A very large Goto GEM built with DIY techniques.

[MarcusH]

5 hours ago, Rusted said:

In fact the worms are clearly of brass and the bearings are plain ball journal bearings.
With no obvious adjustment for end float and held in place by a thin film of shellac! [sic]
The solitary grub screw locked the races solid in the flimsy off-cut of aluminum channel profile, if over-tightened in the slightest.

It sounds like much more has been given up than just an elderly machinist….

[Rusted]

The worm bearings were mentioned earlier:

S6000Z stainless steel, deep groove, twin metal shields. 10x 26 x 8mm. 

No manufacturer's name visible.

Deep groove bearings provide some resistance to side loads but they are hardly taper roller.

 

EDIT: After endless searching it seems that 10mm bore taper roller bearings are as rare as... well, rare things.
They exist, but nobody [so far] lists then in the UK. 15mm is the minimum standard bore.
With the few, small bore exceptions being on Chinese manufacturer's websites.
The logical thing to do, if needs must, is to add thrust ball bearings to the existing bearings.

Edited February 7, 2020 by Rusted

[MarcusH]

Could an angular contact bearing be an option ? Much easier to find with a 10 mm bore, if that's the only critical measurement.

Can take both axial and radial load, but don't know if it's much inferior to tapered roller bearings. I think I've read somewhere that the tapered rollers were preferred in slow rotating assemblies whereas the angular contact was preferred in fast rotating ones (think it was on some CNC-forum....)

[Michele Scotti]

3 hours ago, MarcusH said:

Could an angular contact bearing be an option ? Much easier to find with a 10 mm bore, if that's the only critical measurement.

Can take both axial and radial load, but don't know if it's much inferior to tapered roller bearings. I think I've read somewhere that the tapered rollers were preferred in slow rotating assemblies whereas the angular contact was preferred in fast rotating ones (think it was on some CNC-forum....)

https://simplybearings.co.uk/shop/p88001/Budget-7000-Single-Row-Angular-Contact-Open-Ball-Bearing-10x26x8mm/product_info.html?backstep=1

2tons of axial load - i guess its enough

[Rusted]

Thank you both. The bearings in the link would be a straight swap for the present ones by size.
Though I'd need to check for the availability of seals and all stainless steel construction.
The last thing I want is rusty bearings from exposure to damp in a constantly cycling temperature regime.
Everything suffers from heavy condensation out in my well ventilated observatory.

First I ought to check the forces required at the rims of the wormwheels to get a better idea of the tangential loads on the worm bearings.
The AWR micro-step drives provide an acceleration and deceleration ramp with each movement.
It is certainly very smooth. With zero visible backlash to slews despite the long and heavy refractors. The 180/12 is over 7' long.

Meanwhile, the worm's bearing support structure needs serious reinforcement to gain any advantage from better bearings.
A hinged and sprung, perpendicular, motor/worm mounting arrangement is under serious consideration to minimize [known] concentricity issues.

If the mounting was ever truly finished it wouldn't be so much fun any more. I'd need to start building something else.
If it was a commercial product I'd probably be moaning about its minor foibles.  But then, it doesn't owe me a near fortune for its load capacity.
And, it's only a Mk1. Built from the cheapest [mostly scrap] materials to hand at the time. 

[MarcusH]

1 hour ago, Rusted said:

Though I'd need to check for the availability of seals and all stainless steel construction.

Should be available, but a quick Googling showed that for some reason the shielded bearings in the 10 mm bore range are usually double direction (e.g. double row). So you might have to accommodate for a slightly wider and larger bearing. But if you are going to make significant changes to the worm bearing holders anyway this might not be an issue for you.

1 hour ago, Rusted said:

If the mounting was ever truly finished it wouldn't be so much fun any more. I'd need to start building something else.

Aaah, a true kindred spirit. 😄

A work of art is never truly finished now, is it... 😉

[Rusted]

3 hours ago, MarcusH said:

Should be available, but a quick Googling showed that for some reason the shielded bearings in the 10 mm bore range are usually double direction (e.g. double row). So you might have to accommodate for a slightly wider and larger bearing. But if you are going to make significant changes to the worm bearing holders anyway this might not be an issue for you.

Aaah, a true kindred spirit. 😄

A work of art is never truly finished now, is it... 😉

Appreciation for "Work of art" depends on whether that was irony or sarcasm.

[MarcusH]

21 hours ago, Rusted said:

Appreciation for "Work of art" depends on whether that was irony or sarcasm.

Neither.

