Jump to content


Photo

Can Guitar Center Do Service Work In Jtv?

variax warranty repair repair guitar center service center

  • Please log in to reply
32 replies to this topic

#21 toasterdude

toasterdude

    Gear Head

  • Members
  • PipPipPip
  • 810 posts

Posted 21 February 2014 - 12:58 PM

1392396982_9222147_P1020031.JPG

Above is a JTV-69

Actually each Piezo saddle signal feeds a fixed gain op amp that feeds an AKM5386 Stereo A/D There are three U2,U3,U5 - to cover all 6 strings.

http://www.akm.com/a...et/AK5386VT.pdf

There is no "adjustable Gain" block in the AKM5386

 

The Variax String Gain is accomplished post A/D inside the Freescale DSPB56725F DSP

 

(trivia : The Mag PUs use a separate AK4556 Stereo CODEC (U6) - (for A/D for the VDI interface) and D/A for the 1/4" out  

http://www.akm.com/a...et/AK4556VT.pdf

You seem to know quite a bit about JTV guts. I have a whacky question for you, if you know the answer. Lets say I was building a guitar and didn't want model selector and alt tune knobs on it. If the knobs are physically not there, can I still select model and tuning via VDI?


  • 0

#22 toasterdude

toasterdude

    Gear Head

  • Members
  • PipPipPip
  • 810 posts

Posted 21 February 2014 - 01:02 PM

Well I got my problem sorted out.  The fix was a bit of a duh moment, and I'm almost embarrassed that it took me as long as it did to figure out how to correct it.  

 

What I had to do is change the global string volume settings from the edit menu.  I searched the forums and heard people talking about how this acts as a type of pre gain.  Lowering it by 5-6db did the trick. 

 

I didn't even know that setting was there.  I don't use workbench all that much, probably because I can't connect through my HD500 and have get out a separate interface, and just never noticed that setting in the edit menu.  

 

I had been adjusting the the string volume on the patch, but that didn't do anything to help the problem.  

 

As to why the signal was so hot and clipping I couldn't say.  My JTV-59 never had that issue.

 

So all is well again thankfully!

Hmmm in another thread someone asked about tapping off the ghost system and if a buffer was required. Maybe the fact that the models seem 6db hot, would allow just tapping off and the possible signal loss could be handled by lowering by less than 5-6db in worbench?


  • 0

#23 stevekc

stevekc

    Iknowathingortwo

  • Members
  • PipPip
  • 240 posts

Posted 21 February 2014 - 01:47 PM

 

 

You seem to know quite a bit about JTV guts. I have a whacky question for you, if you know the answer. Lets say I was building a guitar and didn't want model selector and alt tune knobs on it. If the knobs are physically not there, can I still select model and tuning via VDI?

 

Yes 


  • 0

#24 stevekc

stevekc

    Iknowathingortwo

  • Members
  • PipPip
  • 240 posts

Posted 21 February 2014 - 01:56 PM

 

Hmmm in another thread someone asked about tapping off the ghost system and if a buffer was required. Maybe the fact that the models seem 6db hot, would allow just tapping off and the possible signal loss could be handled by lowering by less than 5-6db in worbench?

 

 

If it were me, and swapping out the LR BAggs Piezo saddles for higher output Graphtech - it all depends on the player.

If you are steve vai and use a very light swept pick attack - or if you can actually play a guitar with a full floating floyd rose during your most aggressive solos and not suffer the "boing-boing"of the bridge bouncing on every note  - then you might be OK and not have to resort to the measure explained below:  

 

In my case I have a very aggressive right hand, based on growing up playing non tremolo Gibsons with P90's and Acoustic Guitars with heavy strings. My right hand can deliver a tremendous amount of dynamic range.

 (evidence here)

http://www.vguitarfo....php?topic=82.0

 

I would study the fixed gain opamp input section on the Variax  DSP Board , and swap out 6 SMT Gain set feedback resistors for different values to lower the input gain right at the input - because once the signal is clipping and distorting - lowering the gain down stream post A/D in the DSP is a futile effort that is never going to tame transient spikes or make the signals  clean again..

http://www.radio-ele...lifier-gain.php

 

But remember swapping to Graphtech piezo saddles with higher output signal also means higher susceptibility to adjacent string crosstalk   - this really clobbers DSP generated alt tunings - so higher piezo saddle output signal is often not a good thing  


  • 0

#25 snhirsch

snhirsch

    Iknowathingortwo

  • Members
  • PipPip
  • 284 posts
  • LocationBurlington, VT USA

Posted 21 February 2014 - 04:05 PM

Yes.  That was exactly the point I was trying to make in an earlier post.  In a perfect world, Line 6 would have designed things such that the pre-AD gain could be software controlled.  If the current setup has monster amounts of headroom, it may be physically impossible to overrun the converter.  But I pick harder than hades myself and wouldn't take that bet.  Piezo transducers can put out some serious peaks.

 

Do they still teach analog circuitry in an EE cirriculum?  Sometimes I worry that analog design is a vanishing skill set.


  • 0

40 years of Rock-n-Roll and proud of it!

