Jump to content


Photo

LX80 RA tracking

  • Please log in to reply
1339 replies to this topic

#1326 dksolar3294

dksolar3294

    Sputnik

  • -----
  • Posts: 25
  • Joined: 22 Aug 2012

Posted 06 March 2013 - 03:06 AM

After an inauspicious start, I now have it [909 APM] working on the bench.



Andrew, I received the second APM I ordered after I could not get the first one to work properly. After a couple of nights of trying this new one, it appears it works about the same as the first. Meaning, I continue to see eastward (slower) mount movement for both E and W AG calibration commands. I see the same basic result when using the HC serial link instead. PEC makes no difference. My experience with other mounts is that the AG overrides the PEC anyway. When I try to speed up tracking with the custom tracking rate, I don't see any change to the slower tracking tendency, so custom rate changes are questionable. Another item I have noticed is when the mount has been running for awhile by itself and then I press one of the arrow buttons, the mount abruptly jumps at full slew speed for a split second. This sort of negates making fine corrections.

You mention you got it to work on your bench. How so? Have you tried to guide on real stars? Results?

Don

#1327 OzAndrewJ

OzAndrewJ

    Viking 1

  • *****
  • Posts: 951
  • Joined: 30 Nov 2010

Posted 06 March 2013 - 04:56 AM

Gday Don

Re PEC and overrriding guiding etc, that is not the reason i suggested it.
The way the latest code is written, for polar operation, the scope behaves differently when PEC is ON vs OFF, and this can also affect how pulse guiding is done .
Being AltAz invokes a different mechanism again, and even tho altaz pulseguides can be issued now, they still do not appear to work. I have attempted to patch around part of this, but its not ideal as i still cannot make it work in a proper/symmetrical manner, esp if a motor reversal is required to do the guide in a reasonable timeframe.

I cant test any of this against the sky, as i only have an ETX-125 and it isnt really precise for this sort of work.
However, I have a test bench set of LX90 motors and i can actually watch the output gears on the motors themselves, so i can clearly see what really happens at the motor when a command comes in.
( Thats how i know the guide commands are getting applied correctly )
I have since then patched the firmware i run, and hence can log the raw encoder positions over time, and thus can see the effects of guide commands in plots of position vs time.
All this is done with LX90 cards tho, and as noted earlier, i still suspect the LX80 cards "may" behave differently at times.

Andrew Johansen Melbourne Australia

#1328 dksolar3294

dksolar3294

    Sputnik

  • -----
  • Posts: 25
  • Joined: 22 Aug 2012

Posted 06 March 2013 - 02:20 PM

Andrew, I'm trying to figure out if my AG problem is just my mount or something design inherent. Seems we need a larger sampling base to compare with.

If anyone has attempted to autoguide with the LX80, then please share your experience and results.

Don

#1329 OzAndrewJ

OzAndrewJ

    Viking 1

  • *****
  • Posts: 951
  • Joined: 30 Nov 2010

Posted 06 March 2013 - 03:49 PM

Gday Don

Have you tried to "guide" using a simple manual "button box"?
Ie ST4 guiding operates by shorting one of the four lines to ground.
Whilst this occurs, the guide gets asserted.
Rather than use a program to do this in an unknown manner,
if you use a button box to force the commands,
you can watch in the EP to see if the behaviour is the same.
Ie it may be the controlling program is behaving wrongly??

Andrew Johansen Melbourne Australia

#1330 dksolar3294

dksolar3294

    Sputnik

  • -----
  • Posts: 25
  • Joined: 22 Aug 2012

Posted 07 March 2013 - 02:54 AM

Andrew, I do have a set of switches from my first attempts at autoguiding with an LXD600 mount. In Nov2012, I put those on the LX80 first 909 APM and did not get proper movement. I don't remember exactly what the response was other than it didn't work. Just to cover that base, I should try that again on this second APM and note the response for each switch closure. Don

#1331 dksolar3294

dksolar3294

    Sputnik

  • -----
  • Posts: 25
  • Joined: 22 Aug 2012

Posted 08 March 2013 - 02:55 AM

Andrew, Good idea on the switches. I connected a set to 909 APM and then noted visual star movement with a high power cross-hair eyepiece. To my surprise, the N & S contacts showed immediate proper response. However, for the E & W contacts, I was not able to tell the effect. The eastward drift (slower tracking) that I have mentioned before is rather persistent and dominates movement along the RA axis.

