Reassembling the apron on the Graziano, and I've discovered the powerfeed worm key has a little wear:

IMG_1501_1400x1050.JPG

The problem is, the key is machined as part of the worm. They've bored 22mm (plus a bit) from one side most of the way through, and then probably used a shaper from the other end in conjunction with a dividing head or rotary table to create a bore with a key. This is the layout of the parts:

IMG_1502_1400x1050.jpg

As you can see, where the key is sits right under the bearing seat and threads for the locknuts. There is less than 4mm wall thickness here, and those bearing seats are, slightly surprisingly at first, tight - dead nuts on 30mm. This had me bamboozled for a bit, as it seems illogical given the taper rollers don't go 'home' against the worm, they're preloaded by the locknuts. The fit being a 'light press' fit seems counterintuitive, as if you overdo the preload you've got to give it a whack with a mallet to back the bearing off and try again, but I believe it's been done this way so that oil doesn't weep through between the bearing and the seat, and then leak out the thread of the nuts. Not, in my mind the best design choice, I'm almost tempted to polish down those bearing seats to a sliding fit, and put an o-ring either in the chamfer on the back of the bearing, or put an o-ring groove inside the spacer that the seal runs on.

Nonetheless, I'm trying to figure if I can do anything at all with this key. I do have a torch for the TIG which is capable of getting inside a 16mm tube, so I could potentially build it up with silicon bronze and then use the mill or lathe to 'shape' it. Problem I forsee is it seems pretty likely the heat will at best pull the journal out of round, and possibly warp the whole thing....

The apron clutch lever uses the same arrangement for a key, but it's just a rotating surface on the OD, so someone has milled a slot through from the outside, and pressed a key into the slot. Theoretically plausible here, but the whole thing is hardened. I do have some carbide endmills that are intended for this kind of work, so that may not be a issue. The key would also have to be relieved for the threads to pass over, and some form of sealant would likely be required under the bearing, unless I can get the fit just so and use an o-ring as mentioned before... Not convinced it's overly viable.

The only other options I see are:
1. Don't worry too much about it on the grounds it's taken 60 years in an engineering shop to get to this point, and the worm is providing a 4.16 reduction, in addition to a 6.871 reduction inside the apron up to the rack gear (making for almost a 30:1 reduction in carriage feed mode if my maths is correct). In theory there's not that much load on it, other than the slop may make it 'bounce' under certain cutting conditions (although with that reduction would likely be a small decrease in surface finish?)

2. There is a guy in Italy who supposedly has some parts for these machines, but I believe his prices are higher than Colchester pricing...

3. Speak to someone like Elanjacobs work, and have a new one made.

4. As per number 1, except make a very careful drawing of the thing, and machine a new one out of some form of prehard at some stage in the future anticipating a failure. I don't think I have the changes gears to cut worms on the AL335, so would need the Graz running to do it (which I do have all the gears for, and a bunch of DP/module threads listed on the table)...

If the key DOES fail, providing it doesn't tear up the feed rod, it's possible to make up a kind of bush that slides over the two locknuts on the tailstock side, retained in the locknut slots by set screws. This would protude out toward the tailstock by some amount (say 40mm) with the appropriate 22mm bore, and could carry a replaceable key. The slight catch with that under heavy cuts it may actually tighten up the two locknuts over time and overdo the preload, but it would be A solution, as it would fit in the space and not interfere with anything....
Attached Images




Read the full thread at metalworkforums.com...