 

PRS Custom 24 (1990) w/ GK-3 Hex PU

James Tyler Variax JTV-69(k) w/ Strat Neck

Roland GR-55 Guitar Synth

QSC K10 FRFR

 


#26 stevekc

stevekc

    Iknowathingortwo

  • Members
  • PipPip
  • 240 posts

Posted 21 February 2014 - 04:48 PM

 

 

 
Do they still teach analog circuitry in an EE cirriculum?  Sometimes I worry that analog design is a vanishing skill set.

 

Tell me a bout it! ;)

 

I'm willing to share more Variax circuit tips  - but I want to avoid getting in hot water with Line6 / Yamaha  - but physics and electronics is my day gig and most of the circuits for A/D>D/A and piezo preamps are already out there  - just a google image search away.  


  • 0

#27 toasterdude

toasterdude

    Gear Head

  • Members
  • PipPipPip
  • 810 posts

Posted 21 February 2014 - 05:55 PM

Tell me a bout it! ;)

 

I'm willing to share more Variax circuit tips  - but I want to avoid getting in hot water with Line6 / Yamaha  - but physics and electronics is my day gig and most of the circuits for A/D>D/A and piezo preamps are already out there  - just a google image search away.  

Wait a second. You know all this cool info on the JTVs. . . .AND . . . have a variax with bigsby? damn. . . .


  • 0

#28 snhirsch

snhirsch

    Iknowathingortwo

  • Members
  • PipPip
  • 284 posts
  • LocationBurlington, VT USA

Posted 21 February 2014 - 06:33 PM

I can't see where discussing perfectly visible aspects of the Line 6 internals even remotely compromises intellectual property.  It's not as if anyone's sitting here trying to reverse engineer the firmware. 


  • 0

40 years of Rock-n-Roll and proud of it!

 

PRS Custom 24 (1990) w/ GK-3 Hex PU

James Tyler Variax JTV-69(k) w/ Strat Neck

Roland GR-55 Guitar Synth

QSC K10 FRFR

 


#29 snhirsch

snhirsch

    Iknowathingortwo

  • Members
  • PipPip
  • 284 posts
  • LocationBurlington, VT USA

Posted 21 February 2014 - 06:44 PM

You are Elantric?  Yow.  Quite a background... 


  • 0

40 years of Rock-n-Roll and proud of it!

 

PRS Custom 24 (1990) w/ GK-3 Hex PU

James Tyler Variax JTV-69(k) w/ Strat Neck

Roland GR-55 Guitar Synth

QSC K10 FRFR

 


#30 Charlie_Watt

Charlie_Watt

    Gear Head

  • Members
  • PipPipPip
  • 814 posts

Posted 22 February 2014 - 06:32 AM

Higher gain piezos does not mean more crosstalk.  It means hotter signal, yes but if you reduce the gain, the crosstalk is reduced also.  Crosstalk is much more a function of the mechanical design.  (I do this for a living too but for Hard Disk Drives)  I have used piezo elements for measuring acceleration disturbances.  I would feel comfortable modifying the input gain stage if I had to.  I would first put a scope on the analog signals going into the A/D and see what they look like when I strum hard.  If they are clipping there or exceeding the input parameters for the A/D, then the gain is too high and there is no way adjusting the string volume in Workbench will fix that.  I have no idea how much headroom Line 6 designed into this circuit.  That is always a Signal / Noise and resolution tradeoff.


  • 0

#31 toasterdude

toasterdude

    Gear Head

  • Members
  • PipPipPip
  • 810 posts

Posted 22 February 2014 - 06:39 AM

Higher gain piezos does not mean more crosstalk. It means hotter signal, yes but if you reduce the gain, the crosstalk is reduced also. Crosstalk is much more a function of the mechanical design. (I do this for a living too but for Hard Disk Drives) I have used piezo elements for measuring acceleration disturbances. I would feel comfortable modifying the input gain stage if I had to. I would first put a scope on the analog signals going into the A/D and see what they look like when I strum hard. If they are clipping there or exceeding the input parameters for the A/D, then the gain is too high and there is no way adjusting the string volume in Workbench will fix that. I have no idea how much headroom Line 6 designed into this circuit. That is always a Signal / Noise and resolution tradeoff.


On the graphtec site they have verbage about their piezos compressing so as not to produce clipping. I will have to investigate to see if that is only with ghost system in addition to piezos.
  • 0

#32 Charlie_Watt

Charlie_Watt

    Gear Head

  • Members
  • PipPipPip
  • 814 posts

Posted 22 February 2014 - 12:42 PM

The piezos themselves will not do any compressing.  They respond to acceleration input (from the string vibration)


  • 0

#33 toasterdude

toasterdude

    Gear Head

  • Members
  • PipPipPip
  • 810 posts

Posted 23 February 2014 - 11:32 AM

Yes 

 

Hmmm. . .what about no volume  or tones pots? lol Pickup selector? Ma make a cool lap steel f  can work all of that out.


  • 0





Also tagged with one or more of these keywords: variax, warranty repair, repair, guitar center, service center

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users