Re: Gemini-2 momentarily went haywire
The diagram is not that useful.? Text is easier if you ask me.?? Step 1.? Turn on mount and see whether RA or DEC has a runaway slew
Step 2.? Swap cables at the Gemini plugs.? RA or DEC?
Step 3.? Swap cables at the motors.? RA or DEC?
That's all the diagram says.? Doesn't actually give the logic for deducing what has failed :-). But that's not hard.
So after Step 2, you know whether or not the motor is at fault.? After Step 3, you know whether or not the Gemini is at fault.??
|
Re: Gemini-2 momentarily went haywire
In the Gemini1 the decode of the signals from the motor encoder is done in an ATF16V8 GAL. ? In the Gemini-2 it is done with a PIC (which one depends on whether it is a G-2 or a G3 == G2 Mini). ? David ?
toggle quoted message
Show quoted text
From: [email protected] <[email protected]> On Behalf Of papagordygrapes@...
Sent: 31 March 2021 05:05
To: [email protected]
Subject: Re: [Losmandy_users_io] Gemini-2 momentarily went haywire ? Hi Michael,
What exactly is the detector unit in the Gemini?? I've done work with rotary encoders before and afaik they just use digital GPIO pins.??
Unfortunately the troubleshooting diagram is only useful if the fault is repeatable.??
I do like the idea of buying a spare cable.??
So far I've seen three glitches:
? ?*? DEC stall message during guiding operation
? ?*? RA stall when initiating a slew? (I've since slowed down slew rate to 600 from 800).
? ?*? runaway slew.? Happened so fast that I don't recall whether it was DEC or RA....I think maybe DEC.??
|
Re: Gemini-2 momentarily went haywire
On Tue, Mar 30, 2021 at 11:57 PM, Brian Valente wrote:
hey are all equally plausible, which is why you want to run the test i outlined to determine which?is at fault. I've read your thoughts about "keeping it at bay" but I really think you are living on borrowed time here. you should determine which part is faulty and replace it (hopefully it's just the cable)
Brian,? You mentioned a test and a diagram, but I am not seeing them attached anywhere. Could you repost or direct me to where this is? Thanks, John
|
Re: 76 second RA periodic error fixed with Belleville washer
Reading through the white paper the following can be summarized:
1) A 76 second error is due to the bearings hanging up momentarily
2) A 76 seconds error can't be removed with a PEC curve since it won't overlap with the worm period
3) Some bearings have slightly defective balls or races
4) Some Losmandy worm blocks have a bearing seat which is minutely elliptical to the point where it will deform the bearing causing the balls to hang up
Solutions:
A) Replace the bearings. Use the Losmandy stock bearings or Boca bearings.
B) Change the seating of the bearing in the blocks by polishing down the outside of the bearings or the inside of the blocks so they are not deformed. Or change the blocks.
C) A combination of A and B
Obviously not all bearings and blocks have this issue or every mount would have the problem to point where it would wreck guiding. This is just my quick interpretation, so let the user have at it if they care to.
|
Re: Gemini-2 momentarily went haywire
Pyle PSL92X regulated power supply.? 13.8V @ 5A constant (7A surge).
|
Re: Gemini-2 momentarily went haywire
sorry now all, the first two
Hi Michael,
What exactly is the detector unit in the Gemini?? I've done work with rotary encoders before and afaik they just use digital GPIO pins.??
Unfortunately the troubleshooting diagram is only useful if the fault is repeatable.??
I do like the idea of buying a spare cable.??
So far I've seen three glitches:
? ?*? DEC stall message during guiding operation
? ?*? RA stall when initiating a slew? (I've since slowed down slew rate to 600 from 800).
? ?*? runaway slew.? Happened so fast that I don't recall whether it was DEC or RA....I think maybe DEC.??
-- Brian?
Brian Valente portfolio
|
Re: Gemini-2 momentarily went haywire
these all sound like power issues
what are you using to power your mount?
Hi Michael,
What exactly is the detector unit in the Gemini?? I've done work with rotary encoders before and afaik they just use digital GPIO pins.??
Unfortunately the troubleshooting diagram is only useful if the fault is repeatable.??
I do like the idea of buying a spare cable.??
So far I've seen three glitches:
? ?*? DEC stall message during guiding operation
? ?*? RA stall when initiating a slew? (I've since slowed down slew rate to 600 from 800).
? ?*? runaway slew.? Happened so fast that I don't recall whether it was DEC or RA....I think maybe DEC.??
-- Brian?
Brian Valente portfolio
|
Re: Gemini-2 momentarily went haywire
Hi Michael,
What exactly is the detector unit in the Gemini?? I've done work with rotary encoders before and afaik they just use digital GPIO pins.??
Unfortunately the troubleshooting diagram is only useful if the fault is repeatable.??
I do like the idea of buying a spare cable.??
So far I've seen three glitches:
? ?*? DEC stall message during guiding operation
? ?*? RA stall when initiating a slew? (I've since slowed down slew rate to 600 from 800).
? ?*? runaway slew.? Happened so fast that I don't recall whether it was DEC or RA....I think maybe DEC.??
|
Re: Gemini-2 momentarily went haywire
Hi John
>>>The encoder theory seems plausible to me.
they are all equally plausible, which is why you want to run the test i outlined to determine which?is at fault. I've read your thoughts about "keeping it at bay" but I really think you are living on borrowed time here. you should determine which part is faulty and replace it (hopefully it's just the cable)
FYI the encoder tends to be very repeatable, so based on your assessment that you can 'avoid it', i don't think that's the likely culprit.??
>>>Perhaps if someone unintentionally switches their RA and Dec cables, this could create the same fault?
I never say never, but that is very unlikely. we do this all the time at the shop and never have that issue.?
On Tue, Mar 30, 2021 at 5:58 PM John Kmetz < jjkmetz54@...> wrote: [Edited Message Follows]
The encoder theory seems plausible to me. This problem started for me right after I had swapped the outer RA and Dec drive assemblies (outer plate with the two drive gears, gear box, motor, etc. and disconnected at the Oldham coupler disc). I was seeking better backlash in RA (that worked). But when I first touched the upper Dec control button on the HC, the Dec motor went into a wild spin at something like 2x Slew speed and would not stop unless I turned off the Gemini 2 power switch.
So in this case, perhaps the Gemini could not immediately read the encoders, since the RA and Dec motors were switched. I tried replacing the Dec cable with a spare and put some dielectric grease on the contact pins to no avail. I also switched the motors back to their original drives, keeping the plates were they were, but this did not alleviate. I also replaced the Gemini battery and reset to HGM defaults, again with no luck. Perhaps if someone unintentionally switches their RA and Dec cables, this could create the same fault?
But as I stated earlier, lowering the slew speed to 600, and selecting Guide or Center from the default Slew on the HC right after start up, then briefly pushing the up and down buttons, seems to have fixed this (for now).? Perhaps a slower initial speed allows the Gemini to better read the encoders and not send a errant motor pulse. I am not an engineer, so I will leave the potential electrical causes to the experts. But these steps seem to be keeping it in check for now. After these initial changes, the Gemini seems to run fine from the HC and from the SGP app for the rest of the night.
Best Regards,?
John
-- Brian?
Brian Valente portfolio
|
Re: Gemini-2 momentarily went haywire
The encoder theory seems plausible to me. This problem started for me right after I had swapped the outer RA and Dec drive assemblies (outer plate with the two drive gears, gear box, motor, etc. and disconnected at the Oldham coupler disc). I was seeking better backlash in RA (that worked). But when I first touched the upper Dec control button on the HC, the Dec motor went into a wild spin at something like 2x Slew speed and would not stop unless I turned off the Gemini 2 power switch.
So in this case, perhaps the Gemini could not immediately read the encoders, since the RA and Dec motors were switched. I tried replacing the Dec cable with a spare and put some dielectric grease on the contact pins to no avail. I also switched the motors back to their original drives, keeping the plates were they were, but this did not alleviate. I also replaced the Gemini battery and reset to HGM defaults, again with no luck. Perhaps if someone unintentionally switches their RA and Dec cables, this could create the same fault?
But as I stated earlier, lowering the slew speed to 600, and selecting Guide or Center from the default Slew on the HC right after start up, then briefly pushing the up and down buttons, seems to have fixed this (for now).? Perhaps a slower initial speed allows the Gemini to better read the encoders and not send a errant motor pulse. I am not an engineer, so I will leave the potential electrical causes to the experts. But these steps seem to be keeping it in check for now. After these initial changes, the Gemini seems to run fine from the HC and from the SGP app for the rest of the night.
Best Regards,?
John
|
Re: FINAL-Definitive Answer regarding Start
Just Warm Restart and continue from where you left off, and that could
be any mount position, not just CWD.
