Hold a gear for cutting?

[coulsea]

A small amount of error really shows on a gear with small teeth. everything has some degree of error so it is best to eliminate any unnecessary source. your collet block has the potential for error mounting in the chuck, the collet to the holder, the collet to the arbour and the holder itself. so mounting the arbour directly to the chuck gets rid of it completely. Also do the final sizing of the blank with it mounted on the arbour. I have a mark on the arbour that lines up with the no1 jaw on the lathe so it always goes back in the same position, you could also experiment with a dial indicator to find the best position in the rotary table and mark that too.

 

[Gordon]

Tight spots are because of nicks on the bevel gear drive from the chuck key. Running a chunk facing up lets a lot of chips into the chuck. I have had it apart twice and cleaned it with brake cleaner and a wire brush. I am not sure what is causing the problem. Filed the nicks with a needle file.

 

[Richard Hed]

Tight spots are because of nicks on the bevel gear drive from the chuck key. Running a chunk facing up lets a lot of chips into the chuck. I have had it apart twice and cleaned it with brake cleaner and a wire brush. I am not sure what is causing the problem. Filed the nicks with a needle file.

I hope you get this fixt soon as I am dying of curiousity as to what is the prob. Often it is some simple little overlookt thing that you will kik yourself in the a$$ when you find out.

 

[Gordon]

I found the problem by centering the RT on the mill. Put a3/4 dia pin in the spindle and tightened the jaws of chuck on the pin and secured the chuck to the RT. When I rotated the RT it was not running centered. Put the 6" chuck in the 8" four jaw on the lathe and indicated in on the OD. Jaws are not centered.

 

[Richard Hed]

I found the problem by centering the RT on the mill. Put a3/4 dia pin in the spindle and tightened the jaws of chuck on the pin and secured the chuck to the RT. When I rotated the RT it was not running centered. Put the 6" chuck in the 8" four jaw on the lathe and indicated in on the OD. Jaws are not centered.

You said you chekt that you have the jaws in the correct slot, didn't you? You might want to double check that and while you're at it, make sure the threads on each jaw are in good shape. When is the last time you removed the jaws? I see you have removable/reversible jaws , could you have put one of those in the wrong slots? I have that type on my chuck but don't rememabers if they are interchangeable.

 

[Brian Rupnow]

Gordon--I always cut gears with my rotary table set up 90 degrees to what you have shown. I have a piece of 1/2" round steel turned to a sharp point on one end. I always hold that part in the chuck and be very careful to set the height of the gear cutter exactly in line with the pointed end. You are right, trying to hold a gear-blank that large on a 3/8" shaft is just asking for trouble. It will deflect and cause problems with the gear being cut. My 3 jaw chuck that I have permanently centered and mounted to my rotary table has about .003" total indicated runout, but that is not enough to cause a problem with the gears being cut.

 

[Gordon]

Each jaw and top is numbered and I have had them out and checked. If they were in the wrong slot they be off center more than they are now. Obvious difference in each jaw.

 

[Richard Hed]

Each jaw and top is numbered and I have had them out and checked. If they were in the wrong slot they be off center more than they are now. Obvious difference in each jaw.

And the scroll doesn't have a big dent in it at the distance does it? Just how, exactly, is the chuck connected to the indexer? There's no way the chuck itself is mis-aligned on the indexer, right?

 

[Chiptosser]

This is exactly why I like having independent jaw chucks.

 

[Gordon]

OK I found the problem. There was a burr on the removable top jaw on #1 and it would not seat properly. The removable jaws have always been really tight and they had to be driven off. apparently when I replaced the jaw at some point it caused a burr. I have removed the burr and polished the slots and key so that they do not have to be driven on. I will polish the other removable jaws so that they go on easier.

I also had the cutter at the wrong height. That as I suspected was a math error. I center the cutter by bringing the cutter down to the top of the blank and then raising the table 1/2 the cutter width plus 1/2 the blank diameter. It turns out that it is important where you place the decimal. Too soon old and too late smart.

I will set it up again and do more checking before I start cutting.

 

[Peter Twissell]

Are you aware that Chuck jaws are normally supplied with undersized slots, to be individually fitted to the chuck?
This allows for some wear in the chuck.
Jaws should be a sliding fit, not tight.

 

[ajoeiam]

The reason why I have the setup at 90° is perhaps not a really valid reason but I did that because there was not enough room on the table to put the RT without removing the vise. If I removed that I would have to reset the vise so for the one project I thought that I would just live with the inconvenience of being in a less than optimal position. That being said that is not my problem. I found the problem. The three jaw chuck has .025 runout. I am not sure why it has not been a problem in the past because I have been using it for several years and I am pretty sure that I have checked it before. It is not a sprung scroll because I have checked it with a 1 dia shaft and a 1 1/2" dia shaft. Jaw #2 is about .030 further from center than #1 and #3. That is not enough to indicate the jaws in the wrong slot. I am pretty sure that the chuck is not repairable so I guess that means a new chuck.

Nope - - - means you need to grind the jaws.

Well - - - - you found the issue but if you had found no issues - - - it still wouldn't mean scrapping the chuck.

(edited to include further information from subsequent posts (1st line relates to the first comments from the OP))

 

[Richard Hed]

OK I found the problem. There was a burr on the removable top jaw on #1 and it would not seat properly. The removable jaws have always been really tight and they had to be driven off. apparently when I replaced the jaw at some point it caused a burr. I have removed the burr and polished the slots and key so that they do not have to be driven on. I will polish the other removable jaws so that they go on easier.

I also had the cutter at the wrong height. That as I suspected was a math error. I center the cutter by bringing the cutter down to the top of the blank and then raising the table 1/2 the cutter width plus 1/2 the blank diameter. It turns out that it is important where you place the decimal. Too soon old and too late smart.

I will set it up again and do more checking before I start cutting.

Hallelujah! It's really nice that you found such a simple problem causing such nasty results. -- and HUGE bad results too. Now for yuour other problem. You didn't want to tear down your vice and your couldn't turn the indexer to the right, how about turning the indexer to the left instead? Your setup is so difficult to see as you cut. I woud say you are bound to make cutting mistakes. Also, as I practice, it might be a good idea to at least turn the indexer enough so you can check the four corners of your disk to make sure the cutter touches as it is turned. I do it, not for that reason, but rather make a tiny cut and then count the teeth marks. I know how easy it is to get it wrong.

 

[Gordon]

As usual one problem leads to another. I found that I had to tram my mill head. When the head is not square to the table my method of adjusting the chuck to the table by tightening the chuck to a pin in the spindle does not work when it is not square. With my stack up extending several inches above the table the slight angle of the ram ended up being a larger error.

Now back to the regularly scheduled program.

 

[clockworkcheval]

The major amount of components in clocks tend to be gears. So if you ask three members of our horological society the best way to hold gear blanks you get at least five different answers. The common straight road to succes is mounting the blank on an arbor, like Brian Rupnow already mentioned. However we prefer to hold the arbor in a collet, as in clocks you try to limit the run-out as much as possible, say to 0,01 - 0,02 mm. You obviously need to use thoroughly cleaned and well serviced collets and colletholders to achieve this. The diameter of the arbor should be the maximum diameter the collet can hold to minimize deflection. We try to avoid the use of a three jaw chuck. With larger gears we prefer the independent four jaw chuck. There is an extreme variety of methods to center the cutter to the blank. A nice but expensive method is the use of a Center Finder scope like from SKOAL, which you put in place of your cutter. A much cheaper and arguably at least as precise method is to put a piece of round brass stock with blackened endface in place of your blank. You then put in any which way - with a height gauge or measuring instrument etc - the cutter at the right height. You then make a half-diameter light trial cut along the endface of the piece of brass, rotate 180 degrees and make another trial cut. Adjust and repeat till the trial cuts form one line.
Several members prefer one set-up to turn the gear blank and cut the teeth with the aid of a milling attachment, so as not to loose precision with the transfer of the blank from the lathe to the mill.

 

[Gordon]

For those who did not like my setup. Here is what I ended up with. I Recut the same gear so it is not usable but it proved that everything is now lined up properly.

 

[Richard Hed]

For those who did not like my setup. Here is what I ended up with. I Recut the same gear so it is not usable but it proved that everything is now lined up properly.

Well, the obvious question is: Is it easier to work on?

 

[Gordon]

Well, the obvious question is: Is it easier to work on?

Being easier was not really an issue. Getting the gear cut properly was the goal and once everything was indexed properly The problem was the chuck not holding the part correctly not the position of the chuck. I could have used the previous setup and been just fine. Just a little inconvenient.

 

[Richard Hed]

Being easier was not really an issue. Getting the gear cut properly was the goal and once everything was indexed properly The problem was the chuck not holding the part correctly not the position of the chuck. I could have used the previous setup and been just fine. Just a little inconvenient.

A LITTLE inconvenient. Just a little. I don't think I would have wanted to work over the end of the table with the traversing handle poking me with the chance of moving it. Anyway, it looks like you are on your way and I hope to see the good results soon. HOw many gears are you making? I once had a setup in which I had several of the same gear to make so I simply held a long rod between centers and cut the teeth the whole length then when done, cut the thing into slices. I had to do a little cleanup on them later.

 

[Gordon]

I made nine gears for the drive and two for the timing. They all worked well except I used the wrong material for one of the timing gears and it was not strong enough so it stripped off some teeth. It was oil impregnated brass which I had in the odds and ends pile and I thought that it did not have much pressure so it would be OK. I made all of the gears, including the one this was to replace using my spin indexer. I could have used the spin indexer again but I had just gotten these dividing plates and I wanted to try them.