Interact 720, Z axis encoder replace

We believe that the Z-Axis position encoder on our Interact 720 is defective. (Z-Axis measuring system defective A) This encoder is at the bottom of the Z-Axis screw.

We're trying to remove the covers over the encoder, and cannot figure out how to remove them.

There appear to be some hidden screws at the tops of the brass slide guides which we cannot access. They're behind the spindle head even when the head is raised to the Z+ limit.

Is there any easy way to remove these slide covers to gain access to the encoder?

We'd prefer a non-destructive procedure :)

I have replaced those encoders, so I know it is possible. But it was very long ago. I would also replace the bellows coupling as I have had more of those break than encoders go bad.Check the bellows for cracks.
Since the control is telling you the encoder is bad (high probability it is a 5 VDC bulb inside the encoder) I believe you are on the right track.

George

Thanks, George.

An update:

We found that by lowering the Z-Axis all the way down and removing the shrouds from the tool changer and spindle area, the upper slide screws were accessible.

With those slides and the covers removed, we gained access to the encoder.

The problem was that oil had been running down the ball screw, through its lower bearing, and then down over the bellows and finally through the bearing in the encoder so that the encoder was loaded with oil.

I cleaned the encoder out repeatedly with hexane alternating with blowing it out (carefully) with compressed air until it was relatively dry and oil-free inside. I then cleaned the encoder wheel and the "window" with my usual digital camera sensor cleaning technique.

I used Eclipse 2 solution and pec pads along with plain cotton swabs. It appears to have worked very well.

This encoder has an LED light source (no doubt an infra-red LED). We had been told that it would probably be an incandescent lamp with a very low value resistor in series (something like 18 Ohms total lamp/resistor resistance).

Instead, it has a 50 ohm fixed resistor and a small potentiometer wired as a rheostat in series with the LED. The total resistance ahead of the LED was found to be about 108 ohms. I adjusted it down to 100 ohms total.

I made sure that the optics and encoder wheel were clean and then blew the whole thing out again with “canned air” to hopefully remove any final dust, and put it back together.

The oil was so heavy on the encoder disk that it was bridging totally across the gap between the optics at the LED side of things and the disk. As it turned fast, bubbles might have formed which would have caused unpredictable refraction and very odd results.

Also, petroleum oils absorb in the IR range. So even if the wavy and unpredictable refraction caused by the oil all over the wheel wasn't a problem, the high IR absorption may have been reducing the amount of light seen by the detector array.

So we've got high hopes that having the oil removed from the encoder should restore proper operation of the system.

So then, the question was: Why was there oil inside the encoder to begin with?

It looks like some precautions had been taken in the design to prevent this from happening, but they didn’t completely fix the problem.

We ended up milling some areas on the encoder's housing and the plate to which it mounts to make sure that oil could drain away from it and not pool up at the encoder's bearing (which seems to have been the entry point for it).

We also created a better seal at the top of the "Sleeve" that hangs down over the bellows. Presumably, that sleeve and the piece it mounts to on the shaft are meant to shed the oil down over the bellows and into the groove in the mounting plate so that it can drain away to the back and miss the encoder.

But there was no seal between the sleeve and the piece it attaches to, nor was the seal between the shaft and that piece any good. So when we put it back together, we put a nice fillet of blue RTV around that area to create an "umbrella" to shed the oil down onto the outside of the sleeve. The trough around where the encoder shaft comes through was also machined differently to make sure that the oil can really drain away and have an unobstructed path out the back.

The bellows looked pretty good. No cracks. But it might not be a bad idea to have one on hand for the future in case it does fail. Do you know of a good source for them?

Now that we know how to get into that area, it will be relatively easy to maintain it in the future.

Our preliminary testing shows that the machine works fine now even at high speeds.

Before this, the machine would throw that "Z-Axis measuring system defective A" error any time you moved the Z axis rapidly. And it got worse and worse to the point that it would give that error even when moving slowly.

If the oil will stay out of the encoder, I think it'll last a long time. It was a relief to see that this unit has an LED rather than an incandescent lamp. LEDs last a very long time.

Oily encoder - previous to cleaning:
http://farm4.static.flickr.com/3492/...83d2237e_o.jpg




Encoder after cleaning:
http://farm3.static.flickr.com/2654/...b9db156e_o.jpg



Fillet of blue RTV to help shed oil:
http://farm3.static.flickr.com/2478/...39bc2589_o.jpg

Well done!
Including the macro photography.
I was looking to see if a oil slinging disk could be put somewhere but per your pictures, I do not see where it could be put effectively.
Yours is the first case that I have heard of oil in the encoder at the bottom of the ball screw (Z axis). So I do not believe it is very common, but may well be as machines age.
I wonder if a slight positive air pressure in the encoder can would prevent this. It is common to have this kind of air set up on CNC machines equipped with linear scales.

George

Hi, George.

We contemplated making some kind of disk out of neoprene and then sliding it up onto the shaft above the protective sleeve so that it would sling the oil outwards before it could run down to the encoder. But that shaft has two flats so we couldn't figure out a good way to punch the neoprene to make it have a good seal and fit.

In the end, it looked like just building up a fillet of RTV might do the trick. We wanted to use something that would be removable in the future yet give us a good seal.

I like the idea of a bit of air-purge for the encoder to keep anything from dripping into it.

I also reinstalled the cable to the encoder with a drip loop but I really feel that the oil was entering through the shaft bearing and running down the shaft onto the disk where it was then thrown outward to its edges and then ran down to the bottom side. That disk was totally covered in oil. Even the photos I have of it don't show how bad it was because I'd already wicked a lot of it off with some tissues before I took those pictures.

I've got more downsampled images of the encoder and its area here:

http://www.flickr.com/photos/9803617...7622734641238/

and the full size images here:

http://www.flickr.com/photos/9803617...7622724090474/