Mostly 3d printed EQ5/HEQ5 class mount progress thread ...

[vlaiv]

I started designing and making 3D printed astronomy mount in EQ5/HEQ5 class to be released as open source once I finish - so that anyone can make one themselves.

Idea of the project is to make available affordable, but good quality basic astronomy EQ mount for imaging and observation. I'm hoping to achieve or even surpass precision of common commercial mounts in this class at much lower cost.

I've already posted several times about this - but most of posts where scattered around and not put in single (as pointed out / suggested by @pipnina) progress thread - so this is the thread where I'll post progress updates and we can discuss the idea in detail.

Here is design image of the mount:

And this is where I've got so far with 3D printing (it's been hectic couple of weeks as far as 3d printing goes, I've gone thru at least 3 spools of ASA so far , just received shipment of further 2kg - which will hopefully be enough to at least finish this stage):

That is altitude part of wedge assembly complete and functional as well as bracket that will hold the mount. Whole assembly weighs in at 2Kg so far (but that includes both plastic and some hardware - at least 4 bearings of different sizes, M4, M5, M6 threaded rods cut to length, nuts, washers, 15mm aluminum pipe and numerous sections of 6mm linear rods).

I've also printed assorted bits and bobs - basically everything except the lower housing for azimuth adjustment - which is massive 300g part that will print for at least 15h. I'm hoping to start that print tonight so it will be ready for tomorrow for assembly.

[Dr_Ju_ju]

Well Done !!  looking forward to the completion, and making me sorely tempted to go for the next class up.....

[vlaiv]

3 minutes ago, Dr_Ju_ju said:

and making me sorely tempted to go for the next class up.....

You know how telescopes and astronomy equipment always look larger in person than in ads or on images?

It looks like it might be the case here as well. I was hoping to design a reasonably sized mount in line with the "class" I selected - but it turns out that it might be a bit larger than expected or perceived from the cad rendering

There is ~60cm from "top" point to the "bottom" one. I think it will look at least twice as large when placed next to HEQ5

I have no idea how will this transfer to carrying capacity and performance of the mount though.

While we are on the subject - I'm accepting ideas on how to test carry capacity of the mount without actually attaching heavy telescopes onto it. I'm happy to try with fairly light weight telescopes like 4" F/10 achromat that has ~4Kg in total, but what about 10Kg or 15Kg? I don't even own a telescope that is 15Kg. All my OTAs are up to 10Kg in weight.

[Dr_Ju_ju]

Do you have/access to any counter-weights, as they usually come in 5-25 Kg ranges....

[Vroobel]

I have 17kg 10" F/5 Newtonian, so imagine it with all accessories needed. Believe me, as far as I know strength of 3D printouts, it will give up under the similar load. Of course, it depends on what filament you use. Anyway, fingers crossed. I like your idea and performance. 🙂 🤞

[vlaiv]

8 minutes ago, Dr_Ju_ju said:

Do you have/access to any counter-weights, as they usually come in 5-25 Kg ranges....

CW bar is supposed to be 20mm one - at least that I's what I got (alu bar 400mm / 20mm diameter). I'm probably going to use M8 tap on it and piece of M8 threaded rod to make screw attachment (similarly end stop on the other end).

I'm hoping to adopt regular workout weights to acts as counter weights. These are quite cheap:

https://www.decathlon.co.uk/sports/fitness-gym/weight-plates

They have 28mm center bore, so I'll 3d print adapter to fit them on 20mm shaft and I'll devise some sort of locking mechanism as well.

[vlaiv]

17 minutes ago, Vroobel said:

Believe me, as far as I know strength of 3D printouts, it will give up under the similar load.

I'm not sure about that.

ASA has ultimate yield strength of say around 20Mp. Let's give it a 50% margin - so 10Mp. That is 10 newtons per 1mm2 of load.

20Kg will be roughly 200N so it would take 20mm2 cross section of plastic to hold 20Kg hang from it. That is 4x5mm or 5mm diameter stud (so basically M5/M6 3d printed bolt could easily hold 20Kg hang from it).

I'm printing large structural parts with 2mm wall thickness and 60% Gyroid infill. Every load bearing part has at least x5 that surface cross section.

[Elp]

At 600mm they'd likely be significant load on the EQ wedge. As you've devised, for the test some sort of bar attached to the top. I've used an all thread bar previously for CW use, weight lifting weights attached, large diameter washers either side of weight and locked down hard with pliers with serrated flanged washers (I wouldn't want to use anything larger than 1.5-2Kg plates for this, maybe also have some 3d printed bushes in between each weight so they don't shift due to gravity due to the difference between the bores and the all thread diameter).

