Jump to content


Photo

Wanted: Willing (LX850) Crash Test dummies

  • Please log in to reply
13 replies to this topic

#1 OzAndrewJ

OzAndrewJ

    Viking 1

  • *****
  • Posts: 968
  • Joined: 30 Nov 2010

Posted 11 April 2013 - 01:08 AM

Gday All

As part of my decoding of some of the modified functions in the LX850 code, i had noted that the GPS functionality had changed a bit.
Just for fun ( as its cloudy again ), i have patched the LX850 firmware enough that i can load it onto my LX200 testbed.
As soon as i connected with my std PEC editor to read out some data, the scope went crazy.
I have tracked this back to the std :gps# serial command that reads a GPS string from the scope, and the bug appears to line up with my notes. :tonofbricks:
As i am not really testing a proper LX850, i am not 100% sure ( but i am 99.9% sure ) it will affect the LX850s as well.
This bug can result in totally random errors, from the Hbx getting "modified" without user interaction, up to full speed runaways of the drives.
As i noted, i am pretty sure i know whats happening, but would like to find anyone stout of heart ( and with a quick finger near the power switch ) who could cross check this.
The error doesnt hurt/damage anything per se, it just invokes uncommanded actions on a random basis, or runs off into nowhere and dies.
A reboot fixes the latter, and a keypress on the slew keys stops the runaway slews.
Anyone game ??????

Andrew Johansen Melbourne Australia

#2 David Pavlich

David Pavlich

    Transmographied

  • *****
  • Administrators
  • Posts: 27495
  • Joined: 18 May 2005
  • Loc: Mandeville, LA USA

Posted 11 April 2013 - 08:05 AM

It's going to be difficult to find someone that has a brand new mount that's under warranty using it as a test bed without any guarantee that it's not going to cause a problem.

Wouldn't it be prudent to just wait until there's enough of the 850s in the field and see if what you speculate will happen, does, in fact happen?

David

#3 Pak

Pak

    Vostok 1

  • *****
  • Posts: 184
  • Joined: 15 Sep 2012
  • Loc: The Great Arc

Posted 11 April 2013 - 09:58 AM

How would the "test" be conducted? I don't own one but I know someone who does.

#4 Pak

Pak

    Vostok 1

  • *****
  • Posts: 184
  • Joined: 15 Sep 2012
  • Loc: The Great Arc

Posted 11 April 2013 - 10:14 AM

Wouldn't it be prudent to just wait until there's enough of the 850s in the field and see if what you speculate will happen, does, in fact happen?

David


David,

The problem I see, if I read Andrew's post correctly, is that there is a random problem that may crop up at the most inopportune time. Imagine if you are remotely controlling your mount when it has a run away slew! You could damage not only the OTA and mount but expensive camera equipment too. If there is a way to prove/disprove the existence of the problem *before* more people get their LX850's and learn the hard way, I think it is prudent to find out.

I am curious what is involved in the testing and what risks there are.

#5 Griffin!

Griffin!

    Mariner 2

  • -----
  • Posts: 201
  • Joined: 12 Sep 2010

Posted 11 April 2013 - 10:25 AM

Wouldn't it be better to get in contact with Meade first and have them test it?

#6 OzAndrewJ

OzAndrewJ

    Viking 1

  • *****
  • Posts: 968
  • Joined: 30 Nov 2010

Posted 11 April 2013 - 03:47 PM

Gday David

As yr an admin, i really meant to post this in Mounts,
but forgot i was in this forum, ( it was late ).
I think it would be better in the Mounts forum,
but i can figure out how to move it???

That said

It's going to be difficult to find someone that has a brand new mount that's under warranty using it as a test bed without any guarantee that it's not going to cause a problem.



Yes and no. Lots of people in the past have been happy to try stuff.
I am 100% sure the bug wont hurt the scope or damage the firmware during testing,
just that it may cause runaways or lock up the handbox.
All this is easily dealt with if you are sitting alongside the scope.

Wouldn't it be prudent to just wait until there's enough of the 850s in the field and see if what you speculate will happen, does, in fact happen?