***
This
*right here* is why I say ¡°use Warm Restart!¡± ¨C even though I
*always* park at CWD and Warm Start *should* work.. For ME, warm start doesn¡¯t
¡°work¡± like warm restart.. Essentially, the mount seems to ¡°remember¡±
exactly where it is.. So if you use your scope and everything works great, you
park it, power off, and the next time, with a Warm Restart, it will continue
on.. For me, it is like a pause button!
?
To
me, and it¡¯s just my anecdotal opinion, this is *THE* answer for anyone who
complains ¡°it worked last night, but tonight I did a warm start and the initial
GOTO is way off but nothing changed!¡±
?
My
scope hasn¡¯t moved in forever and last Friday night I used my usual Warm
Start, did a goto to a star I haven¡¯t gone to in months and is not in the
model, and it was dead center of my quite small FOV on the ASI178.. (f/7 910mm
and a 178 yields around a 24 x 17 minute FOV)
?
I
am pretty sure I know why Warm Start is not ¡°perfect¡±, but no one
cares about that.. ?
?
Bottom
line is that Warm Restart remembers where the scope is, not where it is
supposed to be..
?
Derek
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Paul Kanevsky
Sent: Tuesday, March 30, 2021 3:35
PM
To: [email protected]
Subject: Re: [Losmandy_users_io]
FINAL-Definitive Answer regarding Start
?
Here is a more detailed explanation that agrees with the manual:
Warm Restart is
meant to be used when nothing's been touched or moved since Gemini was turned
off. Mount must be in the same position, axis not moved at all between
sessions, polar alignment exactly the same. Just Warm Restart and continue from
where you left off, and that could be any mount position, not just CWD.
Warm Start can be
used if RA or DEC were rotated by hand while Gemini was turned off (for
example, to balance a scope), but nothing else has changed. Requires mount to
start in CWD position. One Sync or Align is required to get better pointing
accuracy.
Cold Start can be
used to start a fresh new model after moving to a new location, changing polar
alignment, moving axis by hand, bumping the tripod, etc.
Regards,
? ? -Paul
On Tue, Mar 30, 2021 at 10:45 AM, Russ wrote:
Ok, seeing mixed messages regarding Start commands here on Losmandy
User Site and what is printed in Gemini manual.
What I have read here on this users group is:
Warm Start:? For permanent locations (observatory) where all modeling and
positioning is maintained.? Parked at CWD.
Warm Restart:? For preserving modeling data but not current telescope
position.? Parked at an assigned Park location.
Now, the Gemini manual (Page 16) has these 2 items reversed (as described
above);? Warm Restart=Permanent installations; Warm Start=Mount not moved
but scope has.
So, can we get this clearly answered once and for all.? Which method is
correct!? ;-)
Thanks,
Russ
|
Re: FINAL-Definitive Answer regarding Start
THANK YOU PAUL!!!? Perfect description and going to use this as my master config now!!!
|
Re: FINAL-Definitive Answer regarding Start
Here is a more detailed explanation that agrees with the manual:
Warm Restart is meant to be used when nothing's been touched or moved since Gemini was turned off. Mount must be in the same position, axis not moved at all between sessions, polar alignment exactly the same. Just Warm Restart and continue from where you left off, and that could be any mount position, not just CWD.
Warm Start can be used if RA or DEC were rotated by hand while Gemini was turned off (for example, to balance a scope), but nothing else has changed. Requires mount to start in CWD position. One Sync or Align is required to get better pointing accuracy.
Cold Start can be used to start a fresh new model after moving to a new location, changing polar alignment, moving axis by hand, bumping the tripod, etc.
Regards,
? ? -Paul
toggle quoted message
Show quoted text
On Tue, Mar 30, 2021 at 10:45 AM, Russ wrote:
Ok, seeing mixed messages regarding Start commands here on Losmandy User Site and what is printed in Gemini manual.
What I have read here on this users group is:
Warm Start:? For permanent locations (observatory) where all modeling and positioning is maintained.? Parked at CWD.
Warm Restart:? For preserving modeling data but not current telescope position.? Parked at an assigned Park location.
Now, the Gemini manual (Page 16) has these 2 items reversed (as described above);? Warm Restart=Permanent installations; Warm Start=Mount not moved but scope has.
So, can we get this clearly answered once and for all.? Which method is correct!? ;-)
Thanks,
Russ
|
Re: FINAL-Definitive Answer regarding Start
For Gemini II, In my experience, the first. ?I suspect the second is just a misprint. By manual, are you referring to the website? ?
