For those who are considerig DIY dobson

[bojan]

Hi,
this solution of Az drive with friction rollers may be interesting for someone...
I designed it this way because I wanted the specific reduction ratio (it is visible on the pic that rollers provide for reduction 1:3) and some degree of "self-aligning" at the same time, to allow for larger tolerances in positioning of the rollers and axles. This solution was sort of forced upon me, because I was using various parts from waste.. and it turned out pretty well.
Why two rollers? I tried with just one (forced between the driver and driven disk by spring), but that solution turned out bad, because when the telescope was driven in one particular direction, the roller was pushed out and the friction was decreased to the point of not moving at all in that direction.
The most important thing here is that both rollers are identical, and the tension between the rollers is sufficient, which is provided with two springs (another one is below the rollers and not visible on this image).

[joe_smith]

Looks good. Do you have more pics of the mount. I love the way you guys can make bartelized type dobson mounts out of stuff you have laying around :) I would love to be able to make one

[bojan]

Here they are.. of course, everything is still being modified from time to time...:whistle:

[netwolf]

Nice work ther Bojan, did you make the gears yourself?

Regards

[bojan]

No, I was lucky to find them on my friend's machine shop waste (he likes to collect those things just like me). You never know when you may need something like that ;-)

[netwolf]

The first picture you posted kind of remind me of my childhood train set. It had similar shiny wheels. And the motion is similar to that of a train wheel on a track, very ingenious method.

[joe_smith]

Now thats a roboscope :thumbsup: very well done

[bojan]

Thank you all for your kind comments :-)...
This is another example of roller coupling that I am using on my small MTO1000 macsutov cassegrain... same principle, but very simple to implement and auto adjustable to the certain degree (so you do not have to drill those holes for axles very precisely), and better than gears, because there is no backlash and no periodic errors.
Bojan

[netwolf]

The picture is a bit dark on my laptop screen. But i think i got it. your using the drive shaft to turn to rollers that are runing in a milled track. Very nice.

I came accross this other similar idea that seems pretty cool as well using friction to drive a worm type flywheel. They used a Threaded rod to move a square pipe along the rod that turned the flywheel. Saw it posted on CN forums will dig it out. To stop the square tube turning the placed a second roller on the other side of it similar to your drive shaft sitting in the middle of two rollers.

[bojan]

Yes, and springs are holding them in place. It is similar to those "shiny wheels" from the first picture but here there is no additional reduction (apart from ratio of shaft and milled track diameters). Also, you can not break anything (like gears) because the drive will slip if forced too hard... very useful feature in dark :-)

[bojan]

Here is better picture :-)

[Lex]

thank you all
Lex

[Lex]

dont the two bearings work against each other ifthey are bothin contact with the milled part ofthe drive

[bojan]

This is how I did it.....
No slippage observed.

[bojan]

Nope... they work together. Of course, they are not supposed to touch each other. Both of them are only in contact with driving axle and the milled part.

[alistairsam]

Hi,
Had a couple of questions on your alt-az drive

are you using any rubber bushes or coating on the metal rollers to aid friction thereby reducing the need for added tension to the rollers and avoid slippage? what would happen after a period of wear on the metal rollers?
when you say bartelized dobsonian, I presume its adopting methods described by mel bartel, but just wanted to understand what it actually does. does it auto track using a secondary guide star, or is it for motorized slew on alt and azimuth axes or is it for normal tracking using fixed rate motion
if for tracking, how do you get around field rotation thats inherent with alt-az mounts?
do you use prime photography or field projection if you're using it for astrophotography, and if prime, did you have to move the mirror?
I was thinking of prime with a reflector, but by making the focusser detachable, and using an alternate support for the camera thats low profile, that may get away with the need to move the mirror forward to achieve focus.

[bojan]

There is no rubber, just metal-to-metal (brass, steel, al.. with additional pressure in a form of springs holding wheels together... and there is no slippage. Or if there is, it doesn't matter enough to be noticed as a problem).
When it wears off sufficiently (big aluminium wheel, brass and steel will be OK), I will be long gone from this world :-) If not, I will machine the new one.
Tracking is as per Mel Bartels' : software running on 100MHz pentium, DOS, controlling both AZ and EL steppers appropriately.
It does slewing, Goto and tracking at sideral or lunar or solar rate.
It also understands LX200 protocol for autoguiding (but I do not use it, not yet).
I do not have field de-rotator yet (I am using this setup only for visual and occasionally for planetary photography), but Mel's software supports FR. Perhaps one day...
My focuser is easily detachable (Carl Zeiss style) and I am working now on focuser for camera at prime focus.
Mirror is mounted such that I have focus ~95mm outside the tube (my own design), and it will be OK with new camera focuser. I think even if field flattener id fitted, the focus is out enough.

For Moon and planetary photos (I tried couple of times, I saw that I can do it and that's it) I used projection, with camera (kyocera M400R or any other with 50mm filter thread) mounted on adapter (inside which was 9mm eyepiece)

[bojan]

After some time spent in storage, I am doing some modifications to my Bartelized Dobson (10"Newt), in encoder area.

