SAMs

[svenmeier]

Hi all,

John Couvillon and I have been investigating solutions to control SAMs
with jOrgan.

It seems we came up with a nice solution - for now I've put the
description in our playground:

    http://jorgan.sourceforge.net/doku.php/playground:playground

I'd appreciate your feedback.

Thanks

Sven

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
jOrgan-user mailing list
jOrgan-user@...
https://lists.sourceforge.net/lists/listinfo/jorgan-user

[eionmidi]

Hi Svenmeier
I am very interested that you are looking at this topic, as I am in the process of designing and assembling a console which needs a combination action for its SAMs.  The method described by Les Knoll:
http://www.nabble.com/jOrgan-and-Stop-Action-Magnets-How-It-Works-td15902643.html
is somewhat complicated although undoubtedly works.  It also has the feature that the magnets are energised for as long as you hold the piston in, like in the original designs using magnetic relays.  This can be seen as disadvantagous but is a blessing if any of your SAMs are 'sticky'.

I would like one point of clarification about your use of 'open coil' and 'close coil'.  I was thinking of these terms as relating to the stop switch.  In which case they appear to be the wrong way round.  Following the red circles description, putting on the tab fires the open coil, which would put the tab off.

The necessary delay between the activator activating and the stop deactivating the appropriate coil would seem to be governed by the speed of processing and the midi message time through the whole system.  A suggestion for the minimum delay value to ensure the SAMs operation is 100 - 200 milliseconds.  I am not sure whether you would get that sort of delay here.

I am going on vacation for a week tomorrow but I will be catching up on this exciting development on my return.
Eion.

svenmeier wrote:

Hi all,

John Couvillon and I have been investigating solutions to control SAMs
with jOrgan.

It seems we came up with a nice solution - for now I've put the
description in our playground:

    http://jorgan.sourceforge.net/doku.php/playground:playground

I'd appreciate your feedback.

Thanks

Sven

[Lynn Walls]

Sven Meier wrote:
> John Couvillon and I have been investigating solutions to control SAMs
> with jOrgan.
>
> It seems we came up with a nice solution - for now I've put the
> description in our playground:
>
>     http://jorgan.sourceforge.net/doku.php/playground:playground
>
> I'd appreciate your feedback.


My only suggestion would be to consider adding a SAM scheme that would
allow for using a combination action (piston) system that is entirely
internal to the console itself.

One variation would be that the console combo system would only feed MIDI
to jOrgan, and only jOrgan would directly fire the SAMS.

The other variation would be that the console combo system would fire the
SAMS directly as in an existing organ console, and the jOrgan combo system
would coexist and be compatible.

The issue with the latter variation might be whether or not the SAM coils
can be wired to both the MIDI decoder SAM drivers as well as to the
internal console combo system without the two SAM driver systems
interfering with each other electrically.

CLW


-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
jOrgan-user mailing list
jOrgan-user@...
https://lists.sourceforge.net/lists/listinfo/jorgan-user

[Roy Radford]

Hi, Lynn,

 

            "The issue with the latter variation might be whether or not the SAM coils
can be wired to both the MIDI decoder SAM drivers as well as to the
internal console combo system without the two SAM driver systems
interfering with each other electrically."

 

    Sounds like a job for lots and lots of diodes, just as you have to with keyboard matrix encoders... Have to be a bit meatier diodes though!

 

      Have fun,

 

         Roy.


Lynn Walls <lwalls@...> wrote:

Sven Meier wrote:
> John Couvillon and I have been investigating solutions to control SAMs
> with jOrgan.
>
> It seems we came up with a nice solution - for now I've put the
> description in our playground:
>
> http://jorgan.sourceforge.net/doku.php/playground:playground
>
> I'd appreciate your feedback.


My only suggestion would be to consider adding a SAM scheme that would
allow for using a combination action (piston) system that is entirely
internal to the console itself.

One variation would be that the console combo system would only feed MIDI
to jOrgan, and only jOrgan would directly fire the SAMS.

The other variation would be that the console combo system would fire the
SAMS directly as in an existing organ console, and the jOrgan combo system
would coexist and be compatible.

