Auto Focus, Rigel nSTEP and QHY367C


#1

Hi all. New to the forum but have been using SGP for the last year or so. I’ve tried off and on to get the auto focus to work as well as other features, but have hit the wall on a couple so hoping that the members can help me out. I have read the SGP manual and the SGP - The First Week countless times but still have some problems.

First, my setup and settings for auto focus.

Takahashi FSQ-130ED, Rigel nSTEP stepper focuser and a QHY367C OSC camera mounted on a CGE Pro.

Rigel focuser max position = 11600, max increment 999, temperature sensor for telescope

QHY367C has only 1x1 binning

Here are my Focus Control settings:

SGP Focus Control and Auto Focus Options.JPG

Manually focused with Bahtinov achieved at 3650 steps

Following the Setup steps: (without Bahtinov mask)

HFR was 2.99 with 75 stars.
Current Position 3520
After Frame and Focus trying for 3-5 times HFR, I could only reach 6.19 HFR.

The Step Size equation gave me a step size of 60. When I tried this, running the Focus Control, resulted in a huge movement of the focuser and a graph that looked nowhere close to a “V”. In addition the Rigel focuser did not like that setting at all. Sounded like gear slip, but that was not the case. I changed the step size to 26, it was better, but the curve was still not a V or U shape.

Compared with the examples in the manual, the HFR seem awful high. I must be missing some adjustment to bring that number down.

Since I was hoping to image on M81/M82 when I looked at the HFR calculation I notice that M82 was being ID’d as a star.

By then I hit the wall and had to call it a night. I went ahead and let the sequence run wondering what the focuser would do, but by then the overnight temp had stabilized and no other focus routines ran.

Any help or suggestions would be greatly appreciated.

Bill


#2

Hi Bill,

I am away from my gear - so can’t get too deep into this. Also, I don’t have a Rigel focuser; but I think that will not make any difference.

You are lucky to have a Bathinov mask - because you can get to, or very close to, focus before you start anything. Assuming it again gives you 3650…

If you have the focuser set then on focus point of 3650 and start with a small step size of say 5. during the auto focus routine the focuser will be sent to 3670 {3650 + (4x5 =20)} and work its way through 9 focus positions (5 steps apart) to 3630. If the steps are too small, the graph will look very flat - you are not moving away from the focus point very far. Gradually increase the step size until you get the V.
You may also find that that step size is big - but most likely not. I use a RoboFocus (on my FSQ130) with max position up at around 27000. I use step size 7 or 10 (as far as I can remember - and depending on whether or not I am using a reducer). That would indicate that perhaps for you, the step size should only around 5.

In any case, I would try with much smaller step size numbers than you have been using - I am confident you will get the graph you want within a reasonable time.

Kinch


#3

Hi Bill,

Takahashi refractors will give a nicely shaped “V-curve”, if

  • the step size is adjusted properly and
  • the initial focus position is near true focus.

Do you use a reducer or the native fl of 650 mm at /f5? If you determine and tell me the focuser resolution (in µm/step) of your system I can specify a proper value for the step size.

It is important to adjust focuser backlash correctly before you begin with the Autofocus runs.

Bernd


#4

Thanks Kinch. Skies are suppose to clear around midnight, if so, I’ll head out and give this try.

Any thoughts on what I think is a high HFR I’m getting? Would reducing the exposure setting help, if necessary.

Also, what are you looking for, for the number of stars being seen?

Bill


#5

Thanks Bernd.

I am imaging at the native FL of 650mm at f/5.

Focuser resolution. You lost me there in obtaining the µm/step. I’ve searched but come up empty.

Bill


#6

The high HFR is because you are out of focus…that should be down around the “1” mark when you do the focus run (less than 2 anyway).When I do the autofocus I generally do 5 sec exposures, through a Lum filter. Not sure how OSC camera will work.

You do not have to focus it on your target initially…while you are figuring things out. Find a nice star field and figure your steps out…then when you have it sorted you can focus on your target area from then on.

As for number of stars…I have never had a problem with my field of view, even when using the extender. I don’t think you need worry too much at 650mm, f/5.

Kinch.


#7

Got it. Surprised the high HFR means it is out of focus. Had not come across that tidbit. I am using a light pollution filter (LPS D1), don’t know if that will change things.

I will do as you suggest and play with those settings and quit sweating the imaging.

I’ve only had the FSQ 130 for less than three months. I’ve considered the extender and may PM you at some point about that.

Bill


#8

I might be giving you wrong information there…but in my experience it comes down with better focus. I found this on the web…“The Half-Flux Diameter also varies linearly with focus position, except very close to best focus. This makes reliable autofocus possible.”

Re the extender - I have a shot done with it on my web site (http://www.kinchastro.com/jones-emberson-1.html). I must say, it is VERY demanding. One needs good seeing and a good mount. I will try it again but I will plan better when to use it. Humidity is my big enemy.

Kinch


#9

The focuser resolution is determined as follows:

  1. Measure the focus travel of 1 revolution of the motor axis.
  2. Divide this value by the number of steps of the step motor.

Example:
Let’s assume that 1 revolution results in 24 mm focus travel, and the motor has 3600 steps/revolution.
Focuser resolution = 24 mm/rev. / 3600 steps/rev. = 6.67 µm/step

Bernd


#10

I also have QHY 367c and you can set it in bin 2 mode… check
I use it with a F7 FLT 132 refractor, initial autofocus routine in “Bin 2” HFR usually between 1.6 and 1.4… then during the sequence in bin 1 , HFR usally under 2.70 i use sesto senso motor focuser… step size 80, exposure 4 sec (under dark sky) number of step 9
Sesto senso gime me 140.000 step for 10cm of travel course…


#11

Kinch, Bernd, and Acsom75. Thanks for all the information and feedback.

Last night did not go well at all. Never achieved a V-curve. Tried changing various settings but nothing gave me a decent focus. Had to call it a night after four hours. Should have CS tonight and maybe tomorrow so will try, try again.

I plan on using a spread sheet to record the various adjustments to help keep track tonight. The .fit auto focus files weren’t much help, at least for me.

Ascom75, you are the second person to tell me today that they are getting higher HFR’s with the QHY367C. That is one less thing to worry about. I will give the 2x2 binning a try tonight also.

Here are samples of the curves I got.

Bill

Capture auto 3570.JPG

Capture auto 3548.JPG

Capture auto 3548 10 step.JPG

Capture auto 3590 10 step.JPG


#12

Hi Bill,

Bummer - I hate when I spend hours without actually achieving anything.

Looking at the info you have in the above images, there are a couple of things worth asking about.
Firstly, did you check roughly the focus point with the Bathinov mask before starting? It looks like you had the focuser at 3570 when you started the first run. Although, some differences would be in play from the 1st info you gave us…that seems a long way away from the 3650 position you mentioned in the first post.
Secondly, your focus position as determined in SGP jumped from 3570 down to 3518 then back up to 3590…in just over half an hour. This of course cannot be right. Your are either too far away from the actual focus position that the routine is not picking it up OR perhaps you have slippage on the focuser. I am not sure how exactly the Rigel affixes to the FSQ130. Any chance that you have slippage between the focus control and the scope. (Perhaps check focus position with the mask, then run your focuser in/out a few times and check again that the focus point is still more or less at the same step point).
Identifying and correcting slipping focus control coupling was my biggest hurdle when I started using the autofocus routine. So…just a thought…something else to check.


#13

Hi Kinch.

I initially started out at 3650 last night. But, the scope’s temperature was was 13º C for that step number and last night it started out at 19.5º C. I attributed the step number difference to the temperature variance.

I have used a Bahtinov mask to manually focus both nights before beginning the auto-focus tests. It just occurred to me that maybe I should be doing a manual focus each time I am making changes in the parameters to avoid out of focus when trying to run the next test. I cannot believe I didn’t think of that before. Dumb. Live and Learn. Of course as the scope cools, my manual focus point will change but that shouldn’t keep SGP from attaining a V-curve.

I will recheck the reference marks I’ve made on the Rigel gears to see if they are still aligned at the center point of the focuser’s travel.

Thanks again for the suggestions. They are very helpful.

By the way, I really like your website. Impressive collection of Takahashi equipment and the resulting images. To say nothing of your observatory. Looks like it beats my POD hands down.

