w7ay

Hi Chen,

Firstly I must say your replies are very informative, detailed and comprehensive, Thank you for you help and I look forward to your other post's.

Coming back to the Guiding topic, I too have had rather fluctuating guiding results but I think its down to my polar alignment, mount (not the best) and breezy conditions (on one night).

Before I continue, please note I've only had two proper outings imaging (set up description below) and first one just earlier this week, I'm just a beginner.

I've found dithering to be rather erratic, so switched it off for the moment and guiding got better about 1.2" with dithering on it wouldn't settle in the given time 2" and then guiding is all over the place but would settle.

• Based on you previous replies could I request you to kindly elaborate on the tuning of calibration steps i.e. how would one do this in ASIAIR pro? I've had calibration go up to 40 steps definitely West and East don't remember North and South.
• Guide rate was set at 0.9x, I'll leave it next time to 0.5x (default)
• I also have access to EQ5 mount, which I use for solar, but will carry it to my imaging location (not home)
• I also got the following error a number of times at one point "RA Calibration Failed : Star Did Not Move Enough", kept failing, then eventually changed target then all good.
• Any suggestions on Dithering and tuning the parameters/ what to look out for?

Would appreciate your feedback/thoughts.

Setup

• WO Red Cat 51
• ASIAAIR pro
• ASI174MM (guide cam) & ASI mini guide scope
• Nikon DSLR D3300
• Explorer Scientific iEXOS-100

Thanks in advance.

keramos

Based on you previous replies could I request you to kindly elaborate on the tuning of calibration steps i.e. how would one do this in ASIAIR pro? I've had calibration go up to 40 steps definitely West and East don't remember North and South.

Calibration steps is simply the way for PHD2 to discover the scale of the mount's movement when it is told to guide. The Calibration process also finds the actual direction of the movement (the guide camera angle relative to the mount's RA and Declination axes).

To calibrate, PHD2 tries to move the guide star by 25 pixels in multiple steps in the camera's horizontal and vertical directions (and relate it to RA and EDC too). PHD2 averages the movement over these number of steps to average out various noise sources (gear, star centroid, "seeing," etc).

If the Calibration finishes in 6 steps, it is too coarse (i.e., not sufficient points) to make a good estimate of the magnitude and the direction (i.e., the vector) of the movement. On the other hand, if it takes more than 30 steps, you have way more than sufficient points, and you will just be wasting time (otherwise, there is no real harm).

In your case (40 steps), you can save some time in the future by increasing the Calibration Steps parameter in the Guide Settings window by a factor of about 3. This will make each step coarser, and thus finish the 25 pixels in fewer steps.

You might be curious why the step size is specified in milliseconds and not arc second or pixels :-).

Well, the reason is this...

... for mounts that support slewing in RA and DEC, PHD2 uses what is called "pulse guiding." When PHD2 wants to move a guide star by 1 pixel, it turns on slewing in the appropriate direction and then turn it off after N milliseconds later. This is the "pulse" in "pulse guiding." "N" is a function of your mount (and even the latency of the commands from the ASIAIR to the mount).

For most mounts, the timing to determine the duration of the pulse is in PHD2. There are some mounts where you can send a pulse duration command, but the most common mount protocols don't supprt that (I think one day they all will, since it will be more precise).

To figure out how long N needs to be to move 1 pixel, PHD2 needs to measure the mount ("calibrate"). I.e., how many millisconds of N corresponds to 1 pixel.

There is a second parameter that determines how many steps it takes to finish the 25 pixel movement. And that is the "guide rate" of the mount. Now, here is the crazy part... you need to go to Telescope Setup window in ASIAIR to find that.

The even crazier part is that there is a recurring bug that keeps cropping up in versions of ASIAIR where ASIAIR uses the slew rate in the GOTO subwindow as the guide rate, instead of taking the rate from the "guide rate" in the Telescope Setup window. If you find autoguiding for your mount running away at times, try to set the slew rate in the GOTO window to 1x. This will use 1x for guide rate instead of 0.5x, but better than nothing. And of course, recalibrate.