I'm beginning to think the tracking speed may be the main problem and not the autoguider. Attempts to speed up the tracking rate with the Custom command were unsuccessful. The LX80 specs says it has 2000 incremental drive rates. I entered multiple rates up to 10000 and saw no effect. One would expect 2000 as an upper limit, but there are no limits to the number entered.

My next move will be to calculate this slower tracking speed. After that, I don't think there is much else I can do.

Don

#1332 OzAndrewJ

OzAndrewJ

    Viking 1

  • *****
  • Posts: 951
  • Joined: 30 Nov 2010

Posted 08 March 2013 - 03:16 PM

Gday Don

To my surprise, the N & S contacts showed immediate proper response



OOOhhh, that surprised me. Normally its N/S that doesnt "appear" to work properly, due to the lag on reversing, however, if you see movement, then it means that its working.

However, for the E & W contacts, I was not able to tell the effect.



That is truly weird, as the E/W merely speeds up/slows down the drive, and it is normally very evident when it happens.
I test by setting the scope up altaz, but telling it its polar.
Then i track along a fenceline ( using a webcam feed ), so i can see N/S moves
as well as speed up/slow downs.
Oohhh , before that
a) What firmware are you running
B) Have you patched it
c) How did you align it

In the new code, ( IIRC ) if you issue an Alt Guide first, you would have to wait 20secs before the system reverts to pure polar tracking.
In non pure polar mode, all bets are off re guiding as the mechanisms change.
Its a truly horrible bit of code used in that scenario.
Also Pre A3S4 uses a totally different mechanism to apply guide commands compared to A3S4.
Sooo we need to clarify whats happening.

I'm beginning to think the tracking speed may be the main problem and not the autoguider.



One of the very first bugs looked at was very slow tracking, but we are talking massively slow tracking here, not just a bit. Again, need to confirm firmware used.

The LX80 specs says it has 2000 incremental drive rates.



Depending on firmware version, it has more than that.
You can adjust the base tracking rate in increments of 0.1% for all firmwares
up to A3S4, at which point it went to 0.01%.
However, that has NOTHING to do with the guiding.
Guiding ( as noted earlier ) can operate 2 ways when pulseguiding.
Pre A3S4, the system reads the guide time supplied, calculates how many encoder ticks this is equivalent to, and then applies this in a "burst". This process is done 10x per second.
In A3S4, the system merely invokes the std guide speed ( ie baserate +/-10arcsec/sec ) and sets a timer.
This is also done 10x per second, but the timer always overruns a bit, hence if the ST4 line is asserted full time, the timer just resets before a stop is required and you keep guiding.
Thus it is smoother than the old ST4 method, but it also has new problems of its own.
Either way tho, you should see the equivalent of a baserate +/- 10arcsec per sec guide
ie any errors in base tracking really shouldnt affect you seeing the guides.

My next move will be to calculate this slower tracking speed.



Yes please :jump:
Also, if yr game, you can apply the latest patch i have put out.
There is a special option in there to forcibly reset the motor speeds when polar.
This should provide a self repairing system if in fact the LX80 cards are behaving differently ( as i still suspect )

Andrew Johansen Melbourne Australia

#1333 dksolar3294

dksolar3294

    Sputnik

  • -----
  • Posts: 25
  • Joined: 22 Aug 2012

Posted 02 April 2013 - 01:11 AM

Andrew, I apologize for the delay in responding back. I didn't mean to leave you hanging this long. I was gone week before last and have been in catch-up mode this week. This was Spring Break time for kids up here, plus it's also income tax calculating time.

I've had several exchanges with Meade tech. I figure they needed to know how my LX80 is incapable of autoguiding. Other than that, I haven't done anything else with it.

To answer a few of your ?s, I do have A3S4 firmware. I haven't tried an of your patches yet, so that looks likely the next step. The slower drift rate I have seen in polar mode is about 30arcsec/minute of time. Sometimes (like after first power up), the mount actually tracks pretty good. Then something (like AG signals) triggers it into the slower track rate - sort of like a subroutine.

