Wiring rotary switches

[matta757]

Hey guys,

I have been dealing with this problem for almost a year. I am trying to turn rotary switches into rotary encoders, using the techniques described in Ian's document on his website. I tried and tried and tried some more, but my rotaries never worked. Assuming it was a problem with my assembly, I contacted Ian, who was gracious enough to mail me example rotaries (which btw DO work). I copied EXACTLY what Ian did with his example switches, but mine STILL won't work.

I am wondering if anyone has experienced this same problem and found a solution for it? Ian had mentioned to me a while ago that he found another way, but presuming that he is very busy I have not gotten the actual solution from him.

I am using C&K Component rotaries from Mauser. Even more specific, I really need my 45-degree indexed single pole switches to work. That being said, the 30-degree dual pole switches also don't work. If anyone knows why these particular switches won't respond to Ian's treatment and can give me a way around the problem, I would really appreciate it!

Regards,
Matt

[fordgt40]

Matt

Assuming that both your switches and Ian`s have the same pinouts, are wired identically, using the same software, then the only thing that I can think of is that one switch may be "make before break" and the other "break before make" versions.

I am no expert! but this seems to be the only variable, assuming all else is identical

David

[matta757]

I realize I should probably have added some details about how exactly they don't work.

Yes, the switches have the same pin setup and were using the same software as Ian's demo switches. When I connect the switches as wired by Ian's direction to the BUO836X board, they only partially work.

For the first 3 positions they function properly. For example, say that the switch is connected as buttons 1 and 2. Button 1 is a turn left, button 2 is a turn right. With the switch all the way to the left, I can make 2 clicks to the right that register as button 2. But the third click to the right will register as button 1. Everything after the switch is past the third position is messed up. Past that third position, a turn to the left might register as button 2 until returning back to the third position, where it will once again register properly as button 1 in the third, second and first switch positions.

I don't have the board hooked up right now, so I can't remember the specific pattern of errors once I move past the third position. I can check it out later if that will help your understanding. However, as you can deduce, even just one error throws the switch completely out. Luckily, most of the rotaries I have programed only need the 3 positions. The problem now remains the rotaries with more than 3 positions, such as the NAV range and the autobrake.

Matt

[Bob Reed]

I am having a difficulty understanding the problem. You wire them just like any switch. Do you have a picture of the switch or a good link to it so I can see it?

[fordgt40]

Matt

I thought that they did not work at all. Given what you now advise, I think that you need to raise this with Ian as he is the "author" of this system

David

[matta757]

I am having a difficulty understanding the problem. You wire them just like any switch. Do you have a picture of the switch or a good link to it so I can see it?

Hey Bob,

Sorry let me clarify. I am trying to turn rotary switches into rotary encoders, so that the 836X board sees the difference between turning the switch to the left (one button assignment) and the right (another button assignment) so that the switch can be given an increase/decrease function.

Matt

[Bob Reed]

Hey Bob,

Sorry let me clarify. I am trying to turn rotary switches into rotary encoders, so that the 836X board sees the difference between turning the switch to the left (one button assignment) and the right (another button assignment) so that the switch can be given an increase/decrease function.

Matt

Ok that sounds better. I never understood how that worked. Try looking for what they call knitter switches... Now understand I used knitters on my first 737 MCP and they are not the greatest....

[Joe Lavery]

Matt, I had the same problem and also discussed it with Ian, we deduced it was the make before break issue. Mine were not the same as Ian's and as you've found out only partially work.
I believe they should be break before make... it's so confusing that Ian is a wiz!

I just used more pins on Leo's card in the end, I had plenty of spares and it works fine. 

Hope that helps.

Joe.

[matta757]

Joe,

I'm confused about the make before break and break before make thing, could you possible explain? Also, what do you mean when you said you used more pins on Leo's board as a solution? I fail to see how that will work because I need the switch to be assigned as an increase/decrease function, so if you could explain that I would also appreciate it.

Regards,
Matt

P.s.- yes, Ian is a GENIOUS

[fordgt40]

Matt

It is as it says

Make before break = when switching the next contact is made (connected) before the old one is broken

Break before make - when switching the old contact is broken  before the next one is made (connected)

David

[ian@737ng.co.uk]

hello matt.......
ok, i know i've been avoiding this but i have been extremely busy with other things.
the clue for me is that you say the fourth indent gives you the issue.
two channels A & B. connection regime is A1B0, A0B0, A0B1, A1B1, where 1=closed and 0=open.
i used Alpha brand rotary switches for what i need, here's the datasheet.....
http://www.rapidonline.com/Electronic-Components/Switches/Rotary-Switches/Rotary-switches/30304
these are 'make before break' also referred to as 'shorting type'.
i have had no issues with these at all.     however, that doesn't  solve your problem.
my guess is that your C & K switches are 'break before make' which means that the fourth contact if
it is attached to the both channels may be confusing the card because it is expecting to see both channels
connected for an increase in position,but because the switch is breaking the continuity to the third contact before connecting to the fourth, the card figures the encoder has been decreased one stage.
hope that made sense (sometimes i confuse myself  )
if the above is not the case, then possibly your rotary switch contacts are 'bouncing'.  that is as they move from one position to the other, the contact points do not make a clean contact. the tension on the switch may be causing the internal contacts to 'bounce' on and off very quickly.
i did make some simple 'debounce' boards for a friend who had a similar issue which cured it for him.
so, i would look for a make before break switch and try your hand with that.
if you find some, i would be more than happy to make them up for you and test them.
hope that helps you out, let me know.......
regards from siberia (i mean a snowy wales).....

ian

[Boeing Skunk Works]

I read that all of Europe was pretty much dumped on the last couple of days. Stay safe!

[ETomlin]

Im learning something very new here- You mean to tell me that you can have a rotary switch act as a rotary encoder? Seems like that's something that is very useful for when you need indexing, but then again, why cant you in this case simply use the indexed rotary SWITCH to increase/decrease the value you need since it apparently is not incremental by small values like a normal encoder would be?

Understand, Im not questioning your desire to do this, only how it would be useful and if it may be of use in other applications in our sims.

[Joe Lavery]

Hi Matt,

To answer your question a few posts ago, I used the rotary switch for the auto brake function, which has six possible positions. So I simply attached one pin of leo's board to each of the contacts on the switch, using just one of the common (ground) pins. The grounds are all connected so you can use any of them.

Then I assigned the offsets in FSUIPC, it works perfectly. It doesn't matter which position you move the switch to, it will always select the assigned setting.

This is not as elegant as Ian's solution, because it uses up more inputs on the Bodnar board. But because I'm using the FDS SYScard for the overhead, I had plenty of spares.

Hope that helps. 

Joe.

[matta757]

'break before make' which means that the fourth contact if it is attached to the both channels may be confusing the card because it is expecting to see both channels connected for an increase in position,but because the switch is breaking the continuity to the third contact before connecting to the fourth, the card figures the encoder has been decreased one stage.
hope that made sense (sometimes i confuse myself  )

Joe and Ian,

Thank you so much for your replys! That makes PERFECT sense about the make b4 break stuff.

Ian - thanks also for your thorough reply and that link (I figured you have been very busy). Altho I hate to ask one more question but I must... does Alpha make 45 degree indexed rotary switches? I can't find any on the site you recommended. I also did a search on Mouser for other switches, but none of them specify make b4 break or break b4 make. I will call Mouser tomorrow to make further inquiries, but if you know of any other sources of 45-degree (8 position) rotaries that are make b4 break I would greatly appreciate it!

Thanks again,
Matt