If you scroll close to the bottom of the Telescope Settings (it is really mount settings) window, you will find the guide rate. This is how much the mount moves per second when it is commanded to move. The usual default is 0.5x the Sidereal rate (i.e., if you keep slewing engaged at the guide rate, it moves at half the rate that a star moves across the night sky). If you set this to 0.75x, each second of guide pulse will of course increase the movement by a factor of 1.5. If you change the guide rate, you will also need to go back to adjust the Calibration Steps (and also recalibrate) in the Guide Settings window so that you again have somewhere between about 12 and 25 steps.

So, the short answer to you question is:

1) select a guide rate (0.5x should work fine with most mounts -- I have never come across your mount, though)

2) go to the autoguide window (tap Guide in main window and then tap on the autoguide popup window) and start the calibrating process. If there are insufficent steps, reduce the length of Calibration Steps in the Guide Settings window. If there are too many steps, increate the Calibration steps parameter. Rinse and Repeat step 2 (each time clearing the calibration by clicking on the "i" that is inscribed in a circle in the autoguide window; then clicking on "Clear") until you are happs with the number of steps it takes. Note all the numbers down in your notebook so you won't have to do all this again until you replace your mount.

PHD2 is actually quite simple to use (it was originally named "Push Here, Dummy"). Unfortunately, it is not integrated well in ASIAIR's user interface. To set up guiding, you need to jump between three or four windows!

The good news is that you will eventually get the hang of it, and muscle memory takes over.

Chen

keramos I also got the following error a number of times at one point "RA Calibration Failed : Star Did Not Move Enough", kept failing, then eventually changed target then all good.

ASIAIR (or you) may have picked a hot pixel as a star?

ASIAIR has no provision for using a Bad Pixel Map or Dark frame for autoguiding.

Chen

keramos Any suggestions on Dithering and tuning the parameters/ what to look out for?

Over time, you wil get a feel of the RMS error of your mount. You then set the Stability parameter in Dither Settings to cover your mount.

If the Stability is set too low, and your RMS guiding error is too high, ASIAIR might not restart autoguiding after a dither. Set the stability to a larger value than the typical guide error.

Chen

w7ay

Afternoon Chen,

Thank you for the detailed response, I shall get back if I have any more questions.

I just need some more clear skies here in UK so that I can go and practise, I've got more than enough information to keep me busy for a couple of nights

w7ay ASIAIR (or you) may have picked a hot pixel as a star?

I did wonder why there was no fluctuation in the star that was picked, thanks I'll make note of this.

Have a nice weekend.

Nihal.

w7ay

Actually I do have couple of questions:

1. If i'm tuning Dither parameters, do I need to calibrate?
2. Same for Aggression, I plan to start with 25% then up it, do I need to calibrate?
3. Is there a relation between Polar alignment error (as seen at the end of polar alignment in ASIAIR) and guiding error (Total RMS)?

keramos I did wonder why there was no fluctuation in the star that was picked, thanks I'll make note of this.

Ha ha, just like everything else in life, if it is too good to be true... :-).

The other way to spot a hot pixel is that the FWHM shown in the little inset image is very small. With typical guide scopes, guide cameras and typical seeing, the FWHM of a star is more likely to be in the 2.5 to 3.5 range. Worse if the optics is crummy.

I thought PHD2 is supposed to reject hot pixels on the fly (quite easy to do). But if you had manually selected a hot pixel, I think it will accept what you had selected :-).

Good luck tonight. I may have aa clear night here too (Pacific Northwest of USA) and may have a chance for first light of an Askar 200/f4 lens on an ASI2600, after mounting a belt driven ZWO EAF on it yesterday.

Chen

keramos If i'm tuning Dither parameters, do I need to calibrate?

