Making a Worm Wheel

OK, we weren't aware of all of your design constraints in the beginning. I am assuming you are trying to make this because the commercial stuff is outside your budget. I don't think anyone was trivializing your project, just trying to help. Here is what I found with another search, but with what you have just said, is outside your budget. I have a feeling there is a reason these things have the prices they are. The third one on the list looks like the most cost effective. I am using a Google search "360 Tooth Worm Gear".

http://www.mathis-instruments.com/Products/Gears.html

Here are some more.

http://www.aeroquest-machining.com/pricing.php

https://www.ebay.com/itm/Ed-Byers-360-Teeth-15-Drive-Gear-Worm-11-16-Thick-2-Bore-Hole-/361794136860

Good luck with your project.
 
A DC, PWM servo motor will turn at 1 rpm. The Baldor you have may be abel to run as a servo, all you would need is a PWM driver, it's worth looking into. You would be able to make very fine adjustments and with a feedback loop it would stay exactly at speed.
 
Eddyde and others are heading in the same direction as my brain is taking me... Specifically; this is not a trivial problem.

If the OP is concerned about the discrete focus misalignment if a stepper motor is used then, we're talking about a non-trivial level of precision needed for the overall mechanism. Assuming the gears are perfect, even the motor would need to be speed controlled with sensors and some form or controller or PID.

Standard design problem here: The RPM needed for this is approximately 0.00069. Significant digits out to the 5th place. Assuming a drive source of 7 or 70 RPM, you'll need 100,000 or 10,000 times reduction. You can design this as 1:10 x 1:100 x 1:100 or, 1:100 x 1:100. Depending on what path you go, the gear tooth count will be finite and from this, the specifications on the design can be roughed-out in terms of parts per million (or some other scaling factor). Once that is known, the finite specs can be determined once the size of the parts are chosen. This is the kind of thing that slight temperature variances during operation will change tracking accuracy.

One thing I will say, is that with 100,000 times reduction, the final gears will need to be beefy. An ant on a treadmill at the input will be able to pull an oak tree out of the ground.

We need to know the desired level of tracking accuracy or, we're going to be shooting in the dark and crippled with analysis/paralysis.

Ray
 
Eddyde and others are heading in the same direction as my brain is taking me... Specifically; this is not a trivial problem.

If the OP is concerned about the discrete focus misalignment if a stepper motor is used then, we're talking about a non-trivial level of precision needed for the overall mechanism. Assuming the gears are perfect, even the motor would need to be speed controlled with sensors and some form or controller or PID.

Standard design problem here: The RPM needed for this is approximately 0.00069. Significant digits out to the 5th place. Assuming a drive source of 7 or 70 RPM, you'll need 100,000 or 10,000 times reduction. You can design this as 1:10 x 1:100 x 1:100 or, 1:100 x 1:100. Depending on what path you go, the gear tooth count will be finite and from this, the specifications on the design can be roughed-out in terms of parts per million (or some other scaling factor). Once that is known, the finite specs can be determined once the size of the parts are chosen. This is the kind of thing that slight temperature variances during operation will change tracking accuracy.

One thing I will say, is that with 100,000 times reduction, the final gears will need to be beefy. An ant on a treadmill at the input will be able to pull an oak tree out of the ground.

We need to know the desired level of tracking accuracy or, we're going to be shooting in the dark and crippled with analysis/paralysis.

Ray
Guys, no need to go deeper into this. I had requested a quote from anyone to build a simple worm wheel using a tap as the hob for the wheel. I included the descriptions of the gearing needed to get to the speed requirements of the drive to show only that it was non trivial issue. I do not expect anyone to design a drive system. My request was to build a simple worm wheel with 360 teeth using a 3/4 -1 thread tap as the cutting tool. I saw it done a number of times on YouTube and they seem to work just fine. I had hoped that there was at least one adventuresome machinist in the membership who would take on the task on making a wheel for me.

As I said, no need to go any further on this. I have several other ways to achieve my goals!. I thank all of ya'll for assisting and providing information to me and to others. This group has helped me tremendously in the past when I was getting my old 1948 Clausing Atlas lathe back among the living albeit it is worn completely out.

Again, many thanks and see ya'll later.

drmayf

ps, regards accuracy, yeah one of the 600 foot diameter dishes has an accuracy requirement of 1 millisecond of arc at some mind bending stellar distance. Closer is easier. Me I am sticking to under 20 light years because having an instellar conversation would take too long for anything further.

dm
 
I had wanted to hob a 360 tooth worm wheel using my old 1948 worn out lathe to do this. Sadly, neither it nor I am up to the tasks of making that wheel.

Well, aside from the obvious comment, that my pet worm won't appreciate teeth on his wheel toy,
it might be interesting to consider using an odd-ratio (cheap) worm drive (like in lots of surplus
gearmotors), in conjunction with a good rotation resolver. Instead of dead reckoning, you can use any old
ratio of drive, and take feedback from the shaft encoder to get it into the right position...

Not sure how useful that option is, all the 'encoder' items I see are high price and mediocre
resolution.
 
Looking at the reduction and accuracy needed, have you looked at harmonic drives? One could be a ready-made solution for the final reduction gearing and they're appearing on the surplus market...

When I looked after fast-moving 60-foot dishes, they used worm and wheel reduction in Az and El from DC servo motors (about 100 HP each...) but the servo feedback was partly derived from the signal at the focus - if I remember right, there was a phased set of probes that could determine whether the received signal was off-axis. That worked pretty well for the fine error once the dish's position and traverse rates were supplied by a PDP11-73, even for low-orbit vehicles that might go horizon to horizon in under a minute.

Dave H. (the other one)
 
if I remember right, there was a phased set of probes that could determine whether the received signal was off-axis. That worked pretty well for the fine error once the dish's position and traverse rates were supplied by a PDP11-73, even for low-orbit vehicles that might go horizon to horizon in under a minute.

Dave H. (the other one)

PDP11-73 OMG. I've been placed into the "Way-back" machine... Como saved my life a few times.

Ray
 
dm
Whatever you decide to do on this project, I for one would love to see what you do and how you do/did it. I almost pursued a project like this
a decade or so ago, and finally decided to just stick with optical telescopes. Keep us informed and good luck!

CHuck the grumpy old guy
 
Back
Top