Michele Scotti

Preparation work for grinding the Altitude rails and achieving correct precision. 
Stacking up the 2 pacmen and fastening them together. 

I'm using two steel beams to overhang the pacmen so that the swiveling 'apparatus' can go around the circumference with ease. 

  

To start with the run-out (departure from ideal circumference) is checked using a digital gauge that can sense microns. 

The dial starts from 0.000 at one end of the pacmen and climbs up to almost 2mm. However that seems to be attributed to the hinge point being off-center by interpreting the plotted measures. Ideal situation is to have only a few tenths of mm of stock removal to grind off.

Next up: drilling a slightly bigger hole for the swivel center to allow recentering and setting up the grinding apparatus.

I'm wondering how many things can go wrong...

Very useful vids - I see 3 things:

1) the wormgear sliding into the bearings - is that what is fixed by the missing washer?

2) the hosing movement - you describe it as flexure but it's not clear from the vid. Another video from the side would clear that. Again I'm a bit skeptical that a 5mm square tube will move much. It's maybe more likely that it's sliding/tilting due to the screw that allows the springed housing?

3) the big plate the housing is mounted on is seen deflecting  small movenet but it's there. It'd be surprise that such plate is flexing - if it's the RA shaft housing to the pier then you might have another problem to check. That should be rock solid imho as it's the fundation of the mount 

18 hours ago, Rusted said:

Without any end loading on the flange bearings I was able to flex the PA by a similar amount.

Can you explain what do you mean with that?

18 hours ago, Rusted said:

The most obvious problem is that the RA worm housing flexing when I manually rock the telescopes in RA.
I used 70mm square tubing with a 5mm wall thickness [from memory.]
I was able to see the flexure without needing any aids.

Can you see where the flexure is stemming from? A close-up video would be a great aid although by looking at your set-up it could be the fixings clearance for adjusting the wormgear housing. I wouldn't rule out the bearings though

The big plate looks like 10mm, the square tubing 5mm, the worm housing is possibly 8mm thick. That is enough material to make a robust system. Should be somewhere else 

10 hours ago, Rusted said:

It would be tempting to release the clutches [or even the worms] and rotate just the wormwheels for measurement.
Though I don't think this would be a realistic [real world] test for eccentricity.

I agree although the clutch plays a role - I suggest to detach the wormgear assy and attach a camera and mark the wheel. You can then focus on freed axis and make it safely rotate around the pier - then you can take notes of the measurements conviniently fom the video.

Something like this set-up: https://youtu.be/OGJyjb-h-G8

Btw I recall you started a topic to explain the build - can you attach the link pls?

On 24/11/2020 at 18:49, Rusted said:

Thank you both for your input.

I have yet to test the eccentricity of my wormwheels with the telescopes unmounted.

It now occurs to me that I am also at the mercy of the accuracy of the [inexpensive] flange bearings which support the 50mm axes.
I really ought to check the shafts in rotation, at the wormwheel position, with a dial gauge before assuming the wormwheels, alone, are at fault.
Again, this requires removal of the telescopes to allow free rotation.
I could load some weights onto the top of the PA shaft to simulate the telescopes without their considerable length causing issues.

Assuming the teeth appear uniform I think I should be able to ignore variations in tooth depth.
Would anyone like to suggest an acceptable degree of eccentricity at the wormwheel rims?

Thanks again.

Brief and rough trigo. --> arctg(um/R)= ca. 1.5arcsec i.e. for every micron (um) of 'detachment' of the worm from the wheel's teeth means 1.5arcsec of backlash and that is the same you'd see at the telescope. 

A more 0.1mm  of 'sloppiness' would give you back 2.5arcmin.

Quick and tad cryptic - let me know if it makes sense though.

Wrt to RA flange bearing I would not be concerned - gravity loads them always in the same position and take away any bearing clearance (but not roundness isssue although not likely to be an issue in this application.

Can you test with the scope on? What's the reason to remove it?

On 22/11/2020 at 18:23, Rusted said:

Now you mention it, I think I remember there being no taper roller bearings in the worm shaft diameter. 12mm from memory.
Larger bearings could easily be accommodated in more substantial worm housings. Or the worm shafts sleeved to a larger size.

The problem then is housing clearance from the wormwheel rims. It would require much wider spacing of the bearings.
Cheeks containing larger thrust bearings would require worm shaft adapters to be turned up.  The supplied worm shafts are very mean in length.
Barely accepting a timing pulley hub. Which, again, required two more radial holes be drilled and tapped for extra holding screws. One grub screw was not enough for security. 

I think your system has high potential - baby steps: if your shaft is 12mm there could be fairly inexpensive angualar bearings (not as good as taper ones but for these loads they would do).  Youd need to pre-load them but it's easily done with your current set-up - if interested I can sketch-up something for you. 

Don't use adapters - you'll add errors. Here you need to remove them oen by one.

https://www.ebay.co.uk/itm/7001-Single-Row-Angular-Contact-Open-Ball-Bearing-12x28x8mm/143541090346?hash=item216bb7382a:g:4NMAAOSwuvpcR2-C

On 16/08/2020 at 06:00, Rusted said:

Are you aware that woodcutting circular saws, with carbide tipped teeth, can cut through aluminium like butter?
The bench type machines and table mitre saws can save hours of tedious work reducing large aluminium sections.
Downside it it makes a hell of a mess in the workshop! Swarf collection should be attended to.
It also needs very firm work-holding to avoid accidents and potentially serious injury. Plywood jigs?
Use odour free, lamp oil to reduce saw tooth clogging. It really helps.

Hi Rusted, I previously said I was giving it a go. Game changer! 

Time went down from 20min to 5min - without the piece to get hot thus the benefit of spraying water to cool it down...and producing a nasty swarf. 

I took the liberty to sketch up how I would do approach this - courtesy my daughters' crayons.

Taking the brave pill I'd split the wheel in an inner flange and a crown - the latter being adjustable. The 2 are connected by a plate (I envisage a 5mm steel plate which needs to be machined at least on one face).

Wrt the investigation with camera check out this post: https://stargazerslounge.com/topic/340154-800mm-telescope-project/?do=findComment&comment=3787165

On a video the resolution could be lower but if you use a mild tele and a camera I'm sure you can crack the 0.01mm. Slow-mo won't help, it reduces resolution and I assume you don;t have high dynamic phenomena giong on.. Here yo'd be more looking for semi-static deflections.

Abou the gauge: if the grooves on the wheel are consistent i.e. same height you might reasonably suppose that the wheel grooves and outer diameter are fairly concentric. Long story to say that you can place the guage more conviniently on the rim.

Lastly on the worm gear bearings - I try to briefly look up some taper bearing but I couldn't find anything with an ID smaller than 15mm. What's the shaft diam?

1 hour ago, Peter Drew said:

There was no spring loaded bracket involved as I took the trouble to ensure that the o/d and bore of the gear blanks were concentric and an accurate size, no real problem if these dimensions were machined in one operation.  There was a back stop plate with adjustment screws to prevent the worm bracket from backing off under load.  Never had any complaints or returns!     🙂

Out of curiosity: straight or helical tap?

Forgive me if I got it wrong by missing some info already stated in this thread.

You have a run-out issue – can you measure that? From an engineering standpoint it would be crucial to quantify the gap to your ideal situation and see what improvement(s) are closing it up.

A 0.01mm dial gauge is pretty cheap nowadays. You could place in the not-hobbed rim or fabricate a probe tip that gets into the grooves (going to take longer to measure but more accurate). Do you have nay mean to fine tune the cantering of the worm wheel? It’s one of the most usual issue and I guess it might have been exacerbated by the50/60mm bore issue?

Also, I’d suggest to put your phone on a stand pointing at the worm gear and take a video while rock the RA back and forth. If you watch the video and see any movement – then you might better guess where the problem(s) are coming from and address that. With mine I can sense less that 0.01mm movements and have spotted stuff I couldn't see by naked eye.

I always thought that springed systems are inherently a ‘compromise’ and I suppose that beyond a level of error they can’t cope with that anymore.

It would be interesting to know from Peter whether the system he was producing were spring mounted or not.

Your worm gear carrier doens't look that flimsy IMHO. More concerend about not tapered bearing which wouldn't be a big installation issue if you want to go for that.

Is that the only botched-up tooth of the wheel?

On 18/11/2020 at 17:07, Rusted said:

Thanks for the continuing feedback.

I've been in the workshop adding angle profile, mutual reinforcement to the RA motor and worm housing.

Still experimental but I'm making progress: Even added a crude, push-off screw adjuster.

I'll be back.

 

 

Hi Rusted, it may seems slightly off topic but is the wormgear supported by ball bearings or conical/taperes ones?

Brief update: had a chat with the sales manager at Renishaw - he confirmed the feasibility to use their magnetic encoding system - 0.24um resolution.