Vertical axis is first, I added 3D printed timing pulley and a bit of up-gearing, to achieve 6600 ticks per 360° (Ek's interface).
The idea is to avoid friction coupling -it was OK, but because the driving plate was also friction-coupled, it was tricky to adjust frictions such that encoder reading is reliable

[bojan]

Today I printed elevation sector for encoder.
It will be fixed on the inner side of the round plate, to replace the existing friction coupled roller transmission.
With this sector (400 teeth for full circle) and 10-teeth pinion, I will have resolution of 6400 tics/rev, which is 3.38 '/arcmin (error due to plastic parts use will probably be larger).

[Wavytone]

...complete with period errors arising from such a small pinion.

Would have been much better if the teeth had been helical, or use a toothed belt between the two.

[bojan]

Yes, I know.. but I had only this one in my drawer....
If ever I stumble on the suitable helical gear, I will print the new matching sector - it is easy enough to do.

And it can't be belt (in conventional configuration) because I keep the OTA and mount separate when not in use (when put together they are too heavy to handle)... and I do not want to complicate assembly - this way, I simply put the OTA in place and encoder (and drive) is automatically engaged.



BTW, period of PE (due to teeth engagement) is a bit less than 1°, not sure about amplitude...

Since encoder inputs are used to calculate the tracking rates for EL and AZ motors, even 1° accuracy is good enough for visual.

If I need any better, I simply sync the system with observed (or photographed) object from data base (which is a bit fiddly process, not wanted when in dark and people around waiting for next object to have a look at).

[bojan]

Prompted by your comment, I went to have a look at Banggood... and found there metal helical gear (https://www.banggood.com/Tarot-450-Motor-Helical-Bevel-Gear-10T-3_17mm-TL45161-p-938175.html?rmmds=search&cur_warehouse=CN), of apparently right size.
Being a metal part, I will be able to "lap" the gear sector (printed with ABS filament) in boiling water, to improve gear matching even further.

And, on Thingiverse this contributor (https://www.thingiverse.com/janssen86/designs) made available designs suitable for helical sector (and much much more).

Thank you for suggestion!:thumbsup:

[sharpiel]

Bojan, I really like the things you do. :thumbsup:

[Wavytone]

:D great. Bojan !

[bojan]

Banggood has suitable pinion gear (https://www.banggood.com/Efite-Blade-130X-Walkera-V120D02S-Brushless-Copper-Gear10T-p-925848.html?rmmds=myorder&cur_warehouse=CN) as well.

[bojan]

After 20 days of waiting, gears arrived.
I am pleased with smooth match with previously printed test gear sector.
Now I have to print the whole sector (110°)... and remove a bit of material from the neck of the small gear (where the hole for grab screw is now).

As the gears are modulo 0.6, I will have up-gearing ratio of 365/10 = 36.5, and with 40 ppr (and quad decoder implemented in Ek' box) I will have 5840 tics/rev which results in 3.7arcmin resolution.

[bojan]

Gear cut to size...

[bojan]

Yesterday afternoon I experimented a bit with two other options - both gears printed, and cheap 360ppr encoders (~$13 at ebay).
Final decision depends on preliminary assemby, maybe this coming weekend.
I am looking forward to holidays... I really need a week or two away from work to sort out all that I started recently...

[bojan]

Encoders transmission in place...
(rusty spring goes on first occasion..)

[bojan]

This is the final solution for elevation encoder.
The load is now on ball miniature bearing (ø3mm), and encoder shaft is parallel to horizontal shapt of th etelescope (not that it really mattered for good functionality...)

[bojan]

After many years of use, I decided to "modernize" this scope.
OnStep (v4.24 running on MEGA2560 + ESP8266 + HC06) h/w is installed and working ok, functionality is very similar to Mel Bartels' software.
However, I had mechanical issue - vertical shaft (ø25mm) is still not stiff enough so the dob frame swings significantly which is quite annoying when windy and when un-intentionally touching the scope while observing.

I tried to use large lazy Susan (ø440 mm, purchased on ebay for that purpose couple of years ago), I tried it over the weekend and the improvement was dramatic, but I am also trying to find and implement the solution with lower resistance.

Small rubber wheels (3~4 riding on (rusty) ring welded on the tripod - I will have to flatten it and polish with angle grinder to make it aligned with vertical shaft), or ball bearings with rubber tubing is coming to mind.. I will try to put together something, but just in case there is some off-the-shelf solution.. I would try that first.

Does anybody knows about something suitable? maybe something like this (https://www.ebay.com.au/itm/185253020855?ssPageName=STRK%3AMEBI DX%3AIT&_trksid=p2060353.m1438.l2649)?

[bojan]

Maybe this solution could be OK ?... ø22 mm bearing with silicone rubber around ø6 mm shaft as dumping ..

[Rod]

Hi Bojan

Would ball transfer units do the job:

https://www.statewidebearings.com.au/linear-bearings/ball-transfer-unit/

Rod

[bojan]

Hi Rod, this could be OK.

