Have I buggered my Renishaw probe? Disassembly, reassembly and alignment.

There's a problem with my probe:

There's been a bit of a delay getting set up to machine the ballnut bracket. Apart from the need to tile the kitchen floor and find a job (trivialities), I had a problem with probing.

The Renishaw probe is a fine piece of kit and so far the ceramic shaft probe tip remains intact. This is largely due to having fitted a brass screw in place of it until I have learned to use it properly. At that point I will revert to using the pukka tip and will then doubtless break it to short order.

Yesterday I thought I'd break out the pukka tip for setting up the ballnut bracket. The WCS origin is in the centre of the base of the counterbore, so this is an ideal application for the "bore" probing function in the Centroid software. This went very nicely and I managed to set my X, Y and Z coordinates but something didn't look right......

You can possibly just about make out that the probe tip is a little to the left of the axis here - it's actually about 2mm or so. You can measure it with a ruler(!) by eye:

Turning the spindle by hand showed the issue - the probe tip is eccentric by a couple of mm. WTF???? That looks like too much for simple readjustment but apart from that, how did it happen? I can't swear blind that it wasn't skew whiff when I bought it of course but surely I'd have noticed??? Anyway, it's clear that there's no simple explanation or simple fix. Surely I couldn't have simply knocked it out of alignment? The body appears to be fairly concentric but the tip isn't, so the problem surely lies inside. Bollocks - that doesn't sound good. Have I managed to bugger this thing already, after only a few hours use at the most?  I paid "only"£200 or so for it, which is considerably less than a new one but still....

I've replaced the tip with my dummy's screw here. 

Disassembly and checking:

Let's go in...

The body is attached to the holder by 4 screws. The adjustment for eccentricity is made by adjusting these. The whole chebanc pivots on the 8mm ball bearing. The sponge sandwich simply keeps swarf, muck and coolant out.

The only thing you can remove is the front disk-shaped body which is held in by 4 really tiny grub screws. Luckily I have a set of really tiny Allen keys - this isn't even the smallest size in the set. You can't see it here but the bellows had a nasty twist it it. The grub screws are so small it's hard to tell that's what they are. And the front body has knurled edge which might make you thing it simply unscrews. I wonder if somebody had tried to unscrew it before??

The front piece simply pulls out. There is an o-ring between it and the body, plus a delicate-looking bellows to seal the insides. I managed to get it off the probe mount without rupturing it.

There's a second bellows, sort of like "a double bagger". As you’d expect, it’s very nicely made but otherwise, no great surprises. These probes almost invariably use a sort of tripod scheme inside ie 3 radial insulated but conductive rods sitting on and spring loaded against 3 pairs of ball bearings connected in a series electrical circuit. The carrier that holds the rods also holds the tip. If the tip moves and any one of the rods lifts off the balls, the circuit opens. There’s a special magical oil in there to stop corrosion and give good contact. 

Note that the cavity is partially filled with oil. This will doubtless be a special electrical contact oil, so best not to lose any.

Everything seems to be in good condition and no obvious damage or distortion. The nose of the ceramic probe is 7mm and although I don't have a 7mm end mill holder, I have a 7mm ER40 collet, so was able to mount this in the machine. The runout of the main body and the axial position of the 3 rods was within a thou or so on this generic DTI:

This is the carrier that contains the 3 rods and the probe mount:

Reassembly: 

Put the key parts back together again. Measuring the runout on the main body. It's about 15 thou max, which is fine. Can't see anything concerning at this stage.

So I reassembled it properly now. I noticed that the bellows were in fact secured to the probe body by means of little o-rings. These hold the neck of the bellows in the grooves in the probe body. Lucky I managed to get both bellows off without tearing them! Easy enough to replace with some fine tweezers. 

You can see the neck on the outer bellows here - and the o-ring.

There's a large diameter o-ring to seal the front body. 

All back together. I managed to retain almost all of the special oil in there. And given that the probe operates with the bellows at the lowest point, it's A Good Thing that I didn't damage them or it would soon drain away and possibly fill up with dirt, crap, swarf and coolant.

Re-alignment:

First, get it in the ball park using my Chinesium generic DTI picking up on the top end of the probe tip:

Then finally use the Baty 0.01mm DTI. This is the best resolution I have in a DTI and the preload force is sufficiently low to allow me to test the tip of the probe. I seem to have got it within perhaps 5 microns or so. One small division is 10um.

Shorter toolholder for the Renishaw probe:

With turret mills like mine (and the commonplace Bridgeport-type machines), limited quill movement is a real issue, so minimising tool stickout is key. I have a problem with the sheer length of this probe, as well as my 3D "Martest" probe (I've recently machined back the holder for that but it's still far too long). 

Looking at the holder for the Renishaw probe, now that it's apart, there's the best part of 55mm I could get rid of straight away.

This is a good opportunity to make some measurements. It's not an obviously nominal dimension, looks like 52.5mm and this is all metric, so it's not actually an imperial dimension. So what, I can reproduce it. 

The steel pivot ball is 8mm. 

I have this ancient looking collet chuck that won't get much use. Looks to me as if I could machine off the collet part and be left with a simple disk shaped body, which would be ideal for mounting the probe on. That would gain me the best part of 50mm. The tricky part will be drilling and hopefully tapping the 4 holes.

It came with a 1/4" collet, so I can make use of that for setting it up, then machine it off. And yes, I've checked that it's metric (ISO40, M16 thread) rather than imperial (NMTB40 5/8-11 UNC). It looks as if it should be reasonably simple to set up....

I'll have a go at this over the weekend if I get a chance.