Shop OBEX P1 Docs P2 Docs Learn Events
Audio Design *Input from PhiPi needed* - Page 2 — Parallax Forums

Audio Design *Input from PhiPi needed*

2»

Comments

  • davejamesdavejames Posts: 4,047
    edited 2012-02-23 11:53
    ...thanks for the diagrams. I'll investigate using a similar approach on my project.

    BTW - a PM is a'watin.
  • average joeaverage joe Posts: 795
    edited 2012-04-15 02:21
    DJ, it was awesome to have a chance to chat with you. Here's the schematic. guitar_pre_boardonly.pdf
    So it goes like this...
    5 wires in, each pickup has a balanced pair. (1-2) =Neck, (4,5) = Bridge. 3=gnd.*those caps should have been from cin1,2,4,5 to STARgnd.
    IC1(a-b) de- balance the signal.
    IC5(a,b) bias amp, adds 2.5v to signal for DigiPot.
    Clamp diodes, digipot, then
    IC6(ab) IC4(ab) buffers for switching.
    IC9(ab) for inverters. 4066's for switching and
    IC7(ab) out buffers.
    That's pretty much it. Let me know how it goes

    *edit*
    those 4066's are "throw - away" components so some or all of the switches may be faulty. Be sure to do an ohms check before using them!
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2012-04-15 09:38
    Hi all, I am a bit late in posting. In any event, Elliot Sound Products out of Australia has a lot of excellent info on contruction of preamps and mixers of all kinds - including guitar preamps. I've built a few of his boards and amy absolutely pleased.

    sound.westhost.com/

    If you are still having construction problems, his site may help.
  • Peter KG6LSEPeter KG6LSE Posts: 1,383
    edited 2012-04-15 19:17
    I too have used ESP's data... If for no other reason .. Bookmark it for a good read at night ..

    Many good insights !



    Peter
  • average joeaverage joe Posts: 795
    edited 2012-04-16 00:47
    Thanks for the info guys! I have looked at ESP's designs as reference before.
    DJ,
    I was able to narrow down the noise issue I experienced at the expo. As it turns out, my power supply is un-grounded. SO, for testing purposes I pulled out the 800Mhz cordless phone. I took the phone "off the hook" and placed in the middle of my guitar cable. The noise floor went from -80dBv to -40dBv. Attaching a ground wire from star ground to mains ground brought the noise floor to -85dBv. Removing the clock signals from the circuit brought the noise floor to - 95dBv. I hope the new analog board will see better performance than the current design.
    So I imagine the problems I had were due to RF noise and poor ground. The guitar cable was acting as a large loop antenna and the RF signals induced on the shield of the cable were dumped to star ground. Since there was no direct path from star ground to mains ground, the signal was superimposed on the signal ground. Or at least this is what I *THINK* was going on.
    Just wanted to let you know what I found out in testing.
  • davejamesdavejames Posts: 4,047
    edited 2012-04-16 08:13
    Joe - great trick using the phone to track down a stinky ground. Perusing Robert Pease's book "Troubleshooting Analog Circuits" in hopes of gleaning some tips from the master, I found a similar reference where he used an AM pocket radio to track down a problem. Interestingly enough, I've noticed a persistant multi-hundred kHz noise everywhere in my lab...even after turning off all other un-used equipment. Guess I'll find me an AM radio and apply that tactic.

    Thanks again for 4066s...I took yesterday off from pretty much all tinkering so I'll start looking at their use sometime this week. Understood about their condition.

    Later man.
  • average joeaverage joe Posts: 795
    edited 2013-03-14 05:34
    Hi all, revived this thread since my goal over spring break is to build a new guitar preamp! Found the problem in the last one... long story short, just because it works once doesn't mean it works right.. Seems in my rebuild for UWEP, I swapped pickup / preamp channel. One pickup is twice as hot as the other... Turns out one preamp channel is over twice as hot as the other... *+2.9dB, +0.1db.

    SO, I grabbed 8 J201's and set out to prototype a TRUE Tillman. Bread-boarded everything out and ran some simple tests... Now I must be doing something wrong since my results are nothing close to expected. With stock component values, all DRAIN voltage measurements are too low *R3-C1-Q1 junction* I'm reading 2v79 to 2v81. SOURCE is 2v11 to 2v13.

    In Tillman's description, he states "R3...is chosen for roughly 6.0 volts at the FET drain." later goes on to say"An easy test is to measure the voltage at the drain of Q1 and if it's between 5.0 and 7.0 volts things are fine."

    Not even close! The only changes to schematic is R1 = 2.2m and R4 is 100K. Have not tried the ear test yet... I think R2 needs to be reduced. My not seeing really any current draw, but can't reliably measure below Milli-amps right now... Not sure what to do yet. Maybe try kicking it up to 15v, where I plan on running. Need to figure new target voltage...
    Any thoughts would be greatly appreciated!
  • davejamesdavejames Posts: 4,047
    edited 2013-03-14 08:20
    Average! Long time...

    I'm told to "watch out" for current stock of J201s, as in they vary wildly in some characteristics.

    Tillman's circuit is a classic! I've used it as a 2X bumper to deal with loss in the guitar cable. And, it just "sounds nice".
  • average joeaverage joe Posts: 795
    edited 2013-03-14 08:28
    DJ, I've been busy in the propeller 1 forum, school and whatnot. Hope things are well!
    *edit*

    After realizing the jfet's I had will not work at all in a stock Tillman circuit, I've begun to crunch the numbers. Hopefully PhiPi can verify I'm on the right track... I'll bug him once I get to an answer, or get stuck ;)

    Step 1. Find JFET's Idss and Vp
    resources here : www.runoffgroove.com/fetzervalve.html#11
    test results:
    JFET           Idss          Vp 
    1              0.43ma      2.96
    2              0.32ma      2.33
    3              0.45ma      2.91
    4              0.43ma      2.82
    5              0.36ma      2.37    
    6              0.45ma      2.94        
    7              0.43ma      2.69     
    8              0.45ma      2.78
    
    As you can see, these vary quite a bit.
    I think I will use 3-6 as a main pair. Then 2-5 as backup. Then I plugged these into a calculator, and got these vaules
                                    calculated values                             -       Actual values
    JFET        Rd           Rs      Vd        Vs        Gain              Rd           Rs           Vd       Vs       Gain
    3-6          18360      5367    11         1.15      4.7dB            18k         5.6k            ??       ??       ??
    2-5          29081      6043   10.63     0.91      7.6dB            33k         5.6k           ??      ??         ??
    

    *aedit*

    Started testing and boy do I have issues, lol. The worst seems to be a ground issue with my bench supply. When I connect circuit ground to supply ground, scope gets UGLY. Almost 1v p-p @60hz....

    Eventually I will need to fix grounding on on the rack-mount, current supply is 2 prong ungrounded and I'd like to remove it and install a standard IEC socket on the back panel. The supply has AG and DG, right now I have these connected together on entry to breadboard. I also have grounds on each 1/4" jack and 2 jacks installed. The final point is ground from my guitar, shield from cable and electrical connection to bridge. I want to star-ground these points to metal chassis, then 10m resistor to supply ground? How do I safely isolate supply ground, so I don't get shocked! I've had this happen with a couple different amps in places with bad grounds... Had to "lift the ground" with a 2 prong ground adapter I kept for just such circumstances!!
Sign In or Register to comment.