The issue with the latter variation might be whether or not the SAM coils
can be wired to both the MIDI decoder SAM drivers as well as to the
internal console combo system without the two SAM driver systems
interfering with each other electrically.

CLW


-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
jOrgan-user mailing list
jOrgan-user@...
https://lists.sourceforge.net/lists/listinfo/jorgan-user

---

Yahoo! for Good helps you make a difference
-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
jOrgan-user mailing list
jOrgan-user@...
https://lists.sourceforge.net/lists/listinfo/jorgan-user

[eionmidi]

Hi Sven

There have not been many comments on the SAMs solution but since I am needing this for my project, I will ask a couple of points.
It is always a bit difficult explaining technical concepts in words, so first I would like to check the definitions of the parts.  I assume:
- the reed switch is essentially the physical stop tab, ie the switch operated by the tab (my SAMs do not have reed switches they simply switch 12V on and off)
- the switch is a new type based on an activator
- the stop is the normal element seen on the monitor screen which you can operate via a touch screen or mouse

So the possible ways of operating the stop on the monitor screen are:
- touching the stop with mouse or finger on touch screen
- via a combination by pressing an on-screen piston
- via a combination by pressing a physical piston
- operating the physical stop tab

I apologise beforehand, if the points are elementary and display only my ignorance, but I do need to be able to implement something like this for my project.

1) Looking at the solution, I do not understand why there are connections between the reed switch and the switch.  I other words, point 2 in the real console list and point 4 in the virtual console list appear to be redundant and thus unnecessary.
In principle, the solution could be summarised as :
- the reed switch operates the stop on or off
- a piston press switches on the appropriate magnet, this operates the reed switch and thus the stop, the stop switches off the magnet

2)  As there are no connections allowing the stop to switch on the magnets, this must mean that pressing the stop with mouse or touch screen on the monitor will not result in the physical stop being operated on the console.  This may not be a drawback in my project as the monitor screen should be redundant when playing, but it is worth mentioning.

Eion.

svenmeier wrote:

Hi all,

John Couvillon and I have been investigating solutions to control SAMs
with jOrgan.

[svenmeier]

Hello Eion,

explaining this setup is very difficult to me, as I've never seen a SAM
in real life and I'm afraid my English skills are missing important
technical terms :(.

- reed switch
The reed switch *detects* changes of the physical tab. It might be an
optical switch or something else, what's important is that it initiates
a Midi message every time the physical tab changes.
- open coil
This coil puts the physical tab on.
- close coil
This coil puts the physical tab off.
- Switch
This new virtual element does nothing else than sending and receiving
messages. So it offers a subset of the Activator's features, i.e. it
doesn't activate other elements.
- Stop
Very important: the virtual Stop is never directly operated from jOrgan.
It only reacts to Midi messages send by the receiver.

I've update the description, please read it once more:

    http://jorgan.sourceforge.net/doku.php/control:sams

... then restate your questions.

Thanks

Sven


eionmidi schrieb:

> Hi Sven
>
> There have not been many comments on the SAMs solution but since I am
> needing this for my project, I will ask a couple of points.
> It is always a bit difficult explaining technical concepts in words, so
> first I would like to check the definitions of the parts.  I assume:
> - the reed switch is essentially the physical stop tab, ie the switch
> operated by the tab (my SAMs do not have reed switches they simply switch
> 12V on and off)
> - the switch is a new type based on an activator
> - the stop is the normal element seen on the monitor screen which you can
> operate via a touch screen or mouse
>
> So the possible ways of operating the stop on the monitor screen are:
> - touching the stop with mouse or finger on touch screen
> - via a combination by pressing an on-screen piston
> - via a combination by pressing a physical piston
> - operating the physical stop tab
>
> I apologise beforehand, if the points are elementary and display only my
> ignorance, but I do need to be able to implement something like this for my
> project.
>
> 1) Looking at the solution, I do not understand why there are connections
> between the reed switch and the switch.  I other words, point 2 in the real
> console list and point 4 in the virtual console list appear to be redundant
> and thus unnecessary.
> In principle, the solution could be summarised as :
> - the reed switch operates the stop on or off
> - a piston press switches on the appropriate magnet, this operates the reed
> switch and thus the stop, the stop switches off the magnet
>
> 2)  As there are no connections allowing the stop to switch on the magnets,
> this must mean that pressing the stop with mouse or touch screen on the
> monitor will not result in the physical stop being operated on the console.
> This may not be a drawback in my project as the monitor screen should be
> redundant when playing, but it is worth mentioning.
>
> Eion.
>
>
>
>
>
> svenmeier wrote:
>  
>> Hi all,
>>
>> John Couvillon and I have been investigating solutions to control SAMs
>> with jOrgan.
>>
>>    
>
>  

