×

INDI Library v1.9.1 Released (26 Jun 2021)

Bi-monthly INDI Library v1.9.1 is released bringing a few new drivers and improvements to existing drivers.

INDI focuser driver for Waveshare Stepper Motor HAT for Raspberry Pi

  • Posts: 257
  • Thank you received: 61
By the way, here's my 3D printed motor bracket, and some photos of it mounted on my scope, a Stellarvue SV80 Access.

That Raspberry Pi case is no longer used, I replaced it with a custom 3D printed one, I just haven't taken any photos of it yet.

I used a DB9 connector because that's what my Moonlite focuser on my other scope uses. My Waveshare driver controls the Moonlite as well as my DIY focuser. I wired it to match the Moonlite, of course.

1 year 4 months ago #51118
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 257
  • Thank you received: 61

Wow, beautiful setup! Love all the red! And your custom mounting solution is super clean. Love it!

And yeah, the Waveshare HAT has overheat protection built it. You are probably right that it overheated and shut down. If you plan on keeping the motor powered full time, you'll have to adjust the current limit. It isn't terribly obvious how to do it. You will need a multimeter and a tiny screwdriver.
1 year 4 months ago #51119

Please Log in or Create an account to join the conversation.

  • Posts: 215
  • Thank you received: 15
Yes...the primary constraint that forced me to gear to the coarse focus was the complete lack of a fine focus mechanism. :)
SkyWatcher 10" SynScan Goto Dobsonian, SV106 Guide, ZWO ASI178MC, Svbony SV105, SV305, Pi3HQcam, Pi4, DIY WaveShare focuser, Linux Ubuntu, KStars/Ekos, Carte du Ciel/CCDCiel, AstroDMx_Capture, Siril, StarTools, GIMP
1 year 4 months ago #51120

Please Log in or Create an account to join the conversation.

  • Posts: 59
  • Thank you received: 3
It looks so clean because there are only two cables in the pic. A rat's nest is not far behind :).

Next is a sidecar clip/mount for the SSD and some additional clips/hooks for cable management. I also need to make some short power cables. Now that I've got the main designs done I can size the cables appropriately. I also need to shorten the stepper cable. I wish I had ordered some connectors before Amazon got overloaded and deprioritized the non-essentials. I'll have to go through Digikey or somebody else like that.

I think I'm going to try to keep to full steps for the focuser operation. That will have multiple benefits. If needed, I can adjust the current limit, though if I'm hitting the limit I may have other issues as well.

I really love the Waveshare stepper hat though. I'm very happy I have been able to get rid of the voltage converter I was using to go from 12v in the PPB to 5v for the rPi4.

There is a benefit to the cloudy skies I've been having. I've been able to pace myself better with this project and avoid the sense of urgency to GET IT DONE NOW... THE SKIES ARE GOING TO BE CLEAR TONIGHT. The downside is plenty of time for 3d model revisions == running out of the black PETG I wanted to print everything with.
The following user(s) said Thank You: Jon Carleton
1 year 4 months ago #51122

Please Log in or Create an account to join the conversation.

  • Posts: 215
  • Thank you received: 15
ESB:
Yes, wire management/reduction is an ongoing problem. I think I've redone my wiring 4 or 5 times since January, when I got the scope. Yours looks good so far! I agree on the design of the WaveShare HAT. I was able to lose ome more wall-wart in the process and am currently down to only 3. One for the Pi/Hat, One for the SynScan/Motors and one for a powered USB hub. I may be able to cut that down to one soon if I can find just the right LED power supply to replace all the wall-wart transformers.

KROSS:
I am jealous of your 3D-ism. I am still old-school, traditional mill, lathe, cutters and fiddly bits. I get the job done, but it takes time. I did save time on the Pi case, however. I found this case made from pieces of cut-out lexan with a nice little fan. So, since they are just layers of cut-outs, I bought 2 and tossed out the extra top and bottom, used longer screws to hold the stack together and presto! A double-tall Pi case with fan. I believe I paid $12.00 for the two cases on eBay.

