Yahoo Groups archive

Fairlight-CMI

Index last updated: 2026-04-29 00:03 UTC

Thread

Series III Keyboard

Series III Keyboard

2011-03-02 by arroncx

So, as I mentioned earlier, Im almost there in full operation :)

One thing that needs fixing is that there are a couple of areas of the music keyboard which dont work : all of the bottom half trigger notes, then about half an octave worth of notes do nothing, then the next few work, and then its dead to the top.

Anyone with any experiance ? Ive never taken the music keyboard apart before.

Cheers,

Arron

Re: Series III Keyboard

2011-03-04 by linear226@excite.com

I've taken them apart a few times, it's been a while though, turn the keyboard upside down, take all the scews out around the perimeter, I think you have to be careful not to remove the large ones which sit underneath the keyboard, I'm pretty sure these release the keyboard seperately, as I say it's been a while..anyway when its loose you'll know, lift the panel from the rear of the chassis, upwards and backwards and the keyboard with the base attached slides through the chassis, be careful though. You can clean the contacts carefully with something like Servisol and a cotton bud. It's not a very elaborate mechanism from what I remember, see how you get on, if that doesnt work it'll be down to the electronics..


--- In Fairlight-CMI@yahoogroups.com, "arroncx" <aclague@...> wrote:
Show quoted textHide quoted text
>
> So, as I mentioned earlier, Im almost there in full operation :)
> 
> One thing that needs fixing is that there are a couple of areas of the music keyboard which dont work : all of the bottom half trigger notes, then about half an octave worth of notes do nothing, then the next few work, and then its dead to the top.
> 
> Anyone with any experiance ? Ive never taken the music keyboard apart before.
> 
> Cheers,
> 
> Arron
>

Re: [Fairlight-CMI] Series III Keyboard

2011-03-04 by Joe Sleator

Hi, Arron,

If you've got the right ROMS in the keyboard, you can pull the power lead off, then hold down * and D on the keypad, and connect the power again, to run some keyboard diagnostics, which will display through the keypad display. Make sure the keyboard is assembled enough to operate safely in this case, but it's not necessary to take it apart at all to run these diags. If PIA 2 has failed, there's your answer. Probably it hasn't. The PIAs are 6821s and you can get them from pinball machine supply places.

Once you get looking at the keyboard underside, you';ll see two buss bars, with the keys holding nickel-silver wire springs which contact the two buss bars at various times. Firstly, make sure the ones on the "bad" keys look the same as the ones on the "good" keys, i.e. none of the contact springs have fallen loose or become disconnected. You can fix the ones that are bad by reconnecting them with tweezers. The keys hold the springs out between the two buss bars, and when the key flies down, the spring breaks loose from one and flies across to the other. The time interval is obviously proportional to the key-strike velocity, and that's how it works. It's very reliable and simple.

Before you start, it's a good idea to count up to where it starts mis-behaving. If the mis-behaviour or dead-key effect starts at the boundary of one of th CMI11 boards, that'd be a dead giveaway that one or more of the decoders (4051 multiplexors) have failed. If you';re just getting sporadic non-functional behaviour, that's more indicative of loose switch springs. You can easily buy 4051 chips cheaply and replace any ones that are obviously bad. In fact you might just piggy-back the chips onto the suspect ones (right way round of course) briefly with the legs bent in to apply tension, to verify that they're the right answer. 4051's are fairly cheap and plentiful, so it's a low-risk activity.

Personally I've seen connections and springs fall off and get filthy way more that I've ever seen 4051s blow up from normal use.

The only other good test I can think of is to verify that the ribbon cable and end connections are healthy. They should be as that internal ribbon won't move much during normal operation.

Also, the voltages are fairly high by modern standards, +, - 5V and silver-oxide is conductive, so they should work unless they9;re very dirty. I'd inspect them before cleaning them to see if the springs have fallen loose with use or transportation.

IIRC, the relevant circuit schematics are in the IIx service manual (that part works the same as the series III). I dunno if the III service manual has been digitized, but it wouldn't be a bad idea.

Good luck!

Joe

Show quoted textHide quoted text
On Thu, Mar 3, 2011 at 9:14 AM, arroncx <aclague@synapse-consulting.com> wrote:

So, as I mentioned earlier, Im almost there in full operation :)

One thing that needs fixing is that there are a couple of areas of the music keyboard which dont work : all of the bottom half trigger notes, then about half an octave worth of notes do nothing, then the next few work, and then its dead to the top.

Anyone with any experiance ? Ive never taken the music keyboard apart before.

Cheers,

Arron


Re: [Fairlight-CMI] Series III Keyboard

2011-03-04 by Martin Selwood

Hello, 

There is a SIII service manual PDF  from the old Non-maskable web site, it can 
be found here:

http://cmi.fairlightus.com/

Best regards, Martin




________________________________
From: Joe Sleator <joe.sleator@gmail.com>
To: Fairlight-CMI@yahoogroups.com
Sent: Fri, 4 March, 2011 15:29:23
Subject: Re: [Fairlight-CMI] Series III Keyboard

  
Hi, Arron,

If you've got the right ROMS in the keyboard, you can pull the power lead off, 
then hold down * and D on the keypad, and connect the power again, to run some 
keyboard diagnostics, which will display through the keypad display. Make sure 
the keyboard is assembled enough to operate safely in this case, but it's not 
necessary to take it apart at all to run these diags. If PIA 2 has failed, 
there's your answer. Probably it hasn't. The PIAs are 6821s and you can get them 
from pinball machine supply places.

Once you get looking at the keyboard underside, you'll see two buss bars, with 
the keys holding nickel-silver wire springs which contact the two buss bars at 
various times. Firstly, make sure the ones on the "bad" keys look the same as 
the ones on the "good" keys, i.e. none of the contact springs have fallen loose 
or become disconnected. You can fix the ones that are bad by reconnecting them 
with tweezers. The keys hold the springs out between the two buss bars, and when 
the key flies down, the spring breaks loose from one and flies across to the 
other. The time interval is obviously proportional to the key-strike velocity, 
and that's how it works. It's very reliable and simple.

Before you start, it's a good idea to count up to where it starts mis-behaving. 
If the mis-behaviour or dead-key effect starts at the boundary of one of th 
CMI11 boards, that'd be a dead giveaway that one or more of the decoders (4051 
multiplexors) have failed. If you're just getting sporadic non-functional 
behaviour, that's more indicative of loose switch springs. You can easily buy 
4051 chips cheaply and replace any ones that are obviously bad. In fact you 
might just piggy-back the chips onto the suspect ones (right way round of 
course) briefly with the legs bent in to apply tension, to verify that they're 
the right answer. 4051's are fairly cheap and plentiful, so it's a low-risk 
activity.

Personally I've seen connections and springs fall off and get filthy way more 
that I've ever seen 4051s blow up from normal use.

The only other good test I can think of is to verify that the ribbon cable and 
end connections are healthy. They should be as that internal ribbon won't move 
much during normal operation.

Also, the voltages are fairly high by modern standards, +, - 5V and silver-oxide 
is conductive, so they should work unless they're very dirty. I'd inspect them 
before cleaning them to see if the springs have fallen loose with use or 
transportation.


IIRC, the relevant circuit schematics are in the IIx service manual (that part 
works  the same as the series III). I dunno if the III service manual has been 
digitized, but it wouldn't be a bad idea.

Good luck!

Joe
Show quoted textHide quoted text
On Thu, Mar 3, 2011 at 9:14 AM, arroncx <aclague@synapse-consulting.com> wrote:

  
>So, as I mentioned earlier, Im almost there in full operation :)
>
>One thing that needs fixing is that there are a couple of areas of the music 
>keyboard which dont work : all of the bottom half trigger notes, then about half 
>an octave worth of notes do nothing, then the next few work, and then its dead 
>to the top.
>
>Anyone with any experiance ? Ive never taken the music keyboard apart before.
>
>Cheers,
>
>Arron
>
>

Re: [Fairlight-CMI] Series III Keyboard

2011-03-04 by Gordon JC Pearce

On Sat, 2011-03-05 at 02:29 +1100, Joe Sleator wrote:


> Personally I've seen connections and springs fall off and get filthy
> way more that I've ever seen 4051s blow up from normal use.


That said, I don't do many synth repairs really but I still buy 4051/2/3
analogue switches in packs of 50...

'Orrible crappy things that they are.

Gordon MM0YEQ

Re: [Fairlight-CMI] Series III Keyboard

2011-03-04 by Joe Sleator

> That said, I don't do many synth repairs really but I still buy 4051/2/3
> analogue switches in packs of 50...

I have 4 spares for 2 CMI keyboards which I've not used yet. Just lucky I guess.

> 'Orrible crappy things that they are.

Agreed, and fairly noisy, too by modern standards. I think maybe part of their general longevity in the CMI keyboard context is that it's a bit of a sheltered workshop for them in there. They only ever see signals well inside their operating spec, are gated very gently, and they sit on a big piece of cool circuit board attached to an even bigger piece of metal, that doesn't get very hot, unlike the system enclosure. The CMI use case is unlike some 4051 uses which come direct from outside world plugs, or nearly so, which I'm sure you've seen in HAM circles. And of course CMOS is way sensitive to ESD, which I think is probably the main degradation/faiure culprit for most old chips that have survived normal use for more than a few hours of burn-in, and the keyboard 4051s never see any static unless you play your CMI in a Van de Graaf room or a lightning storm, like the famous photo. :-D

Joe
Show quoted textHide quoted text
On Sat, Mar 5, 2011 at 2:50 AM, Gordon JC Pearce <gordon@gjcp.net> wrote:

On Sat, 2011-03-05 at 02:29 +1100, Joe Sleator wrote:

> Personally I've seen connections and springs fall off and get filthy
> way more that I've ever seen 4051s blow up from normal use.

