Group Project: Dividing Head - The Build

[extropic]

Suggestions:
Put a thrust bearing w/ washers between the sector arms and the spring to decouple torque. It looks like you have plenty of shaft length to accommodate a TB.
I have used an O-ring, stretched over the gap of a thrust bearing w/ washers to keep swarf out. It's not an idiot proof seal but it's cheap, simple and I've had zero problems using it in similar situations.
I've never used the wavy washer stack springs that you picture. My go to would be Bellville spring washers because force is easily adjustable as required. The wavy washer stack looks like a swarf trap. Not a problem with Bellevilles.

Would the sector arms pictured work if you decoupled the torque?

 

[rwm]

Suggestions:
Put a thrust bearing w/ washers between the sector arms and the spring to decouple torque. It looks like you have plenty of shaft length to accommodate a TB.
I have used an O-ring, stretched over the gap of a thrust bearing w/ washers to keep swarf out. It's not an idiot proof seal but it's cheap, simple and I've had zero problems using it in similar situations.
I've never used the wavy washer stack springs that you picture. My go to would be Bellville spring washers because force is easily adjustable as required. The wavy washer stack looks like a swarf trap. Not a problem with Bellevilles.

Would the sector arms pictured work if you decoupled the torque?

Not really enough room to add thrust bearings and still have reasonable room for the spring. I had to machine a little off here and there to maximize shaft length as it is.
Send me an image of the TB you could see working here.
I agree, I did not think about chips in the wave spring. But if we are worried about that, this whole design is suspect! How about the worm/gear?!
Sector arms would work otherwise although they are not the highest quality.
I still think reducing the spring OD would solve this issue regardless of the spring type.
I could actually solder a bushing inside the sector arm to reduce it's ID to 5/16. The solder line would not really show. I may do this just to have a working prototype. Jeff has sector arms that are Delrin; kind of intriguing....

 

[extropic]

For the TB info click on the "thrust bearing w/ washers", two hot links in reply #581. From your reply, I infer that you're not familiar with Bellville spring washers. Am I correct?

 

[rwm]

For the TB info click on the "thrust bearing w/ washers", two hot links in reply #581. From your reply, I infer that you're not familiar with Bellville spring washers. Am I correct?

Thanks for the links. I was thinking correctly. Those TBs would be much too large to use here. The ID would need to be .875 which means the OD would be very large and would interfere.
I am familiar with Bellville washers. It was my understanding that they generate much larger force loads over smaller distances. In the diameters we are looking at here the force would be much to great and too sensitive to small adjustments. I confess I have not done an exhaustive search.
Anyway, I do appreciate the comments and ideas. They are helpful and fun to consider even if I don't use them. Thanks for following.

 

[extropic]

snip>
I could actually solder a bushing inside the sector arm to reduce it's ID to 5/16. snip>

Thanks for the links. I was thinking correctly. Those TBs would be much too large to use here. The ID would need to be .875 which means the OD would be very large and would interfere.
I am familiar with Bellville washers. It was my understanding that they generate much larger force loads over smaller distances. In the diameters we are looking at here the force would be much to great and too sensitive to small adjustments. I confess I have not done an exhaustive search.
Anyway, I do appreciate the comments and ideas. They are helpful and fun to consider even if I don't use them. Thanks for following.

Is the 5/16" ID, that you specified in reply #582, a typo?
I'm confused why the sector arms need a 5/16" ID while the TB would need to be 7/8" ID?

 

[rwm]

The whole problem arises because of trying to use the ebay sector arms. They have a 21mm center hole. I made a sleeve to upsize the shaft. The spring needs to go over this sleeve.
My comment in #582 was proposing to reduce the ID of the sector arm. I have not done that but I think it would eliminate the problem.
I am typically averse to soldering brass together for a project. I would make new arms eventually. If I use a sector arm with a 5/16 ID I think any spring will work fine without adding a TB.
Let me solder and re-work the sector arms and I will report back. This is definitely a design consideration we want to check out and nail down.

Sector arms over 21mm sleeve:


One nice thing about the sleeve is that it prevents forward motion of the shaft.

 

[extropic]

My comments about the TB and Bellville washer(s) was based on viewing the pictures you have provided, not any evaluation of the "drawings".
Curious why the waters got more muddy, I have looked at, and tried to interpret the "drawings".
The worm Shaft (DH 013) is .561" OD.
The Sector Bottom (DH 010) is .562" ID with 2" OD.
So we're talking about 9/16" not 5/16".
Pressing or soldering a bushing to reduce the 21mm to .562" is the obvious practical, near term solution to the sector arms.
A 5909K951 TB is 9/16 ID with 1" OD (NO problem) x .078 thick and leaves plenty of OD clearance to add an O-ring swarf seal.
Using 2 each 5909K979 (one on both sides of the TB) makes the TB stack .078 + .125 = .203 ish.
The TB washers are available in 1/32" or 1/16" thicknesses, your choice.
After my cursory look at the "drawings", I see that a 9/16" ID Bellville would be too fiddly to set to the low force you're after.
However, McMaster offers a variety of wavy spring washers, curved disc springs and finger disc springs. I'm sure you'd find something to use without modifying the worm shaft. Even a felt washer would suffice.