Or simpler, use some round wood, or aluminium round bar, and lock the weights in place with shaft washers (that's not the proper name, Ive got some for another CW bar, they're reasonably thick in width and have grubscrews in each to lock them to a bar, think they're for stuff like motor assemblies).

[Astronomist]

Very interesting project! How will it be driven? 

BTW @Elp you are refering to shaft collars.

[vlaiv]

3 hours ago, Astronomist said:

Very interesting project! How will it be driven? 

Ok, so I envisioned 4 different reduction systems.

3 of them will be open source, and fourth will be "proprietary" - just because it will require very precise lathe machining and some "tricks" to get it working. Since whole thing is open source - people might even contribute other versions like DIY harmonic drive or DIY cycloidal drive (there are examples of both being made in DIY version for robotics projects on you tube).

Ones that I envisioned are:

1. Split ring compound planetary type reduction system - here whole assembly and gears will be 3d printed, except for bearings.

2. Multistage belt system - here we will probably have choice of 3d printed pulleys or aluminum ones that are available (GT2/GT3) - belts will be stock ones that are available of the shelf.

3. Special type of system that I developed that uses magnetic coupling to provide reduction. This is "maybe" type of reduction since I have prototype of it - which is working as it should when you rotate it by hand, but needs to be checked for number of parameters - like what type of periodic error it has and can it be corrected in firmware, what type of "elasticity" does it have - how much torque can it transfer and so on ...

4. Friction drive. This will be best option, but I'm guessing no way to make it in house. Even sourcing it will be hard / very expensive. I've come up with idea of how to make friction drive much better than existing versions (smaller footprint, higher reduction, more points of contact, no need for special tensioning and so on ...).. If I manage to make one myself, then I'll offer them as separate "add on" that you can purchase for your mount (provided that I can cost effectively manufacture them).

All of this will be coupled to 0.9 degree modern nema 17 steppers with 2209 drivers and raspberry pi pico as micro controller.

At some point, I thought of a "side project" - to create special type of hand controller for the mount. Again DIY version, but question remains if there will be interest for that, or will people predominantly use the mount with computer.

Hand controller will also be very inexpensive project, but much different than regular hand controllers as it will be based on raspberry pi zero computer (running linux) and touch screen (no buttons). It may feature voice recognition as special feature - but this will be largely up to community to develop under my guidance or just initial idea. Idea is to have mount that can be voice controlled - without the need to look at the screen.

You tell it to "go-to NGC 7331" and off it goes and when slew is finished - it will use speaker to inform you that it is done (or beeps or something).

This will allow you to preserve night vision and still locate targets with ease.

However - those are just ideas (that I've contemplated in some depth) - but not immediate goals of the projects.

Immediate goal is to have mount that can carry payload securely, tracks well, has accurate goto and can be controlled via computer.

[vlaiv]

Another small update.,

I managed to put all printed parts together to complete wedge for the mount (which works very nicely and smooth, and I think it's precise enough for polar alignment).

Here it is attached to HEQ5 tripod.

I've started the work on RA axis cage now.

[vlaiv]

I just realized that I mounted top part in the wrong way

It should be facing the other way.... Luckily it is just few M5 bolts to separate the two parts and rotate upper / alt section.

[Vroobel]

Every post I'm more excited. 🤩

[vlaiv]

Seems that I've hit first bumps in the road ...

Well, maybe not first bumps, but first thing that I was really annoyed about.

There seems to be several places where I messed up tolerances and dimensions. In most cases, I was able to do a simple fix to prototype - often drilling out hole to appropriate size - or using soldering iron to squeeze nut into slot that's just a bit too tight for it.

However, I've messed up cage tolerances quite a bit. I mean - I managed to put it together, but I really worked up a sweat in doing that as it was very hard to force aluminum pipes into appropriate holes. I even managed to crack a bit of plastic because I seriously over-tightened a bolt. I think that would not have happened if I designed the feature appropriately - even with very large tightening force.

Above image shows my mistake. Gap is just too large for plastic to bend that much if I put very tight fitting pipe in the hole. Gap should be barely larger than the space around the pipe, so that:

a) plastic around pipe actually closes on pipe as part of tightening process (so it that pipe is held by friction)

b) after that happens - there is minimal gap that can be closed on over tightening and then plastic on plastic on plastic contact will resist further tightening of the bolt.

The way I missed the tolerance - it might as well be that forcing the pipe in the hole even expanded gap a bit.

