Re: Re: PC3K8 Going Out Of Tune

I just wanted to follow up on this.   FYI

Because I needed my PC3K8 to work quickly, I did NOT spend the time for a process of elimination to figure out exactly what was wrong... but just tried to get everything working right as fast as possible.

After determining that the wheel itself is most likely not the problem, I sprayed a little electronic contact cleaner into the pitch bend area and exercised the wheel very thoroughly.  I also went into each Setup and removed pitch bend from every zone that did not need it.

The craziest thing happened, however, just when I thought that everything was coming together with the pitch problems.  Suddenly, many Programs...the individual sound patches built into each Setup, were not working, not in Setup mode nor Program mode!  Why?  What the hell happened?  I went into a full-sweat panic that day, 2 days before the JOVI show.  I thought for sure that the spray must have damaged the electronics and messed it up terminally..  I was praying for lightning to strike me down!

But then I started going through the settings of each Setup/Zone and I found that the “Entry Pan” was set to None for each and every program.  When I adjustied the pan just a little...say to 1 or 2, suddenly the sounds were working again.  Not all of the sounds were affected, though.  Some of them were still working but several important ones were not... like the main sound I use for Runaway... until I set the Entry Pan to a number, just 1 or 2 would do the trick, then it was fine.  I’ve never had this happen before.  I’ve never had to adjust any pan settings for Programs or Zones and I’ve worked with both Stereo and Mono outputs many times.  I’ve never had any problems like that before but I’m glad that I found this “solution”.

The show went off perfectly, without a hitch.   But I’m still very concerned about this keyboard.

Dave



From: [email protected] [KurzList] 
Sent: Tuesday, October 09, 2018 8:53 AM
To: [email protected] 
Subject: [KL] Re: PC3K8 Going Out Of Tune

  

It has taken a a very long time to track the source of this PC3K8 issue, including replacements for most PCBs and the wheel housing to no effect.  The issues were pitchbend activates, dropping the pitch (never upwards), intermittent sustain pedal, so notes would release without letting the pedal go, SW1 and SW2 would not work, although the LEDs would light.  The unit was returned to the distributor for fixes maybe 4 times, to no effect.
I very nearly binned it.

If you have intermittent pitch bend problems, you need to use the diagnostics mode (press and hold 456 for a moment) to see what you can see.  


If the pitch bend sits at approximately 128 when you let it centre itself, the wheel is working fine.  If it centres 126-130, there is a small screwdriver adjustment which can be made through the bottom of the casing, to get it to 128. Wear and tear can cause this precision to fade, so if it settles at on a few different numbers this can indicate wear and the wheel housing unit can be replaced.


Now, if you notice the pitchbend dropping huge and apparently random amounts, but always lower than 128, then what is likely to be happening is the voltage on the scanner is dropping from 5V to a lower value, so the DAC  which reads the potentioneter is reading a lower voltage.  Mine would drop to 100, or 80, even 0 on occasions, then work perfectly for a session, before striking again. 


The root of the bad scan, lies in the way the looms are connected to the main board internally.  The IO, Fader and wheel housing PCBs are connected to the main board by a single armoured loom, which has a ferrite bead around it at the mainboard end.  This loom was the problem in my PC3K8.  The sheer weight of the ferrite, had stressed the tiny wires to the connector to the point of intermittent connection.  I have heard from other techs this loom does not work whether loose or tag tied - the bead can cause impact damage on the mainboard or strain on the cable.    

The solution was to fabriacate a new cable using PC ribbon cable, Molex crimps, and use a series of ribbon ferrites spaced along the cable and away from the mainboard connection.  In the end, we fitted some small posts to the case and tag tied the ferrite beads to those to preventm movement in transit, and left the molex connectors generous room to move unhindered and unstressed.

So far so good.

My Twitter feed has some photos, and the trail of woes - from July 7 through to Sep 2.
State upon opening - https://twitter.com/All97Notes/status/1015673290889875457

My new fabricated loom - https://twitter.com/All97Notes/status/1033007694242881538
Keybed reported wrong - https://twitter.com/All97Notes/status/1030485036548730880

I hope this helps you.  It was extremely hard to find as a fault.
  https://twitter.com/All97Notes/status/1033007694242881538