Guiding a mach1 - and pec?


Ron Kramer
 

I've now had my 2nd night of stars since my Mach1 arrived. I wanted to concentrate on guiding (since that is why I upgraded from the NEQ6).


I carefully balanced the scope.  Calibrated (PHD2) near meridian/ecliptic.


guiding didn't look good at that location. I moved to another area and wasn't much better. 


Seeing I suspect was horrible. (I Could see a hazy icy ring out around the moon). Maybe that's why it was bad?   So I tried guide assistant in PHD2 (I don't think it uses seeing as it doesn't do actual guiding)? 


The results I think? were very good? 


https://groups.yahoo.com/neo/groups/ap-gto/photos/albums/1596808407/lightbox/36747959?orderBy=ordinal&sortOrder=asc&photoFilter=ALL#zax/36747959


Hoping my photo will appear above?


Though that seemed to report good things, guiding was very abrupt and lots of zig zags.  Like OVER correcting...  Though I ended up with Dec agression at  9.  (yea... 9).  Could that be normal? 


I was doing 4 second exposures as was recommended here.  HOWEVER I find it very difficult to do this or longer as my camera's gain is already turned down to 10 in order not to over saturate the guide star (and I still get saturation warnings at that).  Any other way to turn down sensitivity so allow 5 second exposures as recommended by Roland?   (ASI290MM)  ND filter added to camera? 


Frustrated, I wondered about PEC.  It was on, and I read it's "recorded" at the shop and to leave it be... it's set.  Well if I disengaged the worm gears (as I do this to balance with no drag) would not the pec be "off"?  Turning pec OFF - looked better...  I did then click RECORD and watched guiding which was a bit better without PEC.  Later I noticed it was engaged again. (once recorded does it then apply that recorded error?)  


I then was getting a lot of star lost errors (I think high thin clouds) as I was in a small gap in days of rain.  Advice?  Wait and try again? the GA feedback is good?   what about PEC.  I read somewhere on here about uploading it to the mount?  Id I recorded it - didn't it auto apply it? 




Roland Christen
 


Frustrated , I wondered about PEC.  It was on, and I read it's "recorded" at the shop and to leave it be... it's set.  Well if I disengaged the worm gears (as I do this to balance with no drag) would not the pec be "off"?
No, disengaging the worm from the worm wheel does not change the PE corrections.

I think you are grasping at straws. Please read my other post about polar alignment - try to get it as good as possible so that the guiding program does not need to fight your misalignment drift.

PLEASE read the following and try it (for your sake and the rest of us who are trying to help):

If you have any doubts about the tracking accuracy do this: TURN OFF the guide corrections to the mount by turning the aggressiveness to zero. Your guider will still record the motion of the guide star and will show you the slow drift back and forth of the RA axis caused by PE AND the steady drift up or down of both RA and Dec due to polar misalignment. Once you have that data, take a screen shot and save it, or post it somewhere so we can all see what your basic mount is actually doing sans the complications of the guiding software corrections. THEN do it all again but this time turn on the PE corrections. You should see a marked improvement of the RA wiggles back and forth because the PEM will be correcting most of that and you should have only the steady drift due to polar misalignment. Both of these tests should take 10 minutes at most (one complete worm cycle).

Now you actually have something to work with instead of the performance below that is mixed up with corrections that may or may not be correct. The other advantage to the above test is that it can show you exactly what to do to precision align your mount on the pole. If you want, I can explain again how that is done, and believe it or not, it is really quite simple and straight forward. It is well worth doing drift alignment and need not be difficult at all and take very little time.

Rolando


-----Original Message-----
From: ronkramer1957@... [ap-gto]
To: ap-gto
Sent: Mon, Oct 30, 2017 6:09 pm
Subject: [ap-gto] Guiding a mach1 - and pec?



I've now had my 2nd night of stars since my Mach1 arrived. I wanted to concentrate on guiding (since that is why I upgraded from the NEQ6).

I carefully balanced the scope.  Calibrated (PHD2) near meridian/ecliptic.

guiding didn't look good at that location. I moved to another area and wasn't much better. 

Seeing I suspect was horrible. (I Could see a hazy icy ring out around the moon). Maybe that's why it was bad?   So I tried guide assistant in PHD2 (I don't think it uses seeing as it doesn't do actual guiding)? 

The results I think? were very good? 


Hoping my photo will appear above?

Though that seemed to report good things, guiding was very abrupt and lots of zig zags.  Like OVER correcting...  Though I ended up with Dec agression at  9.  (yea... 9).  Could that be normal? 

I was doing 4 second exposures as was recommended here.  HOWEVER I find it very difficult to do this or longer as my camera's gain is already turned down to 10 in order not to over saturate the guide star (and I still get saturation warnings at that).  Any other way to turn down sensitivity so allow 5 second exposures as recommended by Roland?   (ASI290MM)  ND filter added to camera? 

Frustrated , I wondered about PEC.  It was on, and I read it's "recorded" at the shop and to leave it be... it's set.  Well if I disengaged the worm gears (as I do this to balance with no drag) would not the pec be "off"?  Turning pec OFF - looked better...  I did then click RECORD and watched guiding which was a bit better without PEC.  Later I noticed it was engaged again. (once recorded does it then apply that recorded error?)  

I then was getting a lot of star lost errors (I think high thin clouds) as I was in a small gap in days of rain.  Advice?  Wait and try again? the GA feedback is good?   what about PEC.  I read somewhere on here about uploading it to the mount?  Id I recorded it - didn't it auto apply it? 





Ron Kramer
 

thanks I'll try this next clear night.
I did try to set DEC to 0 - as per your suggestion before but it wouldn't stick - it changed to 1  (is that ok?)
You say do this for both axis then. Will do.

I assumed that teh "start" of the pec curve would change if he contact with the worm changed place.
I'll figure that's wrong.

Will do. Don't give me to much at once. My memory is shot. I'll likely print the above out and get that step done.



On Mon, Oct 30, 2017 at 7:24 PM, chris1011@... [ap-gto] <ap-gto@...> wrote:
 


Frustrated , I wondered about PEC.  It was on, and I read it's "recorded" at the shop and to leave it be... it's set.  Well if I disengaged the worm gears (as I do this to balance with no drag) would not the pec be "off"?
No, disengaging the worm from the worm wheel does not change the PE corrections.

I think you are grasping at straws. Please read my other post about polar alignment - try to get it as good as possible so that the guiding program does not need to fight your misalignment drift.

PLEASE read the following and try it (for your sake and the rest of us who are trying to help):

If you have any doubts about the tracking accuracy do this: TURN OFF the guide corrections to the mount by turning the aggressiveness to zero. Your guider will still record the motion of the guide star and will show you the slow drift back and forth of the RA axis caused by PE AND the steady drift up or down of both RA and Dec due to polar misalignment. Once you have that data, take a screen shot and save it, or post it somewhere so we can all see what your basic mount is actually doing sans the complications of the guiding software corrections. THEN do it all again but this time turn on the PE corrections. You should see a marked improvement of the RA wiggles back and forth because the PEM will be correcting most of that and you should have only the steady drift due to polar misalignment. Both of these tests should take 10 minutes at most (one complete worm cycle).

Now you actually have something to work with instead of the performance below that is mixed up with corrections that may or may not be correct. The other advantage to the above test is that it can show you exactly what to do to precision align your mount on the pole. If you want, I can explain again how that is done, and believe it or not, it is really quite simple and straight forward. It is well worth doing drift alignment and need not be difficult at all and take very little time.

Rolando


-----Original Message-----
From: ronkramer1957@... [ap-gto] <ap-gto@...>
To: ap-gto <ap-gto@...>
Sent: Mon, Oct 30, 2017 6:09 pm
Subject: [ap-gto] Guiding a mach1 - and pec?



I've now had my 2nd night of stars since my Mach1 arrived. I wanted to concentrate on guiding (since that is why I upgraded from the NEQ6).

I carefully balanced the scope.  Calibrated (PHD2) near meridian/ecliptic.

guiding didn't look good at that location. I moved to another area and wasn't much better. 

Seeing I suspect was horrible. (I Could see a hazy icy ring out around the moon). Maybe that's why it was bad?   So I tried guide assistant in PHD2 (I don't think it uses seeing as it doesn't do actual guiding)? 

The results I think? were very good? 


Hoping my photo will appear above?

Though that seemed to report good things, guiding was very abrupt and lots of zig zags.  Like OVER correcting...  Though I ended up with Dec agression at  9.  (yea... 9).  Could that be normal? 

I was doing 4 second exposures as was recommended here.  HOWEVER I find it very difficult to do this or longer as my camera's gain is already turned down to 10 in order not to over saturate the guide star (and I still get saturation warnings at that).  Any other way to turn down sensitivity so allow 5 second exposures as recommended by Roland?   (ASI290MM)  ND filter added to camera? 

Frustrated , I wondered about PEC.  It was on, and I read it's "recorded" at the shop and to leave it be... it's set.  Well if I disengaged the worm gears (as I do this to balance with no drag) would not the pec be "off"?  Turning pec OFF - looked better...  I did then click RECORD and watched guiding which was a bit better without PEC.  Later I noticed it was engaged again. (once recorded does it then apply that recorded error?)  

I then was getting a lot of star lost errors (I think high thin clouds) as I was in a small gap in days of rain.  Advice?  Wait and try again? the GA feedback is good?   what about PEC.  I read somewhere on here about uploading it to the mount?  Id I recorded it - didn't it auto apply it? 






Roland Christen
 


I did try to set DEC to 0 - as per your suggestion before but it wouldn't stick - it changed to 1  (is that ok?)
I don't know what you mean by setting DEC to 0.

What I said below was to set the guide aggressiveness in RA to zero in order to see what the actual periodic error is with PEM off. Then do the same with PEM on. Do each one for about 10 minutes which is just over one worm cycle. Then post the guider graph so we can see it. If your guider program cannot set aggressiveness to zero, then just pull the guider connection to the mount out so that guide signals are not sent to the mount. Otherwise what's the point of doing this?

To determine if you are properly polar aligned you can also do the same with Dec - turn off Dec corrections but record the guide star drift in your guider graph. That will tell you infinitely more than just guiding. If your Dec drift is more than 1 arc sec for a 5 minute period, then your polar alignment is not good.

Rolando


-----Original Message-----
From: Ron Kramer ronkramer1957@... [ap-gto]
To: ap-gto
Sent: Mon, Oct 30, 2017 8:39 pm
Subject: Re: [ap-gto] Guiding a mach1 - and pec?



thanks I'll try this next clear night.
I did try to set DEC to 0 - as per your suggestion before but it wouldn't stick - it changed to 1  (is that ok?)
You say do this for both axis then. Will do.

I assumed that teh "start" of the pec curve would change if he contact with the worm changed place.
I'll figure that's wrong.

Will do. Don't give me to much at once. My memory is shot. I'll likely print the above out and get that step done.



On Mon, Oct 30, 2017 at 7:24 PM, chris1011@... [ap-gto] <ap-gto@...> wrote:
 

Frustrated , I wondered about PEC.  It was on, and I read it's "recorded" at the shop and to leave it be... it's set.  Well if I disengaged the worm gears (as I do this to balance with no drag) would not the pec be "off"?
No, disengaging the worm from the worm wheel does not change the PE corrections.

I think you are grasping at straws. Please read my other post about polar alignment - try to get it as good as possible so that the guiding program does not need to fight your misalignment drift.

PLEASE read the following and try it (for your sake and the rest of us who are trying to help):

If you have any doubts about the tracking accuracy do this: TURN OFF the guide corrections to the mount by turning the aggressiveness to zero. Your guider will still record the motion of the guide star and will show you the slow drift back and forth of the RA axis caused by PE AND the steady drift up or down of both RA and Dec due to polar misalignment. Once you have that data, take a screen shot and save it, or post it somewhere so we can all see what your basic mount is actually doing sans the complications of the guiding software corrections. THEN do it all again but this time turn on the PE corrections. You should see a marked improvement of the RA wiggles back and forth because the PEM will be correcting most of that and you should have only the steady drift due to polar misalignment. Both of these tests should take 10 minutes at most (one complete worm cycle).

Now you actually have something to work with instead of the performance below that is mixed up with corrections that may or may not be correct. The other advantage to the above test is that it can show you exactly what to do to precision align your mount on the pole. If you want, I can explain again how that is done, and believe it or not, it is really quite simple and straight forward. It is well worth doing drift alignment and need not be difficult at all and take very little time.

Rolando


-----Original Message-----
From: ronkramer1957@... [ap-gto] <ap-gto@...>
To: ap-gto <ap-gto@...>
Sent: Mon, Oct 30, 2017 6:09 pm
Subject: [ap-gto] Guiding a mach1 - and pec?



I've now had my 2nd night of stars since my Mach1 arrived. I wanted to concentrate on guiding (since that is why I upgraded from the NEQ6).

I carefully balanced the scope.  Calibrated (PHD2) near meridian/ecliptic.

guiding didn't look good at that location. I moved to another area and wasn't much better. 

Seeing I suspect was horrible. (I Could see a hazy icy ring out around the moon). Maybe that's why it was bad?   So I tried guide assistant in PHD2 (I don't think it uses seeing as it doesn't do actual guiding)? 

The results I think? were very good? 


Hoping my photo will appear above?

Though that seemed to report good things, guiding was very abrupt and lots of zig zags.  Like OVER correcting...  Though I ended up with Dec agression at  9.  (yea... 9).  Could that be normal? 

I was doing 4 second exposures as was recommended here.  HOWEVER I find it very difficult to do this or longer as my camera's gain is already turned down to 10 in order not to over saturate the guide star (and I still get saturation warnings at that).  Any other way to turn down sensitivity so allow 5 second exposures as recommended by Roland?   (ASI290MM)  ND filter added to camera? 

Frustrated , I wondered about PEC.  It was on, and I read it's "recorded" at the shop and to leave it be... it's set.  Well if I disengaged the worm gears (as I do this to balance with no drag) would not the pec be "off"?  Turning pec OFF - looked better...  I did then click RECORD and watched guiding which was a bit better without PEC.  Later I noticed it was engaged again. (once recorded does it then apply that recorded error?)  

I then was getting a lot of star lost errors (I think high thin clouds) as I was in a small gap in days of rain.  Advice?  Wait and try again? the GA feedback is good?   what about PEC.  I read somewhere on here about uploading it to the mount?  Id I recorded it - didn't it auto apply it? 








Roland Christen
 

I did try to set DEC to 0 - as per your suggestion before but it wouldn't stick - it changed to 1  (is that ok?)
I don't know what you mean by setting DEC to 0.

What I said below was to set the guide aggressiveness in RA to zero in order to see what the actual periodic error is with PEM off. Then do the same with PEM on. Do each one for about 10 minutes which is just over one worm cycle. Then post the guider graph so we can see it. If your guider program cannot set aggressiveness to zero, then just pull the guider connection to the mount out so that guide signals are not sent to the mount.

To determine if you are properly polar aligned you can also do the same with Dec. Turn off Dec corrections but record the guide star drift in your guider graph. That will tell you infinitely more than just guiding. If your Dec drift is more than 1 arc sec for a 5 minute period, then your polar alignment is not good.

Rolando


-----Original Message-----
From: Ron Kramer ronkramer1957@... [ap-gto]
To: ap-gto
Sent: Mon, Oct 30, 2017 8:39 pm
Subject: Re: [ap-gto] Guiding a mach1 - and pec?



thanks I'll try this next clear night.
I did try to set DEC to 0 - as per your suggestion before but it wouldn't stick - it changed to 1  (is that ok?)
You say do this for both axis then. Will do.

I assumed that teh "start" of the pec curve would change if he contact with the worm changed place.
I'll figure that's wrong.

Will do. Don't give me to much at once. My memory is shot. I'll likely print the above out and get that step done.



On Mon, Oct 30, 2017 at 7:24 PM, chris1011@... [ap-gto] <ap-gto@...> wrote:
 

Frustrated , I wondered about PEC.  It was on, and I read it's "recorded" at the shop and to leave it be... it's set.  Well if I disengaged the worm gears (as I do this to balance with no drag) would not the pec be "off"?
No, disengaging the worm from the worm wheel does not change the PE corrections.

I think you are grasping at straws. Please read my other post about polar alignment - try to get it as good as possible so that the guiding program does not need to fight your misalignment drift.

PLEASE read the following and try it (for your sake and the rest of us who are trying to help):

If you have any doubts about the tracking accuracy do this: TURN OFF the guide corrections to the mount by turning the aggressiveness to zero. Your guider will still record the motion of the guide star and will show you the slow drift back and forth of the RA axis caused by PE AND the steady drift up or down of both RA and Dec due to polar misalignment. Once you have that data, take a screen shot and save it, or post it somewhere so we can all see what your basic mount is actually doing sans the complications of the guiding software corrections. THEN do it all again but this time turn on the PE corrections. You should see a marked improvement of the RA wiggles back and forth because the PEM will be correcting most of that and you should have only the steady drift due to polar misalignment. Both of these tests should take 10 minutes at most (one complete worm cycle).

Now you actually have something to work with instead of the performance below that is mixed up with corrections that may or may not be correct. The other advantage to the above test is that it can show you exactly what to do to precision align your mount on the pole. If you want, I can explain again how that is done, and believe it or not, it is really quite simple and straight forward. It is well worth doing drift alignment and need not be difficult at all and take very little time.

Rolando


-----Original Message-----
From: ronkramer1957@... [ap-gto] <ap-gto@...>
To: ap-gto <ap-gto@...>
Sent: Mon, Oct 30, 2017 6:09 pm
Subject: [ap-gto] Guiding a mach1 - and pec?



I've now had my 2nd night of stars since my Mach1 arrived. I wanted to concentrate on guiding (since that is why I upgraded from the NEQ6).

I carefully balanced the scope.  Calibrated (PHD2) near meridian/ecliptic.

guiding didn't look good at that location. I moved to another area and wasn't much better. 

Seeing I suspect was horrible. (I Could see a hazy icy ring out around the moon). Maybe that's why it was bad?   So I tried guide assistant in PHD2 (I don't think it uses seeing as it doesn't do actual guiding)? 

The results I think? were very good? 


Hoping my photo will appear above?

Though that seemed to report good things, guiding was very abrupt and lots of zig zags.  Like OVER correcting...  Though I ended up with Dec agression at  9.  (yea... 9).  Could that be normal? 

I was doing 4 second exposures as was recommended here.  HOWEVER I find it very difficult to do this or longer as my camera's gain is already turned down to 10 in order not to over saturate the guide star (and I still get saturation warnings at that).  Any other way to turn down sensitivity so allow 5 second exposures as recommended by Roland?   (ASI290MM)  ND filter added to camera? 

Frustrated , I wondered about PEC.  It was on, and I read it's "recorded" at the shop and to leave it be... it's set.  Well if I disengaged the worm gears (as I do this to balance with no drag) would not the pec be "off"?  Turning pec OFF - looked better...  I did then click RECORD and watched guiding which was a bit better without PEC.  Later I noticed it was engaged again. (once recorded does it then apply that recorded error?)  

I then was getting a lot of star lost errors (I think high thin clouds) as I was in a small gap in days of rain.  Advice?  Wait and try again? the GA feedback is good?   what about PEC.  I read somewhere on here about uploading it to the mount?  Id I recorded it - didn't it auto apply it? 








Ron Kramer
 

Will do. (waiting for stars). RA to 0.  If I can't set 0.  If I can't,  I do not have a guide cable.  I have a guide camera, USB to computer and I believe the guiding is then passed
along through software to phd2 and PHD2 controls the mount.  Not sure how else I can 'cut the cord' so to speak.
Sorry for being so naive about this stuff - I just started in this last April. Guiding was around July. Then all new mount and software a month ago.



On Tue, Oct 31, 2017 at 1:11 PM, chris1011@... [ap-gto] <ap-gto@...> wrote:
 

I did try to set DEC to 0 - as per your suggestion before but it wouldn't stick - it changed to 1  (is that ok?)
I don't know what you mean by setting DEC to 0.

What I said below was to set the guide aggressiveness in RA to zero in order to see what the actual periodic error is with PEM off. Then do the same with PEM on. Do each one for about 10 minutes which is just over one worm cycle. Then post the guider graph so we can see it. If your guider program cannot set aggressiveness to zero, then just pull the guider connection to the mount out so that guide signals are not sent to the mount.

To determine if you are properly polar aligned you can also do the same with Dec. Turn off Dec corrections but record the guide star drift in your guider graph. That will tell you infinitely more than just guiding. If your Dec drift is more than 1 arc sec for a 5 minute period, then your polar alignment is not good.

Rolando


-----Original Message-----
From: Ron Kramer ronkramer1957@... [ap-gto] <ap-gto@...>
To: ap-gto <ap-gto@...>
Sent: Mon, Oct 30, 2017 8:39 pm
Subject: Re: [ap-gto] Guiding a mach1 - and pec?



thanks I'll try this next clear night.
I did try to set DEC to 0 - as per your suggestion before but it wouldn't stick - it changed to 1  (is that ok?)
You say do this for both axis then. Will do.

I assumed that teh "start" of the pec curve would change if he contact with the worm changed place.
I'll figure that's wrong.

Will do. Don't give me to much at once. My memory is shot. I'll likely print the above out and get that step done.



On Mon, Oct 30, 2017 at 7:24 PM, chris1011@... [ap-gto] <ap-gto@...> wrote:
 

Frustrated , I wondered about PEC.  It was on, and I read it's "recorded" at the shop and to leave it be... it's set.  Well if I disengaged the worm gears (as I do this to balance with no drag) would not the pec be "off"?
No, disengaging the worm from the worm wheel does not change the PE corrections.

I think you are grasping at straws. Please read my other post about polar alignment - try to get it as good as possible so that the guiding program does not need to fight your misalignment drift.

PLEASE read the following and try it (for your sake and the rest of us who are trying to help):

If you have any doubts about the tracking accuracy do this: TURN OFF the guide corrections to the mount by turning the aggressiveness to zero. Your guider will still record the motion of the guide star and will show you the slow drift back and forth of the RA axis caused by PE AND the steady drift up or down of both RA and Dec due to polar misalignment. Once you have that data, take a screen shot and save it, or post it somewhere so we can all see what your basic mount is actually doing sans the complications of the guiding software corrections. THEN do it all again but this time turn on the PE corrections. You should see a marked improvement of the RA wiggles back and forth because the PEM will be correcting most of that and you should have only the steady drift due to polar misalignment. Both of these tests should take 10 minutes at most (one complete worm cycle).

Now you actually have something to work with instead of the performance below that is mixed up with corrections that may or may not be correct. The other advantage to the above test is that it can show you exactly what to do to precision align your mount on the pole. If you want, I can explain again how that is done, and believe it or not, it is really quite simple and straight forward. It is well worth doing drift alignment and need not be difficult at all and take very little time.

Rolando


-----Original Message-----
From: ronkramer1957@... [ap-gto] <ap-gto@...>
To: ap-gto <ap-gto@...>
Sent: Mon, Oct 30, 2017 6:09 pm
Subject: [ap-gto] Guiding a mach1 - and pec?



I've now had my 2nd night of stars since my Mach1 arrived. I wanted to concentrate on guiding (since that is why I upgraded from the NEQ6).

I carefully balanced the scope.  Calibrated (PHD2) near meridian/ecliptic.

guiding didn't look good at that location. I moved to another area and wasn't much better. 

Seeing I suspect was horrible. (I Could see a hazy icy ring out around the moon). Maybe that's why it was bad?   So I tried guide assistant in PHD2 (I don't think it uses seeing as it doesn't do actual guiding)? 

The results I think? were very good? 


Hoping my photo will appear above?

Though that seemed to report good things, guiding was very abrupt and lots of zig zags.  Like OVER correcting...  Though I ended up with Dec agression at  9.  (yea... 9).  Could that be normal? 

I was doing 4 second exposures as was recommended here.  HOWEVER I find it very difficult to do this or longer as my camera's gain is already turned down to 10 in order not to over saturate the guide star (and I still get saturation warnings at that).  Any other way to turn down sensitivity so allow 5 second exposures as recommended by Roland?   (ASI290MM)  ND filter added to camera? 

Frustrated , I wondered about PEC.  It was on, and I read it's "recorded" at the shop and to leave it be... it's set.  Well if I disengaged the worm gears (as I do this to balance with no drag) would not the pec be "off"?  Turning pec OFF - looked better...  I did then click RECORD and watched guiding which was a bit better without PEC.  Later I noticed it was engaged again. (once recorded does it then apply that recorded error?)  

I then was getting a lot of star lost errors (I think high thin clouds) as I was in a small gap in days of rain.  Advice?  Wait and try again? the GA feedback is good?   what about PEC.  I read somewhere on here about uploading it to the mount?  Id I recorded it - didn't it auto apply it? 









Andy Galasso
 

On Tue, Oct 31, 2017 at 2:16 PM, Ron Kramer ronkramer1957@... [ap-gto] <ap-gto@...> wrote:
 

Not sure how else I can 'cut the cord' so to speak.

In PHD2 there's a check-box labeled "Enable mount guide output" on the Guiding tab of the Advanced settings window (click the brain button).  You can leave your aggressiveness setting alone and just use the check-box to "cut the cord".  Also, the Guiding Assistant in PHD2 will automatically disable guide output and re-enable it when it is done. After running the Guiding Assistant the PHD2 Guide Log will have the un-guided tracking information that Rolando is looking for. Either approach--running the Guiding Assistant or manually clearing the check-box-- will get the un-guided tracking data into the log, and you'll also see it on the graph window display.

BTW, if you were not already aware of it there is a dedicated PHD2 Forum where you can go for PHD2-related questions.

Andy


Eric Claeys
 

I tried Roland's suggestion to check my Mach 1's tracking accuracy.

All runs were with a C11 with mirrors locked and the scope pointing near the zenith.

Camera was rotated 0 degrees (per a plate solve).


With PEM off, the results look pretty bad: +/1 about 6" in RA.


With PEM on, the results are significantly WORSE.  I ran PEMPro earlier (pointing near the meridian and equator, per the instructions).


All screenshots are in the "EricC" folder, called "PE and PEMPro runs.zip":

  • PE run - 1 PEM off, APPC settings: shows APPC and 3D view, prior to starting the PE run per Roland's instructions.
  • PE run - 2 PEM off - PHP graph: PHP graph with PEM turned off.  Roughly 13 minutes.  DEC was doing ok until about half way then went down.
  • PE run - 3 PEM on - PHP graph: the peaks go to about 12" (I left the scale the same as in the prior run).

Screenshots of my PEMPro runs, before and after sending to the mount:

  • PEMPro - 1 Acquire Data, 3 cycles: shot of the "Acquire Data" tab.
  • PEMPro - 2 Analyze - Create PEC Curve: after clicking on "Create PE Curve" in the "Analyze" tab, and changing the Drift Fitting to get the lowest RMS error, then checking all Frequencies below 16 that had at least 0.05 amplitude.  Periodic error is +4.7/-3.6 which is outside the +/-3.5 advertised.
  • PEMPro - 3 AP Raw PE: the curve downloaded to the mount.
  • the next 3 shots (PEMPro - 4, 5, and 6) are the same but with PEC on (and NOT loaded to the mount).

Given that the new PEMPro curve made things worse, I assume I did something wrong (although I followed Roland's instructions).  RA goes left and right in the pictures (if I take a picture when the mount is parked the streaks go right and left).


Any ideas?


Eric



Ray Gralak
 

Hi Eric,

Any ideas?
Yes. :-) I think you need some corrections in the procedure you used.

I ran PEMPro earlier (pointing near the meridian and equator, per the instructions).
First, you only sampled about 3 worm cycles. Mathematically, the longer duration you use the more accurate the analysis results can be. There is a fundamental relationship between duration and frequency/phase/amplitude accuracy that cannot be avoided by any method. Too short duration can produce a poor fit, or even a wrong fit.

Also, with shorter durations you increase the likelihood of false frequencies in the analysis so I recommend at least 5-6 worm cycles for A-P mounts. Up to a point where drift becomes irregular, the more worm cycles collected the more accurate the frequency/phase/amplitude fit will be.

changing the Drift Fitting to get the lowest RMS error,
You want to increase drift fitting until it hardly changes, then back off one level.

then checking all Frequencies below 16 that had at least 0.05 amplitude
Wrong! Use the "Known frequencies only" checkbox and select your mount type. From your images it looks like you wrongly selected some false frequencies.

Given that the new PEMPro curve made things worse, I assume I did something wrong
One of your images shows that you need to invert the PEC curve because measured PE afterwards is worse. This can happen if you didn't run PEMPro's calibration wizard. Nonetheless, I think you should start over, make sure to run the calibration wizard, and collect 5-6 worm cycles.

Best regards,

-Ray Gralak
Author of APCC (Astro-Physics Command Center): http://www.astro-physics.com/index.htm?products/accessories/software/apcc/apcc
Author of PEMPro: http://www.ccdware.com
Author of Astro-Physics V2 ASCOM Driver: http://www.gralak.com/apdriver
Author of PulseGuide: http://www.pulseguide.com
Author of Sigma: http://www.gralak.com/sigma

-----Original Message-----
From: ap-gto@... [mailto:ap-gto@...]
Sent: Thursday, November 2, 2017 12:16 AM
To: ap-gto@...
Subject: [ap-gto] Re: Guiding a mach1 - and pec?



I tried Roland's suggestion to check my Mach 1's tracking accuracy.

All runs were with a C11 with mirrors locked and the scope pointing near the zenith.

Camera was rotated 0 degrees (per a plate solve).




With PEM off, the results look pretty bad: +/1 about 6" in RA.




With PEM on, the results are significantly WORSE. I ran PEMPro earlier (pointing near the meridian and
equator, per the instructions).




All screenshots are in the "EricC" folder, called "PE and PEMPro runs.zip":

* PE run - 1 PEM off, APPC settings: shows APPC and 3D view, prior to starting the PE run per Roland's
instructions.
* PE run - 2 PEM off - PHP graph: PHP graph with PEM turned off. Roughly 13 minutes. DEC was
doing ok until about half way then went down.
* PE run - 3 PEM on - PHP graph: the peaks go to about 12" (I left the scale the same as in the prior run).

Screenshots of my PEMPro runs, before and after sending to the mount:

*
PEMPro - 1 Acquire Data, 3 cycles: shot of the "Acquire Data" tab.
*
PEMPro - 2 Analyze - Create PEC Curve: after clicking on "Create PE Curve" in the "Analyze" tab, and
changing the Drift Fitting to get the lowest RMS error, then checking all Frequencies below 16 that had at least
0.05 amplitude. Periodic error is +4.7/-3.6 which is outside the +/-3.5 advertised.
*
PEMPro - 3 AP Raw PE: the curve downloaded to the mount.
*
the next 3 shots (PEMPro - 4, 5, and 6) are the same but with PEC on (and NOT loaded to the mount).

Given that the new PEMPro curve made things worse, I assume I did something wrong (although I followed
Roland's instructions). RA goes left and right in the pictures (if I take a picture when the mount is parked the
streaks go right and left).




Any ideas?




Eric





Roland Christen
 

Hi Eric,

I looked at your curves of the PEMPro runs and can see immediately that you added the Periodic error backwards (180 degrees out of phase) to the mount. So instead of subtracting the error, you are actually doubling it. I don't know how this might happen, but it is a PEMPro problem, not something the mount electronics did. Ray Gralack can probably help you on this one.

Rolando



-----Original Message-----
From: AstroEric@... [ap-gto]
To: ap-gto Sent: Thu, Nov 2, 2017 2:16 am
Subject: [ap-gto] Re: Guiding a mach1 - and pec?



I tried Roland's suggestion to check my Mach 1's tracking accuracy.
All runs were with a C11 with mirrors locked and the scope pointing near the zenith.
Camera was rotated 0 degrees (per a plate solve).

With PEM off, the results look pretty bad: +/1 about 6" in RA.

With PEM on, the results are significantly WORSE.  I ran PEMPro earlier (pointing near the meridian and equator, per the instructions).

All screenshots are in the "EricC" folder, called "PE and PEMPro runs.zip":
  • PE run - 1 PEM off, APPC settings: shows APPC and 3D view, prior to starting the PE run per Roland's instructions.
  • PE run - 2 PEM off - PHP graph: PHP graph with PEM turned off.  Roughly 13 minutes.  DEC was doing ok until about half way then went down.
  • PE run - 3 PEM on - PHP graph: the peaks go to about 12" (I left the scale the same as in the prior run).
Screenshots of my PEMPro runs, before and after sending to the mount:
  • PEMPro - 1 Acquire Data, 3 cycles: shot of the "Acquire Data" tab.
  • PEMPro - 2 Analyze - Create PEC Curve: after clicking on "Create PE Curve" in the "Analyze" tab, and changing the Drift Fitting to get the lowest RMS error, then checking all Frequencies below 16 that had at least 0.05 amplitude.  Periodic error is +4.7/-3.6 which is outside the +/-3.5 advertised.
  • PEMPro - 3 AP Raw PE: the curve downloaded to the mount.
  • the next 3 shots (PEMPro - 4, 5, and 6) are the same but with PEC on (and NOT loaded to the mount ).
Given that the new PEMPro curve made things worse, I assume I did something wrong (although I followed Roland's instructions).  RA goes left and right in the pictures (if I take a picture when the mount is parked the streaks go right and left).

Any ideas?

Eric




Roland Christen
 


> then checking all Frequencies below 16 that had at least 0.05 amplitude

Wrong! Use the "Known frequencies only" checkbox and select your mount type. From your images it looks like you wrongly selected some false frequencies.
Hi Ray,

I'm the person that suggested the above. I have done a number of experiments and found that checking frequencies below 16th with more than .05 does indeed result in a better PE curve for our mounts, especially the Mach1.

Rolando


-----Original Message-----
From: 'Ray Gralak (Groups)' groups3@... [ap-gto]
To: ap-gto
Sent: Thu, Nov 2, 2017 7:50 am
Subject: RE: [ap-gto] Re: Guiding a mach1 - and pec?

Hi Eric,

> Any ideas?

Yes. :-) I think you need some corrections in the procedure you used.

> I ran PEMPro earlier (pointing near the meridian and equator, per the instructions).

First, you only sampled about 3 worm cycles. Mathematically, the longer duration you use the more accurate the analysis results can be. There is a fundamental relationship between duration and frequency/phase/amplitude accuracy that cannot be avoided by any method. Too short duration can produce a poor fit, or even a wrong fit.

Also, with shorter durations you increase the likelihood of false frequencies in the analysis so I recommend at least 5-6 worm cycles for A-P mounts. Up to a point where drift becomes irregular, the more worm cycles collected the more accurate the frequency/phase/amplitude fit will be.

> changing the Drift Fitting to get the lowest RMS error,

You want to increase drift fitting until it hardly changes, then back off one level.

> then checking all Frequencies below 16 that had at least 0.05 amplitude

Wrong! Use the "Known frequencies only" checkbox and select your mount type. From your images it looks like you wrongly selected some false frequencies.

> Given that the new PEMPro curve made things worse, I assume I did something wrong

One of your images shows that you need to invert the PEC curve because measured PE afterwards is worse. This can happen if you didn't run PEMPro's calibration wizard. Nonetheless, I think you should start over, make sure to run the calibration wizard, and collect 5-6 worm cycles.

Best regards,

-Ray Gralak
Author of APCC (Astro-Physics Command Center): http://www.astro-physics.com/index.htm?products/accessories/software/apcc/apcc
Author of PEMPro: http://www.ccdware.com
Author of Astro-Physics V2 ASCOM Driver: http://www.gralak.com/apdriver
Author of PulseGuide: http://www.pulseguide.com
Author of Sigma: http://www.gralak.com/sigma

> -----Original Message-----
> From: ap-gto@... [mailto:ap-gto@...]
> Sent: Thursday, November 2, 2017 12:16 AM
> To: ap-gto@...
> Subject: [ap-gto] Re: Guiding a mach1 - and pec?
>
>
>
> I tried Roland's suggestion to check my Mach 1's tracking accuracy.
>
> All runs were with a C11 with mirrors locked and the scope pointing near the zenith.
>
> Camera was rotated 0 degrees (per a plate solve).
>
>
>
>
> With PEM off, the results look pretty bad: +/1 about 6" in RA.
>
>
>
>
> With PEM on, the results are significantly WORSE. I ran PEMPro earlier (pointing near the meridian and
> equator, per the instructions).
>
>
>
>
> All screenshots are in the "EricC" folder, called "PE and PEMPro runs.zip":
>
> * PE run - 1 PEM off, APPC settings: shows APPC and 3D view, prior to starting the PE run per Roland's
> instructions.
> * PE run - 2 PEM off - PHP graph: PHP graph with PEM turned off. Roughly 13 minutes. DEC was
> doing ok until about half way then went down.
> * PE run - 3 PEM on - PHP graph: the peaks go to about 12" (I left the scale the same as in the prior run).
>
> Screenshots of my PEMPro runs, before and after sending to the mount:
>
> *
> PEMPro - 1 Acquire Data, 3 cycles: shot of the "Acquire Data" tab.
> *
> PEMPro - 2 Analyze - Create PEC Curve: after clicking on "Create PE Curve" in the "Analyze" tab, and
> changing the Drift Fitting to get the lowest RMS error, then checking all Frequencies below 16 that had at least
> 0.05 amplitude. Periodic error is +4.7/-3.6 which is outside the +/-3.5 advertised.
> *
> PEMPro - 3 AP Raw PE: the curve downloaded to the mount.
> *
> the next 3 shots (PEMPro - 4, 5, and 6) are the same but with PEC on (and NOT loaded to the mount).
>
> Given that the new PEMPro curve made things worse, I assume I did something wrong (although I followed
> Roland's instructions). RA goes left and right in the pictures (if I take a picture when the mount is parked the
> streaks go right and left).
>
>
>
>
> Any ideas?
>
>
>
>
> Eric
>
>
>
>
>



------------------------------------
Posted by: "Ray Gralak \(Groups\)" <groups3@...>
------------------------------------

To UNSUBSCRIBE, or for general information on the ap-gto list
see http://groups.yahoo.com/group/ap-gto
------------------------------------

Yahoo Groups Links

<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/ap-gto/

<*> Your email settings:
Individual Email | Traditional

<*> To change settings online go to:
http://groups.yahoo.com/group/ap-gto/join
(Yahoo! ID required)

<*> To change settings via email:
ap-gto-digest@...
ap-gto-fullfeatured@...

<*> To unsubscribe from this group, send an email to:
ap-gto-unsubscribe@...

<*> Your use of Yahoo Groups is subject to:
https://info.yahoo.com/legal/us/yahoo/utos/terms/


Ray Gralak
 

Hi Roland,

Unfortunately FFTs are not perfect so one must be careful in selecting frequencies with low amplitudes. There is a tradeoff in time versus frequency accuracy so an FFT for data spanning 30 minutes can show an amplitude for a frequency that occurred for just a fraction of that time, but does not repeat. For instance vibration from , walking near the mount, or another source, or even tube currents can add rogue frequencies to the spectrum.

One way to confirm that rogue frequencies are real is to compare the results after uploading a pec curve to results with pec disabled. If the amplitude of a rogue frequency does not improve then it could have been caused by a momentary instance of that frequency or aliasing from a higher frequency, and the reason it hasn't changed is that the new pec curve just added it, making the mount unnecessarily oscillate at that frequency. More often than not I have seen and "demoted" such rogue frequencies to "false" frequencies.

-Ray Gralak
Author of APCC (Astro-Physics Command Center): http://www.astro-physics.com/index.htm?products/accessories/software/apcc/apcc
Author of PEMPro: http://www.ccdware.com
Author of Astro-Physics V2 ASCOM Driver: http://www.gralak.com/apdriver
Author of PulseGuide: http://www.pulseguide.com
Author of Sigma: http://www.gralak.com/sigma

-----Original Message-----
From: ap-gto@... [mailto:ap-gto@...]
Sent: Thursday, November 2, 2017 10:01 AM
To: ap-gto@...
Subject: Re: [ap-gto] Re: Guiding a mach1 - and pec?




> then checking all Frequencies below 16 that had at least 0.05 amplitude

Wrong! Use the "Known frequencies only" checkbox and select your mount type. From your images it
looks like you wrongly selected some false frequencies.

Hi Ray,


I'm the person that suggested the above. I have done a number of experiments and found that checking
frequencies below 16th with more than .05 does indeed result in a better PE curve for our mounts, especially
the Mach1.


Rolando



-----Original Message-----
From: 'Ray Gralak (Groups)' groups3@... [ap-gto] <ap-gto@...>
To: ap-gto <ap-gto@...>
Sent: Thu, Nov 2, 2017 7:50 am
Subject: RE: [ap-gto] Re: Guiding a mach1 - and pec?

Hi Eric,

Any ideas?
Yes. :-) I think you need some corrections in the procedure you used.

I ran PEMPro earlier (pointing near the meridian and equator, per the instructions).
First, you only sampled about 3 worm cycles. Mathematically, the longer duration you use the more accurate
the analysis results can be. There is a fundamental relationship between duration and
frequency/phase/amplitude accuracy that cannot be avoided by any method. Too short duration can produce
a poor fit, or even a wrong fit.

Also, with shorter durations you increase the likelihood of false frequencies in the analysis so I recommend at
least 5-6 worm cycles for A-P mounts. Up to a point where drift becomes irregular, the more worm cycles
collected the more accurate the frequency/phase/amplitude fit will be.

changing the Drift Fitting to get the lowest RMS error,
You want to increase drift fitting until it hardly changes, then back off one level.

then checking all Frequencies below 16 that had at least 0.05 amplitude
Wrong! Use the "Known frequencies only" checkbox and select your mount type. From your images it looks
like you wrongly selected some false frequencies.

Given that the new PEMPro curve made things worse, I assume I did something wrong
One of your images shows that you need to invert the PEC curve because measured PE afterwards is worse.
This can happen if you didn't run PEMPro's calibration wizard. Nonetheless, I think you should start over,
make sure to run the calibration wizard, and collect 5-6 worm cycles.

Best regards,

-Ray Gralak
Author of APCC (Astro-Physics Command Center): http://www.astro-
physics.com/index.htm?products/accessories/software/apcc/apcc
Author of PEMPro: http://www.ccdware.com
Author of Astro-Physics V2 ASCOM Driver: http://www.gralak.com/apdriver
Author of PulseGuide: http://www.pulseguide.com
Author of Sigma: http://www.gralak.com/sigma

-----Original Message-----
From: ap-gto@... [mailto:ap-gto@... <mailto:ap-gto@...?> ]
Sent: Thursday, November 2, 2017 12:16 AM
To: ap-gto@...
Subject: [ap-gto] Re: Guiding a mach1 - and pec?



I tried Roland's suggestion to check my Mach 1's tracking accuracy.

All runs were with a C11 with mirrors locked and the scope pointing near the zenith.

Camera was rotated 0 degrees (per a plate solve).




With PEM off, the results look pretty bad: +/1 about 6" in RA.




With PEM on, the results are significantly WORSE. I ran PEMPro earlier (pointing near the meridian and
equator, per the instructions).




All screenshots are in the "EricC" folder, called "PE and PEMPro runs.zip":

* PE run - 1 PEM off, APPC settings: shows APPC and 3D view, prior to starting the PE run per Roland's
instructions.
* PE run - 2 PEM off - PHP graph: PHP graph with PEM turned off. Roughly 13 minutes. DEC was
doing ok until about half way then went down.
* PE run - 3 PEM on - PHP graph: the peaks go to about 12" (I left the scale the same as in the prior run).

Screenshots of my PEMPro runs, before and after sending to the mount:

*
PEMPro - 1 Acquire Data, 3 cycles: shot of the "Acquire Data" tab.
*
PEMPro - 2 Analyze - Create PEC Curve: after clicking on "Create PE Curve" in the "Analyze" tab, and
changing the Drift Fitting to get the lowest RMS error, then checking all Frequencies below 16 that had at
least
0.05 amplitude. Periodic error is +4.7/-3.6 which is outside the +/-3.5 advertised.
*
PEMPro - 3 AP Raw PE: the curve downloaded to the mount.
*
the next 3 shots (PEMPro - 4, 5, and 6) are the same but with PEC on (and NOT loaded to the mount).

Given that the new PEMPro curve made things worse, I assume I did something wrong (although I followed
Roland's instructions). RA goes left and right in the pictures (if I take a picture when the mount is parked the
streaks go right and left).




Any ideas?




Eric






------------------------------------
Posted by: "Ray Gralak &#92;(Groups&#92;)" <groups3@...>
------------------------------------

To UNSUBSCRIBE, or for general information on the ap-gto list
see http://groups.yahoo.com/group/ap-gto
------------------------------------

Yahoo Groups Links






Roland Christen
 

OK, got it.

Rolando



-----Original Message-----
From: 'Ray Gralak (Groups)' groups3@... [ap-gto]
To: ap-gto
Sent: Thu, Nov 2, 2017 2:36 pm
Subject: RE: [ap-gto] Re: Guiding a mach1 - and pec?

Hi Roland,

Unfortunately FFTs are not perfect so one must be careful in selecting frequencies with low amplitudes. There is a tradeoff in time versus frequency accuracy so an FFT for data spanning 30 minutes can show an amplitude for a frequency that occurred for just a fraction of that time, but does not repeat. For instance vibration from , walking near the mount, or another source, or even tube currents can add rogue frequencies to the spectrum.

One way to confirm that rogue frequencies are real is to compare the results after uploading a pec curve to results with pec disabled. If the amplitude of a rogue frequency does not improve then it could have been caused by a momentary instance of that frequency or aliasing from a higher frequency, and the reason it hasn't changed is that the new pec curve just added it, making the mount unnecessarily oscillate at that frequency. More often than not I have seen and "demoted" such rogue frequencies to "false" frequencies.

-Ray Gralak
Author of APCC (Astro-Physics Command Center): http://www.astro-physics.com/index.htm?products/accessories/software/apcc/apcc
Author of PEMPro: http://www.ccdware.com
Author of Astro-Physics V2 ASCOM Driver: http://www.gralak.com/apdriver
Author of PulseGuide: http://www.pulseguide.com
Author of Sigma: http://www.gralak.com/sigma


> -----Original Message-----
> From: ap-gto@... [mailto:ap-gto@...]
> Sent: Thursday, November 2, 2017 10:01 AM
> To: ap-gto@...
> Subject: Re: [ap-gto] Re: Guiding a mach1 - and pec?
>
>
>
>
> > then checking all Frequencies below 16 that had at least 0.05 amplitude
>
> Wrong! Use the "Known frequencies only" checkbox and select your mount type. From your images it
> looks like you wrongly selected some false frequencies.
>
> Hi Ray,
>
>
> I'm the person that suggested the above. I have done a number of experiments and found that checking
> frequencies below 16th with more than .05 does indeed result in a better PE curve for our mounts, especially
> the Mach1.
>
>
> Rolando
>
>
>
> -----Original Message-----
> From: 'Ray Gralak (Groups)' groups3@... [ap-gto] gto@...>
> To: ap-gto gto@...>
> Sent: Thu, Nov 2, 2017 7:50 am
> Subject: RE: [ap-gto] Re: Guiding a mach1 - and pec?
>
> Hi Eric,
>
> > Any ideas?
>
> Yes. :-) I think you need some corrections in the procedure you used.
>
> > I ran PEMPro earlier (pointing near the meridian and equator, per the instructions).
>
> First, you only sampled about 3 worm cycles. Mathematically, the longer duration you use the more accurate
> the analysis results can be. There is a fundamental relationship between duration and
> frequency/phase/amplitude accuracy that cannot be avoided by any method. Too short duration can produce
> a poor fit, or even a wrong fit.
>
> Also, with shorter durations you increase the likelihood of false frequencies in the analysis so I recommend at
> least 5-6 worm cycles for A-P mounts. Up to a point where drift becomes irregular, the more worm cycles
> collected the more accurate the frequency/phase/amplitude fit will be.
>
> > changing the Drift Fitting to get the lowest RMS error,
>
> You want to increase drift fitting until it hardly changes, then back off one level.
>
> > then checking all Frequencies below 16 that had at least 0.05 amplitude
>
> Wrong! Use the "Known frequencies only" checkbox and select your mount type. From your images it looks
> like you wrongly selected some false frequencies.
>
> > Given that the new PEMPro curve made things worse, I assume I did something wrong
>
> One of your images shows that you need to invert the PEC curve because measured PE afterwards is worse.
> This can happen if you didn't run PEMPro's calibration wizard. Nonetheless, I think you should start over,
> make sure to run the calibration wizard, and collect 5-6 worm cycles.
>
> Best regards,
>
> -Ray Gralak
> Author of APCC (Astro-Physics Command Center): http://www.astro-
> physics.com/index.htm?products/accessories/software/apcc/apcc
> Author of PEMPro: http://www.ccdware.com
> Author of Astro-Physics V2 ASCOM Driver: http://www.gralak.com/apdriver
> Author of PulseGuide: http://www.pulseguide.com
> Author of Sigma: http://www.gralak.com/sigma
>
> > -----Original Message-----
> > From: ap-gto@... [mailto:ap-gto@... <mailto:ap-gto@...?> ]
> > Sent: Thursday, November 2, 2017 12:16 AM
> > To: ap-gto@...
> > Subject: [ap-gto] Re: Guiding a mach1 - and pec?
> >
> >
> >
> > I tried Roland's suggestion to check my Mach 1's tracking accuracy.
> >
> > All runs were with a C11 with mirrors locked and the scope pointing near the zenith.
> >
> > Camera was rotated 0 degrees (per a plate solve).
> >
> >
> >
> >
> > With PEM off, the results look pretty bad: +/1 about 6" in RA.
> >
> >
> >
> >
> > With PEM on, the results are significantly WORSE. I ran PEMPro earlier (pointing near the meridian and
> > equator, per the instructions).
> >
> >
> >
> >
> > All screenshots are in the "EricC" folder, called "PE and PEMPro runs.zip":
> >
> > * PE run - 1 PEM off, APPC settings: shows APPC and 3D view, prior to starting the PE run per Roland's
> > instructions.
> > * PE run - 2 PEM off - PHP graph: PHP graph with PEM turned off. Roughly 13 minutes. DEC was
> > doing ok until about half way then went down.
> > * PE run - 3 PEM on - PHP graph: the peaks go to about 12" (I left the scale the same as in the prior run).
> >
> > Screenshots of my PEMPro runs, before and after sending to the mount:
> >
> > *
> > PEMPro - 1 Acquire Data, 3 cycles: shot of the "Acquire Data" tab.
> > *
> > PEMPro - 2 Analyze - Create PEC Curve: after clicking on "Create PE Curve" in the "Analyze" tab, and
> > changing the Drift Fitting to get the lowest RMS error, then checking all Frequencies below 16 that had at
> least
> > 0.05 amplitude. Periodic error is +4.7/-3.6 which is outside the +/-3.5 advertised.
> > *
> > PEMPro - 3 AP Raw PE: the curve downloaded to the mount.
> > *
> > the next 3 shots (PEMPro - 4, 5, and 6) are the same but with PEC on (and NOT loaded to the mount).
> >
> > Given that the new PEMPro curve made things worse, I assume I did something wrong (although I followed
> > Roland's instructions). RA goes left and right in the pictures (if I take a picture when the mount is parked the
> > streaks go right and left).
> >
> >
> >
> >
> > Any ideas?
> >
> >
> >
> >
> > Eric
> >
> >
> >
> >
> >
>
>
>
> ------------------------------------
> Posted by: "Ray Gralak \(Groups\)" <groups3@...>
> ------------------------------------
>
> To UNSUBSCRIBE, or for general information on the ap-gto list
> see http://groups.yahoo.com/group/ap-gto
> ------------------------------------
>
> Yahoo Groups Links
>
>
>
>
>
>



------------------------------------
Posted by: "Ray Gralak \(Groups\)" <groups3@...>
------------------------------------

To UNSUBSCRIBE, or for general information on the ap-gto list
see http://groups.yahoo.com/group/ap-gto
------------------------------------

Yahoo Groups Links

<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/ap-gto/

<*> Your email settings:
Individual Email | Traditional

<*> To change settings online go to:
http://groups.yahoo.com/group/ap-gto/join
(Yahoo! ID required)

<*> To change settings via email:
ap-gto-digest@...
ap-gto-fullfeatured@...

<*> To unsubscribe from this group, send an email to:
ap-gto-unsubscribe@...

<*> Your use of Yahoo Groups is subject to:
https://info.yahoo.com/legal/us/yahoo/utos/terms/


calypte@...
 

"so I recommend at least 5-6 worm cycles for A-P mounts

Then did I misread something from one of Roland's past posts?  I thought he was recommending no more than three cycles.


Roland Christen
 

You did not misread, I probably did not fully explain the reasons for what I said (although I try).

If you do 3 and the curves all lie on top of each other within +- 1 arc sec with drift box checked, then you can do a few more. The problem occurs if your scope setup is not tight and you start getting mirror flop or tube flex after 3 or more cycles (20 minutes), then you will see it in the way the curves do not line up on top of each other. The more they separate, the worse your results will be. The main result is that the fundamental will not cancel fully and when you run with PEM on, there will be a small fundamental.

This is the main reason I said to limit it to 3 cycles. For most people this will be just hunky. However, for those who have rigid permanent setup and want perfection, then indeed do as many cycles as your system can handle until you see the individual curves start to depart from each other.

I know of one instance using a commercial SCT where I could not get good repeatability due to mirror flop. In that case it is useless to make a curve at all since it will most likely increase the PE error rather than reduce it.

Rolando



-----Original Message-----
From: calypte@... [ap-gto]
To: ap-gto
Sent: Thu, Nov 2, 2017 4:09 pm
Subject: RE: [ap-gto] Re: Guiding a mach1 - and pec?



"so I recommend at least 5-6 worm cycles for A-P mounts

Then did I misread something from one of Roland's past posts?  I thought he was recommending no more than three cycles.



Joe Zeglinski
 

Hi Rolando,
 
    I have been keeping my silence on this topic of PemPro PEC curve cycle overlap, but I have a concern about the discussion on curve overlap.
 
    I always do  10 cycles, and none of my curves EVER overlap, on the horizontal (time base) axis. Vertical drift seems pretty good, but I get a “basket weave” pattern from the overlapping curves, which concerns me. It is almost as if the worm period is not quite the same as the “AP default”, preset option in PemPro (version 3 or beta-4).
 
    The PemPro manual shows one sample graph, where all 10 curves very nicely overlap horizontally, with slight different drift offset vertically. I am seeing something different, no matter whether I used my AP-900 or AP-1200, with the lightweight Questar-7,  or now my heavy weight RC-14.5 neither of which can have mirror flop.   Mount/saddle “Flexure” was unlikely with the AP-900 carrying the Q7 – and I always lock down the knobs with a hex wrench.
 
    I figured that no matter at what point of the worm cycle, the tests start, surely, the curves must overlap, just based on the tooth spacing and constantly stable motor sidereal drive rate. For a constant DEC angle, should I expect a drift in tracking rate, the further the mount is away from the prime meridian starting point, after the first cycle? That might (hopefully) account for the horizontal curves misalignment, but that isn’t mentioned in the manual, and perhaps the graphed example is somehow optimized for explanation. Certainly never saw this on my AP setups.
 
    Anyway, “mirror flop” shouldn’t ever be an issue in such a very short period of just 6 worm cycles, so that can’t be a problem during any PemPro PEC run – Unless by utter coincidence, the mirror cell mechanical system instability,  happens to be exactly at the mount’s prime starting position for these runs.
 
    I sure hope my PEC  “basket weave” cycles pattern is of little consequence, but it is disconcerting to see this, in spite of the good PEC curve that eventually results,  on a good run ... mostly during calm PEC testing nights.
 
Joe


Roland Christen
 


I sure hope my PEC  “basket weave” cycles pattern is of little consequence, but it is disconcerting to see this, in spite of the good PEC curve that eventually results,  on a good run ... mostly during calm PEC testing nights.
 
If you have a good PEC curve why is this disconcerting? If the mount tracks well, what's the problem?
I have a very nice 175 EDF refractor that produces a perfect (well nothing is perfect) Airy disc at high power on calm nights, but it has a chip on the edge of one of the elements and it has a couple of dust spots - very disconcerting. But then i don't worry about it because it certainly works extremely well.

Rolando


-----Original Message-----
From: 'Joseph Zeglinski' J.Zeglinski@... [ap-gto]
To: ap-gto
Sent: Thu, Nov 2, 2017 6:07 pm
Subject: Re: [ap-gto] Re: Guiding a mach1 - and pec?



Hi Rolando,
 
    I have been keeping my silence on this topic of PemPro PEC curve cycle overlap, but I have a concern about the discussion on curve overlap.
 
    I always do  10 cycles, and none of my curves EVER overlap, on the horizontal (time base) axis. Vertical drift seems pretty good, but I get a “basket weave” pattern from the overlapping curves, which concerns me. It is almost as if the worm period is not quite the same as the “AP default”, preset option in PemPro (version 3 or beta-4).
 
    The PemPro manual shows one sample graph, where all 10 curves very nicely overlap horizontally, with slight different drift offset vertically. I am seeing something different, no matter whether I used my AP-900 or AP-1200, with the lightweight Questar-7,  or now my heavy weight RC-14.5 neither of which can have mirror flop.   Mount/saddle “Flexure” was unlikely with the AP-900 carrying the Q7 – and I always lock down the knobs with a hex wrench.
 
    I figured that no matter at what point of the worm cycle, the tests start, surely, the curves must overlap, just based on the tooth spacing and constantly stable motor sidereal drive rate. For a constant DEC angle, should I expect a drift in tracking rate, the further the mount is away from the prime meridian starting point, after the first cycle? That might (hopefully) account for the horizontal curves misalignment, but that isn’t mentioned in the manual, and perhaps the graphed example is somehow optimized for explanation. Certainly never saw this on my AP setups.
 
    Anyway, “mirror flop” shouldn’t ever be an issue in such a very short period of just 6 worm cycles, so that can’t be a problem during any PemPro PEC run – Unless by utter coincidence, the mirror cell mechanical system instability,  happens to be exactly at the mount’s prime starting position for these runs.
 
    I sure hope my PEC  “basket weave” cycles pattern is of little consequence, but it is disconcerting to see this, in spite of the good PEC curve that eventually results,  on a good run ... mostly during calm PEC testing nights.
 
Joe



Joe Zeglinski
 

Sorry Rolando,
 
    Perfectly happy with the mount – just wish my calibration software had displays that matched what was demonstrated in the software manual. Makes me wonder if I, or the equipment,  is having a problem, and the unexpected curves are a clue.. Like you, I just live with it, but it would be nice to know why there is an inconsistency.
 
Guess nobody is seeing the same thing in running their PemPro PEC curve creation.
 
Joe


Eric Claeys
 

Ray,

I followed Roland's instructions exactly (which are also at http://www.astro-physics.com/tech_support/mounts/Making_PE_Curve.pdf).  He suggested 3 curves when using the C11 SCT on the Mach 1.  I had run the calibration wizard beforehand.  His instructions didn't mention inverting the data so I didn't do that.


I just ran the calibration wizard and did a 6-cycle run with an 80mm refractor using my 2016 Mach 1 and saved the data and some screenshots in a file called "PEMPro - 2017-11-02.zip" in the EricC folder.

The .txt file contains the raw data with PEC off.  The "Analyze" graph of it is in "PEMPro - 2a Analyze.jpg".


The "Known frequencies only" checkbox isn't in the PEMPro help file (V2.99.42.0), and the AP instructions don't show the "Filters" group on the "PEMPro - Create PEC Curve" window at all, even though they used V2.99.41.0.

However, per your suggestion I checked "Known frequencies only" and selected my mount.  The list of frequencies in the "FFT Waveform Analysis" window didn't change however.

Is that the expected behavior?

I thought checking that box should only show the 1, 4, and 64.04x frequencies listed for the mount.


Anyhow, I did not check or uncheck any frequencies and created a PEC Curve which did a decent job of matching the PEC, and saved a screenshot in "PEMPro - 2b1 Analyze - Create PEC Curve - per Ray.jpg".  Periodic Error was +4.1/-3.7.


I then went back and clicked on "Show used frequencies only" which resulting in only 1x and 4x showing.  I created another PEC Curve and it was significantly worse - see "PEMPro - 2b2 Analyze - Create PEC Curve - per Ray - Show used freq. only.jpg".


I loaded the first curve (the one that fit the PE better) into the mount an re-ran PEMPro with PEC on.  Periodic Error went down to +2.0/-2.6.


Can you look at the raw data and see if you can get a better fitting curve?  If you give me the resulting .ppc file I'll load it into my mount and compare to what I used.


Thanks - Eric


Ray Gralak
 

Hi Eric,

His instructions didn't mention inverting the data so I didn't do that.
You might want to look at the help file, and also online at the link below, for the basic workflow for PEMPro:

http://www.siriusimaging.com/Help/PEMProV2new/index.html?workflow.html

The "Known frequencies only" checkbox isn' t in the PEMPro help file (V2.99.42.0), and
the AP instructions don't show the "Filters" group on the "PEMPro - Create PEC Curve"
window at all, even though they used V2.99.41.0.
The improvements to the Create PEC Curve window have not been updated in the documentation yet, but I am hoping that they aren't too hard to figure out.

However, per your suggestion I checked "Known frequencies only" and selected my mount.
The list of frequencies in the "FFT Waveform Analysis" window didn't change however.
Is that the expected behavior?
Yes it is. For performance reasons the filter list doesn't get recalculated every time an entry in the drop down lost box is changed. To effect the filter change just enable/disable the Enable checkbox.

BTW, enabling the filters can change the list of frequencies to match the mount, but it might not show a frequency if the frequency is not present or if, mathematically, the data cannot show it.

Can you look at the raw data and see if you can get a better fitting curve?
I loaded the first curve (the one that fit the PE better) into the mount an re-ran PEMPro
with PEC on. Periodic Error went down to +2.0/-2.6.
Off hand, that sounds like there is an image scale error. If you want me to look at your logs I would prefer that you create a Dropbox link as Yahoo's file management is horrible and may not accommodate large files. Make sure you include the debug logs in the debug directory.

Also, I suggest that you make sure you have followed the procedure outlined in the help file that was written specifically for A-P mounts. The procedure says it's for the GTOCP3 but the procedure for the GTOCP4 is exactly the same. The procedure has worked for literally hundreds of A-P users.

BTW, if you are using APCC Pro, you should turn off tracking rate correction.

http://www.siriusimaging.com/Help/PEMProV2new/index.html?astrophysicsgtocp3.html

Best regards,

-Ray Gralak
Author of APCC (Astro-Physics Command Center): http://www.astro-physics.com/index.htm?products/accessories/software/apcc/apcc
Author of PEMPro: http://www.ccdware.com
Author of Astro-Physics V2 ASCOM Driver: http://www.gralak.com/apdriver
Author of PulseGuide: http://www.pulseguide.com
Author of Sigma: http://www.gralak.com/sigma

From: ap-gto@... [mailto:ap-gto@...]
Sent: Friday, November 3, 2017 1:37 AM
To: ap-gto@...
Subject: [ap-gto] Re: Guiding a mach1 - and pec?


Ray,
I followed Roland's instructions exactly (which are also at http://www.astro-physics.com/tech_support/mounts/Making_PE_Curve.pdf). He suggested 3 curves when using the C11 SCT on the Mach 1. I had run the calibration wizard beforehand. His instructions didn't mention inverting the data so I didn't do that.

I just ran the calibration wizard and did a 6-cycle run with an 80mm refractor using my 2016 Mach 1 and saved the data and some screenshots in a file called "PEMPro - 2017-11-02.zip" in the EricC folder.
The .txt file contains the raw data with PEC off. The "Analyze" graph of it is in "PEMPro - 2a Analyze.jpg".

The "Known frequencies only" checkbox isn' t in the PEMPro help file (V2.99.42.0), and the AP instructions don't show the "Filters" group on the "PEMPro - Create PEC Curve" window at all, even though they used V2.99.41.0.
However, per your suggestion I checked "Known frequencies only" and selected my mount. The list of frequencies in the "FFT Waveform Analysis" window didn't change however.
Is that the expected behavior?
I thought checking that box should only show the 1, 4, and 64.04x frequencies listed for the mount.

Anyhow, I did not check or uncheck any frequencies and created a PEC Curve which did a decent job of matching the PEC, and saved a screenshot in "PEMPro - 2b1 Analyze - Create PEC Curve - per Ray.jpg". Periodic Error was +4.1/-3.7.

I then went back and clicked on "Show used frequencies only" which resulting in o nly 1x and 4x showing. I created another PEC Curve and it was significantly worse - see "PEMPro - 2b2 Analyze - Create PEC Curve - per Ray - Show used freq. only.jpg".

I loaded the first curve (the one that fit the PE better) into the mount an re-ran PEMPro with PEC on. Periodic Error went down to +2.0/-2.6.

Can you look at the raw data and see if you can get a better fitting curve? If you give me the resulting .ppc file I'll load it into my mount and compare to what I used.

Thanks - Eric

________________________________________
Posted by: AstroEric@...
________________________________________
Reply via web post

Reply to sender

Reply to group

Start a New Topic

Messages in this topic (20)
________________________________________