Slight crack won't be a problem for functioning of the mount - but I'll have to disassemble the cage at some point to fit reduction mechanism and I feel that is going to be right pain ...

I'll post latest images tomorrow.

[LuckieEddie]

You might want to check out out Slant 3D on YouTube. The channel is run by a commercial print farm and has lots of tips on designing 3D printed parts for mass production including aspects like tolerance and fit.

The only thing is I find the presenter quite annoying but the info is good (I tend to view him muted with subtitles on).

[Clarkey]

1 hour ago, LuckieEddie said:

The only thing is I find the presenter quite annoying but the info is good

I wish I could do this with the better half🤣

It's looking like a really good project @vlaiv. I might even be able to justify buying a decent 3D printer if you can get it to work.

[JOC]

It looks like a big project - are you costing it vs. buying the mount?

[Vroobel]

Sometimes making things from scratch is the value itself. Many people asked me the same question when I was building my ATM EQ fork mount. Finally, it costed not less than a used EQ8, but I made it, it works quite well and I'm proud of it. That's priceless. 

[vlaiv]

3 hours ago, LuckieEddie said:

You might want to check out out Slant 3D on YouTube. The channel is run by a commercial print farm and has lots of tips on designing 3D printed parts for mass production including aspects like tolerance and fit.

The only thing is I find the presenter quite annoying but the info is good (I tend to view him muted with subtitles on).

Yep, seen many of their videos. They are OK (and I can tolerate the guy ).

I'm usually fine with tolerances, but here I actually had few numbers entered wrong and one clear design mistake - that I'll correct. I still have two such parts to print for DEC axis and I'll change design for that.

 

[Elp]

Tolerance depends on filament also as not all print the same. I've found a typical free fitment tolerance around +0.5mm (a side on holes) usually works, for a more free fitment +1mm.

[vlaiv]

1 hour ago, Clarkey said:

I wish I could do this with the better half🤣

It's looking like a really good project @vlaiv. I might even be able to justify buying a decent 3D printer if you can get it to work.

I'm starting to believe that 3d printer is an essential tool in astronomy box. It is however a tool/workshop machine and not an appliance (as many would like to make it - just "load" the model and it pops out), so there is some adjusting and tuning involved to make it produce accurate parts.

If I manage to make this work satisfactory - I think it will even be feasible for someone to invest in 3d printer, make this mount and still be in the same price range as say HEQ5 or similar.

[vlaiv]

1 hour ago, JOC said:

It looks like a big project - are you costing it vs. buying the mount?

It is actually fairly cheap if you print it your self. I'm not sure how much it would cost to have all the parts printed for you by 3d printing service though.

I've spent roughly 100 euro on plastics so far (4 spools at ~25 euro each), might need one more spool.  Aluminum profiles were less than 25 euros (for total of 25 euros I purchased all that and also tube for DIY small 80mm refractor that I'm building). Stepper motors are like 10-15 euro each and we need two, Raspberry PI pico is something like 5-10euro and drivers are 10 euro each. Assorted pieces of hardware are less than 50 euro in total (I'm just eyeballing it).

I believe that bill of material will be around 250 euro if you go with ASA or possibly around 350 euro mark if you decide to use stronger filament like Prusa PC Blend (which has very similar properties to ASA as far as durability / absence of creep goes but is at least double the strength and more stiff - but costs twice as much).

Again guessing, but total with external printing service might set you back 500 euro or so? Still much cheaper than commercially available options in this class.

 

[vlaiv]

11 minutes ago, Elp said:

Tolerance depends on filament also as not all print the same. I've found a typical free fitment tolerance around +0.5mm (a side on holes) usually works, for a more free fitment +1mm.

I've found that I need just 0.2mm in radius to have somewhat looser sliding fit.  With exact dimensions - I need to run a drill of that size just to scrape ridges on layer lines and I have tight fit / need a bit of pressing to push it thru.

[vlaiv]

Here is mini progress report:

And here is the crack:

Sorry about the poor focus / phone shake.

As you see - it started to split along layer line because I forced it to bridge the whole gap and join with the other side.

Making gap smaller and hole larger so that two nicely complement each other will solve this issue. Or simply being careful not to over torque the screw (and it did take some effort to crack it like that).

[Elp]

Having a set of files at hand is useful too, filament files down quite well. It was useful when I made a model cast tool with removable tool inserts and sliding ejection system.

You can get hot tools too I've got one for punching holes, carving etc (it's a bit like a soldering iron with different tip attachments), you could repair that split with it. Its a messy process though as the melted filament tends to stick to the tool you're using (a bit like trying to carve hot melted chocolate).