Topics

Onstep 4.23 beta problem


How Eee
 

Please help me to solve. I made a short video with description of the issue.
https://youtu.be/P_wAmQfabu8
Thanks in advance. 


Howard Dutton
 

I did an approximate test similar to yours.

  • Mega2560.
  • Ramps1.4.
    • No stepper drivers installed (but that is 100% irrelevant to OnStep since it can't tell if they are there or not.)
  • WeMos D1 Mini.
    • No changes to its configuration, just uploaded.
  • Latest OnStep Beta branch.
    • Configured for the Ramps14 pinmap and A4988's in 16x mode on Axis1 and Axis2, no other changes from default.

Same steps in the Android App and the above worked fine here.

Could be hardware, seems like the Mega2560 is crashing.  Power supply?  You could hook up USB and see if the connection drops out there too, seems very likely that it does.

Less likely your different configuration might be triggering it, post the Config.h and I'll try it here to verify that it works.


Howard Dutton
 

On Sat, Feb 20, 2021 at 04:57 AM, Howard Dutton wrote:
Could be hardware, seems like the Mega2560 is crashing.
Be aware that usually "start align" is where the stepper drivers are enabled hence a larger load on the power supply.  If you pull the drivers or disconnect the motors and the problem goes away, this is the answer.


Howard Dutton
 

On Sat, Feb 20, 2021 at 05:33 AM, Howard Dutton wrote:
If you pull the drivers or disconnect the motors
Powered off while doing so, obviously.


How Eee
 

Thanks for reply. 
Stepper motors were disconnected during test. Not a power issue. 
Standard config. Just changed steps per degrees, altazm, ramps1.4. 



сб, 20 февр. 2021 г., 18:34 Howard Dutton <hjd1964@...>:

On Sat, Feb 20, 2021 at 05:33 AM, Howard Dutton wrote:
If you pull the drivers or disconnect the motors
Powered off while doing so, obviously.


How Eee
 

Here config is.

Just tried last release 3.16. Everything works as it should. No issue. But still no spiral search there. 

сб, 20 февр. 2021 г., 18:40 How Eee via groups.io <how.eee.s=gmail.com@groups.io>:

Thanks for reply. 
Stepper motors were disconnected during test. Not a power issue. 
Standard config. Just changed steps per degrees, altazm, ramps1.4. 



сб, 20 февр. 2021 г., 18:34 Howard Dutton <hjd1964@...>:
On Sat, Feb 20, 2021 at 05:33 AM, Howard Dutton wrote:
If you pull the drivers or disconnect the motors
Powered off while doing so, obviously.


Keith Trivett
 

For the record I'm getting the same problem with a Stm32 clone. It appears to be after setting the site , if I then go to start an alignment or even just toutch the "i" on the info window It looses comms or crashes.


On Sat, 20 Feb 2021, 13:40 How Eee, <how.eee.s@...> wrote:
Thanks for reply. 
Stepper motors were disconnected during test. Not a power issue. 
Standard config. Just changed steps per degrees, altazm, ramps1.4. 



сб, 20 февр. 2021 г., 18:34 Howard Dutton <hjd1964@...>:
On Sat, Feb 20, 2021 at 05:33 AM, Howard Dutton wrote:
If you pull the drivers or disconnect the motors
Powered off while doing so, obviously.


Ken Hunter
 

Howard...

Time for some good news?

RE: U8g2 with the new beta and U8g2 Ver 2.28.10

The new beta DOES Compile and work well as long as the IDE is clean.
From reading all the error messages it is looking like the IDE compile
tries to reconcile previous successful compiles with the compile ongoing.
With my flaky directory structure the IDE was getting compile errors.
The new clean load of the Arduino IDE works a charm. THANKS

Ken


Ken Hunter
 

Forgot to mention that I have not proceeded to the current thread issue, only
the errors presented with the U8g2 issue.

