Dividing head math for large # of gear teeth

I'm not scared of making a good plate. I won't divide that. I will drill a hole where the dro says to 127 times. I played with hole size, hole spacing, and circumference. Im pretty sure I can get it on a standard size plate. I am curious what method of indexing when I cut the gear will be more accurate. Do you think direct indexing without the gear train (plate mounted to rotable face) will be any less (or more) accurate than simple indexing with the gear train in tact (plate mounted to dividing set up)? tcarrington, what do you mean about 50 something hole plates?
 
My preference is a little more "brute force", I don't like electronic gizmos doing my thinking for me. I used a "spin indexer", is that the proper name? In any case, I used the indexer and mounted a 127 tooth gear from a Grizzly lathe. The Grizzly was Modulus 1 which equates to 25.4 DP. With this rig, I cut a 16 DP (plastic) gear for an Atlas lathe. Simple, boring as hell, slower than Christmas, but everything came out first try. Do I need a 16 DP 127 tooth gear? No, there are several combinations for an Atlas that get within a partial percentage. But I proved to myself that I could do it with what I had and that's what matters.

For what it's worth, a fellow here recently ran a 127 point circle on a DRO and came up with a 126 point circle on the mill. Something about that particular DRO, I don't recall what actually happened but I simply don't trust an electronic gizmo. I worked 50 plus years building such machines. For my own use, I prefer to do everything manually. I use AutoCAD for convenience, it can divide a circle 127 times. But that little gear from the Grizzly wasn't electronic. It just worked. . .
.
 
I think the DRO I got would have a neurotic breakdown if I asked it to do a 100+ hole bolt circle.
But then, you get what you paid for ;)
 
Bi11Hudson said:
For what it's worth, a fellow here recently ran a 127 point circle on a DRO and came up with a 126 point circle on the mill. Something about that particular DRO,
Click to expand...


This is a classic DRO gotcha. The problem arises when you ask it for N number of holes and set the start and end at the same angle. Let's say you ask for 4 and enter a start and end angle of zero. It'll give you a three hole pattern because you've asked for the last hole to be at the same angle and position as the first. You can either ask for 4 holes and manually calc the ending angle of 270, or you can just ask it for 5 holes, set zero start and end and get the result you're actually wanting with no calc needed. It's an operator issue, rather than an electronic one (meaning no disrespect to whoever was involved) ;)

It's actually because if it's flexibility that you get this issue. If the function was simply "circular pattern", you'd expect to enter the number of holes, start angle and let her rip. As it let's you spread the pattern over an arc shorter than 360 - lets say 90 -you'll see that it works exactly as intended: 4 holes, 0, 30, 60, 90. You need that "ah ha" moment to realise that entering zero as the end angle isn't going to end well!!!

@LEEQ I've successfully 3D printed plates for odd jobs like this. They won't last forever, but for one offs like this it's well worth it, cheap and quick
 
Lo-Fi said:
This is a classic DRO gotcha. The problem arises when you ask it for N number of holes and set the start and end at the same angle. Let's say you ask for 4 and enter a start and end angle of zero. It'll give you a three hole pattern because you've asked for the last hole to be at the same angle and position as the first. You can either ask for 4 holes and manually calc the ending angle of 270, or you can just ask it for 5 holes, set zero start and end and get the result you're actually wanting with no calc needed. It's an operator issue, rather than an electronic one (meaning no disrespect to whoever was involved) ;)
Click to expand...

As a point of reference I would ask if the DRO systems have the capibility of doing a reverse or negative angle. With AutoCAD I want to rotate an angle to 270 degrees, I can enter -90 and get the same result. It's no big deal for the basic compass points, but an angle of negative 13.25* saves me from doing the math in my head to find the inverse angle.
.
 
Sounds like Bill votes for direct indexing. Funny thing ,Bill. I'm looking to cut a 127 tooth gear for my grizzly like you cut for your atlas with the grizzly gear. It's also a can I kind of thing.
 
I'm sure I can chuck up the old sharpie and mark the positions to check the dro.
 
As far as printing plates, that seems to be a good easy way out. I don't have the capability though. I think letting the dro think for me is a plenty good cheat for me, and my plate can have other dividing circles added to it as needed with no worries of longevity.
 
Well I dug through the manual for my el400 dro. It looks like it wants me to input center in x and y, radius of pitch circle, number of holes, and start angle. after that I can mark the first hole, then advance to hole 127 and mark it to make sure its not on top of the first hole. This dro doesn't ask me for finish angle, and it lets me scroll through or select by number the position I want to go to next.
 
I'm sure you'll get a great result from the DRO method. That being said, there are plenty of people here who would be happy to help out printing a plate for you, I'm sure.

Not sure about negative angle, Bill. I suspect it only works "clockwise" and you'll manually calculate the start and end angles.

That's an interesting way your DRO works, Leeq. Sounds pretty sensible to me! Pics as you progress please :)
 
You must log in or register to reply here.
Back
Top