Since I aquired an Losmandy G11 mount without any goto provision I decided to create my own solution. I was a bit reluctant to use a Gemini, mainly because of the cost, but also due to the reliability since the servo motors are somewhat prone to failure. I also like to have complete control over the software/firmware so that no "show stopper" issues will ever be a problem for me. Looking around the 'net I saw other goto systems that seemed capable, but either they had needlessly complex hardware or were commercial systems that cost about twice what they should.
OnStep is a computerized goto system, usually for stepper motor equipped mounts though any step/dir interface motor driver (including servo) should work. It was designed, from the beginning, as a more or less general purpose system and provisions were made in the firmware to allow for use on a variety of mounts including Equatorial and Alt/Az (GEM, Fork, Dobsonian, etc.) It uses an LX200 like computer command set with a few extensions to suit hand controller-less operation.
There's a telescope hand controller App for Android (free in the Google Play Store,) an option to control through a website (Smart Web Server,) and a dedicated physical hand controller (Smart Hand Controller.) Or, on a PC there's ASCOM and INDI drivers for control. These options allow you to setup and control OnStep using a wide range of software including my Sky Planetarium, Cartes du Ciel, Stellarium, SkySafari, KStars, PHD2, etc. You can see most of the control possibilities in the diagram below:
Also, please visit my StellarJourney.com site for updates, links to OnStep resources, my other astronomy software, or to see some Images taken with the assistance of my OnStep operated mounts.
If you find OnStep useful and would like to show your appreciation and support please consider making a small monetary donation. This keeps me motivated and helps cover expenses associated with my projects.
|
Re: FYsetc S6 GPS
Gerry Byrne
My GPS does not appear to function and I note Khalid's comment here about defining the GPS serial port as #define TIME_LOCATION_SOURCE GPS #define SERIAL_GPS Serial3 #define SERIAL_GPS_BAUD 9600 And wired it to PC10 and PC11 on EXP1 Checking my setup I find that my Config h defines the serial_GPS as Serial 2 but that I had been connecting to EXP1 as Khalid mentions above. Is it possible to correct this without re-flashing the board again? For example by connecting elsewhere? Gerry
On Fri, 14 Jan 2022 at 19:11, Chris Whitener <chris.whitener@...> wrote: Kahlid, Did you move the GPS from RX3/TX3 to the softserial pins? --
Copernicus
|
|
|
|
Re: Controller Time Drift
I believe that was typo.
On Fri, Jan 14, 2022 at 03:52 PM, <planetibo@...> wrote: Here is how to finely adjust tracking rate (spoiler: not by editing steps per degree)
|
|
|
|
Re: How to double increase torque with silent stepper driver LV8729
#bluepill
#configuration
John Petterson
A couple of thoughts.
|
|
|
|
Re: Controller Time Drift
John Petterson
On Fri, Jan 14, 2022 at 10:09 AM, Khalid Baheyeldin wrote:
OnStep does need a PPS source unless the board's clock is inaccurate.You might want to read that sentence carefully....
|
|
|
|
Re: Question on Torque, Currents, steppers, voltages, CONFIG settings
Guy...
Do a Google search on Vexta Clean Damper for information on how the "flywheel" actually works. DO NOT get isolation dampers that mount the stepper on rubber gaskets. You need the shaft mounted Dampers. There are many listed on eBay but here is one from the Oriental website for use on a 1/4 inch motor shaft: https://catalog.orientalmotor.com/item/all-categories-components/ac-motor-accessories/d6cl-6-3f Same thing on eBay tinyurl.com/6w7vm6dc And a video: https://www.youtube.com/watch?v=S3m5fEwg9K4 Notice the Torque on this motor: tinyurl.com/3rsdb8wy
|
|
|
|
Re: New User - Encoders
John Petterson
Barry, I am one of the few people that implemented encoders on my mounts. I have a couple of Losmandy mounts with Astro Devices 311,296 step encoders. I am driving them through a Teensy 4.0 that is also my wired Ethernet connection. I originally was using a Teensy 3.2 for that internet and encoder interface, but had to make the 4.0 work because the 3.2 could not keep up with the pulses during a slew. I have some serious doubts if any of these processors will track properly with a 10 million tick encoder (about 30x as many pulses as I see) while also running either Wifi or Ethernet links. Should not be an issue at tracking rates, but if you connect these be certain to test how your slews work carefully. When I was using the Teensy 3.2 the OnStep would think it has moved about 1/2 as far as it actually moved.
|
|
|
|
Re: Question on Torque, Currents, steppers, voltages, CONFIG settings
John Petterson
Guy,
I found a web page with several of those steppers pictured on it. If that is a data, it has to be in the format YY/MM/DD as one of the pictured motors has a date of 10/12/28 and another one is 13/09/11. So yours is probably 7 years old. John
|
|
|
|
Re: Started new OnStep hardware design
John Petterson
I get it - you want to use something you are familiar with and comfortable with and you are not afraid of designing a board for this. It will be interesting to see how that set of parts gets mounted on the telescope mount. Ali Express has the TMC5160s listed at $30 for 2 of them (link below). John
|
|
|
|
Re: Question on Torque, Currents, steppers, voltages, CONFIG settings
I am in communication with someone that has first-hand experience with Ealing mounts.
His comments are that the mount could be ridden by someone and still be able to slew. Guy has stated that the mount NEVER was able to slew and that it was PUSH-TO only. There appears to be a difference in what the facts are so I am hoping to find some way to get down to the basic situation. I'm told that the Ealing mount has a "Clutch knob" at the SOUTH end of the RA Axis but NO Movable weights on that AXIS. ROUGH Balance was the best you could do by installing or removing heavy weights inside the Dec Counter Weight casting. Supposedly the rough balance was good enough since the Drive apparently had plenty of torque to move mountains. He mentioned something I had forgotten and that some steppers need a small weight added to the shaft (flywheel) to keep kinetic energy up to the point where the flywheel can help the motor avoid resonance and stalling. We are going to get this working eventually. My suggestion at this time would be to investigate adding a small flywheel to the motor shaft, extending the ramp time and since you have a 24V power source, give that a try but the motor current is defined by the drivers so I would not expect that the increased voltage would have as much effect as a flywheel on the stepper shaft. I would also disconnect the RA Drive by releasing the clutch to verify the motor can at least drive the gearing up to the big worm gear on the RA axis.
|
|
|
|
Re: Started new OnStep hardware design
skyguynca
The 6 to 15 deg/minute is only a ruff estimate using nema 17 steppers which are limited by the need for more torque (low rpms and torque small stepper need alot of reduction) to start and the mass of mount and scope. The nema 23 steppers i will be using can turn 2500 rpms at 64 microsteps, with the right ratio I should get slewing rates in excess of 120 deg/min with enty of torque to spare.
David Mikesell San Jose, CA
|
|
|
|
Re: Started new OnStep hardware design
John Petterson
On Fri, Jan 14, 2022 at 06:18 PM, skyguynca wrote:
I really can not see a problem on the driver/stepper end of it running a telescope mount in tracking (which I believe is .04 deg/min) to slewing at 6 to 15 deg/s a minute.This is basically the problem. At 15 degrees/minute, it will take 12 minutes to move 180 degrees. Are you ready to wait that long for it to move from the North American nebula to the Orion Nebula?
|
|
|
|
Re: Question on Torque, Currents, steppers, voltages, CONFIG settings
Prasad
Ken Guy is 100% on the mark. I was there when Guy removed all the balancing weights on the RA axis. The DEC axis has no load at all. In this condition, I have tried to turn the worm that drives the RA axis by hand. It was very free in both directions (remember 1:359 ratio) but it did get a bit tight at certain positions. This was due to mechanical binding between the worm and worm wheel. Then we loosened the fasteners that hold the worm shaft and the movement became really free in both directions when turned by hand. (But created some backlash which did not matter for testing the motor). It was very cold that day and we were late. So we wrapped up without completing the adjustment. So, as long as the worm shaft is loose in its engagement with its mating worm wheel (gear) it should be OK. The penny is all that is needed at the moment. Of course, adding a good cold temp lube would help additionally, maybe half a penny would work. Guy has made progress since then because he has coupled the second stage worm shaft in its place during his recent visit. Guy, I hope I am right in my understanding/interpretation. Please correct as necessary. Thanks -Prasad
On Friday, January 14, 2022, 11:20:01 AM EST, Guy Brandenburg via groups.io <gfbrandenburg@...> wrote:
Hi, Ken, Right now, we have taken absolutely everything off of the mount, except for some bolts: all four OTAs, all of the finders, all of the counterweights, the large steel plate all of that was bolted to, and even the cover for the Ealing counterweight box. When the mount was last operational, we balanced everything on both axes and in all positions by adding or subtracting counterweights of various sizes and compositions, in various locations, and also by sliding some of them along special grooved racks we fabricated. And loosening and sliding and reclamping the OTAs as needed. If we attached any imaging gear, we re-balanced. (Or at least I did. I hope other users did so as well.) Right now, during very preliminary trials, it’s not balanced at all in RA because the empty CW box sits on one side. The DEC axis is balanced because the six remaining bolts are evenly spaced. My next stop will be to fabricate and rig up an counterweight of some sort to balance out the empty Ealing counterweight box. Maybe this weekend. I wish it were not 50 miles away. In raising a big weight with a penny, it’s like Archimedes supposedly said: all you need is a long enough (weightless) lever, or (I would add) the proper gear set, and almost no friction…. , which capitalizes Weirdly & misinterprets words just To keep you on your toes
|
|
|
|
Re: Started new OnStep hardware design
skyguynca
External stepper drivers and paired with the right stepper has such a huge speed range. A good example is a nema 34 motor and driver on my router table. I can run at Cutting speeds as slow as .125 inches per minute just as smoothly and accurately as cutting at 6 inches a minute. That is 48x the speed change, well with in the limits of my steppers and drivers. My gantry weighs 328 lbs, when not cutting I can run it at .065 inches a minute without stalling and max rapid speed of 150 inches a minute, again no stalling.
I really can not see a problem on the driver/stepper end of it running a telescope mount in tracking (which I believe is .04 deg/min) to slewing at 6 to 15 deg/s a minute. If there is a problem I believe it would be needing to adjust the coding to allow such speeds and in the physical limitations of the scope and mirror mountings. David Mikesell San Jose, CA
|
|
|
|
Re: Question on Torque, Currents, steppers, voltages, CONFIG settings
Robert Benward
Guy,
|
|
|
|
wSHC
Glenn Sammes
Hi Guys,
I am after some guidance re: wireless SHC. I have bread boarded 2 of them. Intend to make final on VERO (strip board). CPU is ESP32…ESP-WR00m-32 or node MCU-32s Currently using 1.3” displays. Blue pill using STM32F303 transplant. OnStep working well. Latest firmware.
The thing is the wSHC do connect and work OK. THE PROBLEM is they can take 3 to 6 minutes to connect. Everything is 57600 baud.
Starting WIFI :ONSTEP then Restart WIFI: Failed repeated on the display until the connection is made. Android app, SWS or Stellarium are almost instant.
Also… there is in the firmware settings for an analog joystick and a pin map to go with it. Anyone got a hardware schematic to match?
Cheers, Glenn.
|
|
|
|
Re: Started new OnStep hardware design
John Petterson
One issue I see is that these external drivers do not allow changing the microstep rate dynamically for slewing vs tracking, so your slew rate will be limited. I looked at them a year ago when I was starting to build my controllers, but decided against them quickly for that reason. They may handle higher speeds better because of the higher current ratings, but it is still a risk.
For A/P it may not be an issue, but for a visual scope that does not spend hours on one target it could get to be a pain.
|
|
|
|
Re: Question on Torque, Currents, steppers, voltages, CONFIG settings
Robert Benward
Guy,
If you double the voltage, the current will rise to twice the value in a given time. If you terminate the pulse at a certain current, you will reach that time twice as fast (which means more RPM). The higher the voltage, the better the performance at high RPMs & torques. The voltage does makes a significant difference. Use 24V. In your case, you might even go higher. You most likely do not need that regulator near the motor connectors. Unless you are using drivers that can't handle the voltage, just jump out the regulator pin1 to pin3. While playing with slewing, a few things can happen, you are going too fast for the motor, you are not allowing the current to rise high enough before you terminate the cycle. You can also stall the motor (which you did with your fingers), the motor loses step, and you are finished. The motor needs a ramp up time, it can't go from zero to 500RPM in an instant, you need to ramp it up slowly. Once you skip a step, the motor will most likely stall and you need to go to zero again and start over. So, for starters, keep you slew rate in check, start slow. In config.h, this is listed as acceleration time. Missing the target current can cause a fault, a possible cause while ramping up and then stalling (I think you are using TMC5160). Setting a target current too high will cause this. Start low in both current and slew rate, and go back and forth between the current and the slew speed increasing each a little at a time. Eventually you will reach the maximum RPM you are going to get out of the combination of motor and voltage. Look at the curves at the bottom, the faster you go, the less torque you can deliver. At speed, you may only have 10-20% of the holding torque. The bottom graph shows the starting RPM for a given torque and no load torque, and the then the maximum torque at speed. Bob  Once you are through the time constant, the current is simply V/R (note, you may never get there at high RPMs).   
|
|
|
|
Re: Started new OnStep hardware design
On Fri, Jan 14, 2022 at 06:10 PM, skyguynca wrote:
You save time soldering and cost of components with board made for running external drivers. Using my MaxSTM is overkill. Realistically almost half of the board and components are used with just picking up the 3 pins for each driver. I know i will only save about $20 on single electrical components, but as far as drivers and steppers it breaks even.I still don't see why you need a new board at all. Using external drivers has been a feature of OnStep for quite some time now. Again, you can do all of what you said with a normal OnStep board. Just don't buy the LV8729 or TMC5160. In lieu of them, do the following: Use a 10+10-pin adapter board which plugs in the socket of axis1 and axis2 drivers. Don'tsolder any capacitors that are specific to the drivers at all. So you save on the cost of components and time to solder them. Note that the MaxSTM requires an EEPROM chip, otherwise OnStep will not run at all.
|
|
|
|
Re: Started new OnStep hardware design
skyguynca
Yes external drivers, no not like the one you linked. That is a servo driver, anything servo either ac or dc servo is a bit more expensive.
LV8729 - Amazon $7, Ebay $8. Limited to 0.8 amps cont, peak 1.5ams TMC2130 - Amazon only listing 2208 and Ebay $11 each Limited to 1.4amps cont, peak 2amps TB6600 stepper driver Amazon $11, It will run Nema 17 and 23 Stepper Motors, 9 to 40VDC and 3.5amps cont 4amps Max Amazon Here is a stepper driver combo, 4amp driver and nema 17 motor $20 Amazon driver and motor If you shop around even better deals to be had on ebay or ali express. I have used steppers both open and closed loop on cnc machines from vertical mills, router tables, lathes and plasma cutters. I can get them within .001 repeatable accuracy. Perfect smooth lines and cuts. Accuracy and smoothness that the telescope mount will need in tracking and goto. Micro steps from 1/4 all the way down to at least 32 is available on most drivers. The ones I have about about $22 apiece and go all the way down to 256 micro steps. You save time soldering and cost of components with board made for running external drivers. Using my MaxSTM is overkill. Realistically almost half of the board and components are used with just picking up the 3 pins for each driver. I know i will only save about $20 on single electrical components, but as far as drivers and steppers it breaks even. The biggest saving is in time. Probably at my speed I will save about 4 to 6 hours in setup and soldering time not having to install about half of the sockets, caps, resistors and sockets that are not going to be used by using external steppers. When I am done with the new board and have one sitting here. I will post real time assembly and list the total cost and BOM for a direct comparison. David Mikesell San Jose, CA
|
|
|
|
Re: Started new OnStep hardware design
Robert Benward
David,
There seems to be a shortage of the TMC5160. I paid $17 middle of last year for a Bigtreetech TMC5160 V1.2. I believe they are up to V1.3 now, and I do see them from $30 to $47. Khalid mentioned external drivers, I bought one of these on Amazon for $10. Generic Step & Dir controller $10: https://www.amazon.com/Twotrees-Stepper-Driver-TB6600-Controller/dp/B07PQ5KNKR?pd_rd_w=BlZwm&pf_rd_p=ee521540-07c2-4687-9605-13c98e32ab2c&pf_rd_r=VWGZ0KXPHFRVY92572PG&pd_rd_r=96a98394-7544-4851-ab01-e5492cd422a2&pd_rd_wg=Ma8mi&pd_rd_i=B07PQ5KNKR&psc=1&ref_=pd_bap_d_rp_1_t TMC5160 V1.2 Ebay, $58 for two pcs. https://www.ebay.com/itm/393547723888?hash=item5ba145b470:g:-4YAAOSwucphMQVj Bob
|
|
|