Spikes in Dec


Roland Christen
 

Hi Sebastien,

Howard has been going thru your log files and discovered some problems with your PHD setup:

"PHD2 is indeed commanding moves of 2 – 3 arc-seconds, all of them south.  It is also sending many moves of single digit mSecs.  There were 60 instances of  :Ms1#  in the one APCC log file.  That’s calling for a move of ~ 0.015 arc-seconds.  The screen below shows 11 move commands in a row, most or all of which were too small, followed by a ~2 arc-second move -  :Ms130#."

Basically you have PHD set up to respond to every tiny error in such a way that no movement occurs for many cycles, which is then followed by a single large movement. This type of Dec algorithm was developed for mounts that had huge backlash where the mount does not respond to move commands in a timely way, and then finally PHD sends a large correction command in order to get the mount axis moving. That's when the spike occurs.

Apparently you have set up your Min and Max moves wrong as well as choosing the wrong algorithm for a precision mount. never should guiding software send 1 millisecond ( 0.015 arc sec) move commands because the mount basically will not move at all, or so slightly that it will never register in your guide star motion. You may even have set your guide speed wrong - it should always be left at 1x sidereal, never anything else.

I would suggest that you go to the PHD group and let someone there analyze your log files and look at your settings. I am not an expert when it comes to PHD, I have only a rudimentary knowledge, so it would be est to ask the users on that group to help you. I know that the Mach2 will guide extremely well with your setup, even if you have some small unbalance. I put a 3000mm focal length 10" Mak-Cass on my observatory Mach2 last night and guiding was smooth and slick as snail snot on a door handle. I even set it purposely out of balance and still it guided the same.

And thanks to Howard here who sifted thru your log files and spotted the discrepancies.

Roland Christen
Astro-Physics Inc.




--
Roland Christen
Astro-Physics


Seb@stro
 

Hello Roland, Howard,

Thanks for the heads-up. I will hover to the PHD2 forum with this additional info as you suggest.

Strange thing is while I created the guiding profile, I did check the precise encoder mount box (or whatever it’s called). And guide speed was set to 1x. I also followed all recommendations I got from the Guiding Assistant. Maybe this isn’t enough which such highly precise mounts though, as you suggest.

There’s also the fact that I’m using the latest dev5 version (to benefit from the multi-star guiding functionality). I might revert back to the stable version for further testing.

Again, many thanks for your very professional support. Much appreciated.

Sébastien

Le 6 mai 2021 à 15:19, "chris1011@..." <chris1011@...> a écrit :


Hi Sebastien,

Howard has been going thru your log files and discovered some problems with your PHD setup:

"PHD2 is indeed commanding moves of 2 – 3 arc-seconds, all of them south.  It is also sending many moves of single digit mSecs.  There were 60 instances of  :Ms1#  in the one APCC log file.  That’s calling for a move of ~ 0.015 arc-seconds.  The screen below shows 11 move commands in a row, most or all of which were too small, followed by a ~2 arc-second move -  :Ms130#."

Basically you have PHD set up to respond to every tiny error in such a way that no movement occurs for many cycles, which is then followed by a single large movement. This type of Dec algorithm was developed for mounts that had huge backlash where the mount does not respond to move commands in a timely way, and then finally PHD sends a large correction command in order to get the mount axis moving. That's when the spike occurs.

Apparently you have set up your Min and Max moves wrong as well as choosing the wrong algorithm for a precision mount. never should guiding software send 1 millisecond ( 0.015 arc sec) move commands because the mount basically will not move at all, or so slightly that it will never register in your guide star motion. You may even have set your guide speed wrong - it should always be left at 1x sidereal, never anything else.

I would suggest that you go to the PHD group and let someone there analyze your log files and look at your settings. I am not an expert when it comes to PHD, I have only a rudimentary knowledge, so it would be est to ask the users on that group to help you. I know that the Mach2 will guide extremely well with your setup, even if you have some small unbalance. I put a 3000mm focal length 10" Mak-Cass on my observatory Mach2 last night and guiding was smooth and slick as snail snot on a door handle. I even set it purposely out of balance and still it guided the same.

