I think Roland got it right, the sub framing seems to have made this all work just fine. Ray -- I sent you a photo of the physical setup. To say you think something is wrong, after seeing this, is

Well the values are different than they were last night, but for some reason it is working perfectly now.

The models should have about the same values each run. If not, then something is changing physically (a non-repeatable system) or the plate solves were wrong. In either case, no wonder you were

Have you tried taking your failed solve image and uploading to nova.astrometry.net and comparing the results with your settings? If you have the failed fits file we could take a look as well -- Brian

Okay so tonight has been very good. I built a model using subframes and have had no problems at all. 🙂 I think this case is closed! Roland, my Dew Shield thanks you.

I am running the model now and the error terms look much lower. I have subframing on and I looked at all of these frames and there is a very full field of stars in them. Roland, I am hoping you are

You might want to ask Han, the author of ASTAP about what the problem is, especially since ASTAP fails to solve your image directly. From my experience, Han is very responsive. cytan Sent from Yahoo

I had H18 running, still am not successful. Tried to plate solve the failed image via ASTAP and no luck either. Not sure what is next. I’m fairly sure that all the APPM settings are ok.

The change in this field using the subframe is fairly significant. I do plan to test some things out with the first 5-10 points in the model as it runs. :) I think Roland nailed this one though. That

On Sat, Jul 23, 2022 at 10:52 PM, Bill Long wrote: The only thing that changed physically was removing a dew shield and rebalancing the mount. 🙂 Higher error model was the one without the dew

Would increased amounts of light interference influence the measured non-perp?

Roland already answered this. 🙂

The only thing that changed physically was removing a dew shield and rebalancing the mount.🙂 Higher error model was the one without the dew shield.

I believe the answer to why distortion might affect plate solving is that most algorithms group stars into triangles or quads then look them up in a hash table to get the star field rough

Bill, it's not me, it's the two models that indicate something is physically different. I am just the messenger! :-)

Sent you a photo Ray, I dont think its mounted wrong.

Bill, I think you should double check that the OTA is mounted correctly. That's a pretty high level of mis-orthogonality, and completely different than what it was in the other model.

Well in the model with more error, there was no dew shield. �� That, I put back on. I'll do another test with subframes on again tonight.

What happens when you have two cameras connected to a hub that is connected to a single USB port on the computer and both cameras want to download at the same time? Somehow the two traffic streams

Its mounted and balanced to the same degree it was in the first model.