I know there have been a few posts about this here in past months but I'm not finding those posts easily.
It seems that no matter what I do I cannot improve PEC on my PMX. My results seem to always worsen PE. I have done the following:
1. Camera is within 2 degrees of position angle = 0, ie, X axis is aligned with RA.
2. Turned off protrack and I am very well aligned in PA (few arcsec)
3. collected data close to zero degrees dec, near meridian
4. calibrated autoguiding in CCDSoft, then turned off corrections in X and Y
5. collected 10 min of tracking data in .5 sec exposures
6. measured PE in TSX as 2.2 to 2.4 arcsec uncorrected (depending on which night I have collected data)
7. generate PE curve, saved to mount. I have done this for both West checked/unchecked
8. measure error with PEC enabled.
9. cry
I have tried for 3 nights with zero success. My image scale from plate solves is .54 arcsec. I could try binning 2x2 (1.08 arcsec) and longer exposures. Is my problem that I'm just chasing seeing? Would longer exposures average out seeing? Or, say if I collected 20 min of data would the software be better able to extract the true PE curve and possibly be less effected by seeing?
My uncorrected PE seems very good, but it looks to be a quick repeating cycling through of the pinion gears that waves up/down within 2.4 arcsec. In other words, two different curves, one slow and one rapid superimposed over the slow curve. I wonder if the mount is actually able to correct for the pinion error? Is the change too fast? I will attach a photo of the raw data for PE.
Any help would be much appreciated. I used the mount and my TEC140 successfully at f7 and got pretty good stars, but now at f14 I just cannot guide out these errors.
Have you tried collecting the PE data using Sky X rather than CCDSoft?
I don't know if this will make a difference, but it might be worth a try.
It could be something to do with the latest builds of Sky X.
The last build for Mac has made Sky X unusable for me, it has created a whole lot of problems that weren't there three builds ago.
I've got an ONAG on the scope so I might try collecting data into two programs at the same time - from both the guide scope to TSX, and the G2-8300 to CCDSoft and compare the results.
The manual I'm reading says min of 10 min so I assumed that was 4 turns of the worm. But I guess the number of turns is indicated on the graph by the vertical lines (?), so I might have only collected 3 turns. I will certainly try more.
So, question. Is it possible to correct 2 arcsec PE error collecting data when seeing might be 2+ arcsec? r, is one bound to wait and wait for those rare steady nights?
question. Is it possible to correct 2 arcsec PE error collecting data when seeing might be 2+ arcsec? r, is one bound to wait and wait for those rare steady nights?
Peter
According to SB their software is supposed to average out the seeing and you should be able to achieve sub arc second PE.
But if your seeing is over 2 arc seconds is there any point in having sub arc second PE?
My uncorrected PE on the MX is around 3 peak to peak so +- 1.5
I trained the PEC but it did not make a lot of difference, I put it down to the seeing conditions.
I will have another go once the current problems are sorted with Sky X. I will keep you informed.
My image scale is .6 arc seconds per pixel, I still managed to achieve 10 min unguided without PEC.( That was with a good TPoint model and Pro Track.)
According to SB their software is supposed to average out the seeing and you should be able to achieve sub arc second PE.
But if your seeing is over 2 arc seconds is there any point in having sub arc second PE?
My uncorrected PE on the MX is around 3 peak to peak so +- 1.5
I trained the PEC but it did not make a lot of difference, I put it down to the seeing conditions.
I will have another go once the current problems are sorted with Sky X. I will keep you informed.
My image scale is .6 arc seconds per pixel, I still managed to achieve 10 min unguided without PEC.( That was with a good TPoint model and Pro Track.)
Cheers
Hi Phil,
I can't answer your question. I guess I actually don't know what the problem is. I used to be able to guide just fine at f7, but in spite of quite a good model, excellent PA, and Protrack have never had good unguided results. Now at F14, same scope, weight, PA, etc, I just get erratic guiding, or maybe it's something else? Guiding will sometimes look excellent on the graph (say under .3 in TSX) and then it will suddenly start spiking upwards of 1 pixel or more in RA. Maybe that is just seeing, but usually it just comes and goes. Sometimes changing stars helps but more often doesn't. I was advised to try longer exposures to try to even out the seeing and lower aggressiveness. That helped some, but the longer exposures seem to result in distorted stars (6 seconds). Sometimes the stars are not so bright through the ONAG and actually need the longer exposure....so, I figured that if I improved PEC maybe my luck would improve. I guess the bottom line is I don't really have confidence in myself to know if it's my technique that is lacking, or if there might be an actual problem with the mount. I know this is a challenging hobby, but after more than a year at this I had hoped (in vain) to be further along. Sorry for the little rant. I'm just not sure how to sort through these issues. Any ideas appreciated!!
Peter
When I guide, I watch the graph and it is as you described. it will go from close to zero to over 1 sometimes 2 pixels. But I would expect this with the seeing I have here in Queanbeyan. Also imaging at .6 arc seconds per pixel every fluctuation in the atmosphere looks huge.
I haven't done any imaging in the past two months because I have been away.
The last image I did was a twenty minute guided exposure, and I remember watching the graph and it was fluctuating all over the place. The end result was quite acceptable.
What do your final images look like? are the stars eggy?
One thing, when you are capturing your PE data does the guide star stay well centred or does it drift away?
