High Spec Computerized Equatorial Mount : building from scratch.

[FrenchyArnaud]

Hi guys, 
Starting yet a new thread for another diy build - but this time it could well be the diy project of a life time because, let's face it, I am too stupid and arrogant to know when I am out of my league. So, on the strength of "He did not know it was impossible so he did it",  I am tackling the build from scratch of an equatorial mount with very serious ambitions, namely and in no particular order: 

- Must be able to take overkill loads without breaking a sweat, even if one day I can afford a quattro 12" or sthg like that. 
- Must have near perfect tracking speed
- Must have negligeable backlash RA + Dec 
- Must be 100% ascom compatible so as to snuggle lovingly with NINA & PHD2, which also means
- that it must be "pulse guiding capable"

The things I DO NOT need, because it will be on the pier with illimited power supply:

- A weight limit. Even if it ends up weighting 50kg: who cares. 
- Battery autonomy and reasonnable current use (it's on the mains)
- Advanced Goto.  I only need to be able to point roughly in the right direction, NINA will plate solve afterwards. 

With this in mind I have already determined quite a few things. 

Hardware: 
- Motors will be heavy duty steppers, NEMA17 or even 23. With low backlash gearboxes either 30:1 or 100:1 and the transmission will be 100% belts.  First calculations tell me that can move up to 15kg of balance error which means it's effectively unlimited load. 
- The base will be a wedge, much easier to build rock solid than a GEM turret. 
- For now the plan is to have both axis free spining using car wheel bearings (there is pretty much zero play in these things and the load is obviously not a problem) 
- These axis will be hard mounted on a BIG steel barcket that will in effect act as the "chassis" of the mount.
- It will embark its own RealTimeClock and Eprom so that it is effectively plug and play.

Electronics: 
- prototyping and firmware/software dev is now on Arduino Nano and Uno but I will surely need a Mega plus a whole collection of rotary encoders, LCD, temp probes etc. 

For now I have just put 2 tiny steppers on a bread board to confirm that I can get it to be seen as a telescope by a basic ascom driver and get the motors to act as Ra+Dec. So yes, this tiny mess is, technically, a computerized telescope mount.

First estimations tell that the build budget would be surprisingly low - in the order of £200 to £500 at the very, very max. In the other hand, the firmware and ascom driver could be really, really long to write. 

But hey, dream big. 

All ideas, inputs, remarks etc more than welcome. 

[Dr_Ju_ju]

I don't know whether you've come across this site (https://www.astroeq.co.uk/tutorials.php) but I made one to drive an old EQ5 mount, & it works quite well ....

[FrenchyArnaud]

1 hour ago, Dr_Ju_ju said:

I don't know whether you've come across this site (https://www.astroeq.co.uk/tutorials.php) but I made one to drive an old EQ5 mount, & it works quite well ....

Thanls, a few interesting pointers there 👍

[LuckieEddie]

Onstep could be your friend for this project. Its a full featured open source control system that runs on various Arduino systems. Drivers for Ascom and Indi are mature. I've converted two mounts to use it so far an E-Q5 and a G11.

[FrenchyArnaud]

22 minutes ago, LuckieEddie said:

Onstep could be your friend for this project. Its a full featured open source control system that runs on various Arduino systems. Drivers for Ascom and Indi are mature. I've converted two mounts to use it so far an E-Q5 and a G11.

Another good pointer, thanks! 

[windjammer]

Hi 

I have been down this route!  A wonder about your choice of motor drive - I found it very difficult to get more than a few hundred full steps per second out of a stepper without stalling, even when whacking it with huge drive voltages.  Suppose you microstep, x16 at best case, that gives about say 3kHz your max stepping rate.

You want at least one step per arcsecond of RA (1 second of time is 15 seconds of RA, say 20 for the sake of argument).  So normal tracking rate is one ustep every 50ms or 20Hz. The maximum slew speed you can achieve is 3kHz, or 150 times tracking. So slew rate would be 24hrs/150 or about 10 minutes for a whole RA revolution = 36 degrees/minute.