I doubt if I will spend much more time on the LX80. I've got two other imaging mounts to occupy my limited astro time. Don

#1334 OzAndrewJ

OzAndrewJ

    Viking 1

  • *****
  • Posts: 951
  • Joined: 30 Nov 2010

Posted 02 April 2013 - 01:39 AM

Gday Don

If you do try the patch, you need to leave the "fix"
[Fix Pulseguiding] selected
and also select the "enhanced"
[Always Use Pure Polar Tracking]
the latter part will forcibly reset the RA tracking every few seconds.

Andrew Johansen Melbourne Australia

#1335 DuiA1

DuiA1

    Vostok 1

  • -----
  • Posts: 116
  • Joined: 07 May 2012
  • Loc: Ontario, Canada

Posted 02 April 2013 - 11:41 AM

Has Meade conceded that their mount will not autoguide? When I asked Meade tech support they said it works. Can the mount autoguide without the APM module?

#1336 OzAndrewJ

OzAndrewJ

    Viking 1

  • *****
  • Posts: 951
  • Joined: 30 Nov 2010

Posted 02 April 2013 - 03:22 PM

Gday Don

Has Meade conceded that their mount will not autoguide?



No idea.
I know that unpatched, the AltAz "pulseguiding" wont work,
( it just rubberbands back to start )
but in Polar, my bench testing showed that the new 909 was returning the correct data, and it was being sent to the scope correctly.
I have no real LX80 cards to test with tho,
to see if they read/use it correctly.

Can the mount autoguide without the APM module?


The system should be able to pulseguide using the std serial commands sent to the Hbx, ( vs APM909 into Aux port ), however, both methods use the same mechanism once in the Hbx, so if the ST4 isnt working due to a cards problem, that will also affect the serial commands.
Try it and see ( there is a special dialog box in my PEC editor to allow you to test pulseguiding via serial, so see what that shows.

Andrew Johansen Melbourne Australia

#1337 rmollise

rmollise

    Hubble

  • *****
  • Posts: 15404
  • Joined: 06 Jul 2007

Posted 02 April 2013 - 05:29 PM

I'm not overly surprised that pulse guiding doesn't work in alt-az mode. The question I have:

Will regular serial guiding through the serial port work in polar mode?

#1338 OzAndrewJ

OzAndrewJ

    Viking 1

  • *****
  • Posts: 951
  • Joined: 30 Nov 2010

Posted 02 April 2013 - 07:16 PM

Gday Rod

I'm not overly surprised that pulse guiding doesn't work in alt-az mode.


I am sort of.
Alt/Az "pulseguiding" cannot reverse, it is only able to stop the motors and let drift account for the movement, so in certain areas of tracking it wont work "quickly" if a slowdown of the motor is reqd to do the move. That said, the way Meades code works for "general" tracking requires a "modified" target position to be set at the same time the guide is sent. They dont do this for Alt/Az, hence it moves then rubberbands back.
I have patched it to work as best as is possible, but its still an asymmetric function.

Will regular serial guiding through the serial port work in polar mode?



As far as i can tell, all three means of automated guiding should work in polar ( ie ST4 pulseguide, :Mg pulseguide or :M std move )
That said, in polar, if PEC is OFF or an RA pulseguide hasnt been sent within the last 30seconds, the system reverts to AltAz "mode", ( even if you are polar ), and results can get real "woofy" real quick.

Andrew Johansen Melbourne Australia

#1339 dksolar3294

dksolar3294

    Sputnik

  • -----
  • Posts: 25
  • Joined: 22 Aug 2012

Posted 03 April 2013 - 01:36 AM

Rod,

Will regular serial guiding through the serial port work in polar mode?



I have not been able to AG with either serial or APM. If someone does successfully autoguide, please post details of how you did it and a resulting astro image. Don

#1340 DuiA1

DuiA1

    Vostok 1

  • -----
  • Posts: 116
  • Joined: 07 May 2012
  • Loc: Ontario, Canada

Posted 09 August 2013 - 02:51 PM

I'm considering purchasing the Orion Starshoot autoguider. I have the lx80 autoguide module. Has anyone got this to work with the lx80? Want to confirm before buying autoguider. It's been since April with no update on this question...






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics