Old 22-06-2021, 08:34 AM
etill (Elliot)
Registered User

etill is offline
Join Date: Oct 2019
Location: Melbourne
Posts: 111
HEQ5 belt mod


Recently I finished a belt drive conversion of an HEQ5 (non-goto) mount I got from another member here. I managed to get it out last night to do some initial testing.

The brackets were designed in fusion360 and made from some aluminium flat bar offcuts in my mini cnc machine. The rest of the parts I had lying around from cnc machine / 3d printer mods done recently, but they would not cost much if purchased for the project.

There's 2x 60t and 2x 15t GT2 pulleys, 2 x tensioners and some 5mm bearings which are way too small for this, but since i had them on hand..

The motors are 0.9deg nema16. I had hoped for an install without removing the dec axis but that's not possible with these. I did manage to meet another goal though which was to not modify, grind or cut the mount at all. It can be put back the way it was just by swapping all the parts back.

There are a few mistakes I've made that need correcting with some new brackets.

First off the 60t pulleys are upside down (see photos), the grub screws were a little too high up the worm shaft and wouldn't quite hold. I just made some extra spacer brackets to raise everything up, but given the way the motors are mounted the 15t pulley sits right at the end of the motor shaft, and those little 5mm bearings only provide a little extra support. I'll fix this by turning down the center of pulleys in the lathe and then drilling and tapping new threads for the grub screws at a 45deg downward angle. I may have to make the grub screws to suit but that shouldn't be a problem.

The second mistake was fixed belt tensioners. I had planned to make it adjustable or spring loaded but space was tight and I couldn't quite work it out. The tensioners work fine - when the mount is assembled exactly as it was when I measured everything. Once you adjust the mesh of the worm gears they either get a little too tight or a little too loose, not to the point you can't use them, but worth a rethink for the next next version.

The next issue was initially the motors were in contact with the dec shaft. You can see the rubbing in some of the photos. That required another set of spacers with 0.5mm shaved off and slightly bigger holes in the main mounting plate to give me some wiggle room. I doubt you could get a sheet of paper between them now, but they are clear of the shaft.

The drivers and electronics are TMC5160 on an OnStep miniPCB2 board with a Teensy 4.0. Microstepping for tracking and goto is set to 256, couldn't see any point going lower given the Teensy 4 is plenty fast enough. The maximum slew rate was lower than expected, it's not a problem at 1.5deg/sec but double that up to 3deg/sec and there are occasional stalls. This was true with any combo of microstepping for tracking and goto - from 8 all the way to 256. Adjusting the current didn't change anything either. The worm mesh on both axis is super tight, no detectable bashlash and still smooth to operate by hand perhaps some more adjustment there could help, but since I don't mind 1.5deg/s I'll probably leave it alone.

The driver board is mounted inside the mount on a couple of simple brackets I printed. The original cover that holds the power socket and hand controller port has been replaced with a printed one that has USB, power and an external wifi antenna.

For testing I used an ED120 as a 'guide scope' (900mm F/7.5) with a QHY485 and PHD2 for polar alignment and measurement. After alignment and running through the guiding assistant a couple of times to make adjustments I tested guiding in four places, way to the west, way to the east, straight up and pretty close to the SCP. All results were the same, and I'm pretty happy with it for a first attempt. PHD was making some fairly agressive looking corrections but I assume it knows what it's doing after all the calibration and adjustment it recommended.

There was a bit of cloud about now and then in Melbourne last night, and it was very cold with lots of dew issues but I had managed to get the testing done I needed to by about 11pm which is when I had to retire for the night.

I've got a few revisions to do before I'll call it done, but I'm happy to share the drawings for anyone to give it a go, unfortunately not suitable for 3D printing at this point, but it would only take building up some new parts in fusion from the same sketches and export the STLs. I don't think they'd be sturdy enough for long term use but I made a full set of brackets and pulleys for my CGEM conversion in PLA and they lasted several weeks of use before starting to warp so you never know, ABS or something may work ok.
Attached Thumbnails
Click for full-size image (PXL_20210612_230939880.jpg)
159.3 KB41 views
Click for full-size image (PXL_20210613_030134433.MP.jpg)
153.8 KB36 views
Click for full-size image (PXL_20210620_064708024.jpg)
137.0 KB33 views
Click for full-size image (PXL_20210620_064713547.jpg)
143.2 KB33 views
Click for full-size image (b.jpg)
191.1 KB33 views
Reply With Quote
Old 22-06-2021, 09:28 AM
bojan's Avatar

bojan is offline
Join Date: Jul 2006
Location: Mt Waverley, VIC
Posts: 6,431
Very nice work!
Reply With Quote
Old 22-06-2021, 12:43 PM
Registered User

RyanJones is offline
Join Date: Jun 2018
Location: Melbourne,Australia
Posts: 1,233
Very Interesting Elliot and a great write up. Well done 👍
Reply With Quote
Old 23-06-2021, 05:26 AM
Pepper (Steve)
Registered User

Pepper is offline
Join Date: Dec 2018
Location: Tamworth
Posts: 297
Mighty fine work mate. That mini Cnc would be just the thing.
Reply With Quote


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 05:04 PM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.
Limpet Controller
NexDome Observatories
Astronomy and Electronics Centre