ICEINSPACE
Most Read Articles
Moon Phase
CURRENT MOON New Moon
0.2%
The Sun Now
Time Zones
Sydney
4:40 am
Perth
2:40 am
Auckland
6:40 am
New York*
2:40 pm
Paris*
8:40 pm
GMT
6:40 pm




Go Back   IceInSpace > Equipment > ATM and DIY Projects

Reply
 
Thread Tools Rating: Thread Rating: 12 votes, 5.00 average.
  #1  
Old 30-03-2009, 08:01 PM
kinetic's Avatar
kinetic (Steve)
ATMer and Saganist

kinetic is offline
 
Join Date: Jun 2008
Location: Adelaide S.A.
Posts: 2,218
Hobbing a worm gear Part 1 and Part 2

Some pics of the steps I'm taking to attempt to make
a high accuracy worm and worm gear.
(Inspired by a few articles on the internet and also
keenly following Jeff's progress on his Project 24 thread):
http://www.iceinspace.com.au/forum/s...ad.php?t=39135

This is only part 1. Hopefully, if this works, I will have
a part 2

I started with the spindle.
I made it out of a pipe fitting welded to a piece of ms plate.
After turning down the two housings for the bearings, I then
made a shaft and turned a thread either end.
These parts once assembled are mounted on the tool post
cross travel on the lathe.
This gives you cross travel adjustment for the cut and transverse
travel to centre it on the thread cutter. Height is obviously critical
and some homework was done to make the spindle just the right
height. Some fine adjustment can be done with packers under the
'flange' (below).

The flange for holding the raw worm wheel blanks was turned down from
an old VCR head.
These have the most beautiful aluminium I have ever seen. Probably
what they call aircraft grade.

I cut two steps on the flange for locating centre the two types of
worm wheels I'm ever likely to make:
One for fitting on a 1.5" shaft (my R.A. axis)
One for fitting on a 1.25" shaft (my Dec axis)

To lock the flange down on the spindle I needed to temporarily lock
the spindle from rotating.
This was done with a hole drilled through the side of the spindle
housing and shaft to enable a pop rivet to be slid through for tightening
and undoing the flange. And also for pre-loading the bearings to eliminate
any play (very important).

My first test run will be the aluminium platters from a very old hard disk
drive. These are enormous, but also beautiful high grade aluminium.

The worm thread cutter is obviously also critical for accuracy.
As my lathe can cut almost any metric threads, I decided to try
an M12 thread to start with on the test case blank.
My proper cutter (not shown) is an M12 plug tap mounted on an
extension and held between the chuck and a live centre.

For the test, I will just use a piece of M12 thread also mounted in an
extension.
I have read that the best worm thread is an ACME shaped thread , the
type similar to a vice thread or car jack.
If these test runs go well, I will then try making an ACME shaped thread
in the lathe, starting with M12.

more to follow.

Steve
Attached Thumbnails
Click for full-size image (tool_post_removal.jpg)
55.9 KB887 views
Click for full-size image (spindle_bits.jpg)
28.5 KB837 views
Click for full-size image (spindle_complete.jpg)
56.8 KB839 views
Click for full-size image (vcr_head_flange.jpg)
86.2 KB897 views
Click for full-size image (fitted.jpg)
90.9 KB925 views
Click for full-size image (rough_test.jpg)
73.7 KB1282 views

Last edited by kinetic; 04-04-2009 at 07:53 PM. Reason: M12 not M10!!!
Reply With Quote
  #2  
Old 30-03-2009, 08:09 PM
GeoffW1's Avatar
GeoffW1 (Geoff)
Registered User

GeoffW1 is offline
 
Join Date: Sep 2006
Location: Sydney
Posts: 1,835
My word,

I would like to send my applause and appreciation. I started my working life as a Fitting & Machining apprentice, so I can see what this involves.

Cheers
Reply With Quote
  #3  
Old 01-04-2009, 04:08 PM
kinetic's Avatar
kinetic (Steve)
ATMer and Saganist

kinetic is offline
 
Join Date: Jun 2008
Location: Adelaide S.A.
Posts: 2,218
Quote:
Originally Posted by GeoffW1 View Post
My word,

I would like to send my applause and appreciation. I started my working life as a Fitting & Machining apprentice, so I can see what this involves.

