Thoughts on these graphs? Pempro users?

[E_ri_k]

Thought I'd re-do the polar alignment on the G11. Last time I used Pempro, it was easy as, and took maybe 45 minutes. Now over three nights and hours of frustration, I can't get my head around whats going on.

These graphs show the DEC drift plotted in 5 second intervals. They all start off erratic, then level out, then erratic again, in a very periodic way? I unplugged the DEC motor just in case, but no change. I have checked balance, binding, end play, cables, no wind, and the seeing was good.

Even the star image in Maxim was jumping up and down between exposures. Im not sure if I was imagining things or not, but I ended the Pempro session, and did a 5 second continuous exposure and watched the star movement, and it was very steady. As if the software was messing with something, but I don't know how as the DEC cable was unplugged! Thats crazy right:screwy: I think I was going insane at this point:lol:

It looks to me like really bad seeing with the up and down spikes, but it's strange how it has these nice and smooth periods.

Last time I used this method I got pretty smooth linear plots, nothing like this?

I popped in a reticle eye piece and a 2x barlow for a manual go at it.

Any thoughts?

[PRejto]

I'll take a stab at this but I'm just guessing.

OK here is a wild guess! I think you might be seeing PE superimposed onto the DEC axis. Maybe I'm crazy but trying to reason this out is giving me a headache. However, is it wrong to reason that the motion one measures is a vector sum of the motion in both axis? If the polar axis is not set very close then a component of motion in RA would have a small vector in declination. Given how extraordinarily small an arc-sec is I think this little vector in DEC might be measurable by PEMpro. The saw tooth pattern looks quite regular and might be a pinion gear perhaps. Over time the motion goes above and below the average position calculated by PEMpro. Perhaps if you know the period length of the worm you can roughly calculate if the larger scale up/down flux matches. Anyway, this is just a wild guess and I'm probably wrong!! I was never all that great in math. So there. But, on the other hand (!), I know that if we don't have a completely level mount moving the mount in either azimuth or altitude will have a small effect on the other component. The further from level the more one adjustment effects the other.....therefore, if one is not polar aligned movement in RA will also move in declination....and that led directly to this theory.

Perhaps you should just go ahead and ignore the fluctuations and make adjustments according to PEMpro. Perhaps as you get closer and closer to being polar aligned the saw tooth pattern will diminish. Given the steepness of the slope I'd say you are not very close to the pole.

Perhaps your interpretation of the image in MaxIM being "smooth" is influenced by image scale difference between a full frame picture and what PEMpro is doing with probably a very small ROI. After all, PEMpro is capable of measuring fractions of an arc-sec so motion on the PEMpro graph will look quite exaggerated.

What image scale are you at with your scope/camera combo?

Peter

PS. A final thought. I'd post these on the PEMpro forum and get Ray's opinion. I'm most curious now!

[E_ri_k]

Hi Peter, thanks for your input and theory, and sorry for giving you a share of my headache :P I had a similar thought that perhaps PE in the RA was influencing the DEC. I never thought of vectors, math wasn't my strongest point either.

The mount is level and quite closely polar aligned already. I'm pretty sure the slope is so steep because of the fluctuations? A visual check showed no significant drift over a 10 minute period, so I'm not too concerned, just curious as to what is causing it. My reason for running this routine was simply to refine the alignment.

I don't understand why last time I ran it, the plotted line was linear, as in some sections of these new graphs, never random all over the place stuff.

As with last time, the FL is 2000mm and camera pixel size is 6.05 um, so I think that is about 0.6 Arc Sec/Pixel

I will post on the PemPro forum also, and let you know the outcome!

Erik

[PRejto]

Erik,

I keep thinking about this issue...but getting nowhere fast!

A couple of thoughts to consider.

1. I think you are further away from polar alignment than you think. The slope of the average position in your photos shows you should be making an adjustment. Just because you don't see drift is not necessarily an indication that all is well. I say that because PA is absolutely a compromise measurement and so where you observe "drift" makes a huge difference. Perhaps your "zero" drift is only a the particular location you measured it. I personally would trust PEMpro on this. Anyway, the adjustment will probably not be too large.

2. Is it correct to equate an azimuth adjustment with declination? I think they are different things, just as I don't think you can say an altitude adjustment is = to RA. This makes me feel a bit more confident that what you are seeing in the saw tooth pattern is related to some sort of PE.

Glad you will post on the PEMpro forum!

Peter

[gregbradley]

Making a T-point model is the best way to do Polar Alignment in my experience. Its worth the cost of the software.

Greg.

[stevous67]

For accurate polar alignment, Pempro is all you need. Your graphs are indicating hardware issues, and when you have flat sections as you show, you sampling hot pixels. A basic polar alignment in PemPro [to say 0.6 arc minutes of error] should only take 15-25mins [depending on the stability of the atmosphere].

It is a good idea to watch what the star is doing in Maxim like you say, as then you can tell if the drift is too severe for an accurate measurement. Being far off the pole can give you very misleading results too. But I assume you are close as you are revising a previous alignment.

Its important to make a cheat sheet of which way to adjust and for how much per amount of error. Making even marks on your adjustment knobs can be of a big help.

But definitely, you should not get flat spots in your graph when viewing it at 10 arc seconds of drift scale.

Good luck,

Steve

[E_ri_k]

I had a reply from Ray on the Pempro forum, he didn't believe it was anything associated with PE in the RA, but mentioned tube currents. It's interesting, because I didn't have the primary fans running, and I always do at other times, and the days had been consistently quite warm.

I'm really curious now to see what happens with them on?

Last time I ran the routine, I was under 1 Arc-min according to Pempro, 6 months ago.




Hi Greg, never used T Point, I'll have a look though.

Erik

[PRejto]

I agree with Greg that T-Point is terrific. However, to be able to use it effectively one needs to be able to move the mount very accurately. What good is a T-Point result that tells you to move your azimuth adjustment 45 arc-sec west unless you have a mount with this level of refinement in the adjusters? However, in saying that, there are ways to confirm the recommendations photographically. It's fairly straightforward to adjust altitude photographically but more complicated for azimuth given that the move is the cosine of the altitude etc etc etc... PEMpro does a nice job and you get essentially no drift around dec=0. But, T-Point proponents argue that all PA is a compromise and that all sorts of things other than drift need to be considered, such as refraction, where imaging, etc.

Please post your further efforts with PEMpro. Personally I don't see how fans would generate that consistent saw tooth pattern seen in many of your graphs. Also, I'd argue that the flat sections are not due to a hot pix. They don't come and go so if it was a hot pix I think the line would always be flat.

Peter