Yes and no.
The bug will only show if the serial :gps# command is used.
I have no idea how many other apps may or may not use that
command, but my PEC editor "does use it" on every connect.
As such, i wanted to find out very early if it is a problem
as anyone using "MY" app will trigger it.
Im not happy with that,
so wanted to get irrefutable confirmation before i post the bug to Meade,
and also a general heads up on the problem.

Andrew Johansen Melbourne Australia

#7 OzAndrewJ

OzAndrewJ

    Viking 1

  • *****
  • Posts: 968
  • Joined: 30 Nov 2010

Posted 11 April 2013 - 03:56 PM

Gday Pak

How would the "test" be conducted?



Using any application that can connect to the scope
( even hyperterminal )
send the :gps# command.
Send it say once per second or so and you should see
the data coming back is completely trashed and has missing bits.
Anything could happen to the scope whilst you do this,
ie Hbx lockup or runaway slew.
Note, most runaway slews can be stopped by just pressing
the real slew buttons on the Hbx.
I also have a modified version of my PEC editor that can be used
to do ( and log ) these tests in an automated way.

Andrew Johansen Melbourne Australia

#8 OzAndrewJ

OzAndrewJ

    Viking 1

  • *****
  • Posts: 968
  • Joined: 30 Nov 2010

Posted 13 April 2013 - 04:49 AM

Gday All

So i take it, no one is game to try?????? :(

Andrew Johansen Melbourne Australia

#9 CharlesW

CharlesW

    Apollo

  • -----
  • Posts: 1294
  • Joined: 01 Nov 2012
  • Loc: Chula Vista & Indio, CA

Posted 13 April 2013 - 09:05 AM

You might assume that some enterprising employee at Meade has read this by now. But, maybe not.

#10 OzAndrewJ

OzAndrewJ

    Viking 1

  • *****
  • Posts: 968
  • Joined: 30 Nov 2010

Posted 13 April 2013 - 03:28 PM

Gday Charles

You might assume that some enterprising employee at Meade has read this by now



I have already sent in a bug report,
but when it reads like
"Dear Meade, i have mucked around with your LX850 firmware so i can load it into my LX200 and it doesnt work ( i think )"
i dont hold out for rapid reply.
More fun to get independent corroboration from people with real units.
( The LX80 crowd weren't afraid to do a little testing :bow: )

Andrew Johansen Melbourne Australia

#11 herrointment

herrointment

    Aurora

  • *****
  • Posts: 4584
  • Joined: 12 Mar 2011
  • Loc: North of Hwy. 64

Posted 13 April 2013 - 04:59 PM

Sorta like asking a LX850 owner to throw the mount off a cliff but not to worry, I won't let go of the rope.

After reading the LX80 thread I'd be inclined to trust you but I can't say I'd be willing to go first!

#12 OzAndrewJ

OzAndrewJ

    Viking 1

  • *****
  • Posts: 968
  • Joined: 30 Nov 2010

Posted 13 April 2013 - 05:34 PM

Sorta like asking a LX850 owner to throw the mount off a cliff but not to worry,



Not really.
The bug involves the handbox processing routine still running when the GPS is being read.
In this mode, and knowing what data comes back in GPS sentennces, i can see that the GPS data may simulate "keydown" events, but not "keyup" events.
As such, only things that could happen via you pressing handbox buttons will happen.
( the worst i can see is if a "N","S","E" or "W" byte is grabbed by the hbx routine, the scope will start slewing at about 5deg/sec, which is easily stopped by pressing a slew key )
As long as you are near the scope, nothing bad can happen.
Anyway, i wont bother anymore.

Andrew Johansen Melbourne Australia

#13 dmdouglass

dmdouglass

    Viking 1

  • -----
  • Posts: 748
  • Joined: 23 Dec 2007
  • Loc: Tempe, AZ

Posted 13 April 2013 - 08:49 PM

Howdy Andrew....

You know that I would be there for you.....

But I don't have one of these!

Drats !!

#14 Mkofski

Mkofski

    Surveyor 1

  • *****
  • Posts: 1531
  • Joined: 19 Jul 2011
  • Loc: Greenfield, Indiana, USA

Posted 13 April 2013 - 08:52 PM

Howdy Andrew....

You know that I would be there for you.....

But I don't have one of these!

Drats !!


Same here!






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics