Vintage Synth Repair

Never heard of anything like this before.  Do you have the original voice bank that was in the receiving U-20 before the transfer?  Do you have a PC with a MIDI interface?  

  Question one is whether this is a hardware problem, firmware problem, or voice configuration problem.  Most likely it is a voice configuration issue.  If there were a bi-directional connection, then it is possible that it is resulting from the receiving U20 transmitting data back to the tranmit U20.  Were there two MIDI cables connecting the U20s, or only one (Tx U20 MIDI out ---> Rx U20 MIDI in)? 

  Can you edit voices from the front panel or top of the U20?  Is there an envelope parameter for the voice amplitude?  Is there a parameter for delay before the envelope begins to create sound?  After the patch is selected, is there any delay from when a key is pressed and the note sounds?  Does the U20 have seperate banks for voices that are in ROM and voices that in USER static RAM?  The U20 is about twenty years old, maybe the back-up battery for the USER voice RAM is too low to keep the data and the data in the TX U20 was corrupted?  If you make edits on the TX U20, save them to a memory slot, power-down the unit, disconnect the main AC power cable, reconnect the main power cable, and re-power it on, are your edits still in place?  If so, then the battery is good.  Try the same on the RX U20.

  Maybe the TX U20 sent data that was all blank.  The RX U20 gets a patch change message or button press, determines that the patch is bad (a checksum) and tries the next patch.  Maybe all the user patches are bad and it takes 3 seconds to determine this.

  What does Roland tech support say?  This must have happened to hundreds of other people.  They must know something about it.
--- On Sun, 8/2/09, bemccut <bemccut@...> wrote: