Changing resistor values for the R-2R section

I'm currently building a second WTPA based on the bare V 1 board. The r-2r DAC specifies 59k and 118k resistors, which are not easy to find (for a fair price). Is there any reason why I couldn't substitute the values for 10k-20k or 75k-150k combos? None of the r-2r descriptions I read discussed absolute values and their drawbacks)

Comments

  • edited July 2010
    you can solder the values together from resistors you do have, yeah.
  • in germany i couldn't find them at all, got them from this guy (very very fast):

    http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=400014621943&ssPageName=ADME:B:EF:US:1123
  • edited July 2010
    Thanks for the link. I bought both missing values from the guy.
  • For posterity, Erik is right. As long as the ratio is 2:1 and high enough impedance to be driven from the latch, any value is fine. WTPA2 uses 25.5k and 51k resistors.
  • edited November 2012
    i had the same problem and finally used 62k metal film 1% tolerance and 120k mostly metal film 1% tolerance, three or four carbon film 5% tolerance, cause i was out of the metal ones.... works fine!

    what exactly do these resistors do? just asking for a possible bent/hack... havn't been too deep into it, but maybe there are some nice possibilities...
  • it's a r2r ladder, basically you're adding voltages to each other.
  • thanks. :O)
  • The precision of the resistors in that ladder sets the linearity of the sample when it's played back. So, lower precision or odd-value resistors will add distortion. Might sound cool, but it probably takes some pretty significant variation to notice most of the time.
    TB
  • edited January 2013
    i do have a small, nearly non-recognizable, background beeeeep... when i switch on an have nothing recorded yet. it's depending on the positions of the fine & coarse pots, if the beep appears or not. as soon as a sample is recorded and played back, it's gone. does this beep have to do with the R2R section? it doesn't really bother me, but i fear that in a live situation over a bigger PA the beep might be amplified too much(?).

    here's a little demo video with my amp turned to maximum (it would have burst my speakers, if i would have played a sound)... my cell phone's camera even doesn't record the high frequencies in the beginning, but in the second half of the video you can hear it. as said, this is really a volume level one would never use. but still, on a big PA and when pluged into an effects chain, the beep might increment ugly:
    http://vimeo.com/57735416
  • That's clock feedthrough. I think if you search the forum for "clock noise" or "whine" or the like you'll find other people who've addressed this problem.
    It's not the R2R section, it's the sample clock. If you ground the chassis of the pots and keep the wiring short (and away from audio wiring) it will help a lot.

    On WTPA1, it's a bit of a bear to tame IIRC (although it can be improved a lot). WTPA2 is a lot better that way.

    TB
  • ah, okay. thanks for the info. grounding the pots might be worth a try... the cables can't be shortened in this construction :(
Sign In or Register to comment.