Ken


Howard Dutton
 
Edited

On Sat, Feb 20, 2021 at 05:44 AM, How Eee wrote:
Just tried last release 3.16. Everything works as it should. No issue.
I was able to recreate the issue using your Config.h.  Took a few hours to track it down, the Mega2560 was crashing as the timer rates were too high, numeric overflow... has to do with being ALTAZM and having different steps per degree for Axis1 vs. Axis2.  Checks were added and the latest Beta worked for me with your Config.h now.

But still no spiral search there.
Some have reported trouble with that feature anyway, hope it works...
I'll remove the feature from v4/5 over fixing it, if it presents a problem, as I'm working on OnStepX and would rather not be distracted.  I'll revisit that later.


Drew 🔭📷🚴‍♂️
 

On Sat, Feb 20, 2021 at 04:13 PM, Howard Dutton wrote:
I'll remove the feature from v4/5 over fixing it
Howard, I don't think there is any problem with the Spiral Search that would justify removing it. My complaints are just that, annoyances and complaints.


Drew 🔭📷🚴‍♂️
 

I should add that most people who report that it just doesn't work have not moved the telescope into the "operable" zone.

The telescope must be outside of the Polar limits in particular before the Spiral Search is operable. Since there is no error message or beep (unless you are looking at the Webserver, etc.) I myself (and I am sure many others) would just assume it does not work.

Upon learning this I just started moving the scope to a star somewhere off of the NCP before trying to start a Spiral Search.


Khalid Baheyeldin
 

I used spiral search several times, and it works.
At the start of a session, I use it to get to the first star so I can focus using a Bahtinov mask.
No complaints about it, although if I forget to set the speed to fast enough, and my alignment
is way off, it takes a while to find the star. But that is not a problem with the design.

Please keep it.


How Eee
 

Sometimes my mega starts on 4.23 without issues. Somehow. And then spiral search works great. I didn't try it with a real telescope but in skysafari everything looks great.

Its good that you found the reason of the problem. I will try to use another step per degree numbers for bigger microsteps. 

вс, 21 февр. 2021 г., 2:33 Khalid Baheyeldin <kbahey@...>:

I used spiral search several times, and it works.
At the start of a session, I use it to get to the first star so I can focus using a Bahtinov mask.
No complaints about it, although if I forget to set the speed to fast enough, and my alignment
is way off, it takes a while to find the star. But that is not a problem with the design.

Please keep it.


Mike Ahner
 

On Sat, Feb 20, 2021 at 11:14 PM, How Eee wrote:
Its good that you found the reason of the problem. I will try to use another step per degree numbers for bigger microsteps.
Hi How Eee, reading what Howard wrote, he fixed the problem and it was working for him using the latest Beta and your config.h file. So there should not be a reason for you to change your settings, unless you need/want to do so.

-Mike


How Eee
 

Ok. Thanks a lot.
It works now. Stable connection. Spiral search is good. 

But. 

There is another problem. Wrong position at startup or reset at home. It should start at NCP. But it starts Alt0.00 Az0.00.
In version 2.22 and 3.16  Alt41.5 Az0.00 which is my latitude 41.5
With GEM in config it starts at NCP correctly. In ALTAZM it starts pointed at horizon 

вс, 21 февр. 2021 г., 10:58 Mike Ahner <mahner@...>:

On Sat, Feb 20, 2021 at 11:14 PM, How Eee wrote:
Its good that you found the reason of the problem. I will try to use another step per degree numbers for bigger microsteps.
Hi How Eee, reading what Howard wrote, he fixed the problem and it was working for him using the latest Beta and your config.h file. So there should not be a reason for you to change your settings, unless you need/want to do so.