And thanks to Howard here who sifted thru your log files and spotted the discrepancies.

Roland Christen
Astro-Physics Inc.




 

Seb i seem to recall you using lowpass2 in PHD which is the correct algorithm for high res encoders



On Thu, May 6, 2021 at 12:54 PM Seb@stro <sebastiendore1@...> wrote:
Hello Roland, Howard,

Thanks for the heads-up. I will hover to the PHD2 forum with this additional info as you suggest.

Strange thing is while I created the guiding profile, I did check the precise encoder mount box (or whatever it’s called). And guide speed was set to 1x. I also followed all recommendations I got from the Guiding Assistant. Maybe this isn’t enough which such highly precise mounts though, as you suggest.

There’s also the fact that I’m using the latest dev5 version (to benefit from the multi-star guiding functionality). I might revert back to the stable version for further testing.

Again, many thanks for your very professional support. Much appreciated.

Sébastien

Le 6 mai 2021 à 15:19, "chris1011@..." <chris1011@...> a écrit :


Hi Sebastien,

Howard has been going thru your log files and discovered some problems with your PHD setup:

"PHD2 is indeed commanding moves of 2 – 3 arc-seconds, all of them south.  It is also sending many moves of single digit mSecs.  There were 60 instances of  :Ms1#  in the one APCC log file.  That’s calling for a move of ~ 0.015 arc-seconds.  The screen below shows 11 move commands in a row, most or all of which were too small, followed by a ~2 arc-second move -  :Ms130#."

Basically you have PHD set up to respond to every tiny error in such a way that no movement occurs for many cycles, which is then followed by a single large movement. This type of Dec algorithm was developed for mounts that had huge backlash where the mount does not respond to move commands in a timely way, and then finally PHD sends a large correction command in order to get the mount axis moving. That's when the spike occurs.

Apparently you have set up your Min and Max moves wrong as well as choosing the wrong algorithm for a precision mount. never should guiding software send 1 millisecond ( 0.015 arc sec) move commands because the mount basically will not move at all, or so slightly that it will never register in your guide star motion. You may even have set your guide speed wrong - it should always be left at 1x sidereal, never anything else.

I would suggest that you go to the PHD group and let someone there analyze your log files and look at your settings. I am not an expert when it comes to PHD, I have only a rudimentary knowledge, so it would be est to ask the users on that group to help you. I know that the Mach2 will guide extremely well with your setup, even if you have some small unbalance. I put a 3000mm focal length 10" Mak-Cass on my observatory Mach2 last night and guiding was smooth and slick as snail snot on a door handle. I even set it purposely out of balance and still it guided the same.

And thanks to Howard here who sifted thru your log files and spotted the discrepancies.

Roland Christen
Astro-Physics Inc.





--
Brian 



Brian Valente


Seb@stro
 

Hi Brian,

For the DEC, yes, you recalled right, Lowpass2. For RA, I have Hysterersis selected. 

I see there's also the possibility to select lowpass2 for RA, but I believe I read somewhere not to use it for that axis (possibly in the PHD2 manual). Anyway, I didn't seem to have any guiding problems in RA, so I'd leave it there for now.

Just looking around at other settings, I realize that I have the "Use DEC compensation" checked in the Guiding tab,  Calibration section. Help file says that it usually should stay checked unless in "unsual" cases, like when a mount's controller would apply a compensation automatically. Should I assume that a mount with abs encoders correspond to that definition and uncheck the box? Do you think, this could explain the issue reported by Howard/Roland ?

I realize this is more a question for the PHD forum, (sorry about that) but I thought it could be of general interest here as well if you have the answer at hand.



Sébastien


De : main@ap-gto.groups.io <main@ap-gto.groups.io> de la part de Brian Valente <bvalente@...>
Envoyé : 6 mai 2021 17:01
À : main@ap-gto.groups.io <main@ap-gto.groups.io>
Objet : Re: [ap-gto] Spikes in Dec
 
Seb i seem to recall you using lowpass2 in PHD which is the correct algorithm for high res encoders


 

You want to enable DEC compensation here

This adjusts the guiding based on your sky position, which doesn't change (unless you are going to say you have a space telescope)

>>>I realize this is more a question for the PHD forum, (sorry about that) but I thought it could be of general interest here as well if you have the answer at hand.

that's fine - happy to help. Maybe you can shoot me some of your logs direct via email and I can review them. 

I'm not sure I saw what Howard mentioned, but guiding with high quality encoders is a relatively new thing, so anything is possible. lowpass2 was introduced only a few years ago iirc

 
Brian


On Thu, May 6, 2021 at 1:45 PM Seb@stro <sebastiendore1@...> wrote:
Hi Brian,

For the DEC, yes, you recalled right, Lowpass2. For RA, I have Hysterersis selected. 

I see there's also the possibility to select lowpass2 for RA, but I believe I read somewhere not to use it for that axis (possibly in the PHD2 manual). Anyway, I didn't seem to have any guiding problems in RA, so I'd leave it there for now.

Just looking around at other settings, I realize that I have the "Use DEC compensation" checked in the Guiding tab,  Calibration section. Help file says that it usually should stay checked unless in "unsual" cases, like when a mount's controller would apply a compensation automatically. Should I assume that a mount with abs encoders correspond to that definition and uncheck the box? Do you think, this could explain the issue reported by Howard/Roland ?

I realize this is more a question for the PHD forum, (sorry about that) but I thought it could be of general interest here as well if you have the answer at hand.



Sébastien


De : main@ap-gto.groups.io <main@ap-gto.groups.io> de la part de Brian Valente <bvalente@...>
Envoyé : 6 mai 2021 17:01
À : main@ap-gto.groups.io <main@ap-gto.groups.io>
Objet : Re: [ap-gto] Spikes in Dec
 
Seb i seem to recall you using lowpass2 in PHD which is the correct algorithm for high res encoders



--
Brian 



Brian Valente


Roland Christen
 

Basically the way RA guiding happens is the same as a non-encoder mount, except that the guide software doesn't have to fight a periodic error. So the guide pulses can be sent at a slower cadence. (cadence is not equal to guide star exposure).

Declination is definitely different in an Astro-Physics encoder mount. Reversal backlash is gone, so that a 1 arc sec command in the reverse direction will cause the axis to move 1 arc sec in that reverse direction. A non-encoder mount sometimes needs many 1 arc sec move commands before the Dec axis actually reverses direction. That's because the motor shaft must move a significant rotational distance (number of arc seconds) in reverse before the worm wheel begins to move. It could be 5 arc seconds, or 10 or even more on some mounts. It all depends how tight the gearing is and the clearances in the reduction gears or the amount a belt stretches in a belt drive mount. It's never zero in any mount.

There is a caveat however for the 1 arc sec reversal in an encoder mount. The encoder does not eliminate the actual mechanical delay, or the distance it takes for the reversal to happen. The motor shaft still needs to move 5 or 10 arc seconds at the 1x rate, but the encoder makes that happen automatically after only one move command. It means that any move command in Dec might take a second of time if it is a reversal move, rather than being instant. Therefore the best way to guide in Dec is to let the axis settle for a second or two before taking another guide exposure and sending another guide pulse. Too fast cadence will cause the tracking to be less accurate than it could be. Adding a 2 second delay between guide exposures is a good way to to keep the Dec axis stable and prevent hunting and chasing the seeing.