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
jOrgan-user mailing list
jOrgan-user@...
https://lists.sourceforge.net/lists/listinfo/jorgan-user

[eionmidi]

Hi Sven

Thanks for taking the time to re-express the SAM solution.  It is now much clearer in my mind.  
To be truthful, it is probably easier to understand if you list the virtual operation first and the physical operation of the tab as an additional explanation.
Your description now makes clear that the stop can be operated on the monitor with a mouse but that it is the switch which must be clicked and not the stop.

Will this require two new elements to be built into jOrgan to give the switch and a stop which cannot be operated on the monitor screen?
I have built one Midibox  DOUT with ULN2308 darlington drivers so I would hope to be able to try out this solution in the next few days.  I reckon that I can just use an activator for the switch and an ordinary jOrgan stop.

Here are a few photos of one of my SAMs.  Like all of them, they are very old and this one works but has one damaged solder tag on the top: http://www.eion.me.uk/christie/

Eion.



Hello Eion,

I've update the description, please read it once more:

    http://jorgan.sourceforge.net/doku.php/control:sams

... then restate your questions.

Thanks

Sven

[svenmeier]

Hello Eion,

 >Will this require two new elements to be built into jOrgan

yes, each real tab switch requires two virtual elements in jOrgan: a
Switch and a Stop.

You can give the virtual Stop a 'warning' style so that you cannot
accidentally change it with the mouse.

Regards

Sven

eionmidi schrieb:

> Hi Sven
>
> Thanks for taking the time to re-express the SAM solution.  It is now much
> clearer in my mind.  
> To be truthful, it is probably easier to understand if you list the virtual
> operation first and the physical operation of the tab as an additional
> explanation.
> Your description now makes clear that the stop can be operated on the
> monitor with a mouse but that it is the switch which must be clicked and not
> the stop.
>
> Will this require two new elements to be built into jOrgan to give the
> switch and a stop which cannot be operated on the monitor screen?
> I have built one Midibox  DOUT with ULN2308 darlington drivers so I would
> hope to be able to try out this solution in the next few days.  I reckon
> that I can just use an activator for the switch and an ordinary jOrgan stop.
>
> Here are a few photos of one of my SAMs.  Like all of them, they are very
> old and this one works but has one damaged solder tag on the top:
> http://www.eion.me.uk/christie/
>
> Eion.
>
>
>
> Hello Eion,
>
> I've update the description, please read it once more:
>
>     http://jorgan.sourceforge.net/doku.php/control:sams
>
> ... then restate your questions.
>
> Thanks
>
> Sven
>
>  

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
jOrgan-user mailing list
jOrgan-user@...
https://lists.sourceforge.net/lists/listinfo/jorgan-user

[Bruce Miles]

Could someone please confirm that this does work reliably in real life.
Will it cope with SAMs that are in less than perfect adjustment? Compton
SAMs, which have a mechanical (ie sprung) latch can be tricky. There may
be a possibility of a positive feedback loop if, for example, the
positive coil is turned off before the SAM has fully operated and locked.

On the basis that if a thing can go wrong, it probably will, I'm
inclined to regard this solution with some suspicion. We did work out a
(possibly) more reliable and straightforward solution using Midiox with
jOrgan which can be used to introduce a fixed and finite time (say 150
ms) for which the SAM coils are energised. This ensures that the SAM has
fully operated and latched. I think this was last discussed in December
last year but my emails are in a shambles because of (you guessed it)
computer problems.

