Monitor Out noisy with DXR10

    Quote from OhG

    I'm pulling my hair out due to this same issue tonight. I've tried both the DXR10 and DXR15. With both, I used a Mogami 1/4" cable out of the Monitor Out and into the Line 2 input of the DXR. I've got my monitor output level set at -11 dB and the volume knob on the DXR set at "0". The noise from the monitor out is nothing short of ridiculous. But as soon as I use an XLR cable from the Main Output, the speaker is dead silent. Not being able to use the monitor out makes the Kemper practically useless because I need to be able to control the volume for FOH and my monitor separately. And without having to listen to this amount of noise as a compromise.


    Have you tried an in-line balun transformer with a 1/4" on one end and an XLR on the other?

    Quote from viabcroce

    My Monitor Out is as silent as the Main out.
    Your unit might be defective, or in need of a reset.


    What happens if you use the unbalanced cable on Main out?


    I've done the reset and that didn't help. I haven't tried using the 1/4" on the main outs. I've used a couple different cables of various lengths though, so I know it's not the cable itself. I'll try that tonight just to see what happens though.

    Using a normal unbalanced cable will be the same...
    No need to buy an adaptor, unless the DXR doesn't accept 1/4".


    OhG, what reset(s) did you perform?
    Also, I had suggested to use your 1/4" cable on the Main out. IOW, the same cable on a different output and not the other way round.
    sorry if I was unclear :)

    Quote from viabcroce

    Using a normal unbalanced cable will be the same...
    No need to buy an adaptor, unless the DXR doesn't accept 1/4".


    OhG, what reset(s) did you perform?
    Also, I had suggested to use your 1/4" cable on the Main out. IOW, the same cable on a different output and not the other way round.
    sorry if I was unclear :)


    It isn't just an adaptor, it's a matching transformer from HiZ to 600 ohms, to allow him to use the balanced input where the noise isn't an issue. :)

    I re-tested cables tonight with my cable tester just to make sure they're all good (they were). I forgot to try your suggestion @viabcroce, but I then made some more changes on the Kemper and DXR. I dropped the volume on the DXR down to -8dB on the back and turned up the monitor out on the Kemper to about -11dB. This helped quite a bit. It's not as silent as the XLR output from the main output, but it's at least more tolerable now and is at the point where on stage, it shouldn't be an issue.

    Remember you want to maximize your guitar input into the KPA without clipping. The same goes for the final output of the Kpa. Make sure you maximize all presets to just tickle the output clip led. Then for extra insurance back the output down just a smidgen to avoid the clip led. Although if I'm not mistaken if you occasionally see the output clip indicator you still have about 3dBu of headroom left before hard clipping. If both criteria input and output have been met you will have maximized your signal to noise ratio and have as quiet of an output as can be.
    Hope this helps.

    Quote from SgtPepper

    Remember you want to maximize your guitar input into the KPA without clipping. The same goes for the final output of the Kpa. Make sure you maximize all presets to just tickle the output clip led. Then for extra insurance back the output down just a smidgen to avoid the clip led. Although if I'm not mistaken if you occasionally see the output clip indicator you still have about 3dBu of headroom left before hard clipping. If both criteria input and output have been met you will have maximized your signal to noise ratio and have as quiet of an output as can be.
    Hope this helps.


    Thanks @SgtPepper. I think I need to go back and make sure I have everything set up correctly in terms of input/output levels. Because as of right now, the lights on my Kemper never hit red. Even if I'm on my bridge or neck humbucker pickup and strumming really loud. So you think increasing my input a smidge would help? I noticed you said to do this for all presets. So you're saying this is not a global setting, and needs to be done for each profile, correct?


    For my output, I think I've got it about as good as it can get by setting the DXR to the -8dB setting (9:00 position) and bringing the Monitor Output to around -11dB. I could lower the DXR volume a little more and increase the Kemper Monitor Output a bit more, but the DXR is effectively "off" when the volume knob reaches around the 7:00 position or so.


    After using the G-System for the past few years (and getting it to sound REALLY good), you'd think I'd be a noise/hiss expert by now. Those things are far more touchy than the Kemper has been so far (a positive for the KPA).

    For input adjustment you set it and then lock it
    That will be applied glibally. If you have other guitars that vary then you can create input presets.


    For final outputs because of a presets effects chain this is not possible so currently you must twiddle things to maximize the output gain.
    Now a neat little feature request would be to have a soft button a user could press once they are ready to maximize a presets output level that would sample the player hitting hard chords for say 5 secs and automatically maximize the presets output level .
    Can I get a resounding KPA amen for that request?

    I think for most users coming from the analog amp world into the digital one these concepts of maximizing signal levels are not the easiest things to get a handle on. What I described above as a feature request would help 99.5 % Of all KPA users even if some have KPA degrees. Also to my knowledge no other product in the digital guitar amp arena has a feature for maximizing presets. You could name this feature PRESET or RIG FINALIZATION ©2016 KPA.