No, you do not.

Calibration calculates how much to move the mount by.

Dither will just use thhe same calibrated parameters to determine how much to (randomly) move the mount after N exposures.

Chen

keramos Same for Aggression, I plan to start with 25% then up it, do I need to calibrate?

"Aggressiveness" needs to be redone after calibration, since the "loop gain" of the feedback system depends on the scale (pixels per millisecond of guide pulse).

A low aggressiveness (thus low loop gain) will cause the mount to move slower and also can have a residual error.

For example, if you see the Declination graph stay pretty constant, and not tracking back to zero, it means there is a bias (perhaps balancing problem, perhaps crummy mount with a lot of backlash, perhaps other reasons), it usually points to too low of a loop gain. I.e., you should try bumping up what PHD2 calls aggressiveness (not "aggression;" there is no anger involved :-)

On the other hand, if you "live at the edge" and increase the aggressive parameter to just before the system oscillates, you will need to back off from that aggressiveness (and live with a larger tracking error) when seeing is poor. If you are not using long focal lengths, I recommend not raising the aggressiveness to the point autoguiding almost oscillates -- IMHO, it is simply not worth it. With a 400mm or so focal length, a 1 arc second RMS noise is just going to bloat a star by perhaps 15% -- focusing error usually bloats a star by more than that.

Chen

w7ay

Haha, meant aggressiveness not aggression.

keramos Looks like have a couple of nights next week but due to bonfire night.. lots of fireworks.

Ha ha, you guys [sic] are still blowing up Parliaments for fun, eh? :-)

We, in the colonies, have a different kind of fireworks on the night of November 3.

Good luck Nihal, hope autoguiding works out well.

Chen

It might be a bit much to ask, but if these changes in setting parameters can be discerned and adjusted for, could not the Guiding itself be given some basic AI to do its own learning?

heyjp it was suggested that one should perform guiding calibration on a star that is close to DEC 0. Is that a good idea? Why?

Pull up a planetarium program (like SkySafari). Display an FOV on it. Perhaps a circle or a rectangle that has dimensions of about 2 degrees.

Move that FOV close to 0 degree declination. Note the size of your FOV in degrees of Right Ascension.

Next, move the FOV close to the pole. Notice that the same FOV (your camera sensor) now covers more degrees of Right Ascension. In fact, if your FOV includes the pole itself, it will cover the entire 24 hours in RA!

Now imagine trying to calibrate near the pole. Because of the above, the guide star will barely move on your sensor for each step of the polar axis of the mount direction.

Calibration is a process where you issue a slew command to the telescope mount at the guiding rate, and then turn off the slew after a predetermined time (this is shown in ASIAIR as the "calibration step size" -- that is why "step size" is in milliseconds of time). This then tells the guide algorithm what the "scale" is, for example, how may pixels the guide star has moved for 1000 milliseconds worth of "pulse" given to the RA motor and to the Declination motor. If the scale is large (as in the case of the RA direction near the pole, the same 1000 milliseconds will produce a much larger movement in the declination direction than in the hour angle direction.

It is usually perfectly fine to calibrate near the pole, and then autoguide where you have calibrated. However, if you calibrate near the pole and then move closer to the equator to autoguide, the RA axis will have poor tracking accuracy.

You can minimize the error by making the guide camera angle orthogonal to your mount's axis (i.e., the red and blue lines in ASIAIR guide info (that circle with the inscribed "i") fall right on the ordinate and abscissa). But it is much easier to just calibrate near declination 0, or recalibrate ever time you change to a different target.

Chen

It was mentioned in this post: "Don't forget to do a flip in the Guide Info whenever a GOTO crosses the Meridian. Or redo calibration."
Is this really necessary? Can ASIAir do it automatically?

kvastronomer
Does this happen automatically in AutoRun mode?

kvastronomer It should flip automatically. We will test it again.