The reluctance to call a project finished, even though said project already had met it's design goals or even surpassed them, is something that I can easily relate to.  😄

[windjammer]

Re: >>The last thing I want is rusty bearings from exposure to damp in a constantly cycling temperature regime.

try screwing power resistors (like these) to your metal work:

https://www.rapidonline.com/aluminium-clad

You could power them with a dew heater.  A small amount of heat would keep the damp off.  I use four 6.8 ohm 50W resistors in parallel at 12V on heatsinks with a small fan to blow warm air under the telescope covers.  It works a treat.  I have mounted a greenhouse remote RH and temp sensor (ThermoPro TP60s Wireless Thermometer):

https://www.amazon.co.uk/ThermoPro-Thermo-hygrometer-Thermometer-Hygrometer-Temperature/dp/B07G2YM494?ref_=s9_apbd_otopr_hd_bw_b5zxl5j&pf_rd_r=M1YG8Y42H5YYTJ6DWBY5&pf_rd_p=dd27ab4b-5b90-57e6-98da-609b80414dcd&pf_rd_s=merchandised-search-10&pf_rd_t=BROWSE&pf_rd_i=5496263031

on the scope.  Whenever the RH goes above 70% I turn the heater power on.

On worm gears - have you thought of turning one on your lathe ?  I understand (don't have lathe!) hobbing a wormwheel is quite hard, turning a worm screw is a lot easier of you have a screw cutting lathe.  You could choose your material of choice, brass or stainless, with shafts as long as you like.

I recommend this magazine, Model Engineers Workshop and forum if you need info - they are always running articles on gear hobbing etc which will be in the digital archive if you subscribe.

https://www.model-engineer.co.uk/editorial/page.asp?p=250

Yes, projects are never finished !  But how can you avoid long periods of downtime while you make and install new stuff ?  So you need TWO versions - one up and running so you can use it, and a second in development to swap over when commissioned, then start all over again!

Simon

 

 

 

 

[Rusted]

Episode 14: Worm drives:

I am strictly a "make it up as you go along" designer these days.
My earliest trials with the worm drive layouts was using wood blocks for support.
Wood is easy to work with and avoids wasting irreplaceable aluminium scrap.
What turns up at a scrap yard is fleetingly available and once gone is well and truly gone.
If you find something useful you buy it at the price they ask.

The RA wormwheel sat at the top of the Polar Axis shaft for a year or more before I finally moved it down to the bottom.
To do this, I had to make room on the south side of the huge timber and plywood pier.
That meant shifting the whole [massive] mounting southwards on the pier.

Every change I made required the hefty, builder's, folding stepladders be carried up and erected on the observatory floor.
Once bridged above the mounting, into a stable A-frame and lashed firmly together at the top I could use my chain hoist.
This is not without its dangers, despite the stabilizer bars resting on the floor.  So I always guyed the stepladders to the dome perimeter for safety.
I have lifted the massive mounting up through the stairwell to the observatory several times now.
For the lift off the ground floor I use a massive, square steel tube bridged across the base ring and pier.
This is an early photo of the lift while the observatory was still under construction.

Scrap metal box section turned up just when I needed it to house the stepper motors and support the worm bearings.
It was spot on for size but needed a lot of cosmetic cleaning up.
The clearances for the motor were minimal which required CSK fixing screws for the worm bearings.
I had to invent a way to countersink the the holes on the inside of the boxes!

Assorted earlier images:

[Rusted]

Thank you Simon. I am well aware of wormwheel and worm and wheel gashing, hobbing and machining.
I have even made a few smaller wormwheels an worms myself over the decades. I have a 9" Ø S&B toolroom lathe.

I once made a huge dividing wheel in polycarbonate and a fly-cutting shaft with belt drive, high speed motor.
All in anticipation of doing lots of gear cutting.
Then I became sidetracked on other projects and trusted Beacon Hill to supply the goods.

Heating several hundreds pounds of aluminium, iron and stainless steel in a mounting is asking for the impossible in a deliberately draughty dome.
Everything is literally dripping with dew in the mornings and often for most of the day.

I could indeed cover the mounting with an insulating cover overnight but that can't possibly extend to the telescopes as well.
Though I have recently purchased some heating straps and controller. And, a secondhand hair drier from a charity shop.
A panic reaction to hours of precious solar imaging time being lost to objective dew after recent cold nights.
Even then it takes ages to rid the large lenses of dew.  I am very wary of leaving anything running overnight due to the risk of fire.

[windjammer]

Thats a very nice lathe! - actually I never bought one (though came close many times, and regretted many times) because I didn't want to be sidetracked from doing astronomy !  Perhaps you should go into business supplying us all with pukka worm gear sets - seriously, how much ?!

Agree on the fire risk overnight, not so much at the scope, more inside the house where the power units are.  The resistor idea is a very low intensity method  to provide local heat to the bearings.  I read about them used to heat lathe beds in damp workshops to keep the rust off - so it could work for you.  A cheap idea to try anyway.

Dew heaters (mine are homemade, 330ohm 1/4W resistors in parallel wrapped in duct tape ) work well but you are right, they do take an age to clear once the dew has condensed, but very good at stopping it in the first place.  So I tend to leave them on all day when I am present and the RH is high.

Condensation is a big problem I worry about a lot - I have previously found the dew to be quite damaging to the lens coatings.

Simon

 

 

 

 

 

[Rusted]

Thanks Simon. I wasn't aware that lens coatings were vulnerable to dew damage.

I made a half-hearted job of wrapping the objective ends of the OTAs in camping mattress foam.
Not ideal as the closed cell stuff is too stiff to go over objective cells.
Something softer would help the wrapping process.

Online research suggests that insulating the dome roof would help to reduce radiation to the night sky.
I'm using agricultural fleece as a stand-off lining but it's probably more cosmetic than thermal.

Lining the dome with foam is doable but the plywood seams leak in places.
I am torn between re-covering the dome in full gores of heavy PVC tarpaulin.
Or re-cladding in full length aluminium gores.  Access is poor because of the working height.

Making large wormwheels demands expensive materials. Thick rounds of free machining aluminium.
One mistake, and you have to use the blank for a smaller wheel. You can't arbitrarily change the tooth count.
The teeth ought to be gashed at the correct angle before hobbing to ensure the correct tooth count.
Which means some way of dividing the blank while the gashing is going on.
Either using a fly cutter or a slitting saw of the correct diameter and angle to match the worm thread.
The wheel needs to be mounted on a hub which is never removed from its concentric [between centers] holder until completed.

Ideally the worm should be a common [if rather large] thread for which [machine] threading preferably spiral taps are available and affordable.
A purpose built, wheel cutting "engine" or specialist jig is probably necessary because of the poor access to the wheel circumference on most lathes.
The length of the tangent must allow for the hobbing tap to be supported rigidly from both ends. 

The investment in amateur-built equipment could still be quite considerable at start-up and probably slow to pay for itself.
At a professional tooling and equipment level it would be a foolish investment due to lack of demand and the high price putting off customers!
I vaguely remember Peter Drew mentioning that he has some experience in wormwheel manufacture.

 

[Peter Drew]

@Rusted.  Yes, I must have made hundreds of worm and wheel gear sets during my career.  They were all hobbed on a 4.5" lathe.  I always took great care to make sure gear blanks were turned with all details perfectly concentric, not difficult just needs care.  How big players such as Meade can often get this wrong escapes me!  My worms were turned from solid in one operation which ensured complete concentricity of bearing journals, drive shafts and worm teeth.  I still wince when I see composite designs.

Getting the correct number of teeth was a mixture of always using the same tap, same blank diameter, a few tricks of the trade plus a bit of luck.  On expensive large diameters it was prudent to pre gash the blank with a dividing head to guarantee the correct number of teeth.

I've no idea how good my gear sets were but I never had any complaints, users got good imaging results and some owners brought their mounts back for minimal adjustment after "donkey's years" of reliable use.   🙂

[Rusted]

Thanks Peter and welcome back!

I'd love to see a photo of your wheel hobbing arrangements if such exists.

There are a great many potential questions:

But: Did you use straight fluted taps or spiral? Threaded "machine" taps or the "manual" type?

Unfortunately I never found a dividing head with my name on it. Far too expensive back then!
So I used large, home made, dividing disks on the outer end of the lathe mandrel for [gear] wheel cutting.

[Peter Drew]

I always used straight fluted machine taps.  The tool post was removed and replaced by a matching item that had a vertical spigot.  The gear blanks were then fitted to the spigot with spacers to get the correct cutting height.  A thin PTFE washer above and below the blank allowed it to be held firmly against tipping but free enough to be advanced by the tap.  A very light first cut was used to confirm that the teeth were not going to chew themselves up after turn one.

No pictures I'm afraid, images of me doing anything useful are rare.       😀   

[MarkAR]

An excellent thread, I do a little hobby engineering occasionally and can really admire the work and thought that goes into a project like this.

 

[Rusted]

1 hour ago, MarkAR said:

An excellent thread, I do a little hobby engineering occasionally and can really admire the work and thought that goes into a project like this.

Thanks. 

The plates were mostly cut out with a powered jigsaw, lubricated with lamp oil [UK: paraffin] and then filed straight by hand.
Only later did I obtain a sliding, 12" DeW, miter saw which could manage aluminium with far greater speed and accuracy.
Cut-offs up to 6" [or so] were relatively painless, even in 20mm thick alu. But I never had the nerve to saw the long sides on my table saw.
Besides, DeWalt specifically warns against such use. The problem with sawing aluminium with a circular saw is the swarf.
It flies everywhere and can't be easily collected. Non-magnetic and a shop vac with "turbo" sawdust separation might be damaged.
So I had to use a dustpan and brush over several months until it was mostly all gone.