I fund this (https://www.ebay.com.au/itm/203303687213?hash=item2f55d8142d:g: TsUAAOSw4XRgS07i) on ebay, it looks similar.. worth a try. No need to alignment in direction.
Or this (https://www.ebay.com.au/itm/133018233712?hash=item1ef8812b70:g: V~8AAOSwhBtcrT7W&frcectupt=true)..
What I need is vibration dumping.. load is secondary.

[bojan]

I will try this silicone tubing (https://www.ebay.com.au/itm/303970112130?_trkparms=ispr%3D1&hash=item46c607d282:g:UI0AAOSw-UxggO0R&amdata=enc%3AAQAGAAACoPYe5NmHp%252B 2JMhMi7yxGiTJkPrKr5t53CooMSQt2orsSz aVbXtmGxqmWnpn8RLN2q2O8tZ6vxOCZIcKK Ta6Z1wNrQAt9YEdnSKreGnoL9T%252Fy%25 2BUT9JlH3TB93OB7jYPzwFuoXfGfkhV7sFO XWVy9ONEQ3bW32smRv5B99Jaf5QFfz2ZyBf 9cajKHaOs2TTrn9tM7J1qpCzicFxruZjbAQ p%252BJWBuwmhA8RMEnWNMWoh973LubLlkL 7z2n33qqentxBFmntZtxqiV6YyHOmpP%252 BtSJBqYu6Udiovp1Q8qaX2SD7hqOhCY9jmk eWj2R2gXhmQ2BpVqrk2NIRwA79p9XZ8CnMx j0PyoTNOh5fTk9kieW6tDZRUkerYgfKiik% 252FRBHaYZQntk%252BwpLWo0%252FZKMk7 Rzq5g9Cbma4AD7xEqQQd9K20FgIZkgfJMlg tdOGUy0s2mjKcEfHmqDK8Cyfo3TQu94ztMa 1KqWyE7PGDSGUNfMiJdG759jZIf4RWEgUBc NmduZ3C4DtFHWQHja%252BajCwPn57sC75b NbACysharksDM7HPw91LyD9MWQUV69VBkHg SkQmCPrz3cScw3Ix%252BgMUGdjDpPrj7UX T%252B2qPjIguWAbMVl9QHb4yBs%252BQJB ecZbOju8syltNXbPy13Is9LM%252BII0lU5 n86u0RRLAHfER7uMi9fCONzinU6cbv5%252 FtAP4gTishHk0%252FPj%252BRMlt8HRiO6 2dHJ4uFR2fwSxdf2wvCpqvcQrGIuW4tZogg 1jH1JqD5naLuQqye7mGNsbs2JEvbWgjuGBv BqyjwsrjnmOCIFLMzNUOirkE4gTpxdyImW2 n06qJr30umFiIzgTIIhulhtv1S9oTQ5GLbe Vs3em2kyuEurjQWqse3B83UyjPtHOxTKXrt 9vw%253D%253D%7Cclp%3A2334524%7Ctkp %3ABFBMuOTE5ONf) and bearings first (I have them in my stash).

[bojan]

Tube arrived today... I hoped it would have a tad thicker wall, but this should be OK as well.
I can try to glue the tube with silicone rubber...which will fill the gap (~0.1 mm) Or.. I can try with cone spacers (that will force rubber tighter into bearing)

Anyway, I think 1 mm layer of silicone rubber may be OK for damping and also to accommodate tolerances of the metal ring.. (which I intend to grind flat and perpendicular to shaft by mounting angle grinder on the frame. I started this couple of years ago but stopped.. because I needed scope for something. Now it's the time to finish this idea.

[bojan]

Shown below is the jig for grinding the rail for damping casters, to be flat & perpendicular to vertical shaft...

[bojan]

And.. I found that the timing belt for AZ encoder became sticky and some teeth are now missing.. the rubber deteriorated and soon it will turn into a paste-like goo.. so it's time to change timing belt (420 mm, pitch 2.5 mm, 168 teeth).

I found it on AliExpress (10 pcs for ~$35, free delivery) and RS-components (1 pcs for ~$35, belt + postage).
Also on MBA (https://www.smallparts.com.au/store/item/b025t01370040psa/?v=4069) for 7.66 AUD (+ postage and GST)

[Maberly1959]

thats really very inventive, i hadn't thought of a mechanical intervention to prevent my dobsonian from moving

[bojan]

Thank you..
I build this dob mount from available parts, in hope it will be strong and stiff enough (as it was for the original purpose - front loader washing machine drum). It is.. sort of, but this way it will be much better.
Stability to wind blow and touch varies from case to case.. and here the obvious problem was vertical shaft - too "springy".

It is also pretty loud when stepper motors are working ( I have them mounted on rubber brackets). But, the tube resonates like the cavity of the acoustic guitar. When I remove the OTA from the mount, motors can't be heard at all.

Luckily, those vibrations are not visible even at high magnification (~400x).

[bojan]

Wheel is sort of ready for test on scope.. ( I need yet to cut off the excess length of shaft ø4.75 mm).