Sven,

Don't apologise for your English -  it's far better than my German
language skills will ever be. This in spite of several years in Germany
- but this was in the far North, (Jever - Oldenburg - Schleswig). Up
there they speak a different language and in the market you can still
buy your tomatoes by the pound (pfund) - not this wretched European kilo
business.

Regards,

Bruce


Sven Meier wrote:

> Hello Eion,
>
> explaining this setup is very difficult to me, as I've never seen a SAM
> in real life and I'm afraid my English skills are missing important
> technical terms :(.
>
> - reed switch
> The reed switch *detects* changes of the physical tab. It might be an
> optical switch or something else, what's important is that it initiates
> a Midi message every time the physical tab changes.
> - open coil
> This coil puts the physical tab on.
> - close coil
> This coil puts the physical tab off.
> - Switch
> This new virtual element does nothing else than sending and receiving
> messages. So it offers a subset of the Activator's features, i.e. it
> doesn't activate other elements.
> - Stop
> Very important: the virtual Stop is never directly operated from jOrgan.
> It only reacts to Midi messages send by the receiver.
>
> I've update the description, please read it once more:
>
>     http://jorgan.sourceforge.net/doku.php/control:sams
>
> ... then restate your questions.
>
> Thanks
>
> Sven
>  

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
jOrgan-user mailing list
jOrgan-user@...
https://lists.sourceforge.net/lists/listinfo/jorgan-user

[Roy Radford]

Hi, Bruce,

 

            ... Still can't get my head round buying wood in them thar new-fangled Centigrade units! 

 

      Have fun,

 

          Roy.

Bruce Miles <bruce@...> wrote:

Could someone please confirm that this does work reliably in real life.
Will it cope with SAMs that are in less than perfect adjustment? Compton
SAMs, which have a mechanical (ie sprung) latch can be tricky. There may
be a possibility of a positive feedback loop if, for example, the
positive coil is turned off before the SAM has fully operated and locked.

On the basis that if a thing can go wrong, it probably will, I'm
inclined to regard this solution with some suspicion. We did work out a
(possibly) more reliable and straightforward solution using Midiox with
jOrgan which can be used to introduce a fixed and finite time (say 150
ms) for which the SAM coils are energised. This ensures that the SAM has
fully operated and latched. I think this was last discussed in December
last year but my emails are in a shambles because of (you guessed it)
computer problems.

Sven,

Don't apologise for your English - it's far better than my German
language skills will ever be. This in spite of several years in Germany
- but this was in the far North, (Jever - Oldenburg - Schleswig). Up
there they speak a different language and in the market you can still
buy your tomatoes by the pound (pfund) - not this wretched European kilo
business.

Regards,

Bruce


Sven Meier wrote:

> Hello Eion,
>
> explaining this setup is very difficult to me, as I've never seen a SAM
> in real life and I'm afraid my English skills are missing important
> technical terms :(.
>
> - reed switch
> The reed switch *detects* changes of the physical tab. It might be an
> optical switch or something else, what's important is that it initiates
> a Midi message every time the physical tab changes.
> - open coil
> This coil puts the physical tab on.
> - close coil
> This coil puts the physical tab off.
> - Switch
> This new virtual element does nothing else than sending and receiving
> messages. So it offers a subset of the Activator's features, i.e. it
> doesn't activate other elements.
> - Stop
> Very important: the virtual Stop is never directly operated from jOrgan.
> It only reacts to Midi messages send by the receiver.
>
> I've update the description, please read it once more:
>
> http://jorgan.sourceforge.net/doku.php/control:sams
>
> ... then restate your questions.
>
> Thanks
>
> Sven
>

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
jOrgan-user mailing list
jOrgan-user@...
https://lists.sourceforge.net/lists/listinfo/jorgan-user

---

Sent from Yahoo! Mail.
A Smarter Email.
-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
jOrgan-user mailing list
jOrgan-user@...
https://lists.sourceforge.net/lists/listinfo/jorgan-user

[eionmidi]