[Michele Scotti]

20 hours ago, Peter Drew said:

@Rusted.  Yes, I must have made hundreds of worm and wheel gear sets during my career.  They were all hobbed on a 4.5" lathe.  I always took great care to make sure gear blanks were turned with all details perfectly concentric, not difficult just needs care.  How big players such as Meade can often get this wrong escapes me!  My worms were turned from solid in one operation which ensured complete concentricity of bearing journals, drive shafts and worm teeth.  I still wince when I see composite designs.

Getting the correct number of teeth was a mixture of always using the same tap, same blank diameter, a few tricks of the trade plus a bit of luck.  On expensive large diameters it was prudent to pre gash the blank with a dividing head to guarantee the correct number of teeth.

I've no idea how good my gear sets were but I never had any complaints, users got good imaging results and some owners brought their mounts back for minimal adjustment after "donkey's years" of reliable use.   🙂

Next time make sure you shot a video - worm gear hobbing is fascianting.

Would you reckon you had a similar set-up to any of these? 

https://www.youtube.com/watch?v=19jKlq8Ofd4&list=PLJtiP4DlrmJROzFYnk0rYQ9fnJPO9KyCn&index=16&t=0s

https://www.youtube.com/watch?v=eKJe1JnvRc0

Hope not to detour the thread here....

 

[MarkAR]

2 hours ago, Rusted said:

Thanks. 

The plates were mostly cut out with a powered jigsaw, lubricated with lamp oil [UK: paraffin] and then filed straight by hand.
Only later did I obtain a sliding, 12" DeW, miter saw which could manage aluminium with far greater speed and accuracy.
Cut-offs up to 6" [or so] were relatively painless, even in 20mm thick alu. But I never had the nerve to saw the long sides on my table saw.
Besides, DeWalt specifically warns against such use. The problem with sawing aluminium with a circular saw is the swarf.
It flies everywhere and can't be easily collected. Non-magnetic and a shop vac with "turbo" sawdust separation might be damaged.
So I had to use a dustpan and brush over several months until it was mostly all gone.

Dedicated metal working chopsaws run at a lot slower rpm than woodworking ones, swarf will still get everywhere but not so extreme. A powerful vacuum cleaner definitely saves time, I use a 1400w one that really sucks 🙂

Regarding the poorly machined worm wheel, I hope you kicked up a big enough stink to at least get a partial refund and also well done for making it right. 

[Rusted]

Thank you both for your posts.

Michele,

The milling machine example of hobbing shows a typical problem of clearance along the tangent of the wormwheel.
The chuck is almost rubbing on the largest wormwheel.  Some sort of rigid and accurate extension is necessary.
It must help to clear the chuck as the wormwheel diameters grow larger while still providing the drive to the tap.
If a lathe is being used then the tailstock is usually much narrower and an MT center can provide extra clearance.
A typical 3-jaw chuck would be even larger and force the need for an even longer extension.

Markus,

I couldn't afford a metal cut-off saw as well as the mitre saw. So I just set the speed to the lowest available.

I didn't want a refund or a discount. I just wanted a much better wormwheel.  It didn't happen.
So I no longer protect the vendor from the hideous truth about his trading standards.
I'm too far away to have started a "Sale of Goods" case against him.

 

[Michele Scotti]

On 10/02/2020 at 16:28, Rusted said:

Thank you both for your posts.

Michele,

The milling machine example of hobbing shows a typical problem of clearance along the tangent of the wormwheel.
The chuck is almost rubbing on the largest wormwheel.  Some sort of rigid and accurate extension is necessary.
It must help to clear the chuck as the wormwheel diameters grow larger while still providing the drive to the tap.
If a lathe is being used then the tailstock is usually much narrower and an MT center can provide extra clearance.
A typical 3-jaw chuck would be even larger and force the need for an even longer extension.

I always fancied giving it a go - I might actually need to do that for the derotator wheel.

Ideally I think that lathe+tailstock -as you confirm- it's the must sturting point. I reckon that using the gear already on its on bearing i.e. in its final confirguration, will improve the result in terms of run-out. On top of that, running on the same set-up I'd do the lapp[ing after swapping the tap with the threaded shaft 

[Rusted]

In theory you can invert the hobbed wheel still on its accurate spigot.
To run the tap against it again.
Which should smooth out the sharp edges.
There are always burrs when you cut metal from one direction.
So reversal of the cutting action should smooth things nicely.
It should also halve any errors in the tooth form.