#1  
Old 26-01-2014, 08:32 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Odd guiding behaviour

Guiding with the PMX, RC12 and stxl integrated guide camera I am seeing odd guiding activity. The dec axis is guiding quite nicely in maximDL. The RA on the other hand looks odd. Over several minutes it looks like a sine wave travel over and then below the centre line. And; looking exactly like the position on the PEC graph (which is 3.5 arc seconds below and above the line). The PE was originally 7.6 below and above the centre line. So the correction is the right way around.

Guiding aggression has been tried from 3 to 8 and nothing made much of a difference. I have even tried it without PEC enabled. PA is out by 3.7 tics on MA and 4.7 on ME. I would not have thought this would make much of a difference, though could be wrong at 2440mm.

Stars are round but large though.

Anyone got any ideas? This is one of the last hurdles to go now.
Reply With Quote
  #2  
Old 26-01-2014, 09:59 PM
ericwbenson (Eric)
Registered User

ericwbenson is offline
 
Join Date: Sep 2009
Location: Adelaide, Australia
Posts: 209
Hi Paul,
How long are the guiding integrations? Try to keep them around 0.5 to 1 sec for testing.

Are you saying with PEC off, going from aggr 3 to 8 does nothing? Try going from aggr 0 to 5 (with PEC off to keep things simple). You should see _some_ improvement. Otherwise I would question if the guider signal is actually getting through. You do get an L shape when calibrating?

Try turning simultaneous corrections off (advanced guider settings), just in case, but I don't think it's that.

Best,
EB
Reply With Quote
  #3  
Old 26-01-2014, 11:14 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
I have had it happen where the guide corrections start to oscillate - ging +ve then -ve and it appeared to be correcting its last correction.

What it boiled down to (apart from probably a slight amount of backlash) was the scope was too well balanced. I adjusted the balance slightly so the weight made the scope tip slightly to the east to load up the gears more.

It then stopped. I think also the cam adjusmtent with the threaded tension rods probably need a bit of tweaking as you may have them a bit loose or they go loose over time. This 3 way switch, cam, threaded rod system of the PMX is one of the weak spots of the mount design and I think they may go a bit loose over time and may need tightening occassionally.

Anyway its worth a try in case its something else.

I use from 1 to 6 second guide exposures. If its windy I go 1 sec. If its still I go around 5 seconds with aggressiveness set to around 5-7.

Greg.
Reply With Quote
  #4  
Old 27-01-2014, 04:56 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Hi Paul,

Are you saying your uncorrected PE is over 7 arc-sec peak to peak. If so, you really might want to get a new worm from Software Bisque, or try the worm you already have for declination (just swap them!). 7 really seems too high. My MX, even with it's weird slew problem, had under 3 arc-sec peak to peak, and corrected to well under 1 arc-sec. It may be that something needs adjusting which would correct both your PE and guiding issue. Personally I'd have a real close look at the 1/4-20s (the two hex bolts on opposing ends of the worm). These hold the worm against the ring gear and are not too hard to adjust -though best done by two people. First loosen both bolts, then press upwards to engage the worm as deep into the gear as possible. Then simultaneously tighten both hex bolts. After that take Greg's advice and readjust the cam-stop. You shouldn't be able to feel any free play if done correctly. Also, while you are at it check the positioning of the plungers. They should be 2.25-2.5 turns out.

Re PEC....I don't know what program you used to measure and implement this but the jury is still out re TSX (in the Southern Hemisphere!). For sure PEMpro works and I've had terrific results. You might want to give the free trial a go.

Anyway, perhaps these ideas will fix up your guiding mystery.
Good luck,

Peter (mountless and waiting for SB to send back a repaired unit!)

Last edited by PRejto; 29-01-2014 at 05:08 AM.
Reply With Quote
  #5  
Old 28-01-2014, 07:58 AM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
Quote:
Originally Posted by PRejto View Post
Hi Paul,

Are you saying your uncorrected PE is over 7 arc-sec peak to peak. If so, you really might want to get a new worm from Software Bisque, or try the worm you already have for declination (just swap them!). 7 really seems too high. My MX, even with it's weird slew problem, had under 3 arc-sec peak to peak, and corrected to well under 1 arc-sec. It may be that something needs adjusting which would correct both your PE and guiding issue. Personally I'd have a real close look at the 1/4-20s (the two hex bolts on opposing ends of the worm). These hold the worm against the ring gear and are not too hard to adjust -though best done by two people. First loosen both bolts, then press upwards to engage the worm as deep into the gear as possible. Then simultaneously tighten both hex bolts. After that take Greg's advice and readjust the cam-stop. You shouldn't be able to feel any free play if done correctly. Also, while you are at it check the positioning of the plungers. They should be 2.25-2.5 turns out.

Re PEC....I don't know what program you used to measure and implement this but the jury is still out re TSX. For sure PEMpro works and I've had terrific results. You might want to give the free trial a go.

Anyway, perhaps these ideas will fix up your guiding mystery.
Good luck,

Peter (mountless and waiting for SB to send back a repaired unit!)

I was referring to the same thing not the cam stop. Those 2 bolts control the tension on the worm. Too loose and you get slop and slippage on the gears with little pressure. Too tight and you will get stalls.

As you say 2.25 to 2.5 turns out from fully closed is SB advice for tension.

In the real world that is just an estimate and you need to do practice slews with various tensions until its just before it starts to stall sometimes. It feels firm, no gear slippage under reasonable pressure and little backlash with no oscillating guiding.

Greg.
Reply With Quote
  #6  
Old 29-01-2014, 02:59 AM
frolinmod's Avatar
frolinmod
Registered User

frolinmod is offline
 
Join Date: Dec 2010
Posts: 573
Whoa, if raw uncorrected pe was originally 7+ as above and below (14+ peak to peak) then you're waaay out of specifications and either need some adjustments or need a new worm. SB say raw uncorrected pe should be no more than 7 as peak to peak in order to be within acceptable specifications. Corrected pe should be sub-as.

What's the raw uncorrected pe over on your me? Since you have both me and mx, you can compare them to see which has the better worm. Might be an interesting comparison.

If your mx worm is really that bad, if I were you I'd ask SB to replace it under warranty. :-)
Reply With Quote
  #7  
Old 29-01-2014, 05:25 AM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Paul,

Saw your post at SB. Re the graph you posted there, is that graph raw PE data or a graph of "corrected" PE? As Frolinmod pointed out, and I missed it when I read your first post here (possibly semantic misinterpretation) did you really have over 14 arc-sec Peak to peak before correction? I'm not sure that is what you meant but what you wrote could certainly be read that way! If so that is a huge error. The bottom line, however, is that corrected 3.5 arc-sec is just not good enough. If you have achieved good PE measurements and the correction is applied properly you should be able to do much better. There was absolutely some sort of Southern Hemisphere phase issue with TSX properly applying PE corrections (but, only in the Southern Hemisphere). This was supposed to have been fixed months ago but I have yet to read reports confirming that it now works OK. I've been content to continue using PEMpro because it works and I've been lazy (and distracted by other issues). Perhaps another TSX user in the S.H. will confirm if they have had success with PEC/TSX.

Peter
Reply With Quote
  #8  
Old 29-01-2014, 05:53 AM
frolinmod's Avatar
frolinmod
Registered User

frolinmod is offline
 
Join Date: Dec 2010
Posts: 573
And don't forget to make sure you have already applied the latest daily build (Help->About TheSkyX showing at least build 7398 or higher) before testing it.
Reply With Quote
  #9  
Old 29-01-2014, 06:42 AM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Thanks guys for the thoughts.

The attachment is the raw data. The data is only smoothed and not applied. The camera was at around 9.9 degrees and within tolerance fro PE collection. Apparently the PE is within spec. I was wrong in that I was looking at the RAW which was sliding away down hill due to the PA of the mount. Still the main problem here is that the PEC is not being applied and the guiding remains crap at best in the RA. I will check to see if PEC is worse or better but I took this data on Canopus when it was just coming to culmination and the scope was pointed in the East.

Yes Ernie I have the latest daily build. Ernie the Raw on the ME is a freak and does not have PEC applied. It is from top to bottom 1.8 seconds. Meaning 0.9 peak to peak. I don't know why it is that good but I am not going to mess with something that works. It guides like a demon without any PEC or Protrack applied. So it is an unfair comparison.
Attached Thumbnails
Click for full-size image (PEC curve.jpg)
106.3 KB65 views
Reply With Quote
  #10  
Old 29-01-2014, 08:12 AM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
Paul there have been several threads about PEC corrections on PMX using the Sky X. Mine works but others haven't. I am not sure mine is working as well as it should but have no attention on it as I am getting round stars at moderate focal length.

I would forget using Sky X for PEC and just go with Pempro.

Greg.
Reply With Quote
  #11  
Old 29-01-2014, 11:15 AM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
So, Paul, why is your polar alignment off so much that things are "sliding downhill." Isn't that just going to make your guiding even tougher?

Perhaps off topic but I have lately become quite enamoured of PA using PEMpro. The drift routine works really well and you can make adjustments in real time. I do this right around dec=0, then run a large T-Point model and ignore the PA report. Run the super model and you get great pointing and good Protrack. I'm not saying anything is wrong with the T-Point PA report. Only that it takes some time to run, make an adjustment, run to confirm, run another adjustment, etc...then finally you run a really large model and possibly find out it's saying something different about PA. As long as PA is close (and it's a compromise no matter how you do it) T-Point will do the rest.

For me, with a very obstructed view to the east, this routine gives me quite consistent results. Some testing I did showed me that my PA report out of T-Point would change quite a bit depending on how far to the west I collected points. When I collected points eqidistant from the meridian the T-Point report was quite similar to the PEMpro drift alignment. Even the author of T-Point indicated that setting PA in this manner (i.e., drift aligning around dec=0) was a reasonable compromise.

Peter
Reply With Quote
  #12  
Old 29-01-2014, 04:27 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Quote:
Originally Posted by gregbradley View Post
Paul there have been several threads about PEC corrections on PMX using the Sky X. Mine works but others haven't. I am not sure mine is working as well as it should but have no attention on it as I am getting round stars at moderate focal length.

I would forget using Sky X for PEC and just go with Pempro.

Greg.

Hmmm so that means it might not work at all??? I might have to invest in pempro.

I am getting round stars but the guiding looks like it is following the PE curve in RA. I do need to test to see if the PEC has taken effect and do another run on the PEC.

Quote:
Originally Posted by PRejto View Post
So, Paul, why is your polar alignment off so much that things are "sliding downhill." Isn't that just going to make your guiding even tougher?

Perhaps off topic but I have lately become quite enamoured of PA using PEMpro. The drift routine works really well and you can make adjustments in real time. I do this right around dec=0, then run a large T-Point model and ignore the PA report. Run the super model and you get great pointing and good Protrack. I'm not saying anything is wrong with the T-Point PA report. Only that it takes some time to run, make an adjustment, run to confirm, run another adjustment, etc...then finally you run a really large model and possibly find out it's saying something different about PA. As long as PA is close (and it's a compromise no matter how you do it) T-Point will do the rest.

For me, with a very obstructed view to the east, this routine gives me quite consistent results. Some testing I did showed me that my PA report out of T-Point would change quite a bit depending on how far to the west I collected points. When I collected points eqidistant from the meridian the T-Point report was quite similar to the PEMpro drift alignment. Even the author of T-Point indicated that setting PA in this manner (i.e., drift aligning around dec=0) was a reasonable compromise.

Peter
Yes interesting point. Over 15 minutes the star drifted about 10 pixels. So the PA seems reasonable. All the way through the 400 point run it said no need to make any adjustments to either axis. However once the super model was applied it now says move the MA tighten right 3.2 tics and ME lower 1.6 tics.

Pointing is very good. at 42 arc seconds that is good for the focal length over that large a model. Using a smaller model it was tighter but all sky pointing is probably more important in the end.

I am tempted to just do a manual PA run and then do a model. Pempro might be worth investigating. I have a lot of CCDware stuff so one more programme will not hurt either.

Tell me more about this phase issue for the southern hemisphere?
Reply With Quote
  #13  
Old 30-01-2014, 12:21 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Paul,

Re the phase issue. I started what became a huge thread at Software Bisque about this issue. After numerous attempts by many to get PEC to work in the S.H. Phil (in Canberra) and I were able to prove to S.B. that there really was a problem. Phil ran numerous experiments on his MX. It turned out that his uncorrected PE curve was very symmetrical with respect to the mid point of the worm cycle. What he did was to initiate collection of data but starting at different points in the worm cycle. What he noticed was that if he started at the mid point he could achieve a PE correction when applied by TSX, but only at the mid point. We were discussing this by phone and I suddenly realized that this could be explained if the software was reading the worm phase backwards. Clearly in the S.H. the worm turns in the reverse direction compared to the N.H. Now, whether my theory is correct or not is not known. The other "evidence" was that I could get a good correction using PEMpro. So, it made sense to compare the correction curve in the Bisque TCS from PEMpro to the curve generated from the same data generated by TSX. The curves were very similar but completely out of phase with respect to each other. In other words I could slide the two curves in photoshop over each other but the high and low points of the curves would not align unless one of the curves was moved along the x axis considerably. S.B. finally put up a fix in the daily build but there have been few if any confirmations as to whether the fix works or not. I was so tired of trying to get TSX to work for PE that after I tried PEMpro I just gave up. When I get my mount back perhaps I will test it.

By the way, PEMpro has a very generous trial period (I think 30 days or more) so you've got nothing to lose about trying it. If you have a camera that runs in CCDSoft it's quite easy!

Peter

PS I'm thinking your PA is probably fine. I'd just leave it if your model is that good!
Reply With Quote
  #14  
Old 30-01-2014, 02:57 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
thanks Peter for the explanation.

I will try pempro. I can probably get that done this week. I am thinking this must be the case with the phase. Although would it work if I flipped the curve over and then it would maybe in the right phase. Perhaps some tests either way will give me an answer. Bottom line I need the PE to be corrected and guiding to work better than it has been. I will get back to you when I know.
Reply With Quote
  #15  
Old 30-01-2014, 03:38 PM
SkyViking's Avatar
SkyViking (Rolf)
Registered User

SkyViking is offline
 
Join Date: Aug 2009
Location: Waitakere Ranges, New Zealand
Posts: 2,260
Paul, although my mount is a Losmandy I've always had the impression that PEC was not to be used together with autoguiding. I've never used the PEC function on my mount.
However it may be different for your mount of course - so just a thought.
Reply With Quote
  #16  
Old 31-01-2014, 01:17 AM
Joshua Bunn's Avatar
Joshua Bunn (Joshua)
Registered User

Joshua Bunn is offline
 
Join Date: Oct 2011
Location: Albany, Western Australia
Posts: 1,459
Quote:
Originally Posted by PRejto View Post

.....Perhaps another TSX user in the S.H. will confirm if they have had success with PEC/TSX.

Peter
Hi Peter,

Using my PME with TSX and PEC, my PEC curve now reads 0.7as Peak to Peak. so it works, but maybe there is a difference between the PMX and PME??

Josh
Reply With Quote
  #17  
Old 31-01-2014, 07:37 AM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Hi Josh,

Well, that is good news! I guess people only post (or complain) when they actually have a problem. I doubt that there would be a difference between the ME and MX wrt to PEC working or not in the S.H. I take it, however, that you also couldn't do PEC with TSX and your ME until after the fix was posted?

Peter
Reply With Quote
  #18  
Old 31-01-2014, 07:52 AM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Last night I tested the curve. It came back as 5.1 peak to peak. I flipped it and it was 8.4 peak to peak. However this morning I see that it is 2.8 peak to peak but it has large spikes in it.
Attached Thumbnails
Click for full-size image (new curve.jpg)
108.4 KB44 views

Last edited by Paul Haese; 31-01-2014 at 08:18 AM.
Reply With Quote
  #19  
Old 31-01-2014, 09:24 AM
frolinmod's Avatar
frolinmod
Registered User

frolinmod is offline
 
Join Date: Dec 2010
Posts: 573
Paul, that screen capture says Paramount ME. Did you forgot to set it to Paramount MX? Another thing for you to double check.
Reply With Quote
  #20  
Old 31-01-2014, 11:36 AM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Paul,

When you are computing the data I assume you are connected to the MX. But, in the morning when I presume you took the screen shots I'm guessing you are not connected and the software has defaulted to the ME setting. Maybe that is why you are seeing a difference.

It's pretty hard to read the graph because the spike(s) seems to have increased the y scale so much that all the rest of your data is compressed. To me it looks like the spikes maybe came just from one of the cycles and the rest look normal. Is that right? Maybe some wind? Or maybe you have some crud on the ring gear that started to take effect in one of the cycles. I say ring gear not worm because if latter I think you would see the spikes all the time, no?

How would you rate the seeing when you collected your data? I think it needs to be pretty good! And what about your exposure time? I've seen some comments saying shorter is better. Most of my good results with PEMpro came from quite steady seeing after midnight and short exposures (.5 to 1 sec), 5 cycles. Ray of PEMpro actually said that taking too many cycles can actually cause a worse result.....something along the lines of any more than 4-5 cycles can confuse the algorythym. Something to ponder. Personally I think you should be seeing results of 1 arc-sec or better.

Peter
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 10:25 PM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Testar
Advertisement
Bintel
Advertisement