Bill


#14

Thanks re the comments on the gear. I remember when I bought the first Takahashi - second hand FSQ I06 - I felt like I might be buying a motor cycle with a name like that.:laughing:
I had heard so much talk of Taks shifting focus with temperature that when I was having problems keeping focus during the night and saw such large movements from one focus run to the next, I did not think it other than what people had talked about. But I realised that from night to night, there was a big difference in the range of focus positions, even though not such a huge difference in the temperatures. However, it was quite some time before I realised the coupling between the RoboFocus motor and the Tak focuser shaft was slipping. Since I fixed that, I never had a problem running the SGP autofocus routine on either the 106 or the 130. So, it just makes me suggest - eliminate that as a possible problem area before going further with testing.


#15

Thanks. Good to know. That will be my first task tonight.

Bill


#16

Hi Bill,

Was reading the thread and thought this might be of some help. I have a Rigel nSTEP system on a HD11 Scope with a 0.7 FR, my FL is 1960. I use an ASI071 OSC camera. For my setup I have my step size setup for 300 and I also bin the camera 4x4 with a 5 second exposure.

Focus Settings

Focus Plot

This setup works very well. However, and this is hard to explain but these new cameras are very fast and I found an issue with the way SGP runs the focus. You see once the focus number is reached, SGP kicks off the image. What I found is that the mirror still has not settled and there’s some “Ringing” in the mirror from the vibration of the stepping motor. this will cause all kinds of issues when trying to run the autofocus. Gene Nolan wrote an updated driver that allows the user to put in a delay that will wait for the delay period of time before telling SGP that the position has been reached. This user input delay allows for the system to settle down. I’ve been using this driver every since and depending on the carmera I’m using I will adjust the delay period. For my ATIK383 camera I only need about .5 second however for the ASI071 I need 2 seconds of delay The driver version is 6.0.75Special. I don’t think Gene every released this driver since most folks didn’t appear to be having this problem.I think the issue has to do with your telescope FL as well as the Camera type.

nSTEP

You’ll notice the extra box called “Delay Is Moving” that what I adjust. currently the delay is set to 50 which is 0.5 seconds. I’m currently using my ATIK383 camera so that’s why it only 50. If I was using my ASI071 I would have that set to 200 for 2 seconds of delay.

This may or may not be some of your issues with the Autofocusing but though I would at least chime in to see if this might be of any help to you.

Thanks,
Mark


#17

Mark, thanks for the heads up and the info. Gene has been helpful with the Rigel software. I will reach out to him and see if this might help with my problem(s).

Bill


#18

So I was able to run 26 focusing sequences last night and keep a spreadsheet of the changes and details along the way. I also grabbed screen shots of each of the results.

I took suggestions from everyone’s feedback and ideas to setup where I would start. It was very helpful to have the replies to read as I was doing this. Here are my initial setup settings.

image

Focused with a Bahtinov mask before starting the auto-focus tests. Also rechecked as scope temperature dropped.

This is about the best V-curve I got. For some reason nearly every focus run resulted in the top right line of the curve having the last point angle off the line. The only setting I changed during all the runs was the Step Size. From 5 to 15 steps I changed one at a time. After that, since the curve was still pretty much flat, I jumped the Step Size in 5 step increments. I took it up to 75 steps which this image reflects. The 80 step run was botched since I snipped the image too soon and ran out of time to repeat it. But at least I have a starting point to work from.

25 Auto 3550.JPG

Any suggestions would be appreciated. The last point on the right line is bugging me.

Thanks.

Bill


#19

Well done Bill.

The ‘kink’ at the top of the right hand curve is most likely related to the Backlash setting of your focuser. You will need to set a value that removes the kink that you see. Experiment to see what suits your setup, but perhaps start with a value of 35 IN and see what effect it has on the curve.


#20

Yes - well done indeed Bill. Obviously I gave you a bum steer with a 5-10 step size and I think I know why. Not being at home at the moment, the number I gave you, 27000, for my focuser full travel is most likely 2700…I’ll have to check that now next week. (I guess I am more forgetful than I realise).
Your perseverance has paid off.


www.mainsequencesoftware.com