Dividing head restoration

[Dabbler]

@Canadium I haven't see this. Perhaps on cheaper wood-style aluminum oxide sanding paper, but I use cloth-backed or mylar sheet abrasives. I think some grit goes astray, but always use a fluid lubricant liberally to wash any away. I used to use Varsol, then low-odorur Varsol, but now I'm using 90%+ isopropyl.

Another trick is to use die stones for the roughing work. The stones ablate - they form to the curve, and remove stock as evenly as the pressure of the operator. In this case I'd use ultra fine on the curves surfaces, and take it very easy. the flat surfaces get very fine files and high grit wet and dry paper - all well lubricated with isopropyl.

I'll look up my supplier. I think it was HH Roberts, but I'll post below.

 

[Dabbler]

See post #11

Truing a L0 chuck spindle bore

Just bought a 14" four jaw chuck that just fits my nominally 14" Celtic Mondiale lathe with an L0 spindle. The chuck is used, and it looks like the previous owner didn't bother to clean the spindle and bore many times. The chuck body is a one piece casting. The L0 taper bore has lots of...

canadianhobbymetalworkers.com

 

[Canadium]

More important than how you are removing metal, is how you are measuring where to remove it. And deciding which piece to remove it from.

Are you referring to bluing pastes; eg Permatex Prussian Blue???

 

[Dabbler]

I think making your own plates would be difficult...

3D printing them is also a good approach, adn def accurate enough.

Are you referring to bluing pastes; eg Permatex Prussian Blue???

That is one way. I looked back at the whole thread looking for the contact area between the cradle and the body. What I'm trying to describe is a step before Prussian Blue... On my dividing head, I have 4 side contact surfaces -before- the circular flatway touches. (if you like, 4 ears perpendicular to the curved flat surface).

They have to be fitted by hand and by feel. In sliding the body down into the saddle is very hard, you can do a scratch test instead. You coat all 4 surfaces, or rather 8 that might be making contact. the places of contact will scratch the magic marker away. Depending on your skill and the orientation of the parts, you can relieve the contact by taking the *tiniest* amount of metal from one or the other side. This can be done by file, wet and dry, or using a scraper. Once you do, you have to make it planar again. This needs a "straight" edge that can show you if you are carving out a pit.

If you have the material, you can turn a fitting model that is close to dimension of one side of the flatways on the saddle. If you make it really accurate, you can test the fitment one side at a time. This might help you find the twist. In addition, you can fixture the saddle and body on a surface plate and indicate the contact areas on the ears to look for twist.

You also need to measure that your alterations are perpendicular to the axis.. This can be done using a fixture on the lathe with a tenths indicator if you lathe is big enough. (or surface plate)

Once you have the alignment ears working so that they are moving without jamming, you can use prussian blue to print the curved flatway. It is probably very good. If you have a high spit or other anomily, this is the time to use 1000 or 2000 grit on a stick to take that out. An ablative die stone also works wonders on curves like this.

 

[Susquatch]

I confess I am troubled.

I've never disassembled one of these or put one together. But I'm having trouble trying to figure out why taking one apart would subsequently require refitting in order to be able to put it back together.....

Shouldn't this be just a matter of aligning it the way it was and then sliding it together? Why is all this fitting required when it was originally already fitted?

What am I missing? Help me understand.

 

[Canadium]

I confess I am troubled.

I've never disassembled one of these or put one together. But I'm having trouble trying to figure out why taking one apart would subsequently require refitting in order to be able to put it back together.....

Shouldn't this be just a matter of aligning it the way it was and then sliding it together? Why is all this fitting required when it was originally already fitted?

What am I missing? Help me understand.

Its a very old DH that I got cheap at an industrial auction. It was badly abused and was probably just kicking around unused for a yonks age before I got it. I'm guessing the only reason it didn't go in the scrap bin is because it cost them a fortune to buy it in the first place and they couldn't admit to themselves that they had ruined it. It has deep hammer marks all over the cradle which explains why the cradle was deformed. The parts were out of spec by about 11 thou. I came to the conclusion I could not force them back together again without breaking something. Besides the body is supposed to rotate inside the cradle for adjusting angles. It was totally seized before and I had to pry like heck to get them apart. To make matters worse it looks like someone injected grease into it when it was supposed to be oiled. I always was a sucker for a challenge but I think this sits at the heart of what I love about this hobby so much. I'm just completely taken with the potential for turning an old, abused, abandoned, and forgotten piece of precision machinery and bringing it back to a state where it can do something for me again.

 

[trevj]

3D printing them is also a good approach, adn def accurate enough.

Yep. remember that errors in the accuracy of the holes, are reduced by a factor of the gear reduction of the head's worm gear.

