New G11 OnStep Conversion


Lloyd Simons
 

I was finally able to get my OnStep converted G11 out under a clear sky for some testing. Of course there were a few hiccups, but I was able to get it to guide reasonably well. However, there was some obvious periodic error present in the guide graph (the last three runs in the attached log) and upon frequency analysis, the major spikes were related to the worm period (80', 120', and 240'). My conversion is similar to the one from Rockmover (see pic 1) with a 16t pulley on the stepper and 48t on the worm (3:1 GR). I have the older steel worms. I would like some help in figuring out how to eliminate this periodic error as those large spikes are going to cause some issues while imaging. I have purchased new bearings and Belleville washers for the worms but have not installed yet. The advice from the Losmandy group was to adjust the belt tension and the worm mesh. Any other comments would be appreciated.


biggs@...
 

This is a long shot, but make sure your pulley gears are GT2 and not MXL.  I had a weird periodic error with GT2 belts but didn't realize my large gear was MXL.
I couldn't find the right pulley on Amazon, but did find Rubber Belts Online (no affiliation), which works great.
https://www.rubberbeltsonline.com/6mm-Bore-Timing-Pulley-2mm-Pitch-60-Teeth-6mm-Wide-p/6mm-60t-6mm-gt2-pulley.htm


Drew 🔭📷🚴‍♂️
 

On Mon, Aug 23, 2021 at 11:15 AM, Lloyd Simons wrote:
The advice from the Losmandy group was to adjust the belt tension and the worm mesh.
Worm mesh, yes absolutely. Belt tension, probably not. A timing belt drives by the "tooth face" and is very tolerant of different belt tension. That is why it is called a "timing belt". Any overtightening will just stress the bearings, a too loose belt will skip. I check the tension by twisting the belt between the two pulleys. If I can twist it 90 degrees then the tension is within tolerance.


Lloyd Simons
 

The belts and pulleys are both GT2 (at least that is what I ordered). I don't think it is belt tension either but I will try your twist test Drew. My understanding is to get the worm as parallel as possible (using feeler gauges to set both sides of the worm blocks to be the same distance from the mount head), and insure the worm is "level" (left to right, z axis). I'll try making some adjustments and try to get it out the next clear night. I wish there was a way to test this indoors during the day...


Lloyd Simons
 

So I was able to fix this issue by replacing the pulley. Either the original was not seated properly or it was defective. Replacing it led to a night of guiding with no RA spikes and an RMS between 0.5 and 0.6. SOLVED.


Howard Dutton
 

Nice, thanks for the update!


bjaffa Jaffa
 

Lloyd,  can you go into more detail on how you adjusted the worm and pulley setup? I have tried in the past but was not very successful.
Also do you have any binding issues with temperature variation or backlash?
Can you send a link to where you purchased your components?
Did you have to remove the adjustment knob for altitude like rockmover?

 Thanks I am impressed with our work -- Brent


Lloyd Simons
 

Brent,
I plan to write up a build summary for the wiki, so I can include much of this info there. I have been using feeler gauges to help set the worm mesh. I haven't noticed any temperature variations yet but I have been only using it in a narrow range so far (12 to 25°C). Ideally, the worm would be 5mm longer to fit the 48T GT-2 pulley better, but it does seem to work. It is a bit challenging to get everything lined up with the motor plate and worm gear being held on with the same screws. I found that mounting the motors last was the best solution. Yes, I did remove the altitude knob. The RA motor plate hits it. 


Drew 🔭📷🚴‍♂️
 

Glad to hear you were able to resolve the problem and get good results. Do you have any idea what was wrong with the pulley? I think worrying about the pulley parallelism is secondary to correct worm/gear meshing. The timing belt will always track to the flange. A parallel belt will switch flanges with a direction change. I consider it acceptable as long the belt does not seem to be bunching up on the flange. I don't know what the worm/gear adjustment procedure is for the G11, but I would make sure it is good to prevent future mechanical issues. With your lack of spikes I would look for it being too tight, if anything.


Lloyd Simons
 

I purchased a Losmandy G11 with a 492 clock drive in December last year with the intent to convert it to goto with OnStep. My Vixen SP build was a fun project, but I had already nearly maxxed out the payload on that mount and I have always admired the G11. I decided to follow in the footsteps of Rockmover and try his flipped worm with aluminum plate mount for the stepper motors (https://onstep.groups.io/g/main/message/20803). I also wanted to build a new controller based on the Fysetc S6 that I could use to control several other accessories (focuser, weather, dew heaters, power distribution). I printed a box for the controller and kitted it out with everything I needed. I wanted to use the 12V outs from the board to power my imaging PC and my QHY 268M camera (more on that in a moment). 
 
I ordered 2 aluminum plates from McMaster to minimize the amount of cutting I needed to do. I decided to use aluminum motor brackets instead of printed ones to prevent any warping or flexing. I remixed a motor cover design to accomodate the RJ45 connectors and make a nice clean look. I had to remove the altitude knob so the aluminum mounting plate would fit. I also used the scraps from the aluminum plates to shim up the saddle and prevent the saddle knobs from hitting the DEC motor (I didn't want to spent $300 for the DEC extension).
 
Controller BOM:
Fysetc S6
TMC 5160 drivers (RA and DEC)
LV8729 driver (focuser)
BME280 SPI temperature, pressure and humidity sensor
Wemos D1 mini Wifi module
RCA female plugs (dew heaters)
2.1x5.5mm female 12V barrel connectors
One wire temperature probe (focuser temp)
RJ 45 connectors (RA, DEC and focuser)
2 pin aviation connector (24V power)
on/off switch
switch for boot jumper
various connectors and LEDs
24V power supply (150W)
 
Mount BOM
Nema 17 44n/cm 1.68A 0.9° stepper motors (2x)
Pololu L bracket motor mount (2x)
3"X6"x3/16" aluminum plates (2x)
48t GT2 worm pulley (2x)
16t GT2 motor pulley (2x)
76t (152mm) GT2 timing belts (2x)
 
I used some aluminum extrusion to elevate my OTA so I could mount the controller underneath it. I also printed a mount for my imaging PC and have it piggy-backed on the OTA as well. I added a few cable guides etc to clean everything up.
 
I did encounter a couple of small issues during testing. The most disappointing was that I could not power the camera through the S6 board. Whenever I turned on the TEC the board would freak out and I would have to reset it. I guess the amp draw was just to great. It's not a big deal to run an additional wire to supply the camera, but I was really hoping for a nice clean look with only 1 power supply for the whole system. I also had to replace the worm pulley on the RA axis. The first one had some significant periodic error. 
 
I had to increase the goto amps for the DEC axis to near the max (1600ma) in order to prevent stalling. I worked on balance for a couple of weeks and it would always stall when the scope was trying to go passed vertical. There may be some additional z-axis balancing that could improve this. The motor does not get hot at all, so I don't think it will be an issue. I can also replace it with a higher torque model if need be.
 
Overall, I am quite happy with the result. The mount was guiding at 0.59 to 0.60 RMS on the second night out. I may try to record a PEC curve for grins. I mentioned in another post the issue I am having with the guide rate at the start of the session (https://onstep.groups.io/g/main/topic/85327469). Other than that, it is working well. Now I need to work on tilt and spacing for my new imaging system...
 
Thanks to Howard, Khalid, Dave and all the others that have helped along the way. Clear skies!


Khalid Baheyeldin
 

Lloyd,

That is a nice conversion, using many of the S6 features.

I find that 1600mA for IGOTO too high, and no idea why this is so.


Khalid Baheyeldin
 

Oh, forgot to say that your conversion is in the show case now.


Jesse Lichtenberg
 

This looks great! How has it been working out for you with astrophotography? I'm using a converted Exos-2 at the moment but I'm going to run into payload issues soon. I really like the idea of finding a G11 like Howard, you, and a few others have done.


Alexander Varakin
 

Lloyd,
Great conversion!
You have a huge 1:3 reduction, DEC stalling should not be happening. 
Most likely you are trying to run the motor too fast. Reduce the speed of it and stalling should stop.
Once it is stable, you can reduce the current.