-Mike


How Eee
 

It seems to me that I found mistake in
Astro.ino 4.23
String 330

// the polar home position
#if MOUNT_TYPE == ALTAZM
  homePositionAxis2=AXIS2_HOME_DEFAULT;
#else
  if (latitude < 0) homePositionAxis2=-AXIS2_HOME_DEFAULT; else homePositionAxis2=AXIS2_HOME_DEFAULT;
#endif

Astro.ino v3.16
String 313

// the polar home position
#if MOUNT_TYPE == ALTAZM
  homePositionAxis2=fabs(latitude);
  if (latitude < 0) homePositionAxis1=180.0; else homePositionAxis1=0.0;
#else
  if (latitude < 0) homePositionAxis2=-90.0; else homePositionAxis2=90.0;
#endif


вс, 21 февр. 2021 г., 14:24 How Eee via groups.io <how.eee.s=gmail.com@groups.io>:

Ok. Thanks a lot.
It works now. Stable connection. Spiral search is good. 

But. 

There is another problem. Wrong position at startup or reset at home. It should start at NCP. But it starts Alt0.00 Az0.00.
In version 2.22 and 3.16  Alt41.5 Az0.00 which is my latitude 41.5
With GEM in config it starts at NCP correctly. In ALTAZM it starts pointed at horizon 

вс, 21 февр. 2021 г., 10:58 Mike Ahner <mahner@...>:
On Sat, Feb 20, 2021 at 11:14 PM, How Eee wrote:
Its good that you found the reason of the problem. I will try to use another step per degree numbers for bigger microsteps.
Hi How Eee, reading what Howard wrote, he fixed the problem and it was working for him using the latest Beta and your config.h file. So there should not be a reason for you to change your settings, unless you need/want to do so.

-Mike


How Eee
 

Have just tested it. Seems to work well. 

вс, 21 февр. 2021 г., 16:08 How Eee via groups.io <how.eee.s=gmail.com@groups.io>:

It seems to me that I found mistake in
Astro.ino 4.23
String 330

// the polar home position
#if MOUNT_TYPE == ALTAZM
  homePositionAxis2=AXIS2_HOME_DEFAULT;
#else
  if (latitude < 0) homePositionAxis2=-AXIS2_HOME_DEFAULT; else homePositionAxis2=AXIS2_HOME_DEFAULT;
#endif

Astro.ino v3.16
String 313

// the polar home position
#if MOUNT_TYPE == ALTAZM
  homePositionAxis2=fabs(latitude);
  if (latitude < 0) homePositionAxis1=180.0; else homePositionAxis1=0.0;
#else
  if (latitude < 0) homePositionAxis2=-90.0; else homePositionAxis2=90.0;
#endif


вс, 21 февр. 2021 г., 14:24 How Eee via groups.io <how.eee.s=gmail.com@groups.io>:
Ok. Thanks a lot.
It works now. Stable connection. Spiral search is good. 

But. 

There is another problem. Wrong position at startup or reset at home. It should start at NCP. But it starts Alt0.00 Az0.00.
In version 2.22 and 3.16  Alt41.5 Az0.00 which is my latitude 41.5
With GEM in config it starts at NCP correctly. In ALTAZM it starts pointed at horizon 

вс, 21 февр. 2021 г., 10:58 Mike Ahner <mahner@...>:
On Sat, Feb 20, 2021 at 11:14 PM, How Eee wrote:
Its good that you found the reason of the problem. I will try to use another step per degree numbers for bigger microsteps.
Hi How Eee, reading what Howard wrote, he fixed the problem and it was working for him using the latest Beta and your config.h file. So there should not be a reason for you to change your settings, unless you need/want to do so.

-Mike


Howard Dutton
 
Edited

On Sun, Feb 21, 2021 at 03:05 AM, How Eee wrote:
It seems to me that I found mistake in
That is intentional, not a bug.


How Eee
 

Maybe you left it intentional. I don't know all algorithms behind that, but which way should I align telescope after assembly in field? I used to point it to polar then start onstep. 

вс, 21 февр. 2021 г., 16:34 Howard Dutton <hjd1964@...>:

On Sun, Feb 21, 2021 at 03:05 AM, How Eee wrote:
It seems to me that I found mistake in
That is intentional.