On a separate subject. After looking at the spring loaded indexer pin assembly, I'm curious why the design includes the complexity of anti-rotation features for the knob and pin? The Index Body (DH 032) has a 3/4" hex area which is the obvious installation/removal feature. Why should the knob/pin not rotate? I understand that if the index pin had runout of the pin end relative to the shank, that runout would add variability to the indexing. But the runout control is so much easier than creating the anti-rotation features, I'm perplexed. Do commercial indexers use non-rotatable index pins?

 

[extropic]

I just saw the photo you added to reply #586 and see the LONG bushing you added, which is causing much of our disconnect. Please, .5625" is 9/16", not 5/16".

I understand your comment about the long sector arm bushing controlling the axial play of the worm. my suggestions were based on staying as close as possible to the current "drawings". A lot of ways to skin that cat. The high axial force of a Belleville washer, bearing on a TB, would hold the shaft's collar against the Gear Shift Housing (DH 022) at the expense of pressure/wear. Another option is to go back to the "as designed" sector arms. Add a TB w/ washers on the 9/16" shaft and add a spacer tube between the nut and the TB stack. The length of the spacer tube would set axial clearance without any preload. A spring washer would be chosen with ID to clear the spacer tube OD.

I dislike this long range attempt at collaboration. I'm too slow a typist.

 

[rwm]

I just saw the photo you added to reply #586 and see the LONG bushing you added, which is causing much of our disconnect. Please, .5625" is 9/16", not 5/16".

Oh crap. I missed that typo every time. Sorry!
I like your idea of a thin spacer tube to set the axial play. I'm gonna try to implement that.

On a separate subject. After looking at the spring loaded indexer pin assembly, I'm curious why the design includes the complexity of anti-rotation features for the knob and pin? The Index Body (DH 032) has a 3/4" hex area which is the obvious installation/removal feature. Why should the knob/pin not rotate? I understand that if the index pin had runout of the pin end relative to the shank, that runout would add variability to the indexing. But the runout control is so much easier than creating the anti-rotation features, I'm perplexed. Do commercial indexers use non-rotatable index pins?

It is not anti-rotation. It is a mechanism to lock the pin in the retracted position so you can make rotations without falling into a hole.
I actually have something else in mind entirely! Here is a teaser:

This purports to be a sleeve that will go over the index pin assembly that will allow free rotation. There will be a separate retract lock.

I dislike this long range attempt at collaboration. I'm too slow a typist.

Agreed. But still fun and helpful!

 

[extropic]

Aha! I now I understand the intended purpose of what I thought were anti-rotation features on the index pin assembly. I didn't imagine that one would retract the pin about 3/4" to lock it out. The thickness of the sector arms + the index plate is only .500".

The whole issue of what type and force of spring to use has soaked in, I think. The spring force is the way the sector arms are held stationary against the index plates. Isn't that correct? You need enough spring force to secure the sector arms against any inadvertent bump. Too little spring force and the sector arms are not securely fixed. High spring force transmits torque (without a TB), preloads the worm shaft and makes it more difficult to reset the sector arms for each subsequent index. It will be interesting to see what configuration you end up with.

Here is alternative concept. Grow a snout on the index plate face of the Gear Shift Housing (DH 022). The snout would be long enough to extend through the index plate (.250.") plus the Sector Arms (.250") plus be OD threaded for a knurled nut + washer. No spring or thrust bearing needed.
Advantages: The Sector Arms are mounted to the same part as the Index Plate and they are secured or free to reposition by using a knurled nut that doesn't perturb anything else.
Hopefully, the worm shaft could stay as currently designed. The current design IDs of the Index Plates and Sector Arms and probably the bolt circle of the Index Plate mounting would need to be increased to accommodate the snout OD. If you use about a 21mm OD snout, it will work with the sector arms, and I assume the Index Plates, you bought.

If I had access to a suitable CAD system, all this chatting would become more fun and more clear. Sorry about that.

PS: I don't know why anyone (without a fine tolerance CNC mill) would want to make their own Index Plates. If one has a fine tolerance CNC mill, why would one want to build this indexer? Altering this indexer design to use easily available Index Plates and Sector Arms makes perfect sense to me.

I've gotta go engage real life for a while.