It would even be better if PHD would wait 1 second after a move command before actually taking the exposure of the guide star so that the axis has a chance to reverse and fully settle. But perhaps that is not in the cards at this time. Maxim DL does insert a small delay of that size after issuing a guide command, and so I have found that it will guide very accurately even at the longest focal lengths.

Rolando



-----Original Message-----
From: Brian Valente <bvalente@...>
To: main@ap-gto.groups.io
Sent: Thu, May 6, 2021 4:58 pm
Subject: Re: [ap-gto] Spikes in Dec

You want to enable DEC compensation here

This adjusts the guiding based on your sky position, which doesn't change (unless you are going to say you have a space telescope)

>>>I realize this is more a question for the PHD forum, (sorry about that) but I thought it could be of general interest here as well if you have the answer at hand.

that's fine - happy to help. Maybe you can shoot me some of your logs direct via email and I can review them. 

I'm not sure I saw what Howard mentioned, but guiding with high quality encoders is a relatively new thing, so anything is possible. lowpass2 was introduced only a few years ago iirc

 
Brian


On Thu, May 6, 2021 at 1:45 PM Seb@stro <sebastiendore1@...> wrote:
Hi Brian,

For the DEC, yes, you recalled right, Lowpass2. For RA, I have Hysterersis selected. 

I see there's also the possibility to select lowpass2 for RA, but I believe I read somewhere not to use it for that axis (possibly in the PHD2 manual). Anyway, I didn't seem to have any guiding problems in RA, so I'd leave it there for now.

Just looking around at other settings, I realize that I have the "Use DEC compensation" checked in the Guiding tab,  Calibration section. Help file says that it usually should stay checked unless in "unsual" cases, like when a mount's controller would apply a compensation automatically. Should I assume that a mount with abs encoders correspond to that definition and uncheck the box? Do you think, this could explain the issue reported by Howard/Roland ?

I realize this is more a question for the PHD forum, (sorry about that) but I thought it could be of general interest here as well if you have the answer at hand.



Sébastien


De : main@ap-gto.groups.io <main@ap-gto.groups.io> de la part de Brian Valente <bvalente@...>
Envoyé : 6 mai 2021 17:01
À : main@ap-gto.groups.io <main@ap-gto.groups.io>
Objet : Re: [ap-gto] Spikes in Dec
 
Seb i seem to recall you using lowpass2 in PHD which is the correct algorithm for high res encoders


--
Brian 



Brian Valente

--
Roland Christen
Astro-Physics


Seb@stro
 

Thank you (again), Brian. 

Will forward my PHD2 logs of the entire last session privately as well as what Howard has pointed me out.

Sébastien


De : main@ap-gto.groups.io <main@ap-gto.groups.io> de la part de Brian Valente <bvalente@...>
Envoyé : 6 mai 2021 17:58
À : main@ap-gto.groups.io <main@ap-gto.groups.io>
Objet : Re: [ap-gto] Spikes in Dec
 
You want to enable DEC compensation here

This adjusts the guiding based on your sky position, which doesn't change (unless you are going to say you have a space telescope)

>>>I realize this is more a question for the PHD forum, (sorry about that) but I thought it could be of general interest here as well if you have the answer at hand.

that's fine - happy to help. Maybe you can shoot me some of your logs direct via email and I can review them. 

I'm not sure I saw what Howard mentioned, but guiding with high quality encoders is a relatively new thing, so anything is possible. lowpass2 was introduced only a few years ago iirc

 
Brian


Seb@stro
 

Very interesting. Posted that over the openPHD forum as well. 
Will let know how that turns out.

Sébastien


De : main@ap-gto.groups.io <main@ap-gto.groups.io> de la part de Roland Christen via groups.io <chris1011@...>
Envoyé : 6 mai 2021 18:19
À : main@ap-gto.groups.io <main@ap-gto.groups.io>
Objet : Re: [ap-gto] Spikes in Dec
 
Basically the way RA guiding happens is the same as a non-encoder mount, except that the guide software doesn't have to fight a periodic error. So the guide pulses can be sent at a slower cadence. (cadence is not equal to guide star exposure).

Declination is definitely different in an Astro-Physics encoder mount. Reversal backlash is gone, so that a 1 arc sec command in the reverse direction will cause the axis to move 1 arc sec in that reverse direction. A non-encoder mount sometimes needs many 1 arc sec move commands before the Dec axis actually reverses direction. That's because the motor shaft must move a significant rotational distance (number of arc seconds) in reverse before the worm wheel begins to move. It could be 5 arc seconds, or 10 or even more on some mounts. It all depends how tight the gearing is and the clearances in the reduction gears or the amount a belt stretches in a belt drive mount. It's never zero in any mount.

There is a caveat however for the 1 arc sec reversal in an encoder mount. The encoder does not eliminate the actual mechanical delay, or the distance it takes for the reversal to happen. The motor shaft still needs to move 5 or 10 arc seconds at the 1x rate, but the encoder makes that happen automatically after only one move command. It means that any move command in Dec might take a second of time if it is a reversal move, rather than being instant. Therefore the best way to guide in Dec is to let the axis settle for a second or two before taking another guide exposure and sending another guide pulse. Too fast cadence will cause the tracking to be less accurate than it could be. Adding a 2 second delay between guide exposures is a good way to to keep the Dec axis stable and prevent hunting and chasing the seeing.

It would even be better if PHD would wait 1 second after a move command before actually taking the exposure of the guide star so that the axis has a chance to reverse and fully settle. But perhaps that is not in the cards at this time. Maxim DL does insert a small delay of that size after issuing a guide command, and so I have found that it will guide very accurately even at the longest focal lengths.

Rolando



-----Original Message-----
From: Brian Valente <bvalente@...>
To: main@ap-gto.groups.io
Sent: Thu, May 6, 2021 4:58 pm
Subject: Re: [ap-gto] Spikes in Dec

You want to enable DEC compensation here

This adjusts the guiding based on your sky position, which doesn't change (unless you are going to say you have a space telescope)

>>>I realize this is more a question for the PHD forum, (sorry about that) but I thought it could be of general interest here as well if you have the answer at hand.

that's fine - happy to help. Maybe you can shoot me some of your logs direct via email and I can review them. 

I'm not sure I saw what Howard mentioned, but guiding with high quality encoders is a relatively new thing, so anything is possible. lowpass2 was introduced only a few years ago iirc

 
Brian


On Thu, May 6, 2021 at 1:45 PM Seb@stro <sebastiendore1@...> wrote:
Hi Brian,

For the DEC, yes, you recalled right, Lowpass2. For RA, I have Hysterersis selected. 

I see there's also the possibility to select lowpass2 for RA, but I believe I read somewhere not to use it for that axis (possibly in the PHD2 manual). Anyway, I didn't seem to have any guiding problems in RA, so I'd leave it there for now.

Just looking around at other settings, I realize that I have the "Use DEC compensation" checked in the Guiding tab,  Calibration section. Help file says that it usually should stay checked unless in "unsual" cases, like when a mount's controller would apply a compensation automatically. Should I assume that a mount with abs encoders correspond to that definition and uncheck the box? Do you think, this could explain the issue reported by Howard/Roland ?

I realize this is more a question for the PHD forum, (sorry about that) but I thought it could be of general interest here as well if you have the answer at hand.



Sébastien


De : main@ap-gto.groups.io <main@ap-gto.groups.io> de la part de Brian Valente <bvalente@...>
Envoyé : 6 mai 2021 17:01
À : main@ap-gto.groups.io <main@ap-gto.groups.io>
Objet : Re: [ap-gto] Spikes in Dec
 
Seb i seem to recall you using lowpass2 in PHD which is the correct algorithm for high res encoders


--
Brian 



Brian Valente

--
Roland Christen
Astro-Physics