Roland Juno 6 no DCO waveforms

[belzrebuth]

Hey all,
I've got a Juno 6 with 2 major issues :
1) half of the keyboard is not working (top two octaves)
2) I see no waveforms anywhere
I don't know how these issues could relate; they may be totally unrelated.
I can hear keyboard tracking of the filter when it's self oscillation so part of the synth is working.
I also have the LFO and the noise and envelopes so one could say that all the analog stuff (VCAs, filter etc) are okay.
I have no output waveforms on channels 1 and 2 on the 8253 counters. CH3 has a signal but it doesn't seem to reach the 4013 flip flops so I get no initial waves and the rest of the voice path seems dead because of that.
I'm puzzled as to why one would all voices at once, I would suspect a bad CPU but I see activity on the data bus and the reset is inactive (high) so it's not like it's stuck or something.
Anybody might a have a clue as to what might be going on?
I really hope the keyboard would be somehow related to this weird no waveform issue..

[commodorejohn]

Have you tried replacing the 8253? They're very common parts on account of being used in the IBM PC and clones (up 'til the point where those functions were integrated into larger-scale motherboard ASICs.)

[belzrebuth]

I've had my doubts on whether the CPU is working okay because I can't explain why part of keyboard is not functioning so the counters were my #2 thing after the CPU.
But I could indeed replace those first because they're so cheap and easy to find.

[madtheory]

Most technicians would advise against shotgun replacing chips before properly testing that it is actually broken. It's a lot of work to desolder and replace a multipin IC, and without proper soldering technique it's easy to cause further damage. Even if you don't have a logic analyzer you can check CPU power startup and clock with a scope.

So yea, the chip is cheap but replacing it is not, in terms of labour, time and risk.

[belzrebuth]

I've got experience in soldering and desoldering that's no issue in this case although I fully agree with you.
I've already put the CPU IC in a socket while trying to decide if it's broken.
My main concern is that most of the keyboard is tracking pitch so the CPU is not entirely dead.
On the other hand there is no chance for both the counters (8253) to have failed and I'm not sure if it's possible to lose the DCOs alltogether without the CPU or the counters getting involved hence the thought that one of those is faulty.I'm still not entirely sure about the keyboard though.
Clocks are there btw and reset is not engaged.