I think you might find that slew rate a bit slow, and that is if everything works.  On the gearing, tracking rate at 20Hz, that means one turn of the stepper every 48steps*16ustep/20 = 38 seconds assuming 7.5 degree steppers.  The final RA drive is one turn per 24hrs or 86400 seconds, so gear ratio of the entire RA train is 86400/38 = 2,300:1.  DEC axis is basically the same again.

You might have a look at the stepstick stepper driver modules - a pack of 5 for less than 10 quid, based on the Allegro A4988 stepper controller (comes in1 and 2 amp versions).  This will do all the microstepping current control, 20Khz current limiting, and also enables you to switch your stepper with anything up to 35V external supply, while only needing a 5V supply for the controller.  All you need to do is supply direction, pulse, and ustep setting for it to go.  This would take a lot of work off your arduino - I actually built my controllers out of discrete logic using 4988s.

I hope this is helpful - apologies if you clocked this already!

Simon

[skybadger]

++for onstep and stay away from the A4988 drivers, they are not man enough or as well controlled as other step controllers that the onstep is known to work with. The difference being the level of current control they support and the speed they can drive at. 

My vixen clone mount slews at 3 Deg/sec using on step and tda6210 drivers with nema14 steppers . That has the 9 Deg stepper belt driving a 3:1 external  gear and a final 144:1 worm. 

[FrenchyArnaud]

10 hours ago, windjammer said:

Hi 

I have been down this route!  A wonder about your choice of motor drive - I found it very difficult to get more than a few hundred full steps per second out of a stepper without stalling, even when whacking it with huge drive voltages.  Suppose you microstep, x16 at best case, that gives about say 3kHz your max stepping rate.

You want at least one step per arcsecond of RA (1 second of time is 15 seconds of RA, say 20 for the sake of argument).  So normal tracking rate is one ustep every 50ms or 20Hz. The maximum slew speed you can achieve is 3kHz, or 150 times tracking. So slew rate would be 24hrs/150 or about 10 minutes for a whole RA revolution = 36 degrees/minute.

I think you might find that slew rate a bit slow, and that is if everything works.  On the gearing, tracking rate at 20Hz, that means one turn of the stepper every 48steps*16ustep/20 = 38 seconds assuming 7.5 degree steppers.  The final RA drive is one turn per 24hrs or 86400 seconds, so gear ratio of the entire RA train is 86400/38 = 2,300:1.  DEC axis is basically the same again.

You might have a look at the stepstick stepper driver modules - a pack of 5 for less than 10 quid, based on the Allegro A4988 stepper controller (comes in1 and 2 amp versions).  This will do all the microstepping current control, 20Khz current limiting, and also enables you to switch your stepper with anything up to 35V external supply, while only needing a 5V supply for the controller.  All you need to do is supply direction, pulse, and ustep setting for it to go.  This would take a lot of work off your arduino - I actually built my controllers out of discrete logic using 4988s.

I hope this is helpful - apologies if you clocked this already!

Simon

I clocked a lot of this but I am always grateful for people giving their real life feedback and experience   A4988 are in my understanding the real reason why you experienced stalling issues and snail slew rates... The OnStep wiki mentions it specifically and it can even get worse if you use an 8bit controller like me (Arduino Mega2560) 

1 hour ago, skybadger said:

++for onstep and stay away from the A4988 drivers, they are not man enough or as well controlled as other step controllers that the onstep is known to work with. The difference being the level of current control they support and the speed they can drive at. 

My vixen clone mount slews at 3 Deg/sec using on step and tda6210 drivers with nema14 steppers . That has the 9 Deg stepper belt driving a 3:1 external  gear and a final 144:1 worm. 

Yes, after looking in depth into it at the suggestion of @LuckieEddie I decided to take that route because, even though it's not the most user friendly project ever, it is very, very well documented and seems really well developped and mature.  I see no point in reinventing a wheel others have spent years perfecting before me when I don't even really understand the science of circles and disks.  It will be difficult enough without discarding solid bases that are handed to me!

Anyways! 
There has been quite some progress in design since the original post!

