Almost all of my voices are fucked. Please help.

Discuss Nava construction and related issues
  • User avatar
    Sawtooth
    Posts 24
    Joined: Dec 13th, '16, 04:03
    Location: Bendigo - Australia

    Does the ride distort only when your altering the pitch? Is it stable ( no distortion ) if you don't move the pitch pot?
  • backdoorranky
    Posts 68
    Joined: Jun 21st, '16, 10:33

    Sawtooth wrote:Does the ride distort only when your altering the pitch? Is it stable ( no distortion ) if you don't move the pitch pot?
    Yes. It only distorts just below high tune. In the audio clip I'm just raising the pitch from low to fully high, then high to low, then back high again. So if it's left at low tune it is stable.
  • backdoorranky
    Posts 68
    Joined: Jun 21st, '16, 10:33

    Update:

    I think I may have found the problem in the Cymbal and Hihat section..

    I have measured the 2SA798 transistors in those sections (Q68, Q70, and Q84) and I can only get a measurement on 3 of the 5 pins. These should read as PNPNP, correct? The other 2SA798 transistor in the Handclap section (Q38) measures on all 5 pins.

    I believe 3 of my 2SA798 transistors are dead. That is some bad luck! :lol:

    Edit: nevermind i am wrong. fuck
  • BastienL
    Posts 90
    Joined: Oct 7th, '14, 13:40

    Yeah, it seems theres something wrong around the DAC multiplexer, composed of IC118 (DAC), IC111 (demux) and IC110 (op-amp). I had a short that damaged the DAC on mine and the result was that all instruments where distorted, but here it seems that some are playing well and the other don't, but since the 3 digital voices are wrong I would think that IC110 or IC111 could be the culprit but if you swapped them, it's very unlikely.
    Maybe you can check the different accent out on IC110 with a scope and I can check too to compare, to see if the dac multiplexer is working as it should.
    For the TM2 trimming, did you enter in the "TM2 adjustment process" (by holding "enter" when turning the Nava on)?

    But yeah my guess is that the 3 digital voices sounding wrong share a same problem that comes from the sequencer and for the analog voices, the problem must come from the analog instrument circuit directly. Or they can be all related to a problem from the seq too. So if you check IC110 pin 7/8/14, you can also check IC111 pin 13/14 and IC114 pin 1/12/13/14/15 with a scope to see what comes out.
  • backdoorranky
    Posts 68
    Joined: Jun 21st, '16, 10:33

    Thanks for the input BastienL,
    BastienL wrote:For the TM2 trimming, did you enter in the "TM2 adjustment process" (by holding "enter" when turning the Nava on)?
    Yes. Though, it was originally set at something like +10.5V, so I had to trim down quite a bit. I read your other post saying that this potentially destroyed some ICs, being that the voltage was so high.

    Also, I realize I should have tested each transistor before installing. Naive rookie mistakes...

    I'll check those pins as you suggested. Thanks.
  • backdoorranky
    Posts 68
    Joined: Jun 21st, '16, 10:33

    Also, anyone happen to know how i could "thicken" a molex wire to fit more sturdy in this housing? http://www.taydaelectronics.com/housing ... -pins.html

    My connections (other than J8-J5) are not very stable by just fitting in the molex on their own. The molex moves around and has a lot of play in the housing, causing my connection to go in and out very easily.

    I have only crimped the molex to the wire, I will solder the crimp to the wire today.

    Not sure if i should try tinning the outside of the molex with solder to fit better in the housing? Any tips?

    I don't mean to blow up the forum with all these questions, but i don't really know where else to ask. Sorry.
  • udenjoe
    Posts 13
    Joined: Jan 5th, '14, 23:19

    For your sampled voices, I think you need the unbuffered CD4011s viewtopic.php?f=24&t=1131&p=6945&hilit=sequencer#p6945

    For the sequencer I believe it's the EEPROM which wasn't compatible with the micro, and was shipped with the kit. This will exhibit unpredictable behavior. I used a 24LC1025 and this sequencer worked.

    This might help with other problems. https://derekwarren.net/nava/pages/nava ... endum.html

    I still have two voices that don't work (HC & MT). This project might have been rushed. My Yocto worked right away but the Nava is kind of a disaster and disappointing. I've done quite a few DIY projects and this has been the worst.

    People have had lots of problems and the board that I have 1.01l has some additional hacked fixes included in the instructions, and I still have problems.
  • backdoorranky
    Posts 68
    Joined: Jun 21st, '16, 10:33

    udenjoe wrote:For your sampled voices, I think you need the unbuffered CD4011s viewtopic.php?f=24&t=1131&p=6945&hilit=sequencer#p6945
    Thanks for this. I bought my ICs from the Tayda BOM and yeah, they are the BE, and not UBE. Did you have this same problem?

    Time to pay more Mouser shipping charges....but I hope this fixes it! Will update if so.
  • udenjoe
    Posts 13
    Joined: Jan 5th, '14, 23:19

    I had a problem when the pitch of some of the sampled voices was low enough or high, it would start distorting. It sounded cool too, as you said, but not 909. Some voices sounded fine. Some I just didn't notice the problem right away. I can't remember where I ordered the 4011UBEs from. There seemed to be many in the UK but not much in the US.

    Using the UBEs fixed all of the problems I had with those voices.
    • Post Reply

    Return to “Nava Hardware”