Looking for a replacement or way to fix my smoked turret encoder on the CHNC II

Are you saying there is another magnet and sensor under the turret, or is the magnet that mounts above this position switch going to be all that handles that duty? And I believe you're right on the lack of encoding ability of it, my brain keeps going back to that word because it's simple and I'm not sure what to call it.
Also, I'm listening to suggestions on how to get the soft epoxy or whatever it is off so I can get intimate with the parts inside. You can see I tried cutting it, not an easy task.
"Should be do-able to buy a PCB perf board & some hall switches (or reeds) and mount them at the same radius as the original part.  Wire/solder them to a header and connect a cable back to your control - can test the arrangement on the bench using a magnet and a power supply."
What about the rest of the components on that board? It sounds like you're saying a new setup doesn't need them?

spumco wrote:

Loren wrote:

Repair tech said they are reed switches? 
[assuming someone who'd been inside one of these things doesn't come along...here's my thoughts]

Either way, there's a magnet in the turret plate (sub-plate that engages the Hirth coupling) that triggers one of the switches when the plate is rotated to the correct position.  And likely one wire/conductor per switch, plus a power or ground wire.   My guess is that there's no 'encoding' function - just a simple one-to-one position per switch scheme.

Check your cable and see if there are 9 or 10 conductors.  If they're reeds you may even be able to do some continuity checks with a magnet (don't need power for reeds) and see which one(s) are dead.

If you don't want to try to dig out the dead switch(s), it looks like a circuit board potted in epoxy or similar.  Should be do-able to buy a PCB perf board & some hall switches (or reeds) and mount them at the same radius as the original part.  Wire/solder them to a header and connect a cable back to your control - can test the arrangement on the bench using a magnet and a power supply.

If it works, a 3d printed box or mount can be modeled/printed based on the potted board, and the whole thing coated with epoxy for oil/coolant resistance (3D printed parts aren't watertight).

By the way, if anyone has a oem wiring diagram for this, I'd love to have a copy.

After I somehow get under the coating. That stuff is pretty tough.

tommylight wrote:

The small chips have 4 pins (if i have not gone blind), so dual output hal sensors or 2 or reed per chip?
A DVM and 2 minutes of time would answer that.

Unless you have a wiring diagram, disregard what I wrote earlier about fabricating something up quickly.  My response was a little hasty, and @Aciera posted that great link while I was still writing my response.

The thread @Aciera linked to shows some photos of the assembly without the potting compound, and yes - there are a bunch of other components under there that weren't obvious from your pic.

The big burned spot at the lower left of your board appears to be the same component (same location) that blew up on JT's (I think) board years ago.

As Tommy pointed out, those look like 4-pin hall sensors, and the other post indicated they were halls, too.  But I suspect that your sensors - reed or hall - aren't the problem.  Or aren't the only problem.

Potting compounds can be a pain, but I'd try heating it with a heat gun to see if it softens then start picking it off.  MEK and xylene can also dissolve some potting compounds.  If you can get to the burned component and replace it it may work without having to strip the whole board.

Regarding the magnet, read the other thread. It looks like there's a post with a magnet on it - not embedded in the underside of the turret plate like I first suspected.  The magnet appears not to be keyed to the shaft, so it's possible to adjust the trigger point.  The other thread had a post where the author advanced the magnet a bit to tweak the sensing time in relation to the locking mechanism teeth.

Fundamentally, you could fabricate your own sensor but you need to know what signals the controller expects to see.  5V? 0V? 24V? Sinking or sourcing?  i.e. you need a wiring diagram of some sort.

Loren wrote:

This tiny print on the phone is too hard to follow, I'll get back to you on the pc later, but there are 6 leads out of this unit.
 

And there are two IC chips in the earlier post, too.

Eight halls (or reeds), plus two IC's, plus six wires means that the assembly is an encoder of some sort - not a direct input-output configuration.

More evidence that my first post was way off base and hacking something up isn't really feasible without knowing something about the signals the controller expects to see.

Is it common to find only one component on a board burnt, fix that and you're good to go? I would have never thought that, but BigJohnT's thread leads me to believe that?

Sure, sometimes it's just a single part, although in your case you may have to replace the adjacent part as well, the value of which you might be able to read off of a photograph in the other thread.

[edit]
seems to be another, smaller capacitor

What is the wsum mentioned in that thread? All these acronyms aren't for newbies to follow.

'wsum' is a HAL component in LinuxCNC:
www.linuxcnc.org/docs/devel/html/hal/bas...sub:hal-weighted-sum