Cheers
Indeed, thanks Geoff, but the hard work has all been done by others.
A few really good articles on the web about this subject.
I thought I'd dive in and have a go myself.

Progress:

Tried the first test blank today. Old Hard disk platter, aluminium
mounted to the spindle flange.
M12 test cutter with splines cut with a Dremel.
(Makes it cut better just like a thread tap).

With the lathe on lowest gear I wound the platter in until it just
made contact with the cutter.
The cutter immediately began to drive it's way around the blank.
Nice cutting action. When it had driven all the way around the blank
and made a nice initial set of cuts, it didn't exactly meet up with the
first cut.
This is to be expected.
Unless I had made the blank an exact circumference to match a
theoretical thread pitch, this was never going to meet up.

All of the web articles I have read mention this.
What I then did was to just continue and watch it attempt to
'self index' by cutting deeper.
Eventually, a second set of 'mountains and valleys' appear as the
cutter made it's way around several dozen orbits.
The second set are eventually eroded away until the main set
sort of 'take control'.
The trick is to stop adding depth to the cut once the self index finds
exactly the right circumference for the start to meet the end.

It went better than I expected!

Pics attached.

Steve
Attached Thumbnails
Click for full-size image (testblank1.jpg)
61.2 KB446 views
Click for full-size image (testblank2.jpg)
60.9 KB650 views
Click for full-size image (testblank3.jpg)
15.2 KB740 views

Last edited by kinetic; 04-04-2009 at 07:52 PM. Reason: M12 not M10!!!
Reply With Quote
  #4  
Old 01-04-2009, 08:39 PM
Gama's Avatar
Gama
Registered User

Gama is offline
 
Join Date: Dec 2005
Location: Melbourne
Posts: 1,121
Maybe look at using a friction drive method here and a worm/wheel setup on the drive shaft.
Low to no PEC this way.

Theo.
Reply With Quote
  #5  
Old 02-04-2009, 10:03 AM
kinetic's Avatar
kinetic (Steve)
ATMer and Saganist

kinetic is offline
 
Join Date: Jun 2008
Location: Adelaide S.A.
Posts: 2,218
Quote:
Originally Posted by Gama View Post
Maybe look at using a friction drive method here and a worm/wheel setup on the drive shaft.
Low to no PEC this way.

Theo.
Hi Theo,

not exactly sure what you mean?. Do you mean setting up
a motor driven wheel while cutting?

Or give away the machining a gear alltogether and steer towards
a friction drive/worm , a la motor driven Alt/Az Dobs?

They too have PEC, the most noticeable being non-concentric worms
and slippage/non concentric friction wheels.

All drive related PEC is measurable and manageable in my Bartels
Stepper drive system. I just rarely have it turned on!

Steve
Reply With Quote
  #6  
Old 02-04-2009, 01:14 PM
kinetic's Avatar
kinetic (Steve)
ATMer and Saganist

kinetic is offline
 
Join Date: Jun 2008
Location: Adelaide S.A.
Posts: 2,218
More pics

Last night I tried two old hard drive platters screwed together
to simulate a thicker test blank.

This will almost be the thickness of an actual blank if/ when
all of this works ok.

I also shaved about 2mm off the diameter to make a fresh
start point to attempt the self index.

Self indexing went beautiful this time.
The start met the finish. By pure luck, I suppose,
the circumference must have been just right.
I noted the diameter!

I gradually gave the cutter more depth until the valleys of the
thread had just the right looking amount of wall left between
them.
Once a nice looking depth was reached I just let it drive itself
around for a good half hour with some cutting compound to
smooth over the roughness.
After a bit of a polish up with fine wet and dry it looks like
quite a good serviceable worm gear!
Apart from the fact that this blank is made of two platters,
(you can see they don't quite sit flat together because of trapped
grit and burrs etc), I think this gives me confidence in
attempting my first real full thickness blank.

Steve
Attached Thumbnails
Click for full-size image (2platter_test1.jpg)
125.2 KB403 views
Click for full-size image (2platter_test2.jpg)
87.8 KB490 views
Click for full-size image (2platter_test3.jpg)
36.7 KB651 views
Reply With Quote
  #7  
Old 02-04-2009, 04:57 PM
Gama's Avatar
Gama
Registered User

Gama is offline
 
Join Date: Dec 2005
Location: Melbourne
Posts: 1,121
Friction drives use 2 rollers pressed firmly together or by a band.
They provide very low or no PEC.
The larger mount mobs all use this method.
You can use the worm setup to drive the smaller roller. A few images of one here http://www.observatoryscope.com/prot...operation.html

Theo
Reply With Quote
  #8  
Old 02-04-2009, 05:54 PM
kinetic's Avatar
kinetic (Steve)
ATMer and Saganist

kinetic is offline
 
Join Date: Jun 2008
Location: Adelaide S.A.
Posts: 2,218
Quote:
Originally Posted by Gama View Post
Friction drives use 2 rollers pressed firmly together or by a band.
They provide very low or no PEC.
The larger mount mobs all use this method.
You can use the worm setup to drive the smaller roller. A few images of one here http://www.observatoryscope.com/prot...operation.html

Theo
thanks Theo for the advice.
I do know a thing or two about PE and different drive methods.
Are you sure you're not confusing PE for backlash?
Any gearbox drive train will have PE.
The final drive method turning the RA axis obviously contributes
the most 'error' to the final drive shaft.
I know drive belts (your example shows belts on both axes) would have
almost no backlash, but the biggest PE component would be the
accuracy of the smaller pulley wheel driving the big RA pulley.
If this is dead on concentric, then the next biggest PE would be the
accuracy of the worm/wheel driving this small wheel.
You're probably correct that this whole combination has less PE
than a rather large , well machined worm/worm wheel driving a
GEM.

But it would still have some PE.

I have measured the PE in several systems over the years (using K3
Drift Explorer) and for gear trains it is usually a sine wave.
The main sine wave is repeating/cyclic fast-then-slow PE of the very
last gear in your drive train.
This is the machining errors of that gear.
Superimposed on that sinewave is a smaller cyclic sine wave which is
the next gear back in the drive train.
These are measurable and able to be zeroed out in the Bartels system.

As there is a worm involved, albeit one gear back in the drive train of
your example, it would have a measurable error and contribute PE.

Although I fully appreciate your advice, I thought it would do no harm
to attempt to machine a worm/worm gear.

Steve
Reply With Quote
  #9  
Old 03-04-2009, 02:44 AM
Gama's Avatar
Gama
Registered User

Gama is offline
 
Join Date: Dec 2005
Location: Melbourne
Posts: 1,121
Well, nothing is 100%, you will always get a measurment, but in imaging youd be hard pressed to see it. Heres the site with some images, plus you may want to see the images taken by the scope. All done without guiding.
http://www.observatoryscope.com/

Theo
Reply With Quote
  #10  
Old 03-04-2009, 08:08 PM
kinetic's Avatar
kinetic (Steve)
ATMer and Saganist

kinetic is offline
 
Join Date: Jun 2008
Location: Adelaide S.A.
Posts: 2,218
Final part of the first test was to separate the two HD platters,
clean off all burrs and smooth them down with wet and dry.

Re-joined them and pop riveted them together.
Then I did another half an hour bedding in the
worm cutter with some polish to average out the cut.
Teeth are looking pretty clean and consistent now.

This worm gear could easily go straight on a DEC axis and do
a reasonable job.

The next step I have in mind is to turn up a worm with as close
a representation to an ACME thread as possible.

I'll use another two HD platters and see how the thread shape turns out.

Steve
Attached Thumbnails
Click for full-size image (pop_rivets1.jpg)
85.8 KB245 views
Click for full-size image (pop_rivets2.jpg)
137.4 KB381 views
Reply With Quote
  #11  
Old 03-04-2009, 08:52 PM
Karls48 (Karl)
Registered User

Karls48 is offline
 
Join Date: Jan 2006
Location: Sydney
Posts: 753
Hi Steve. Great effort. I have 7X12 lathe and mini mill machine. I have tried to hobble worms and it kind of worked. I also have small rotary table so I can do reasonably accurate indexing cuts. But my biggest problem is how to calculate blanks diameter. All what I can find on Web requires mechanical engineering degree to understand. Unfortunately Iím electronics guy and last time I worked on real lathe was about 45 years ago. So how do you calculate diameter? Iím casting my blanks from scrap aluminium.
Reply With Quote
  #12  
Old 03-04-2009, 09:02 PM
Karls48 (Karl)
Registered User

Karls48 is offline
 
Join Date: Jan 2006
Location: Sydney
Posts: 753
Steve,I think that hard disk platers are too soft to be good for telescope drive. Easy to work with yes, but I think it will wear out too quick.
Reply With Quote
  #13  
Old 03-04-2009, 09:54 PM
hikerbob's Avatar
hikerbob (Bob)
Registered User

hikerbob is offline
 
Join Date: Jan 2008
Location: Redlands, Australia
Posts: 252
For brissy based readers (and maybe elsewhere) Action Aluminium sell offcuts of aluminium and will cut to size but I don't know if there are limits on that. They often seem to have thin offcuts from quite large round stock.

I'm planning to try some hobbing so last time I was over there I picked up a number of offcuts. One is 255mm diameter and about 15 mm thick. Hardness may be a factor, I'm thinking of practicing hobbing on aluminium and once I've got the hang of it either trying to buy an offcut of brass/bronze of casting a blank. I've got a small stockpile of old brass fittings around that I've accumulated over some years. My casting skills are still at rank beginner level but that's something practice can fix.

I'm considering using the thread out of an old car jack for the worm but have not done anything to test how regular the thread pitch is. I've also looked into ACME rod, by local bolt place can get in 3/4" in 12 foot lengths for about $140 (I think, not quite what I need). It's supposedly very high quality.

I'm still thinking about what to do to hold a blank on for hobbing. I've got a 6" rotary table and I suspect that it might be a good base for the process in free running mode. Any thoughts from more experienced machinists?

Bob
Reply With Quote
  #14  
Old 04-04-2009, 08:11 AM
kinetic's Avatar
kinetic (Steve)
ATMer and Saganist

kinetic is offline
 
Join Date: Jun 2008
Location: Adelaide S.A.
Posts: 2,218
Quote:
Originally Posted by Karls48 View Post
Steve,I think that hard disk platers are too soft to be good for telescope drive. Easy to work with yes, but I think it will wear out too quick.
This one is purely just a test Karl.
I don't intend to ever use it. The alu is quite soft, you're right, but
it could always be anodised?

Quote:
Originally Posted by Karls48 View Post
But my biggest problem is how to calculate blanks diameter.
So how do you calculate diameter? Iím casting my blanks from scrap aluminium.
All guesswork this end Karl!
Some machinists try to aim for a 359 or 360 tooth worm.
I think that's because the motor at the start of the reduction does
1rpm or something....I vaguely recall that was why.
With microstepped steppers (my Bartel drive) this isn't an issue.
I only need a total reduction somewhere in the order of 1:10,000
for the stepper to 'sing' at a nice high pitched , smooth rate.

This test worm ended up having 230 teeth.
The diameter was purely a guess.
Some articles have the same problem calculating the theoretical
starting diameter. As I'm from the same type of background as you
it's all a bit over my head too!
I think indexing on a rotary table is a way better start though.
There is no way to know how accurate my cutter drove it's way
around the circumference to start with.
Some teeth might be stretched and some might be squashed etc.
Testing will show how accurate it ends up.

Steve
Reply With Quote
  #15  
Old 04-04-2009, 07:49 PM
kinetic's Avatar
kinetic (Steve)
ATMer and Saganist

kinetic is offline
 
Join Date: Jun 2008
Location: Adelaide S.A.
Posts: 2,218
Part 2: Making a worm

Today I attempted to machine a nice concentric M12 worm.
I first turned a nice std M12 thread and turned a clear
shank both ends. I decided to make it standard rather than
ACME to make it match the gear. ACME can come later.
Bored it out to fit snugly on this little worm assembly that
had some purpose built thrust/end play bearings.
You can see the original worm sitting beside it.

The idea here was to set it all up on my shed bench and
run it for an hour or so and feel the play/slop/eccentricity
in the whole setup.

I screwed the spindle and HD platter gear down to the bench first
then meshed in the worm until it was on centreline and snug.
I drove it around a bit first using a battery drill to check
concentricity of everything and see if it hadn't bound up
anywhere.

Next I bedded in the M12 worm with the worm wheel
using a 24AC motor/reduction gearbox and small chain and sprocket.
Some cutting compound every now and then was added.