Also as Marcus has said the longer the better, I use at least 20min worth of data for PEC.
You mentioned you have tried with both west or east clicked on but generally if PEC worsens it means its pushing when it should be pulling.
So the curve is upside down. Perhaps this is a build issue that got corrected or perhaps there was some other bug on PEC but at one stage for sure PEC was bugged in Sky X.
Tracking my AP140 (similar focal length) with a MMOAG and SBIG STi guider I get guide errors of around .1 to .6. Typically under .4 and sometimes under .2.
2.0 would be huge and would result in eggy stars. The occassional 1.0 isn't so bad.
I am using 5 second guide exposures and aggressiveness set to 4 and CCDsoft.
When I guide, I watch the graph and it is as you described. it will go from close to zero to over 1 sometimes 2 pixels. But I would expect this with the seeing I have here in Queanbeyan. Also imaging at .6 arc seconds per pixel every fluctuation in the atmosphere looks huge.
I haven't done any imaging in the past two months because I have been away.
The last image I did was a twenty minute guided exposure, and I remember watching the graph and it was fluctuating all over the place. The end result was quite acceptable.
What do your final images look like? are the stars eggy?
One thing, when you are capturing your PE data does the guide star stay well centred or does it drift away?
Also as Marcus has said the longer the better, I use at least 20min worth of data for PEC.
Cheers
Phil,
Here is one of the only nights that has kind of worked for me. 10 7 min subs. The next night the seeing was better and I couldn't get anything acceptable. This is just a combine in CCDStack with some deconvolution:
Capturing PE data I have no problem staying in the guide window. In ten minutes it has hardly moved. I should easily be able to get 20 minutes so I will try that for sure. And, thanks for telling me of your guiding experience. At least I now know I'm not seeing something too out of the ordinary!
You mentioned you have tried with both west or east clicked on but generally if PEC worsens it means its pushing when it should be pulling.
So the curve is upside down. Perhaps this is a build issue that got corrected or perhaps there was some other bug on PEC but at one stage for sure PEC was bugged in Sky X.
Tracking my AP140 (similar focal length) with a MMOAG and SBIG STi guider I get guide errors of around .1 to .6. Typically under .4 and sometimes under .2.
2.0 would be huge and would result in eggy stars. The occassional 1.0 isn't so bad.
I am using 5 second guide exposures and aggressiveness set to 4 and CCDsoft.
Greg.
Hi again Greg,
Yes the TEC140 would be similar to your AP140 but I'm using a 2X AP barlow, so close to 2000 mm. I have mostly been guiding through TSX but I may try reverting to CCDSoft. I do like the graphs in TSX and miss that info in CCDSoft. But, with all the recent reports of buggy behaviour and crashes in TSX of late I think I will try CCDSoft again. I am only using it to collect PE because my main camera (Moravian G2-8300) runs much better in CCDSoft than TSX. I will try your settings in CCDSoft. Many thanks!
I found that I had to do images at 0.1 seconds and over 25 minutes. I don't think that graph looks right to me. Can you check to make sure the camera is pointed to 0?
Your image scale is .57 pixel per arc sec not .54. I am not 100% sure with the Sky X if it automatically takes into account binning (I think it might) but if you are using 2x2 then that may be one source of error and your image scale would be 1.14. I know SB Precision PEC did not take into account binning.
The curve seems too large. Mine on both PME and PMX is very small and only a bit above and below the centre line. This could also be a result of a wrong image scale being entered ie. not taking into account binning.
I did use a 20 minute run on a night of good seeing (most nights the seeing is good at my dark site, sometimes its exceptional, rarely poor, occassionally its just OK). The manual does not say to do 20 minutes though so that would only help not make or break it.
Your image though looks fantastic. At 2 metres everything does get exaggerated. With my CDK17 guide errors are usually much more than AP140 on PMX gets. But I get round stars every time. As pointed out 2x focal length of autoguider (not scope). You are using an OAG right, I think you mentioned that? Another possibility is flexure. How stable is that OAG? I had a look one on the net and it had knobs and adjustments everywhere and seemed huge. Can you just put the guide cam in the eyepiece holder of the scope without all the OAG stuff?
Also worth asking, in the pictured graph you do not show the apply corrections box as ticked in the top right corner. That needs to be checked for PEC to activate on the mount otherwise it does nothing.
Greg, I'm not sure where you are getting the .57 arcsec figure. If you plate solved my M83 image that image was taken with the barlow at a slightly different distance from the CCD. The photo below, taken at 2x2 binning, was taken just before the PE run I posted in my first post. When I collected data I didn't bin.
The plate solve shows 1.08 arcsec at 2x2, and it shows a position angle of 2 degrees 57 min from North. The MX manual states to have the camera within 5 degrees. Am I incorrect about this?
Yes, when I measured error after saving the curve to the mount he box was checked! And I did clear the previous curve before saving.
Josh, I did save the curve with the west box checked and unchecked, and measured hoped for improvement in both situations.
Paul, I appreciate hearing that you used such short exposures over a long period of time. It does seem clear to me that I have not collected enough data and this really gives me something to work on. It shouldn't be too hard given I have very little drift while collecting.
Thanks again everyone. Your help is greatly appreciated!! )Please let me know if I've got the camera positioned incorrectly by looking at the plate solve. That would explain everything!)
