Sealed Chinese optical shaft encoders are cheap as chips on eBay. The late John Stevenson used to swear by them. OK too many counts to use directly, but easily divided down in software.
Sealed is important in your application I reckon
BTW @eSCHEn , I got the macro manual from the Chinese, it clears up a few things for me and makes me none the wiser
Truthfully though it does seem to be a very simple language which I seem to be able to follow.
Everything is easy in software if you know what you are doing, sadly I don't
The reflective option that I seem to be heading in will be housed inside the body of the turret so should be relatively safe.
Pester away champI may be pestering you regarding the macros, I am ever so slightly stuborn (or so I have been told, not really sure why people say that ) so I like to try and work things out for myself. I often fail though so keep your PM page open
Lots of interesting discussion here, just skim read it.
If you use binary combinations to infer multiple positions, make sure you use gray coding for the states. This will save you a whole load of trouble with transient states, if the sensors aren't perfectly aligned, which they won't be.
I don't think it will really matter in this application as there is a large angle between positions and thus a large angle between on/off states. Also the signals are only required whilst the turret is clamped and the clamped positions will be exact due to the clamping being done via a Hirth coupling.
Elucidate please?
A form of binary where only one bit changes state at any one time. Stops transient states existing due to misalignment of sensors, etc.
https://en.wikipedia.org/wiki/Gray_code
No, in this case it is only to confirm to the control that the position of the turret is the same as was requested.Aaah okay, so you don't need to know whilst it's moving?
Yeah I kind of see your point but you could still end up with 1 (or more) faulty signal giving you a faulty reading, so I don’t see what your gaining by adding an extra position, other than using 1 extra sensor adding more components. Not that 1 extra sensor would make much difference I guess but i thought it worthYes but it would mean that if something went wrong with the wiring or whatever position 1 could be falely reported, I much prefer it to be seeing at least 1 signal at all positions.
Of course something could go wrong with certain signals and they might not be seen but the control would cross reference and see there was a problem.
Yeah - you are ensuring that 1 failure can’t occur but there are multitude other ways of failing so the gain is quite small, but 1 extra sensor is not really a big issue in your application I guess, so probably worth doing.The only time it would make a difference is when at position 1, if it was called to go to position 1 and for some reason it clamped in between positions then the control would think it is at position 1 as there are no input signals. If an wire broke or a sensor failed then as long as the control is expecting to see a signal it will know something is wrong when it doesn't see the expected signals.
It is not very likely to happen as if a belt snaps or the coupling breaks then the likelyhood is the turret would not clamp as the teeth on the Hirth would not mesh but there is a slight chance they would. If I had still been using the Curvic coupling I had on the previous turret I made then there would be very little chance of the turret clamping if the belt or coupling broke.