Understanding plate solve with astrometry.net

I am not sure what to do to achieve the following process for which I used Astrotortilla previously

i don’t have a pointing model in my Gemini controller, so I do this to get on target:

tell Gem to goto say m15 - it usually manges to get within a few degrees depending upon how accurate it was set at ‘counterweight down’

solve the plate - AT solves, works out the error and repositions the scope iteratively, so it points at M15

I think SGP can do this using ANSVR, but I’ve read the instructions and got confused.

thanks for any help.


First, is ansvr set up and able to solve?

If you’re able to solve via asnvr/local astrometry.net, then all you need to do is do a “solve and sync” ONCE and your mount will know where it is pointed. After that if you have your target settings set to “Center on when target starts” SGP will center the target when you start the sequence. You can also manually center the target by pressing “Center Now” in the target settings. This assumes of course that you have set your target coordinates so SGP knows where to look.

So first go to Target Settings:

Then check “Center On” when target starts or “Center Now” to center manually.

Here’s what I do with my G11 Gemini2: cold start from the counterweight down position, slew to somewhere else in the sky (usually a star in the vicinity of my imaging target), solve and sync and that’s it. When the sequence starts SGP plate solves to center the target. It usually takes 2-3 plate solves to get it within my threshold (40 pixels).

I’m not sure if this exactly answers your question but it’s a start.

Hi Joel, and sorry to break into Paul’s thread, but I have a question on your technique.
Why is it necessary to do the first part (that is up to “solve and sync”)? Will not the second part of your routine (“center now” or “center on”) use the plate solve to arrive at the same conclusion?
My apologies if this is a dumb question.

Yes Mark, you are correct that simply doing the “center now” or “center on” will get you to the target. What I’m not sure about is if that will also “sync” the mount. It’s really just habit that I first “solve and sync” to tell the mount where it’s at.

“Center on when target starts”, Center Now, and Center Here definitely do sync the mount.

My process is similar to Joel’s:

At the start of an evening’s imaging, I power on my Gemini-2 in CWD position do a cold start. Then I slew to somewhere away from the pole (it doesn’t matter where!) and do a Blind Sync in SGP. That gets the mount syncd to the sky, but with no model (I never bother with any Gemini-2 modeling.)

I have both “Slew to target” and “Center on target” checked for all my targets. As SGP runs through the targets, all the slewing and centering is automatic, and it just works. That includes centering each target when the target starts and re-centering targets after a meridian flip.

I get my coordinates for each target with the Framing and Mosaic Wizard.

It’s really amazingly cool how automatic it is!


The last few times I’ve imaged, I have not been there. Ahead of time I set everything up and tick “Center on when target Starts”. When the time comes that I set to start, an image its taken from my current position - usually the pole. It then plate solves and automatically slews to my target and solves within my set requirement- usually after a couple of tries. This is using astrometry.net local version. I don’t use a pointing model.

I don’t understand why you guys are doing an extra synch away from the pole. I don’t think it is necessary.

I don’t understand what this means. We slew, validate and repeat until we are within your defined tolerance.

Syncing at the pole will totally confuse my G11!

So I use “Slew To” and “Center On” together. SGP will slew first and then Center. This gets me away from my park position (the pole) and makes it so my G11 doesn’t “lose it’s mind”.

Coincidentally Joel and Andy both have G11s…and I’m assuming they have the same issue even though they both have the Gemini 2.

With Andy’s example above he wouldn’t need to do the initial manual slew but it’s possible he has his own reasons for doing this.


thanks to everyone for chipping in with this one. It seems to have sparked some discussion and it’s helped me a lot.

I have just one more question to which I think I may know the answer - Do I have to manually type in the coordinates of the targets i wish to view - sorry I know it sounds like a dumb question, but i am used to using the gemini databases from which I pick a target and goto, plate solve and image.

What is ‘reference data’ for? Does SGP use this as the target coordinates? If so, no typing into the ‘location’ boxes required?

btw, i am still getting to know sgp, but it’s a great product - comprehensive and with a good UI.



Although you can manually put in the target’s coordinates, there are some easier ways.

One way is to grab an existing image of your target and plate solve it. There are a couple ways to do that in SGP:

  • open the image, right click, Plate Solve, then check “use these results as the reference image for target”, or,
  • in the target settings, click Browse, open the image file, click Solve. SGP fills in the location from the solution.

But I think the best way to get the target coordinates is to use the Framing and Mosaic Wizard. Although it is a paid add-on it has a free trial period, you should definitely give it a try!


There are a lot of options for populating the coordinates. Some of which Andy has pointed out. You can also use images from Flickr or AstroBin that have been solved. You can find more info here:


And I agree the Framing and Mosaic Wizard is probably the best and most straightforward method. Not just because it’s an additional purchase…but this is what I use as well. I think most people that have used it don’t look back! You can find a short demo here:


Thanks both - i’ll get the framing and mosaic wizard - was thinking about mosaics anyway with my fov.

Paul Kirk