*** This documents one of my successful attempts to autoguide the NexStar 5i. I did work on it a couple of times more, but found that I just couldn't get it to guide while pointed East. Clearly, balance was an issue, and I just never did the work necessary to solve the problem. According to what I've heard, Celestron has guided the mount successfully, but I have no information on for how long, or what equipment or anything they did to alter balance to get it to work. Further, another fellow I know has been able to guide the mount. So, it is probably doable. It will just require work. Surely there is someone out there willing to put in the effort.
Things to try -- set backlash compensation in the mount, despite the advice from Cyanogen, especially in an attempt to get calibration in Y. And be sure to read Mike Dodd's document analyzing autoguiding.
And something else to consider is that an awful lot of objects eventually spend time in the West. Even if you can only guide there, with patience, you should be able to capture a number of decent images, especially if you do a stack of images captured over multiple nights.***
Celestron NexStar
5i
F6.3 EQ Mode on Wedge
MX916 CCD camera - Autoguiding the NexStar 5i mount
November 4, 2003
** CAMERA : The MX916 when coupled with the STAR2000 autoguiding dongle allows the single chip camera to both guide and image. The S2K dongle plugs into the autoguide port of the telescope mount.
** SOFTWARE : Maxim DL/CCD controls imaging and autoguiding.
After getting some advice from the Starlight Xpress Yahoo group as follows :
Question : Quick question for those of you who know something about autoguiding and Maxim -- when I left off experimenting the other night, I had pretty good control in Y but the star was jumping back and forth between two points in X. So, my 6 minute Star2K guided image (3 mins exposure) looks like little snowmen instead of stars. If you saw this, what would your first thought be? Backlash compensation set too high?
Answers :
(1) If you're using MaximDL to guide, open the settings page on the guider tab of the camera control dialog box...set the "aggressiveness" to "1". Then see how it guides, and possibly try some different backlash settings. With the agressiveness turned all the way day, Maxim won't "chase" the guide star so much, and with the kind of gears you have (as Len mentioned), chasing it will often cause it to overshoot. Then it winds up bouncing back and forth...:)
(2) The mount is oscillating in RA which means there is too much correction being applied. Assuming you ran a calibration run in Maxim (Guide/Calibrate) to obtain the sensitivities, I'd suggest: 1) Lower aggressiveness to 4 2) Insert a larger value for the X-sensitivity – multiply current value by 1.5 It is generally felt that backlash compensation is not required in RA guiding because the direction is never supposed to be reversed. However, Celestron and Meade SCTs have rather large-amplitude periodic error curves with random spikes, so you may in fact need backlash compensation in RA. But, only one compensation, not one in the mount software and another in Maxim. Weighting the lower arm of the mount with a Velcro wrist weight so the mount is always pushing against a steady load is a very good practice. It tamed my AP1200QMD, which is not supposed to require such antics. As a final tip, check the Guide/Settings/More/Enable Log box and save the track log. I or someone else can analyze it for periodicity which would clearly indicate overcorrection.
(3) It looks like you have too high a speed or have set your mount backslash too high. Have you run the calibrate routine? When I do (I've got an LX200GPS), I almost never touch the number and if I do, I increase them by a few decimal places. I also use aggresiveness set between 9 and 10. Try not to set the backslash from your mount (if it allows you to do that) and let Maxim compensate for it.
So using the above as well as noting that you can tweak the individual speed in X and Y by changing the settings derived from the calibration routine, I finally arrived at the following settings :
- 1 second guiding
- Turned off backlash in hand control
- Set Y backlash in Maxim to 10 and none in X
- Set X calibration time to 10s
- Set Y calibration time to 60s (the max) :-(
- Set aggressiveness to 6 (that corrects 60% of the errors)
- after calibration, changed X speed from -2.4 to -1.4 (going the other direction
was worse)
- Guided near Vega which was 20-25 degrees above horizon in NW and was able
to do multi-12 minute self-guided sessions
- Moved to NGC 6888 Crescent Nebula, recalibrated and went from X speed -2.3
to -1.3 and was able to do the same about the same #degrees over horizon
- however when I moved overhead to M33 I couldn't get a bright enough guide
star in 1s guiding and I didn't want to futz with messing with the guiding slice
- could not get Y to move enough to calibrate while looking at M42 and then
Maxim hung anyway so I gave up
I suppose if I can't get any movement in Y in some parts of the sky, I should try guiding in X only. I may also need to add weight to the front of the OTA when I'm shooting overhead because of the angle of the camera being almost perpendicular to the ground.
Here is one of
my 720 second self-guided images of Vega. That's the equivalent of 6 minutes
unguided as the MX916 uses half the chip to guide and half the chip to image.
Light stretch in histogram. The reason I have no other images for the night
is due to sky glow and no flats and my corrector is filthy!
The next one is the crescent which just barely shows up in this 720s autoguided image. (6 minute exposure) The stars aren't too bad. If I had any sense, I would have tried to hit as many of this target as I could before it set.
These are my raw notes from the session if anyone cares.
-----------------------------------
Below was from an much earlier session in which I hadn't gotten things working quite right, but it gives you some idea where I started.
This 360 second (3 minute exposure) image of M27 from July is almost OK. I had gotten a 240 second image (2 minute exposure) not shown here which looked pretty good in fact. Several of those stacked would have been worth pursuing. Not surprisingly, M27 was in the SW at the time.