    My olive branch to you all.

    This would not be as versatile as you probably think tho :)


    if you compare a clean and a distorted sound they have a very different peak Vs. average energy ratio.
    If you equalize the peaks the clean tone will always sound lower, if you follow the other path the clean, if maximized following your suggestion, would clip the output.


    I am also not sure I'm getting the reason why maximizing the output is need? I realize you'll exploit all the depth of the d\A converter, but the order of magnitude of the S\N improvement you get from this is irrelevant to the overall noise ground IMO.


    Olive branches from here too :)

    Quote from OhG

    So you think increasing my input a smidge would help?


    No, definitely not. Even if you don't maximise the input stage by raising clean sens, S/N ratio will never be a problem with the profiler.
    Also raising rig volume just before hitting thered in the output LED will definitely not have any siginificant impact on the S/N ratio.


    Just make it a habit to raise the Profiler's master volume (for the monitor output) a bit to be able to attenuate the input volume of the DXR a bit and you should be good to go.

    Ingolf,
    I think you better talk to somebody there at Kemper that knows a little something about digital conversion and learn about what the benefits are in maximizing the signal path before A-D and before D-A along with the impact it has on signal noise not to mention dynamic range before making a such an generic assumption. You may actually learn something useful.

    Quote from SgtPepper

    Ingolf,
    I think you better talk to somebody there at Kemper that knows a little something about digital conversion and learn about what the benefits are in maximizing the signal path before A-D and before D-A along with the impact it has on signal noise not to mention dynamic range before making a such an generic assumption. You may actually learn something useful.


    What you wrote is indeed correct. OTOH

    • IME, the noise margin doesn't grow so much to be noticeable in most applications;
    • the issue represents every time you enter a digital device with an analog signal, for example Profiler -> CLR. In this case you can't optimize the volume before the CLR A\D converter, because that's actually your Master volume control, and you have gain-staged your rig function your highest signal.

    IOW, this seems to be a mandatory workaround in many applications. Mr. Mitchell (the designer of the CLR) himself answered a related question according to these considerations:


    Quote

    Q: Can one get optimal performance (signal/noise ratio) from the CLR by using the output level control of the Axe-FX as the overall level control of the system?
    A: Absolutely. That is the intent of the design. Just follow the instructions. First, calibrate the input for the strongest signal you will ever send the CLR. Next, set the Master level for the maximum volume you will need. After that, you can control your volume with whatever device is feeding the Axe-Fx. Just don't exceed the level you used to calibrate the Input level control on the CLR.

    Some people say potato and others potato.
    I have read what Mr. Mitchell has expounded to you in 2014 and he's writes in affect that to paraphrase if you do not fill the bit width of an A-D or D-A S/N and dynamic range suffer.
    So he has some algorithm based on using FEEDBACK information obtained at the output Master. Quasi soft input expansion and output ducking, That might be and Oximoron by I digress...
    Fine and dandy .... I'm not convinced.., Besides are you implying that Kemper is employing a similar technique ???? I'm just stating what has been true since the digital audio revolution appeared . If you throw away bits , compand them or in anyway not utilize the full extent of the real bit resolution there is a price to pay.
    All 24bit converters are not the same They may say 24bits but deliver 20 usable bits.
    Anyway the moral of all the digital marketing hype that some like to use there are some things that can not be changed.


    Just remember the analog recording addage.
    Crap in = Crap out. You can't fix it in the mix but you can put lipstick on a pig.


    Maginfying low analog or digital bit streams amplifies noise ! Ofcoarse you can always try adding dither to mask the noise nevertheless it's noise.
    End of story ....
    That should be never be taken for granted.

    All I am saying (and Mr.Mitchell implies) is that it is impossible to fully exploit a converter's resolution, because the signals you feed it in is variable.
    If you gain-stage your rig in order for the louder peak to not clip the converter, all the other sounds will use smaller figures.This is simply related to the variable nature of the sounds a musician generates.
    I am not saying I don't see the (theoretic) issue, I am saying that there's not much you can do about it apart from properly gain-stagin what you use.
    I'd also note that the noise margin in a not-properly gain-staged analog chain or recording may be orders of magnitude higher than the residual noise coming from a not-fully-exploited digital registry. Back in the times, if your signals did not reach almost the -3 dB on analog tapes, after just a few multitrackings you residual band could go south very quickly unless you used the usual tricks (companders, compensated EQ etc.)... With a 40-€ digital device you can record tens of tracks @ -12 and still have less noise. That's why I wrote that the level issue in the digital realm is less critical (from a sonic OV) than in the analog world.
    Hope this clarifies :)