Pro: it should be cheaper than the optical one 

Cons: magnetic has some hysteresis when changing direction. The amount is less than 2um which is less than an arcsec in the 800mm project set-up. On top of that - possibly to simply convince myself!- i'd say that hysteresis would affect mainly the pointing accuracy rather than tracking provided an exposure that doesn’t change the direction of the axis.

5 hours ago, Jonk said:

According to this setup, you would need about 20m of PLA, and the print would take <8 hours using a 0.8mm nozzle, no infill (it doesn't need it) and print speed of 45mm/s.

The resolution won't be fantasitc, as it's a large nozzle. It also won't be smooth - if you want smooth, the best thing to do is print ABS, then chemical polish the surfce by placing in a sealed container with acetone vapour.

The main thing to take into consideration is shrinkage as it cools.

This may or may not be an issue, so you can either design it in when modelling (adding 0.2-0.3mm for example) if you know the outcome will be correct, or you can account for this in the slicing software (Cura in this case).

Another point is holes - if you know where these holes need to be, draw them in and if you need to reem or drill out when printed to suit, then this will be a lot easier than drilling new holes.

Drilling PLA is like any other plastic, it's very easy to get it wrong and melt rather than drill the material.

The best way to achieve this part in my opinion is turn down some ABS pipe. A good machinist will have no problem with it.

Jonk thanks a lot for looking into that - I understand that it's pretty complicated and time consuming given the volume.

My concern about turning stems from the fact that when I had the Aluminium parts turned in a CNC shop they initially rejected the job due to a 3mm wall thickness - similar OD, similar lenght. TBH I need to have the Alu parts pretty accuracy - say +/-0.01, nowhere near this plastic bit. I might just giving it a try as I have the plastic pipe around.

My second choice would be making it out of carbon fiber. I used this method to make the central hub of the spider. 

2 hours ago, johninderby said:

You can get 120mm plastic air duct pipe.

https://www.amazon.co.uk/Manrose-Round-Pipe-White-120mm/dp/B00NJT40NW

 I bought a 6mm wall thickness pipe some time ago but when planning the process on the lathe it just looks like a nightmare. You'd need to make flanges to support the turning of the OD alone and then you are left with a  big ovehang on a thin  wall to turn it down to 2.5mm.  Not impossible but slim chances to do it rigth tbh.

The item is suppose to be sort of sleeve bearing for one hundred 3mm ceramic ball bearings I already got.

4 hours ago, Jonk said:

Try PLA, it’ll be strong and drillable.

I’ll check how much material is needed tomorrow at work, and how long the print will take as I may be able to do it for you.

That would be just super! Thanks a lot  even just for checking.

In case let me know - I need to fine tune the dimensions (by tenths of mm) of the item and generae a new model

I've checked online in a few website, thanks to the STL from Jonk. The cheapest quote is £140 - way more than I expected - I guess it goes with the overall volume....

I guess I need to make friends with somebody tha owns a printer! 😅 

Any suggestion for on the material? FDM or PA are those more basic. The only characteristic that I need it's probably that I can drill it easily, all the rest doenst count

Is height the main parameter that makes it not achievable for 'satandard' printers? Would it be easier i.e. easier to find a more common printer, to split the  item in 2 and glue it later? it would be 

Having to do smal drilling is there a recommended material?

4 hours ago, PeterW said:

If you FDM it I would be tempted to add a few longitudinal ribs inside (if the design allows), just to lower any risk of delamination from flexing. If you use a tough material and print thick layers you won’t need to bother.... I always like to make sure prints aren’t too fragile... stops end users questioning the technology.

There's virtually no room for any ribs unfortunately.

Here's the overall design if anyody is curious:

6 hours ago, Jonk said:

3dhubs was the first one I used.

If you send me the dimensions I've drawn the model and created the .stl file for you (attached).

 

Fantastic! Thnx a lot!

2 hours ago, markse68 said:

The Renishaw encoders i know rely on the grating being bonded to the moving part with epoxy- much easier to unscrew a read head and the grating would likely be ruined trying to remove it- could explain your experience.

I wouldn’t worry too much about accuracy of the different technologies as i’d imagine you’ll exceed their accuracy just with the inevitable sag you’ll get going from vertical to horizontal with your ota?

I agree about the grating - i guess they are glues inside the Alu casing -although I'd be ok to try and debonnd a scale. The only source of an optical grating in a reel is from China and it's £400 a meter....that won't do for me. I have the impression that magnetic scales are cheaper.

Sag is inevitable but I have a couple of considerations:

1) the design is somehow taking that into consideration - see post:

2) sag in a properly designed and executed mount -not saying it's my case!- is predictable and proporsional i.e. it could be mapped out

3) the sag for given an amount of tracking time seems -at first glance- manageable. IT should be less than 1arcsec each Altitude degree. Aiming at 10min integration time I'm not that worried about - we'll see iwhen in operation if all of this ijust theory....

4 hours ago, Jonk said:

Whilst perfectly feasible, can you find a piece of plastic pipe that is close and use that? If you know someone with a lathe, they can make it fit.

I have access to an Ultimaker 3 Extended and this can produce amazing parts.

I always tell people at work, if you can draw it, we can print it (within size restrictions)!

I actaully have bought such a pipe but the turning is just a nightmare - and I'd need to make flanges to support the whole thing. While pondering on all those aspects I  got my Eureka moment.... 3D printing.!

Glad to see you all  cleared my doubt about thinkness.

Are there online businesses that can print out on;y one piece? Do they need a specific CAD file (like STEP) or do I need to create the model  witha  special  software? Any idea of the cost?

Cheers 

12 hours ago, markse68 said:

In your application i’d have thought magnetic or capacitive or inductive (Posic do linear scale versions) would be better than optical purely from the point of view that it’ll be difficult to contain and shield the read head/grating from dirt and humidity which the other options would be more immune to.

Mark

From what I got so far the resolution resides mainly in the readhead  - i.e the digitalization of the sinusoidal information -. whereas the accuracy/repeatibility is in the grating of the scale -regardless if it's magnetic or optical. 

Apparently magnetic scales are """"fairly""" affordable - the optical ones (with 20/40um grating- are nowwhere to be seen...that's weird because the matching readheads are plentiful on Ebay...just like the one I bought.

1 hour ago, Synchronicity said:

It's absolutely feasible. Roundness won't be an issue either but strength might depending on what it was used for. 

What about smoothness? Finer surfaces take much longer to print.

Michael

1 hour ago, MrFreeze said:

A 2.5mm wall thickness is perfectly reasonable. I made a dew shield for my Mak127 (the one on Thingiverse - not recommended - too small and needs filing to fit) and that is 150mm OD, 175mm height with wall thickness about 2.2mm. I wouldn't want to sit on it, but it's perfectly strong enough for it's purpose - it hasn't broken yet despite it being too tight and being forced into place with quite some effort.

Your size is too large for most small resin printers like the Anycubic Photon which make much stronger parts. It would be possible (just) to print this on an Elegoo Saturn if you know anyone lucky enough to have bought one. (The last batch of 200 produced sold out in about 20 minutes!)

David

Thanks Michael and David,

 smoothness and robusnes are not important for this application - it's just a sleeve carrier for ceramic ball bearing of the focuser.

Thanks for the prompt feedback! 

Hi there - question for the much- envied 3D printer experts.

The focuser/derotator for the 800mm project needs a part -a simple cylinder - that can be made of plastic material. The rest of the focuser is made of Aluminium.

As said earlier, a cylinder with OD 120mm, height 135mm, wall thickness 2.5mm

Is it feasible or the wall is way too thin? I don;'t have very stringent need for roundness as long that it doens't look oval! 

Cheers, Michele

Just for info, I've started a thread on a CNC forum - great chaps there too- and  getting a lot of suggestions.

All in all it looks like a tape scale + encoder readheadwith a 0.1um is the way to go.

The challenge is to find the hardware in the used market to contain the expense and stay true to the spirit of the project.

Read-heads seems fairly attainable on ebay ( like less than 100$/£ for a RGH22/RGH24Series if you're lucky and patient).

If I could only find linear scales that are inexpensive then bingo! but I seem not to find any of them – the optical ones with 20/40microns gratings, at least. Still puzzled about how to salvage them from a CNC machine.

Also, what is not entirely clear is if and optical encoder is better than a magneic one. The both are available to 0.1um resolution

5 hours ago, DaveS said:

The ones on my DDM85 are Renishaw (I think) absolute encoders running at 0.02" per "tick".

That's the same as Planewave i.e. 50 ticks per arcsec anyway I just talked to the chap that developed the tracking system and he says that I should aim at 10 ticks/arcsec 

HIgot this one - according to the data sheet it's a 1um resolution with digital output. for phase A and B in quadrature. But I don't think there's a "quadrature" signal. However the SiTech II has separate inputs for A and B so I suppose it performs the quadrature?