JMD
|
492 units can use 12-18v DC power input
you can use the current Losmandy AC/DC 15v power supply, for example
Brian
|
Re: Gemini-2 momentarily went haywire
>>>. That's when runaway slews occur: when there is some feedback from the encoder, but it's either wrong or inaccurate.
or missing i.e., "still waiting?for it to hit that encoder mark"
On Tue, Mar 30, 2021 at 11:27 AM Paul Kanevsky < yh@...> wrote: On Tue, Mar 30, 2021 at 11:09 AM, Michael Herman wrote:
That is true....if the Gemini does not sense the motor optical encoder wheel move it will keep sending motor power to see if it can move it...and the motor will appear to run away.? ?It could have been programmed to stop and report a stalled motor or motor signal not received...but c'est la vie.??
Thankfully, that really is how it is programmed. When Gemini doesn't sense movement after applying maximum power to the motor, it will stall pretty quickly. That's what the stall message from Gemini represents: it couldn't move the mount or at least, couldn't get feedback from the encoders that the mount has moved.
What may get one into trouble is if there are intermittent or poor signals getting in from the encoders. In that case, Gemini may continue to try to move the mount, applying whatever power is needed to get it to the desired position. That's when runaway slews occur: when there is some feedback from the encoder, but it's either wrong or inaccurate.
Regards,
? ?-Paul
-- Brian?
Brian Valente portfolio
|
Re: Gemini-2 momentarily went haywire
Thanks, Paul!
A you say: a missing cable (like disconnecting it) the gemini says "motor stalled".? And that is what I observe.
A runaway motor would have to be caused by these: 1.? bad optical encoder transmitter in the motor is damaged (HEDS unit), 2. the cable has a bad or noisy wire 3. the detector unit in the Gemini is bad 4. the cable connector has a weak or noisy?connection. ? I have seen that on damaged Gemini and damaged Motors too.??
If the motor did a runaway as reported, and then the system came back to normal behavior... you would have to rule out a bad motor, and a bad detector, and say? 1. either the cable is bad,? or? 2. a connector that the cable plugs into could have a weak connection.
This suggests the owner with this problem should buy a spare cable (might as well be a long DEC cable in case the problem cable is either RA or DEC).?
Best, Michael
On Tue, Mar 30, 2021 at 11:27 AM Paul Kanevsky < yh@...> wrote: On Tue, Mar 30, 2021 at 11:09 AM, Michael Herman wrote:
That is true....if the Gemini does not sense the motor optical encoder wheel move it will keep sending motor power to see if it can move it...and the motor will appear to run away.? ?It could have been programmed to stop and report a stalled motor or motor signal not received...but c'est la vie.??
Thankfully, that really is how it is programmed. When Gemini doesn't sense movement after applying maximum power to the motor, it will stall pretty quickly. That's what the stall message from Gemini represents: it couldn't move the mount or at least, couldn't get feedback from the encoders that the mount has moved.
What may get one into trouble is if there are intermittent or poor signals getting in from the encoders. In that case, Gemini may continue to try to move the mount, applying whatever power is needed to get it to the desired position. That's when runaway slews occur: when there is some feedback from the encoder, but it's either wrong or inaccurate.
Regards,
? ?-Paul
|
Re: Gemini-2 momentarily went haywire
On Tue, Mar 30, 2021 at 11:09 AM, Michael Herman wrote:
That is true....if the Gemini does not sense the motor optical encoder wheel move it will keep sending motor power to see if it can move it...and the motor will appear to run away.? ?It could have been programmed to stop and report a stalled motor or motor signal not received...but c'est la vie.??
Thankfully, that really is how it is programmed. When Gemini doesn't sense movement after applying maximum power to the motor, it will stall pretty quickly. That's what the stall message from Gemini represents: it couldn't move the mount or at least, couldn't get feedback from the encoders that the mount has moved. What may get one into trouble is if there are intermittent or poor signals getting in from the encoders. In that case, Gemini may continue to try to move the mount, applying whatever power is needed to get it to the desired position. That's when runaway slews occur: when there is some feedback from the encoder, but it's either wrong or inaccurate. Regards, ? ?-Paul
|
Re: Mount is Unparked when power is restored
toggle quoted message
Show quoted text
It was set to cold boot.? I changed it to ¡°Prompt if not Started¡±.? Thanks?
-- Brian?
Brian Valente portfolio
|
Re: Mount is Unparked when power is restored
It was set to cold boot. ?I changed it to ¡°Prompt if not Started¡±. ?Thanks?
|
David C. Partridge
Paul Kanevsky
Michael Herman