Hi Bruce

I hope to be able to tell you all if this or some variation of this works within the next few weeks.  I have about 150 old Christie stop tabs which I will use with a jOrgan disposition including a combination action.
I can see problems due to sticky tabs.  If a simple feedback loop from switch to stop (see Sven's diagram) is used, there is the problem of the tab not quite making it and the coil energising current staying on.  so if I follow the method as it stands, I will certainly put in a panic button which will send out messages to switch off all the coils!  I have only got as far as one stop tab being set by the switch/actuator and hope shortly to set up a few (three?) to prove the whole principle.
I am still keeping in mind the method described by Les Knoll, which has the advantage that you know that the current is only applied for as long as you hold in the piston.  The Christie stop tabs have the feature that the coil current is automatically switched off as the tab flips.  So if the power supply is not so beefy, the easy stops will flip (and disconnect) allowing the stickier ones the possibly higher voltage.
Eion.

Bruce Miles wrote:

Could someone please confirm that this does work reliably in real life.
Will it cope with SAMs that are in less than perfect adjustment? Compton
SAMs, which have a mechanical (ie sprung) latch can be tricky. There may
be a possibility of a positive feedback loop if, for example, the
positive coil is turned off before the SAM has fully operated and locked.

[Lynn Walls]

Roy Radford wrote:
  >             ... Still can't get my head round buying wood in them thar
> new-fangled Centigrade units!

Fortunately, we in the U.S.A. were able to survive the tyranny of the
"metric-heads".   They took a run at us back in the '70s (I think).  We
had a few gasoline stations try to sell in litres, and for a while some
street signs were "bi-lingual" (Miles/kilometers, etc.).

But we eventually wore 'em down!  Only a few global businesses (General
Motors, for example) keep trying to do it all in metric.  But I haven't
seen a metric street sign in years!

Now if only we can survive multi-culturalism as well!  I'll be damned if
I'm going to learn Spanish just to be able to order fast-food.

CLW

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
jOrgan-user mailing list
jOrgan-user@...
https://lists.sourceforge.net/lists/listinfo/jorgan-user

[eionmidi]

Hi Bruce and All

Well I set up a breadboard with four stoptabs.  I am using midibox for ins and outs and 12V relays to isolate the stopswitch voltage.  I have to confess it was a bit roughly set up but it did the trick.  You can see the setup here along with the disposition file: http://www.eion.me.uk/christie

Using version 3.2.2 I was able to utilise the switch elements.

Well the software certainly works.  I had worried (from the brief discussion there has been about it) that it would not behave 'intelligently' and try to energise the coils of stops which were already in the correct position for the combination but it does not.

The problem is that as Bruce has mentioned if you are using older SAMs, what happens if they do not flip when the piston is pressed?  Well the coil current stays on even though you have finished pressing the piston.  This is a real problem and is happening with one of the four tabs which I used.  I have tried upping the voltage to 14V and it was still unpredictable.

If you have brand new stop tabs, then I am sure that it will all work fine.  But for the rest of us utilising older ones, especially the Compton tabs with their division off second touch feature, this may not be the whole answer.

Mulling over possible fixes, it would be possible to use the piston release message to de-energise the coil but then this becomes complex and must then be similar to the solution given by Les Knoll.  Was that solution for jOrgan version 2.4 only?

My fall back position is to use lighted led stop buttons for everything and not just the couplers as I had hoped.

Eion.

Bruce Miles wrote:

Could someone please confirm that this does work reliably in real life.

[svenmeier]

Hello Eion,

Les' solution doesn't work with the current version of jOrgan :(.

I'm afraid that if the proposed new setup doesn't work either, we have to come up with something new.

Sven


Hi Bruce and All

Well I set up a breadboard with four stoptabs.  I am using midibox for ins and outs and 12V relays to isolate the stopswitch voltage.  I have to confess it was a bit roughly set up but it did the trick.  You can see the setup here along with the disposition file: http://www.eion.me.uk/christie

Using version 3.2.2 I was able to utilise the switch elements.

Well the software certainly works.  I had worried (from the brief discussion there has been about it) that it would not behave 'intelligently' and try to energise the coils of stops which were already in the correct position for the combination but it does not.

The problem is that as Bruce has mentioned if you are using older SAMs, what happens if they do not flip when the piston is pressed?  Well the coil current stays on even though you have finished pressing the piston.  This is a real problem and is happening with one of the four tabs which I used.  I have tried upping the voltage to 14V and it was still unpredictable.

If you have brand new stop tabs, then I am sure that it will all work fine.  But for the rest of us utilising older ones, especially the Compton tabs with their division off second touch feature, this may not be the whole answer.

Mulling over possible fixes, it would be possible to use the piston release message to de-energise the coil but then this becomes complex and must then be similar to the solution given by Les Knoll.  Was that solution for jOrgan version 2.4 only?

My fall back position is to use lighted led stop buttons for everything and not just the couplers as I had hoped.

Eion.

Bruce Miles wrote:

Could someone please confirm that this does work reliably in real life.

[eionmidi]

Hi Sven

thanks for the info about Les's solution.
From my demo system, I am fairly happy that new method operates as designed and it is less complex in implementation than Les's one.  My only unease is that if a SAM does not operate perfectly, then the coil current is on indefinitely (actually until the stop tab is operated manually).  Releasing the piston does not send a message to turn off the current, which is rather counter-intuitive.
My plan is to go ahead and re-build my console, using the new method for now.  I will include a panic button in the disposition which will send messages to de-energise all the coils.
I have got to say that I am having a great time with jOrgan.  I have managed to build a four manual 36 rank theatre  organ disposition (using GSO3) pretty much from scratch after looking over the example offerings.  Thank you so much for your generous donation of time and talents in developing it.
Eion.
PS Habe nicht gewusst dass sie Deutscher sind - Ihre Englischschreibkenntnisse sind grossartig.

svenmeier wrote:

Hello Eion,

Les' solution doesn't work with the current version of jOrgan :(.

I'm afraid that if the proposed new setup doesn't work either, we have to come up with something new.

Sven

[John|C]

Eion,
Congratulations,  you are the first (AFAIK) to test out the SAMS solution!  
Concerning your SAM unit that doesn't switch, can you be more specific as to what happens?  does it stop in the neutral position?  Several things you could check:  Spring is weak.  coil is weak (turns shorted),
Too much load on the armature due to the switch on top.  I suspect that you have already tried graphite on the armature bearing points.  Is there a noticable, physical difference between the SAMs that work, and the one that doesn't.  What happens when you press the gen cancel when the SAM sticks?
Please continue to keep us abreast of your progress.  Its great that the solution will work for newer SAMS, perhaps it can be tweeked to work on older ones also.
Johnc


Hi Bruce and All

Well I set up a breadboard with four stoptabs.  I am using midibox for ins and outs and 12V relays to isolate the stopswitch voltage.  I have to confess it was a bit roughly set up but it did the trick.  You can see the setup here along with the disposition file: http://www.eion.me.uk/christie

Using version 3.2.2 I was able to utilise the switch elements.

Well the software certainly works.  I had worried (from the brief discussion there has been about it) that it would not behave 'intelligently' and try to energise the coils of stops which were already in the correct position for the combination but it does not.

The problem is that as Bruce has mentioned if you are using older SAMs, what happens if they do not flip when the piston is pressed?  Well the coil current stays on even though you have finished pressing the piston.  This is a real problem and is happening with one of the four tabs which I used.  I have tried upping the voltage to 14V and it was still unpredictable.

If you have brand new stop tabs, then I am sure that it will all work fine.  But for the rest of us utilising older ones, especially the Compton tabs with their division off second touch feature, this may not be the whole answer.

Mulling over possible fixes, it would be possible to use the piston release message to de-energise the coil but then this becomes complex and must then be similar to the solution given by Les Knoll.  Was that solution for jOrgan version 2.4 only?

My fall back position is to use lighted led stop buttons for everything and not just the couplers as I had hoped.

Eion.

Bruce Miles wrote:

Could someone please confirm that this does work reliably in real life.