So, first, the electronic base : 
I went for Arduino Mega2560 as the main controller, with a Ramps1.4 to support the connexions with no soldering. On that Ramps1.4 I have elected for drivers TMC2209. (The very active support group online made it perfectly clear that there are only 3 valid choices, 2130, 5160 and 2209 ; I originally ordered a pair of 2130, which never arrived ; then discovered that the 2209 are in fact more capable)
The motors are 2 Nema17 with a native 0.9deg per step (400 full steps) at 4.4Newtons. (To be exact they are : JK42HM48-1684-16A and supposedly the most capable ones) 

Then, the mechanics: 
The drive train as it stands now will be 100% belts ; I am designing a 316:1 to 1250:1 gearbox that will totally eliminate the need for a gearworm and give a theoretical resolution of 0.04" and the torque will into the 10s of N. I spare you the maths 🤪  but on paper it is better than an eq6 and could carry 50kg without breaking a sweat even with 15% unbalanced. Both main axis will be marginally overdimensionned (car wheel bearings)  and the whole structure will be assembled through massive bolts and a welded structure of square tubes. 

Right now the electronic part is on the bench, fully functional; if I plug it to the computer, NINA, PHD2 etc all say as a happy choir "Oh! A telescope is connected!". I can slew, track and goto. Of course, I will not be able to check that the pulse guiding fully works before it is actually in service but I don't have much concerns there. 

I am waiting for the delivery of a whole lot of mechanical / car parts and my brand new welding station because guess what, I never touched one of those in my life so I will need to learn this dark art before attempting to build the chassis. 

The bill for now is about £75 for the whole of the electronics and about £100 for the mechanical bits.  My guess is that I will end up well within my £500 estimated budget and if it works as it looks like it should, it will be one hell of a piece of kit! 

Nope, say nothing, let me indulge in my dreams - I KNOW there will be major and unforeseen difficulties  




 

Edited March 28, 2023 by FrenchyArnaud
4.4N, not 44! I wish :D

[windjammer]

I can see that 256 ustepping and 0.9degree steppers is a factor of 150x over my example!  Technology moves on...

The A4988s may be a problem, but I have had losing lock issues with various stepper controller iterations with different chips  over the years, and looking at motor specs it did not seem surprising.  It would be interesting to see how many turns per second you can get out of your steppers in your little test jig ?  I've never managed more than about 4 or 5 turns per second out of any of my stepper collection - mostly motors salvaged from old printers so I can't vouch for the quality.  But you have plenty of room at the bottom with 256/0.9 motors.

Progress is v rapid!  Interesting doing away with a worm - will you have a brake on each axis ?  It would be handy if when you took a belt off the scope did not swing round (massive inbalance was one of you design goals!)

Good luck with the welding - some pics when you have a chance plse

Simon

 

 

[FrenchyArnaud]

12 minutes ago, windjammer said:

It would be interesting to see how many turns per second you can get out of your steppers in your little test jig ?

At 64usteps and 19V (old laptop charger) around 1500rpm (20rps+) But that is obviously the higher limit. They are rated for 3200rpm but the 8bit arduino mega is the weak link - it just can't process the pulses fast enough. To be fair even if i limit it drastically to avoid overheating the motors and the drivers, it does not really matter. Even if the slew takes 5 minutes, I will take precision over speed any time of the day. 

 

18 minutes ago, windjammer said:

Interesting doing away with a worm - will you have a brake on each axis ? 

The mechanics are not designed yet (just general ideas) and even the electronics are still evolving. For instance I am considering ditching the mega2560+ramps1.4 for a  MKS Gen-L v2.1 ; it's technically equivallent but the MKS has a much better onboard clock which would avoid fiddling with a separate rtc. 

My idea is to have a brake directly on each axis, like a massive bolt pushing a friction pad, because it is much easier to design / build than an actual clutch. 

[AstroKeith]

Interesting project.

Have you considered a Pi Pico instead of the Arduino? Much more powerful. Modern bipolar steppers can be stepped at very high rates (your little ones are unipolar and hate micro stepping). There is also the option of micro stepping while tracking and full stepping for slew.