That said, I don't do many synth repairs really but I still buy 4051/2/3
analogue switches in packs of 50...

'Orrible crappy things that they are.

Gordon MM0YEQ


Re: [Fairlight-CMI] Series III Keyboard

2011-03-04 by Joe Sleator

> There is a SIII service manual PDF from the old Non-maskable web site, it can be found here:

Ahh yes, I'd forgotten about that one, thanks! I have a beloved physical one, but it's heavy and doesn't fit in my phone, and the pages fall out and blow around the joint sometimes.

Sadly, just as my physical manual, it's missing the schematic for the CMI319, which is the left hand side bender, mod-wheel, switch, lamp, and pot circuit on the CMI III keyboard. I have one-too-few of those, and a background project is
fabbing the circuit board and metalwork, as for some reason there seem to be fewer of those than there are III keyboards, and as plentiful as the other spares are, round here, those are practically nonexistent. :-/
There are plenty of I/II/IIx (3 slider) left-hand-side controllers around.

The 319 is not a complex board, but it's crucial to make up a functional series III keyboard. It has a lot of components for what it does, mostly trimpots and diodes, and needs adjusting periodically to centre the bender and correct its range.

The other extremely rare bits seem to be Rev 3 CMI01-At voice cards, but for some strange reason there are way-plenty of 16 bit CMI31 voice cards by comparison.

Joe

On Sat, Mar 5, 2011 at 2:45 AM, Martin Selwood <martythevampire@yahoo.co.uk> wrote:

Hello,

There is a SIII service manual PDF from the old Non-maskable web site, it can be found here:

http://cmi.fairlightus.com/

Best regards, Martin

Show quoted textHide quoted text
From: Joe Sleator <joe.sleator@gmail.com>
To: Fairlight-CMI@yahoogroups.com
Sent: Fri, 4 March, 2011 15:29:23
Subject: Re: [Fairlight-CMI] Series III Keyboard

Hi, Arron,


If you've got the right ROMS in the keyboard, you can pull the power lead off, then hold down * and D on the keypad, and connect the power again, to run some keyboard diagnostics, which will display through the keypad display. Make sure the keyboard is assembled enough to operate safely in this case, but it's not necessary to take it apart at all to run these diags. If PIA 2 has failed, there';s your answer. Probably it hasn't. The PIAs are 6821s and you can get them from pinball machine supply places.

Once you get looking at the keyboard underside, you';ll see two buss bars, with the keys holding nickel-silver wire springs which contact the two buss bars at various times. Firstly, make sure the ones on the "bad" keys look the same as the ones on the "good" keys, i.e. none of the contact springs have fallen loose or become disconnected. You can fix the ones that are bad by reconnecting them with tweezers. The keys hold the springs out between the two buss bars, and when the key flies down, the spring breaks loose from one and flies across to the other. The time interval is obviously proportional to the key-strike velocity, and that's how it works. It's very reliable and simple.

Before you start, it's a good idea to count up to where it starts mis-behaving. If the mis-behaviour or dead-key effect starts at the boundary of one of th CMI11 boards, that'd be a dead giveaway that one or more of the decoders (4051 multiplexors) have failed. If you';re just getting sporadic non-functional behaviour, that's more indicative of loose switch springs. You can easily buy 4051 chips cheaply and replace any ones that are obviously bad. In fact you might just piggy-back the chips onto the suspect ones (right way round of course) briefly with the legs bent in to apply tension, to verify that they're the right answer. 4051's are fairly cheap and plentiful, so it's a low-risk activity.

Personally I've seen connections and springs fall off and get filthy way more that I've ever seen 4051s blow up from normal use.

The only other good test I can think of is to verify that the ribbon cable and end connections are healthy. They should be as that internal ribbon won't move much during normal operation.

Also, the voltages are fairly high by modern standards, +, - 5V and silver-oxide is conductive, so they should work unless they9;re very dirty. I'd inspect them before cleaning them to see if the springs have fallen loose with use or transportation.

IIRC, the relevant circuit schematics are in the IIx service manual (that part works the same as the series III). I dunno if the III service manual has been digitized, but it wouldn't be a bad idea.

Good luck!

Joe

On Thu, Mar 3, 2011 at 9:14 AM, arroncx <aclague@synapse-consulting.com> wrote:

So, as I mentioned earlier, Im almost there in full operation :)

One thing that needs fixing is that there are a couple of areas of the music keyboard which dont work : all of the bottom half trigger notes, then about half an octave worth of notes do nothing, then the next few work, and then its dead to the top.

Anyone with any experiance ? Ive never taken the music keyboard apart before.

Cheers,

Arron




Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.