The gear was SOOOO smooth and had absolutely nil slop.
After driving itself around in both directions for about an hour there
was no slack or backlash present.
With some light grease on it, it was very quiet too!

Steve
Attached Thumbnails
Click for full-size image (M12_worm.jpg)
160.2 KB445 views
Click for full-size image (M12_worm2.jpg)
105.8 KB515 views
Click for full-size image (bedding_in.jpg)
143.7 KB763 views
Click for full-size image (24vdc_gearbox.jpg)
68.6 KB431 views
Reply With Quote
  #16  
Old 04-04-2009, 08:25 PM
hikerbob's Avatar
hikerbob (Bob)
Registered User

hikerbob is offline
 
Join Date: Jan 2008
Location: Redlands, Australia
Posts: 252
Steve beautiful little setup. That looks a lot more controlled than trying to do it all on the lathe. Food for thought in that.

Bob
Reply With Quote
  #17  
Old 08-04-2009, 03:58 PM
DaveGee's Avatar
DaveGee (Dave Gault)
Occultation Observer

DaveGee is offline
 
Join Date: Aug 2006
Location: Blue Mountains, Australia
Posts: 216
Yep, making worm wheels is fun!

I turn a groove around the circumference of the wheel with a tool radiused to the root diameter of the screw. Here is a link to some photos...
http://users.tpg.com.au/users/daveg/drives.html

This allows for deeper meshing of the worm/wormwheel. it also allows a straight cut cutter to keep meshed during the cutting.

The first set of drives I made used stainless worms and aluminium Wwheels. I had the Wwheels hard-anodised by Hawker deHaviland at Bankstown. After 3 years of use there was no sign of wear. although the mount wasn't goto.

The second set were made from stainless worms again and naval brass (almost as hard as steel!) Wwheels they are still in use today. It's goto.

The latest set are using naval brass worms and Delrin Wwheels. Here is a video of them being cut...
http://www.youtube.com/watch?v=mWVqs...eature=channel
I've yet to finish the mount. Maybe this Easter!

I hope this helps Steve etal
Reply With Quote
  #18  
Old 08-04-2009, 05:21 PM
kinetic's Avatar
kinetic (Steve)
ATMer and Saganist

kinetic is offline
 
Join Date: Jun 2008
Location: Adelaide S.A.
Posts: 2,218
Hi Dave,

what a gem of a post.....thanks for sharing!
I will have to read and digest your great site and
images.
Thanks!

I was wondering about a good clutch method for both the
RA and DEC...
And keyways too! You machined the keyways as well!?

Awesome.

Steve
Reply With Quote
  #19  
Old 08-04-2009, 09:39 PM
DaveGee's Avatar
DaveGee (Dave Gault)
Occultation Observer

DaveGee is offline
 
Join Date: Aug 2006
Location: Blue Mountains, Australia
Posts: 216
Quote:
Originally Posted by kinetic View Post
I was wondering about a good clutch method for both the RA and DEC...
And keyways too! You machined the keyways as well!?
Hi Steve,

Yes a clutch is a must to prevent an accident stripping a tooth. Not a good look!

For the hard anodised set I had the wormwheel sandwiched between two clutch faces and thin cork placemat material as the friction material. A screwed collar applied the pressure. That set of drives now lives in the Netherlands.

The current setup has a friction roller as the final drive. see picture below. it will slip if some drongo (usually me) knocks the scope.

Yes, I broached the keyways on my lathe.
Attached Thumbnails
Click for full-size image (t190to1worms.jpg)
41.0 KB497 views
Click for full-size image (tGEMwithGoToSteppersA.jpg)
49.0 KB708 views
Reply With Quote
  #20  
Old 08-04-2009, 10:29 PM
MrB's Avatar
MrB (Simon)
Old Man Yells at Cloud

MrB is offline
 
Join Date: Jan 2007
Location: Rockingham WA
Posts: 3,408
Dave, that mount is inspiring!!
I have to get back out into the workshop one of these days.
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 04:40 AM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
OzScopes Authorised Dealer
Advertisement
Tasco Australia
Advertisement
Meade Instruments
Advertisement
FLI Cameras and Imaging Accessories
Advertisement
Atik 16200
Advertisement
Lunatico Astronomical
Advertisement
SkyWatcher Star Discovery
Advertisement
Astronomy and Electronics Centre
Advertisement