SkyWatcher 10" SynScan Goto Dobsonian, SV106 Guide, ZWO ASI178MC, Svbony SV105, SV305, Pi3HQcam, Pi4, DIY WaveShare focuser, Linux Ubuntu, KStars/Ekos, Carte du Ciel/CCDCiel, AstroDMx_Capture, Siril, StarTools, GIMP
Last edit: 1 year 4 months ago by Jon Carleton.
1 year 4 months ago #51127
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 257
  • Thank you received: 61
Jon,

I come from a family of machinists, so I can relate! I myself have a 12x36 lathe out in the garage. No mill yet. 3D printing can be fun, but it can also be a huge time sink. But as a machinist, you would find the lack of precision in the mechanics of hobby-grade 3D printers would drive you nuts. :)
1 year 4 months ago #51132

Please Log in or Create an account to join the conversation.

  • Posts: 257
  • Thank you received: 61

Ah yes, my scope would look neat, too, if I removed all the cables. :)

May I ask what stepper motor you are using? I actually went for a .9 degree stepper instead of the more usual 1.8 degree, which gives twice the resolution.

And yes, good call on using PETG. PLA doesn't really do all that well when left outside, especially in the sun.
1 year 4 months ago #51133

Please Log in or Create an account to join the conversation.

  • Posts: 59
  • Thank you received: 3
I'm using an SM24240 from Waveshare. It is a 1.8 degree / step motor with .45N-m holding torque. I just grabbed it when I ordered the stepper hat. Fortunately it looks to have a decent amount of torque for that size stepper. Good idea on the .9 degree stepper. I just did some research and it looks like they don't have quite as much torque as the 1.8 degree versions. I couldn't find any that had more torque than the model I'm using and I'm pretty sure torque is a limiting factor for my use. I did try adding a step delay and it may have worked a bit better, though it ended up losing the smoothness and seemed to be starting / stopping for each step (microstep). It was also a lot slower. Given the amount of inertia / friction, I'm not sure its a help to start/stop the rotation like that.

While researching I came upon another option that seems like it might work much better for my use. I found steppers with built-in gearboxes. Specifically, I found a model with a 5.18:1 ratio. That will increase torque (to 2N-m) and decrease the step size to .35 degrees / step. If it ends up too slow, I can decrease the micro stepping ratio. I've ordered one, so I'll see how it works.

As for PETG, I have conflicting requirements. The base / stepper mount needs to be as stiff as possible (and not deform over time while under strain), but it also is going to be outside and very likely may spend time in the sun. So far PETG seems to be working, but we'll see if it holds up over time.

For the base, I definitely could have used a mill and metalworking skills. Its not much more than a flat bar with a couple of slots milled into it to adjust tension (plus the stepper mount). If the 3d printed version fails over time, I'll have to pursue a metal version.
1 year 4 months ago #51138

Please Log in or Create an account to join the conversation.

  • Posts: 257
  • Thank you received: 61
For your use, it sounds like a geared stepper motor is the answer. They are more pricey, and can introduce backlash, depending on the quality.

I'm thinking that using millisecond resolution for the delay between steps is too much. I think adding even just 1 ms per step is not fine enough. I'm thinking I should change it to microsecond resolution.
1 year 4 months ago #51171

Please Log in or Create an account to join the conversation.

  • Posts: 59
  • Thank you received: 3
I ordered a Wantai, so I'm hoping is should be relatively decent. I did realize after I ordered that I didn't check that the shaft had a D shaped profile. They're sending me the mechanical drawings. Worst case I'll have to file it down a bit.

Smaller resolution for the delay would probably be a good idea. I wonder if it would be helpful to add a delay between enabling the motor and starting the stepping.
1 year 4 months ago #51172

Please Log in or Create an account to join the conversation.

  • Posts: 257
  • Thank you received: 61
I've gone ahead and updated the driver to use microsecond (us) instead of millisecond (ms) for the delay per step. Since it's raining outside, I don't want to take the cover off of the telescope to test it, so I'll let you be my beta tester. :)

Just go to the source folder, and:
git pull
cd build
make
sudo make install

And restart your INDI drivers. Let me know if that works better.

Thanks!
-- Kevin
1 year 4 months ago #51188

Please Log in or Create an account to join the conversation.

  • Posts: 257
  • Thank you received: 61
Okay it wasn't raining today, so had a chance to test out my changes. I can say this is a vast improvement. I have found the sweet spot in the delay per step adjustment to be in the 20 to 100 microsecond range. Much smoother and faster than the old method of adding a full millisecond (or more) per step, and still plenty of torque (for my setup).

Give it a try, I think you'll like it!
1 year 4 months ago #51262

Please Log in or Create an account to join the conversation.

Time to create page: 1.145 seconds