What sort of bearings are you considering? You're seeking high precision so this needs care. I would go for either plain, or tapered needle roller.

To remove backlash in belts, requires very high tensioning, especially true for smaller drive pinions. Your high ratio gearbox will require many belts stages.

You will have read of belt problems with commercial mounts. To achieve low backlash thin flexible belts are often used, but these wear and stretch. Your final output stage belt will need to be very substantial and require very high tensioning.

I've been trying out the new generation of high precision planetary gearboxes that fit Nema motors. I'm seeing typically 15 arcminute backlash at the output of a 50:1 gearbox. Then I use a final roller or belt to deliver a further 30:1 reduction. Thus a final backlash or around 0.5 arc minute.

[vlaiv]

What is your designed total reduction ratio in RA and DEC and what is stepper resolution that you aim for - "/step?

[FrenchyArnaud]

10 hours ago, vlaiv said:

What is your designed total reduction ratio in RA and DEC and what is stepper resolution that you aim for - "/step?

I can't exactly answer these questions until I have finished my tests and mechanical design. The total reduction will depend on the microstepping acceptable (probably 32) and the guiding resolution (both could be adjusted a posteriori by flashing a new firmware, aiming for <0.1") but it looks like it will be in the region of 625:1 total.  

[FrenchyArnaud]

I just confirmed that all slewing/tracking functions are working as planned and the whole gizmo is identified automatically through ascom. Progress 🙂

 

 

[FrenchyArnaud]

12 hours ago, AstroKeith said:

Interesting project.

Have you considered a Pi Pico instead of the Arduino? Much more powerful. Modern bipolar steppers can be stepped at very high rates (your little ones are unipolar and hate micro stepping). There is also the option of micro stepping while tracking and full stepping for slew.

 

No I haven't. I jumped on this project because I loved the ease and efficiency of the DIY focuser based on an arduino nano so I stayed with something I had some familarity with. These steppers can also be stepped at high rates (256ustep >3200RPM) and I would not qualify a nema17 400steps 4.4Nm 2A of "little ones" 😁 The torque behind the gearbox will be more than enough to crush fingers beyond repair 😅
 

12 hours ago, AstroKeith said:

What sort of bearings are you considering? You're seeking high precision so this needs care. I would go for either plain, or tapered needle roller.

...

I've been trying out the new generation of high precision planetary gearboxes that fit Nema motors. I'm seeing typically 15 arcminute backlash at the output of a 50:1 gearbox. 

As it stands, my choice goes to SUV wheel bearing assemblies.  They are not expensive and can take ridiculously high load, even excentric, with pretty much 0 lateral play. If it can take a RAV4 surely it can take a 130PDS

15arcmin is nowhere near acceptable for me and neither is 0.5arcmin in fact. I am designing the whole thing to have an actual resolution inferior to the guiding resolution (say, 20arcsec), if the drive has 1/4 of a degree of play out of the motor shaft that totally defeats the purpose, my CG5GT, which is absolutely beyond battered, does noticeably better than that  


EDIT : I may have not made clear that I am not trying to replicate a EQ5 or 6 : I am not limited by weight, have no requirement for transportability, the whole thing will be fixed on a pier... So I don't care if the axis assemblies are 4 kg each  As long as they don't move, I am good! 

 

 

Edited March 29, 2023 by FrenchyArnaud

[vlaiv]

35 minutes ago, FrenchyArnaud said:

I can't exactly answer these questions until I have finished my tests and mechanical design. The total reduction will depend on the microstepping acceptable (probably 32) and the guiding resolution (both could be adjusted a posteriori by flashing a new firmware, aiming for <0.1") but it looks like it will be in the region of 625:1 total.  

Have you done the calculations on that reduction?

I'm not sure you'll be able to achieve such a fine guide resolution with such a coarse stepper resolution.

With 32 micro steps, ratio needed to achieve 1"/step is 202.5:1 - you are looking to boost that by x3 (with 625:1) - which will only reduce stepping precision to about 0.33"/step.

I'm not sure how this impacts RA precision, since there is mass on the mount and inertia does its thing to smooth out steps, but I'm sure you won't get <0.1" total RMS with this approach because of DEC.

In theory DEC should not move if PA is perfect, but it does move for the host of reasons - PA is not 100% precise, there are influences like wind and mechanical things, and after all, seeing influence will cause apparent DEC motion at some point. All of this means that guiding will from time to time issue correction in DEC - and it can only have 0.33" steps for its position. Any change in position in DEC will instantly raise DEC RMS error above 0.1"  and total RMS can only be large than this.

If you want to have any shot of guiding in 0.2-0.3" range - you'll need step resolution below 0.1"/step - which means reduction of at least 2000:1.

Take a look at mounts that are capable of being guided at 0.2-0.3" RMS range - like Mesu 200. It has around 3000:1 reduction if I'm not mistaken.

Second thing that I'm concerned about your design and that low guiding performance is high ratio mechanical gearing right next to stepper motor.

With mechanical gearing you need to trade off some things. It will introduce some level of backlash and it will have some amount of periodic error.

Depending where that gearing "sits" in complete reduction train - you can trade off backlash for smooth PE and vice verse. If gearing is close to motor - you reduce amount of backlash. It will be same mechanical backlash - but it won't be "amplified" with other stages before it (like using belt before mechanical reduction stage), but this also speeds up periodic error, and fast periodic error is tough to guide out.

Putting mechanical gearing "further down the line" - makes PE much slower and easier to guide out - but of course increases amount of backlash for the same gear set.

Have you considered doing friction gearing instead? Mounts that guide well with small errors use that.

[vlaiv]

Ok, so here is an idea - how you might be able to get sufficient (I'm not claiming you'll hit 0.1" guide RMS) resolution and minimize PE and backlash at the same time.

You'll need multi stage belt reduction - like 3 stages of 5:1 or similar.

Get stepper motor with mechanical reduction in say 1:10 range and get 16bit absolute encoder.

Stepper has 200 steps per turn and with 32 micro steps that is 6400 steps per turn. With 10:1 reduction - it is 64000 steps per turn. 16 bit absolute encoder will deal with PE and backlash to precision of 65536 parts in a circle (so you can even go with 256 micro steps and have accurate position for each 7-8 steps as measured by encoder).

Total reduction with 3 belt stages of 5:1 will be 5x5x5x10 = 1250:1 (so you might even consider using higher belt reduction ratios).

If your first pulley in belt system has 18 teeth - that is 1/18th of a turn for fastest component of periodic error on belt system.

With 1250:1 reduction single "precise" step (or measurement of encoder) will be 0.162"/step, so whole turn of motor / gearbox / encoder will be 10616.832" and 1/18th of that will be ~590". Sidereal is 15.041"/s so that is 39s.

Still too fast to be honest, but better than say belt modded HEQ5 which has 11s period on belt tooth (If I remember correctly) - and that is nightmare to guide out if it starts to "behave".

Anyway - consider using encoders on motor shaft for first few stages to avoid issues of fast PE and backlash.

It looks like backlash with those mechanical reduction systems can be significant.

I've found quotes of "less than 1 degree" of backlash - but that is huge. That is 6400/360 = ~18 steps of backlash - and if you have 0.333"/step - that is 5.92 arc seconds of backlash. Again at sidereal that is 0.3s of backlash (and at lower guide speed - even more).

 

[FrenchyArnaud]

4 minutes ago, vlaiv said:

Have you done the calculations on that reduction?

I'm not sure you'll be able to achieve such a fine guide resolution with such a coarse stepper resolution.

Actually I spent 2 weeks discussing it and calculating it with some OnStep devs...  part of your reasoning is biased.

First, <0.1" total RMS is not on the table. I speak only of <0.1 tracking resolution. With mechanical imperfection, electronic limitations, that hopefully will put me somewhere in the .4" to .8" RMS. 

Second :
The tracking, guiding and slewing res are actually led by different rules for microstepping. So the gearbox is set for, say 500:1 (for the sake of argument) and it happens to be sidereal rate at say, 10RPM (figures made up here)

The microstepping does not come into account : 10rpm is 10rpm. What will come into account is the need for torque (and that is influenced by usteps) 
Now, the guiding is different because it is influenced by microstepping (for resolution), torque (for acceleration) and gear ratio (because it's fixed, 500:1 is 500:1) so if you need more resolution for ultra thin guiding, the ustepping is changed on the fly, say from 16 to 128 (that's 8x thiner) and the the RPM are adjusted too. 

It boils down to : 
there are 1296000 seconds in a revolution.  400steps * 32 = 12800 actuations. 
1296000/12800 = 101.25 sec per actualisation. 
To get at, say, 0.5" per actualisation that is 101.25/0.5 = a 202.5:1 gearbox. It is not that much and defo nowhere near 3000:1. Three stages of belts 12-160 (x13.33) will give much more than that straight away, about 2400:1. Two stages would put you in the ball park actually. 
Taking into account mechanical imperfections and counting on an actual efficiency below 50%, say, 30%! That is still only 600:1 to ensure a guiding <0.5"... even at 32ustep. And the firmware will in fact guide at 64 or 128. 

For reference, the classic combination with these modern drivers and 400-full-step-motors is 32ustep + 360:1 ; that competes with pretty much any sub £2000 mount, even when the 360:1 involves a worm. 

I was a bit confused because I could not see at first how 8us*1440:1 is different from say, 128us*90:1 but yeps, it turns out it is! 

So the microstepping is really not the issue, mostly because modern driver boards are able to change it on the fly, we are not at the glorious epoch of A4988's anymore... The real challenge will be to limit all mechanical imperfections. Other guys did it before me - quite a few actually, and even though I don't have their skills obviously, the precision they achieve with belt gearbox and stepper is beyond anything a mount cheaper than a small car could only dream of. 

[FrenchyArnaud]

6 minutes ago, vlaiv said:

Anyway - consider using encoders on motor shaft for first few stages to avoid issues of fast PE and backlash.

Yes, that is absolutely planned. On both axis, straight of the motor shaft to benefit from the gear ration in terms of precision. The firmware makes the correction in real time to eliminate motor backlash, step slipping and PE as they occur. 

[AstroKeith]

9 hours ago, FrenchyArnaud said:

No I haven't. I jumped on this project because I loved the ease and efficiency of the DIY focuser based on an arduino nano so I stayed with something I had some familarity with. These steppers can also be stepped at high rates (256ustep >3200RPM) and I would not qualify a nema17 400steps 4.4Nm 2A of "little ones" 😁 The torque behind the gearbox will be more than enough to crush fingers beyond repair 😅
 

As it stands, my choice goes to SUV wheel bearing assemblies.  They are not expensive and can take ridiculously high load, even excentric, with pretty much 0 lateral play. If it can take a RAV4 surely it can take a 130PDS

15arcmin is nowhere near acceptable for me and neither is 0.5arcmin in fact. I am designing the whole thing to have an actual resolution inferior to the guiding resolution (say, 20arcsec), if the drive has 1/4 of a degree of play out of the motor shaft that totally defeats the purpose, my CG5GT, which is absolutely beyond battered, does noticeably better than that  


EDIT : I may have not made clear that I am not trying to replicate a EQ5 or 6 : I am not limited by weight, have no requirement for transportability, the whole thing will be fixed on a pier... So I don't care if the axis assemblies are 4 kg each  As long as they don't move, I am good! 

 

 

The 'little' motors I was referring to were the tiny ones in your very first photo. Not the nema17's.

15 arc min is at the motor gearbox output shaft. Divide by 30 for the final drive. So 30 arc sec at the mount shaft. As I said, belts will produce backlash in practice, in my experience of this same order. With your own mount, it is easy to calibrate the backlash and apply a software correction, although it is load dependent to an extent. You might find your CG5GT is applying some.

I dont build EQ mounts, but have designed, built and commissioned about 20 different Dobsonian drives. These range from 10 to 80kg OTA weight. I recently moved from toothed belts to roller drives to help reduce backlash. The final belt drive was producing backlash x2 what the planetary gearbox was. Rollers effectively reduced it to zero.

[vlaiv]

10 hours ago, FrenchyArnaud said:

It boils down to : 
there are 1296000 seconds in a revolution.  400steps * 32 = 12800 actuations. 
1296000/12800 = 101.25 sec per actualisation. 
To get at, say, 0.5" per actualisation that is 101.25/0.5 = a 202.5:1 gearbox. It is not that much and defo nowhere near 3000:1. Three stages of belts 12-160 (x13.33) will give much more than that straight away, about 2400:1. Two stages would put you in the ball park actually. 

0.5"/ustep is really rather coarse - it is comparable to AZGti mount and about double that of EQ5 and quarter of HEQ5/EQ6 mounts

EQ5 has 0.287642"/ustep and

HEQ5/EQ6 have 0.143617"/ustep

All of those mounts have ~700:1 reduction (704:1 and 705:1) and difference is that later two use 64 usteps rather than 32.

With regular 200 degree motor and 32 usteps to get ~0.14"/ustep one would need ~1400:1 reduction.

Now, I misunderstood your aims, I thought that you were aiming for <0.1" guide RMS, but you were talking about <0.1" per ustep, right?

Again, if you want that sort of tracking resolution - you need more than 2000:1 with regular 200 degree motors and 32 usteps.

Even with those sort of reductions - you should be able to slew your mount with reasonable speed if that is your concern. Regular nema 17 steppers can achieve almost a thousand RPM, but let's say we limit it to 300 RPM (most will be able to do that). 300 RPM is 50 revolutions per second or 20ms per revolution. 20000us / 6400usteps = 3.125us/ustep

You should pulse every 3.125us - maybe that is too high for Arduino based board - but why not use RPI Pico?

It has much higher clock and dual core, so it is much more powerful. I can easily get 20us resolution with micropython on it (I'm developing small 3d printed star tracker) - which is more than enough for sidereal. In C - I'm sure I could go sub microsecond in resolution with ease, but like you said - there is no need for micro stepping - once sufficient speed is reached - one can switch to quarter, half or even full steps for fast slews.

With 300 RPM, slew speed will be in above configuration, if we assume 0.1"/ustep tracking resolution - ~8.9 degrees per second?

 

[vlaiv]

10 hours ago, FrenchyArnaud said:

Three stages of belts 12-160 (x13.33)

What type of belts are you planning on using? Regular GT2 belts are probably going to be a bit too weak for larger loads close to "exit" stage?

If you use something more robust, then you'll be looking at rather large pulley? Say you use GT5 - that is 5mm per tooth. With 160 teeth - you'll get 5x160 = 800mm circumference for that.

That is 10" wheel just there, and you'll be needing two of them. I know size is not an issue for you, but just wondering if that is a bit too much?

[FrenchyArnaud]

 Yes, your maths stand - 200 steps at 32ustep would require 2000:1 for the necessary resolution, that I agree with. And yes, I always thought in terms of <0.1arcsec/ustep. If I could guide at 0.5" RMS i'd be over the moon (my pixel scale is currently 1.83" and the ota upgrade i am planning will make it 1.25" so 0.5"rms would be AWESOME)

But my motors have 400 steps (full,  natively, 0.9deg/step, not 1.8 so double the resolution for starters)  and the guiding itself is pulsed at 64 or 128 so you technically only need 1/4 of that, somewhere near 500:1. 

I plan to go a bit more to compensate for motor imprecisions, maybe 650:1 OR 700:1 tops (I can't really say until I have tested the actual power and precision of the motors)

A 160th gt3 (11mm belt) is about 6.5" wide. Yes it's big - not overly so. Also, it was just an exemple to point out that I probably would not need 5 or 6 gearbox stages, I should be able to get away with 2 or 3. I'll consider that when I am at that stage! I am confident that with tunable rollers and absolute encoders I can bring the belt backlash really, really low. 

The real concern is slewing time. I simulated a 1250:1 in my config.h yesterday night and it took almost 4 minutes to slew horizon to horizon... But I suppose compromises will need to be found. 

I am taking on board all remarks made and that will come handy when i do the actual gearbox assembly. 

[vlaiv]

6 minutes ago, FrenchyArnaud said:

The real concern is slewing time. I simulated a 1250:1 in my config.h yesterday night and it took almost 4 minutes to slew horizon to horizon... But I suppose compromises will need to be found. 

Try to switch to full steps after some time - or to lower number of usteps as you speed things up in a slew. That should help, shouldn't it?

If you have issues with micro controller speed and ability to turn outputs on and off - do consider RPI Pico. I don't have much experience with either - arduino nor pico, but it took me like half an hour to get everything connected on breadboard (I'm using DRV8825 with nema 17 and PI Pico) and to write my first piece of code to move the stepper.

Most Arduinos have 16mhz clock speed, while Pico has 133mhz - so there is considerable speed increase just with that.

Btw, I'm building my thing with 3d printed parts, and in process I built very high reduction gear box based on split ring compound planetary gear mechanism. First iteration of design was too big for star tracker and probably over kill, so I instead opted to go for pure belt reduction system - 3 stage, 5:1, 5:1 and 10:1 - that will give me 250:1 or 0.81"/ustep resolution. More than enough for low cost 3d printed star tracker.

I still plan to explore how well that planetary thing behaves, and I'll probably design 3d printed EQ5 class mount around it at some point (it won't be completely 3d printed - it will use some aluminum tubing for the frame and axis).

[windjammer]

On 29/03/2023 at 08:53, FrenchyArnaud said:

At 64usteps and 19V (old laptop charger) around 1500rpm (20rps+) But that is obviously the higher limit. They are rated for 3200rpm but the 8bit arduino mega is the weak link - it just can't process the pulses fast enough. To be fair even if i limit it drastically to avoid overheating the motors and the drivers, it does not really matter. Even if the slew takes 5 minutes, I will take precision over speed any time of the day. 

 

The mechanics are not designed yet (just general ideas) and even the electronics are still evolving. For instance I am considering ditching the mega2560+ramps1.4 for a  MKS Gen-L v2.1 ; it's technically equivallent but the MKS has a much better onboard clock which would avoid fiddling with a separate rtc. 

My idea is to have a brake directly on each axis, like a massive bolt pushing a friction pad, because it is much easier to design / build than an actual clutch. 

Re the brake - my first brake iteration was as you suggest, massive bolt pushing a pressure pad, but it generated some unappealing creaking noises from the bearing blocks when tightened!  So I changed to a floating caliper type brake, much happier.  My particular design has straight cut gears axis to motor so an overtight brake can strip the gearboxes (4 chamber epicyclic 1296:1 - as you say even a feeble stepper can generate huge torques through these ratios).  After a few annoying accidents I bought a 0.05-5Nm torque wrench to get consistent brake pressures and to stay out of the danger zone.  Your belt drive system may avoid this issue, but something to bear in mind!

>>I will take precision over speed any time of the day.

Hehe - in despair I took the opposite approach.  I didn't think I could get more than agricultural precision out of my home workshop, so went ahead hoping that fancy electronics would make up for the shortcomings.  Which is more or less where it ended up - my current bottlenecks are elsewhere than the mount.

BTW, are these the kind of SUV bearing assemblies you are talking about ?

https://www.autodoc.co.uk/ridex/8054987?gshp=1&gclid=EAIaIQobChMIo-nezdqD_gIVPIBQBh38ZAF_EAQYASABEgI-2vD_BwE

They look very interesting indeed.  Makes one wonder if you need any shafts at all.  Just bolt two together at right angles and off you go (I simplify :)).  You must supply some sketches!

Have you any ideas about getting rid of the meridian flip ?  Some GEMs seem to have managed that.  A neat trick if you can pull it off with your about to be acquired welding skills. I think it is done by extending the pier along the RA axis toward the pole and fixing the GEM at the end of the extension.  I suppose a similar effect would be to extend the RA axis well away from the pier and put the DEC axis at the end of that.  Neither of those speak well to stability, but if it could be cracked....

Simon