Cam follower switch for position sensing? 1

[jdunfee5]

I don't have any experience designing this type of cam system, an am wondering if I may be wanting to misuse this type of micro switch.

It happens to be a McMaster 7539K5, but I am asking about this general type of switch, rather than this specific model.

My application is for a fairly large machine that I want the user to be able to position at 45Deg increments, and also be able to disengage a clutch, so he can move the load manually. Because of space constraints and other issues, I am considering "rolling my own" rather than purchasing an off-the-shelf absolute encoder. My current design is to use a Gray-Code type of absolute positioning system, based on a 1994 design by Bruce Spedding. He came up with a way to use a single track, and multiple sensors to encode the position. Here is an image of the concept.

It uses 5 sensors, which will provide 30 positions.

In my case, I am hoping to attach a 6" dia encoder wheel to the output shaft of my gearmotor, and use the snap-action roller switch I showed at the beginning of my post. However, I wonder if this type of roller system is really designed to be used as a cam follower. My encoder will be rotating no more than 2 rpm, and will rotate both ways.

While I don't need high precision (+/- 1 deg is sufficient), I am concerned about how to get these switches properly located, and most important, set so that they trigger at the same max diameter on my cam. The roller is 3/8" Dia, and I obviously need to provide a proper ramp-in/out to the peak of the cam point. Of course, the mounting holes can have slots, but ideally I would want to just have accurately located holes to position the switches, so I don't have to spend as much effort adjusting the switches.

I realize there are other types of sensors, but I would prefer the mechanical switches for several reasons.

Are these types of switches commonly used in bi-directional cam following situations?

-Joe

 

[itsmoked]

Yes. It should work (with reservations) unless you're going to exceed the switch's rated life-time in less than a few years.

Normally the switches are mounted with one fixed hole and one slot. Yes, you want some sane ramping. You can have problems with getting just the right switch point with the ramping, if everything isn't exactly identical. Often you want the ramps adjustable and not the switch for a multi position scheme like you're considering.

I would NOT do it the way you are however. I would use an inductive pickup to detect very narrow non-ramped markers. Inductive pickups are usually superior to mechanical switches. They have indicator lights, are liquid-proof, (unlike the switch you propose), and have decade lifetimes.
Inductive Transducers

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jdunfee5]

It has been many years since I was was involved with a project involving inductive sensors, and my memory of the high the cost had put me off, since I would need 5 of them for each axis. But, your message prompted me to re-investigate them. I see now that they are available for as cheap as $17 USD for 8mm dia ones. So, I think this makes it an easy decision to use inductive sensors.

Thanks for your reply,
-Joe

 

[itsmoked]

You are most welcome and I think you are making the right decision. They are a great value these days - one of the few. Microswitches are almost that expensive these days and that's before you add any waterproof boots and such.

Keith Cress
kcress -

http://www.flaminsystems.com