In or the FX Send so those settings are irrelevant right now).

 

A couple thoughts about "line level" - there really is no agreed upon voltage that is defined as "line level", as far as I can tell. There's a range. I doubt that L6's line level is the same as what Mesa Boogie thought line level was in 1989. But in this case it's irrelevant because I'm looking for a Helix VOLUME setting that will give me as close to -14.6 dB at the input of my interface as I can get. But more importantly, "line level" is also a definition of impedance. I don't know what impedance the EFFECTS RETURN on my amp expects, what the Helix 1/4" Out sends, nor whether the LINE/INST setting has any affect on the impedance of the signal coming from the Helix at all. I suppose if there is a difference, it will show up in the frequency response graph?

 

So first thing's first - this is the VOLUME setting on the Helix that gave me 14.6 dB at the input of my interface -

imageproxy.php?img=&key=183b093d1325c78720210324_163857.thumb.jpg.06d32dac09da457bf19c0799d010fa75.jpg

I don't know if I have an opinion on this, or if I need to have an opinion on this. Having the required VOLUME setting be at full or even 12 o'clock would make me feel like I've been doing everything right and actually got a volume setting that I could use and maybe EVERYBODY ELSE could use reliably. But until I see a compelling reason why my entire thinking is wrong on this, I'll accept that this is the VOLUME setting that is going to give me the same level as if I was plugged into the EFFECTS RETURN of the amp with a DI box. And so I put a little green triangular piece of tape on my Helix to mark that setting for future reference.

 

In case anybody's interested, here's the test results - 

1765044780_HelixReturntoOutinto50caliberpoweramptestinvertedpolarity-frequencyandphase.png.b1770b9c6fb8a83eef3b65dd0cdce4fc.png

669557947_HelixReturntoOutinto50caliberpoweramptestinvertedpolarity-distortion.png.468c1b23071cef31cbdd1acb416c7093.png

No major changes. I'm thinking that if the impedance were wrong, there would be a big difference in the high end content, which there isn't, so I'm going to assume that there's no point in investigating the INST setting for the connection between the Helix 1/4" Out and the EFFECTS RETURN of my amp.

 

As far as the question of level settings for the preamp of a guitar amp in the loop of a multi-fx, I don't really see how this kind of testing is relevant to this application. In the case of trying to find an output volume to feed your amp's effects return, it's a question of trying to maximize fidelity and volume (even if, in the case of my amp, the power stage of your amp isn't particularly fidelitous at all). When it comes to a guitar preamp, fidelity isn't even a remote concern. It's a tool for shaping the tone of your dry electric guitar signal (which nobody likes anyway), so even if there is a slight difference in gain or output levels between plugging your guitar straight into the front panel of your amp and putting the preamp of your amp into the FX loop of a multi-fx, you're setting that GAIN control to taste anyway, so that kind of accuracy in settings is less important than if the user can find a setting that "feels" the same as plugging straight into the front panel. And that is a question of impedance, since a low-impedance line-level signal will result (in my experience, at least) in an unbearably "screamy" sounding preamp, I'm assuming because a high-impedance instrument-level signal comes with an inherent high-frequency roll off that a guitar preamp is designed to account for. And so the Helix has a setting that seems to be intended to deal with that - the Line/Instrument settings in Global Settings > Ins/Outs > Send/Return. I've asked the question a few times but never really got a definitive answer as to whether or not this setting actually does anything to the impedance of the signal coming from the FX Send, or if it just dials down the voltage. Back in the X3 days, I started using a reamp box to convert the line-level signal coming from the FX Send of my X3L to an instrument-level signal and that made my preamp "feel" like I was plugged into the front panel when it was actually in the loop of the X3L, and so I continue to use the reamp box with the Line setting with my Helix because it just feels better than the Instrument setting. Whether or not I got the same amount of gain for a given GAIN knob setting is irrelevant, or at the very least the difference was too small for me to notice. So I don't know what there is to gain by running pink noise or a frequency sweep through a guitar preamp for the purposes of finding the "correct" settings, because there isn't a correct setting on a guitar preamp. Right? Maybe someone can give me a compelling reason why it would be worth the time and effort to do such a test. I think that if the question is whether the Line/Instrument settings actually do anything in terms of impedance, it's probably better to answer that question with a multi-meter or something, and I'm not knowledgeable enough to do that test.

 

I might not even be knowledgeable enough to do the tests I just did. Haha.

 

Anyway, I hope this is interesting to SOMEONE and I'm happy to see and respond to any CONSTRUCTIVE comments or questions or criticisms.imageproxy.php?img=&key=183b093d1325c787imageproxy.php?img=&key=183b093d1325c787

 

I think this could be summarized into shorter conclusions and a more digestible form but the LOE is appreciated. How would you translate this into practical suggestions to help get a good sound for those using 4CM (and their ears)?

  • Upvote 1
Link to comment
Share on other sites

15 hours ago, HonestOpinion said:

 

I think this could be summarized into shorter conclusions and a more digestible form but the LOE is appreciated. How would you translate this into practical suggestions to help get a good sound for those using 4CM (and their ears)?

My only conclusion is that this seems to be unity gain on the Helix, with no Blocks enabled and Input and Output Block gains at unity -

20210324_163857.thumb.jpg.10488fab6b4233a700b3bb67ca018431.jpg

... at least, that was the setting that gave me the same level of signal as if the Helix wasn't part of the signal chain at all.

 

The more I think about it the more I wonder if it was worth the effort to include the graphs, as I was really only using them to spot major issues that would suggest a bad connection, etc. At the end of the day, every power amp is going to have its own frequency response and distortion characteristics, and that's part of the trade off with 4CM (which is the same reason I took the speaker out of the equation entirely) and it was never my intention to find a way to counteract that, and I wonder if including the graphs gave the wrong impression of what I was looking for. I was only ever looking for the VOLUME setting on the Helix. 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...