I made a 100/127 hole plate during lunch at work, on a drill press. I used large gears from another machine, that had the requisite tooth counts. I used a mandrel that held the gear to my plate blank, and that was able to pivot around a bolt through the center. Banged out a very fast and dirty spring catch to direct index off the gear teeth, little more than some scrap flat stock, and a couple bends, it was held in place with a C-clamp on the table of the drill. Enough preload that it held in place.
Swing the table to position your hole circle, set the depth stop, and it is done in about 10 minutes for each circle. One hand runs the drill press handle, the other rotates the catch from one notch to the next. Use a small center drill and you get a countersink as well.

 

[Canadium]

How do you remove a badly stuck key from a shaft??? I've tried every method I could find on the net except welding a long handle on to the key for leverage to pry it off. With my crappy welding skills I'm reluctant to try that.

 

[YYCHM]

How do you remove a badly stuck key from a shaft??? I've tried every method I could find on the net except welding a long handle on to the key for leverage to pry it off. With my crappy welding skills I'm reluctant to try that.

Mill it out?

 

[trevj]

How do you remove a badly stuck key from a shaft??? I've tried every method I could find on the net except welding a long handle on to the key for leverage to pry it off. With my crappy welding skills I'm reluctant to try that.

Hammer and a punch is my usual.

May graduate to a cape chisel if the details dictate.

Access to a mill? Mill a slot in some flat stock, and weld through the slot! A pull of aluminum foil off the kitchen supply, will do a great job of shielding the work from spatter. Fold and wrap.

 

[Rauce]

How big is the key?

One way is to drill through the key and tap threads. Thread in a bolt to act as a jacks screw. Might not work if it’s a really small key.

 

[Canadium]

Hammer and a punch is my usual.

May graduate to a cape chisel if the details dictate.

Access to a mill? Mill a slot in some flat stock, and weld through the slot! A pull of aluminum foil off the kitchen supply, will do a great job of shielding the work from spatter. Fold and wrap.

I've hammered, punched, and chiseled and only deformed the key without budging it. The welding option is definitely under consideration!

 

[Canadium]

How big is the key?

One way is to drill through the key and tap threads. Thread in a bolt to act as a jacks screw. Might not work if it’s a really small key.

The key is about 5X17 mm. Pretty small IMO.

 

[John Conroy]

I would tig weld a piece of 1/8" thick by 1/2" wide steel to the key. Small welds with cool down periods between. The heat will aid in loosening the key but too much could distort the shaft

 

[Susquatch]

Heat the shaft and put a tiny amount of supercool on the key?

Also, do you have a press and if so, can you get the shaft and a long collar into it?

 

[Canadium]

Heat the shaft and put a tiny amount of supercool on the key?

Also, do you have a press and if so, can you get the shaft and a long collar into it?

I have a press but because the key is stuck I haven't been able to remove the shaft from the big and heavy nose piece of the dividing head.

 

[Susquatch]

I have a press but because the key is stuck I haven't been able to remove the shaft from the big and heavy nose piece of the dividing head.

Can you rig up a Puller instead then? Make and Broach a collar to fit over the key, and then turn it over 45 degrees from the key and then pull the collar off with a good sized Puller. It might pull the key off with it. If not, it might shear the key and then you can get the shaft out to machine out the rest of the key.

If the Puller doesn't work at all, make a small broach to fit the collar and broach it out.

I feel your pain with welding. I wouldn't be doing that either. Basically, my welding sucks.

 

[historicalarms]

to be that tight, that key must have been tapers a smidge on the corner of the sides and then forcibly pressed into the slot....YYC's idea of milling the key itself sounds like the most viable remedy offered up so far. I'm thinking that if it is so tight that chiseling/prying wont budge it, welding something to it will just eventually pull off anyway.

 

[Canadium]

Can you rig up a Puller instead then? Make and Broach a collar to fit over the key, and then turn it over 45 degrees from the key and then pull the collar off with a good sized Puller. It might pull the key off with it. If not, it might shear the key and then you can get the shaft out to machine out the rest of the key.

If the Puller doesn't work at all, make a small broach to fit the collar and broach it out.

I feel your pain with welding. I wouldn't be doing that either. Basically, my welding sucks.

I'm guessing that you are assuming the keyway extends to the end of the shaft. Unfortunately this is not the case. It may be a woodruff key altho I think a woodruff key should come out more easily by simply hammering on one end. Nothing seems to work here tho.

 

[Dabbler]

How about reserving your options until a later stage of the rebuild?

If this is stopping you in your tracks, this is what I'd do:

(the key is a total loss already)
I'd take a die grinder and carefully remove all the metal I could in the centre of the key. With care, it should be easy as keys should be soft. It does not matter if you go a tiny bit too deep, as long as the sides of the keyway on the shaft aren't compromised. It looks like you will have to reach in beyond the face plate to get the last edges of the key.

Perhaps someone press fit the key and some coolant created rust? if that is the case a die grinder is the only way I know of getting ti free again.