Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

anybody else guide with maxim? results seem better than the graph

  • Please log in to reply
8 replies to this topic

#1 adamphillips

adamphillips

    Ranger 4

  • ****-
  • topic starter
  • Posts: 381
  • Joined: 21 Sep 2014

Posted 16 October 2019 - 12:18 AM

ive had such a hard time guiding well with my MX+ which is a shame because it should do great. and actually it can do say 7 minute exposures no problem, but with narrowband I do have to guide for 20 minute exposures.

 

ive had the best luck, thank god, with Maxim, it seems to do pretty well, not quite perfect always but still. I have one very weird thing I cant figure out.

 

the graph, I can not get it to portray accuracy. I have tried the site and optics tab, entered the focal length. it does not seem to adjust it. basically I can see a graph, I assume if it is in pixels its probably accurate. seems like it should know how to read the pixels. but when I change to arcseconds. well basically the graph just visually looks a little bit better to put it shortly. the lines shrink, or the numbers get further apart.

 

no matter how much I bin the camera it does the same thing. except, the less I bin, it will error more on pixels, so it seems its reading the pixels.

I remember now, I had an idea that because I have to use an ascom driver, I cant find ZWO drivers. maybe it cant read the size of my pixels. that was my idea.

another random thing about theskyx. I did notice that turning protrack off and turning PEC off helps. it must interfere with guiding in maxim. if you use directguide bisque says they work together, but I do believe running a separate guiding program they will interfere with each other.

 

 

the other thing is. I am binning my guide cam 3x3. it seems to work the best so far. but. that image scale is about 1.7"/pixel. not ideal for guiding. now my RMS is about .4 pixels, I assume, would give an error of about .7" in each direction. which is about 1 pixel on my imaging camera. which I don't consider great but a lot of times my stars come out perfect. not quite always. but it does not really seem right to me. maybe it is because sometimes when I zoom in and count the pixels across a star, sometimes it will be say 12 pixels across and 11 up and down. 

either way. maxim is working the best for me and I have to stick with it, just weird that the graph is so off, is it because maxim cant read the size of my pixels?

 

 



#2 rgsalinger

rgsalinger

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 5170
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 16 October 2019 - 12:45 AM

First of all there is a drop down in the graph (click on the little triangle) that allows you to select either pixels or arc seconds. If you've selected arc seconds then the axes will change and you will see (") just after X Error and Y Error. So, that's how you tell what's going on.

 

I am certain that the graph itself is not all that accurate. I'm not sure how accurate it can be given that there's just maybe 20 pixels between the zero line and 2 arc seconds. I pay attention to what the graph shape looks like and the RMS that's displayed. Normally with my MX+ loaded up with my CDK12.5 I get .3 to .5 arc seconds in both axes - depending on conditions and target altitude. I am permanently mounted.

 

ZWO only has an ASCOM driver. So, that's what you use. It may be confusing because if you guide with the SKYX there's an X2 driver that you can select. That's built by ZWO just to work with the SKYX. 

 

Without knowing more about your system I can't give you specific guiding settings advice. With my Paramount MX+ I use aggression of 5 or 6 and minimum move of .03 seconds. My normal guiding cadence is between 3 and 5 seconds with that mount. 

 

Binning or not binning a guide camera depends on the camera from my experience. I am using these days a ZWO290 mini and I bin that 2x which gives me around .7 arc seconds per pixel. 

 

Rgrds-Ross


  • psandelle likes this

#3 adamphillips

adamphillips

    Ranger 4

  • ****-
  • topic starter
  • Posts: 381
  • Joined: 21 Sep 2014

Posted 16 October 2019 - 10:31 AM

First of all there is a drop down in the graph (click on the little triangle) that allows you to select either pixels or arc seconds. If you've selected arc seconds then the axes will change and you will see (") just after X Error and Y Error. So, that's how you tell what's going on.

 

I am certain that the graph itself is not all that accurate. I'm not sure how accurate it can be given that there's just maybe 20 pixels between the zero line and 2 arc seconds. I pay attention to what the graph shape looks like and the RMS that's displayed. Normally with my MX+ loaded up with my CDK12.5 I get .3 to .5 arc seconds in both axes - depending on conditions and target altitude. I am permanently mounted.

 

ZWO only has an ASCOM driver. So, that's what you use. It may be confusing because if you guide with the SKYX there's an X2 driver that you can select. That's built by ZWO just to work with the SKYX. 

 

Without knowing more about your system I can't give you specific guiding settings advice. With my Paramount MX+ I use aggression of 5 or 6 and minimum move of .03 seconds. My normal guiding cadence is between 3 and 5 seconds with that mount. 

 

Binning or not binning a guide camera depends on the camera from my experience. I am using these days a ZWO290 mini and I bin that 2x which gives me around .7 arc seconds per pixel. 

 

Rgrds-Ross

thanks for the response. I know how to change to arcseconds but I know its not doing it accurately. but either way I am just happy something is working well. 



#4 rgsalinger

rgsalinger

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 5170
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 18 October 2019 - 10:11 AM

"but I know its not doing it accurately" --- is with all due respect likely some form of user error. I've used MDL for 10 years with maybe 15 different guiding systems and never once seen an error when the guide camera and focal length of the guiding system are correctly defined. I mean it's just some simple arithmetic. Maybe what you are really getting at is that the graph display could use improvement. What I find is that the bigger I make it the more closely it seems to reflect what's being shown in the Camera 2 Information window that I have up when guiding. If you really think that you have an error you should let Diffraction know over on their registered user forum. 

Rgrds-Ross


  • psandelle likes this

#5 adamphillips

adamphillips

    Ranger 4

  • ****-
  • topic starter
  • Posts: 381
  • Joined: 21 Sep 2014

Posted 20 October 2019 - 02:21 PM

"but I know its not doing it accurately" --- is with all due respect likely some form of user error. I've used MDL for 10 years with maybe 15 different guiding systems and never once seen an error when the guide camera and focal length of the guiding system are correctly defined. I mean it's just some simple arithmetic. Maybe what you are really getting at is that the graph display could use improvement. What I find is that the bigger I make it the more closely it seems to reflect what's being shown in the Camera 2 Information window that I have up when guiding. If you really think that you have an error you should let Diffraction know over on their registered user forum. 

Rgrds-Ross

what I mean by incorrectly is this. the pixel graph might be right, probably is, I don't see why that would be wrong in any case. but if I switch to arcseconds. it changes slightly but I know its not right because I can easily find my image scale. another reason I know its incorrect is because if I change the binning, literally changing size of pixels. the ratio stays the same. yes the error for pixels will be different, but when I switch to arcseconds the ratio is the same as before. that is impossible if I double or triple the size of the pixels. if 1 pixel is .5" and then I bin 3x3 and it makes 1 pixel 1.5". I should easily be able to see that on the graph when switching, in directguide with theskyx it does this for me perfectly, its easy to see that the graph is right. but maxim doesn't do the same thing.

 

you might be right that the information isn't inputted correctly. I tried to do it with the sight and optics tab. that's how I was told.

my only guess is that it cant read the size of my pixels due to the ascom driver. or somehow the focal length is just inputted wrong or in the wrong spot.



#6 rgsalinger

rgsalinger

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 5170
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 20 October 2019 - 06:26 PM

I'm not sure I'm following along. To be clear, no matter how you change the binning, if you are displaying the graph in arc seconds, then the graph should not change. The arc second delta will never change because you binned only the pixel size will change (be converted into arc seconds). 

 

At the same time, it might be worth asking on the Diffraction forum what exactly they are graphing when they are graphing pixels. If you go from bin 1 to bin 2 then if they are graphing the binned value the delta number in pixels will change. If they are disregarding the effect of your binning then the numbers will stay the same. Note also that there is a setting to graph "delta" which may well enter into the equation. 

 

You certainly entered the value in the right place. However, if you are using a guide scope, make sure that you are using the guide scope focal length and not the focal length of the main telescope. 

 

One other thing is worth mentioning. I have seen, on one occasion, a guide camera report the wrong pixel size to MaximDL. So, it's always possible that there is something off with the guide camera. It's a long shot and I've only seen it once.

 

Rgrds-Ross 



#7 adamphillips

adamphillips

    Ranger 4

  • ****-
  • topic starter
  • Posts: 381
  • Joined: 21 Sep 2014

Posted 20 October 2019 - 11:37 PM

its tough to explain here. but roughly this is the image scale. .5" at 1x1, which gives 1.5"/pix binned 3x3. therefore if the graph in pixels is say 1 pixel of error. then at 1x1 the arcsecond error would be .5". now if you are binning 3x3, you see an error of a pixel. if you switch it to arcseconds it should read 1.5". that is not what happens. i guess its a little too hard to explain here. but i do know that direct guide on theskyx. the graph does exactly what im describing, because thats exactly what it should do. 

honestly it cant even be reading the size of the pixels wrong unless its all over the place with it, because like I tried to explain earlier the ratio keeps on staying the same between arcseconds and pixels, and it should not. the ratio of the graph and numbers its representing: pixels and arcseconds. 

 

but either way I do know that the graph is a good representation of whats actually going on it just might not be to scale. and maxim seems to guide better than anything else I can find so I will just stick with it. its not terribly important to have the graph to scale.



#8 rgsalinger

rgsalinger

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 5170
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 20 October 2019 - 11:46 PM

If you go from a guiding error equal to one pixel and bin the guide camera 3x3 then you will then have a guiding error of 1/3 of a pixel instead. So, when you translate the 1/3 or a pixel error back to the graph it's going to show as 1 arc second. In the unbinned case you will be translating that pixel back to exactly the same value 1 arc second because now 1 pixel equal 1 arc second (in this example).

 

Maybe you are thinking about the scale numbers on the graph. Guiding errors expressed in arc seconds will not change if you bin your camera.

 

If you can't follow that post over on the Diffraction forum and my bet is that they will tell you the exact same thing.

 

Rgrds-Ross 



#9 adamphillips

adamphillips

    Ranger 4

  • ****-
  • topic starter
  • Posts: 381
  • Joined: 21 Sep 2014

Posted 21 October 2019 - 08:24 AM

we don't need to beat a dead horse because it doesnt really matter. but in my real life case. as I was saying the ratio between the graph lines size never changes, binning or not. which as you said would not be the case be the arcsecond value would never change but the pixel number value would




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics