PHD and Eggy Stars, Balance issue ?

[badgers]

Hi All,

Last night was pretty hit and miss, I didn't want to set up for a serious imaging session and instead was trying to assess my guiding performance.

I pointed at M57 as its in a fairly dense star field.

I'm getting Eggy stars again on 10 minute subs.

Given that the egginess is along the Dec axis am I right in assuming this is probably a balance issue ?

The sub on the left is a 10 minute luminance sub of a small area of the image, this one was about the best star shape I managed out of 10 subs.

The sub in the middle is more typical, stars are eggy and according to astrometry they look eggy along the Dec axis.

I'm really hoping my setup should be able to deliver 10-20minutes of round stars.

This was the PHD graph (from the log) for the last 20-30minutes of imaging.

I was getting nice flat graphs, then the Dec would wander down and then over correct. Fiddling with guiding settings helped a little (max dec duration increased from 150 to 230).

The RMS was usually around 0.05 0.06 and the Osc. Index was varying between about 0.3 and 0.5.

The scale here makes it look worse than it did on the actual PHD graph.

I still got a rather nice image in the end (can just about make out the outer shell of M57),

but after spending a lot of money on kit recently, I'd like to get my setup as tight as possible.

---

PHD Settings:

RA Agg: 120

RA Hys: 17

Max RA: 1000ms

Search Region: 15

Min Motion 0.05

Calibration Step: 4000ms

Dec guide: Auto

Dec Alg: Reset Switch

Dec Slope weight: 5.0

Max Dec (150 then tried 230).

Star mass 0.5

NR: 2x2 mean

[Digz]

I'm certainly no expert but how is the collimation of the scope? I think this can sometimes affect star shape. However the first sub looks pretty good so I would expect collimation to affect all subs......

[badgers]

Good point. But I checked everything with a Laser and Cheshire before imaging, CCD Inspector listing collimation at 5.9" average, which is pretty much as good as I could hope for.

Should also say that I rotated the camera 180 degrees for 2 of the shots and observed trailing in the same direction, i.e. the trailing is independent of camera rotation/angle.

[Digz]

Ok that rules that out which is good. Could be a balance or backlash issue then......sorry buts that's a far as my knowledge goes

[Starlight 1]

guide star is over saturated will do that as well.

[badgers]

Perhaps, I'll try a less bright guide star, I think I tried two different stars, one bright and one more faint, but had similar results.

I think it might be worth only correcting DEC in one direction. Possibly it was badly balanced and then with backlash was getting into cycles like the ones I saw.

[Starlight 1]

Was your guide star bouncing about .

[kookoo_gr]

here's my input since i had similar problems with similar equipment with yours that drove me nuts.

1. from my experience (and from various readings over the internet) if you have trail at dec then it is possibly bad PA. check it with drift align

2. if you have serviced your mount and the gears are proppely meshed then you don't need to east heavy, a porfect balance will be fine, but it can't hurt to try it out even if you

haven't serviced your mount.

3. when it comes to balance how do you balance the scope here is a guide for proper balancing a newton telescope

4. what kind of puck does your mount have? if you have the vixen type then it is prone to flexure, i changed mine to an adm dual saddle.

5. try piggybacking your finderscope it will have better guiding results since it won't be affected by any oscilations at the end of the tube, also don't use its standard mounting ring since the spring screw adds flexure to the system.

6. is the focusing correct throughout the whole imaging proccess? i found out that if focusing is a bit off it can give fake trail. when focused lock the focuser drawtube and if

possible try to check it between subs every now the then

7. what kind of guiding do you use st4 or ascom? the best guiding is via ascom as you can fine adjust it rather than the standard sidereal rates that st4 has

8. what is the sampling time you use at phd? lower times can make you chase seeing, try times at about 2 secs also if possible try to lower the min dec motion form 0.05

[badgers]

I imagine one will always expect DEC trailing, not even a perfect Polar Align will eliminate it., my issue seems to be overcorrection of the drift, probably down to the DEC motor having to reverse directions and backlash coming into play.

I'll definitely check out the Astro-shed balancing tutorial, I'm still thinking its the biggest culprit, but flexure and other issues are undoubtedly at play too.

The mount has been stripped, cleaned and regreased, it's working about as well as I could hope for on a HEQ5.

My scope is on a vixen type dovetail, but the finder scope is piggybacked via a second vixen dovetail bolted on top of the tube rings with very solid guidescope rings on that.

The finder guider is very solidly mounted, I can't budge it.

Focus has been very good checked at start middle and end of imaging.

The guiding goes via the ST4 port on the lodestar direct into the mount.

Guiding was 2.0 seconds so it wasn't seeing I'm hoping.

Thanks for all the tips, that's really helpful!

[badgers]

That astroshed video is excellent! Thanks for the tip.

[Starlight 1]

You have put !!! ("The mount has been stripped, cleaned and regreased") may be a bit of grit in the worm .

[badgers]

You have put !!! ("The mount has been stripped, cleaned and regreased") may be a bit of grit in the worm .

There may have been grit in there before, but those gears were clean as a whistle when I put them back together

I can see now that my balance was pretty far off, especially in DEC and for the angle I was imaging at (West of meridian). It's beautifully balanced now.

If the clouds ever clear I'll update you on how I got on!

Thanks again guys

[badgers]

Glad I'm not visual, because my eyepiece would now require a contortionist to access it.

[badgers]

After doing the full astronomy shed balancing act and moving my counterweight slightly east, my tracking was spot on last night for M57.

Still a little bit of coma and some tilt issues, but overall it was hugely improved.

Thanks for flagging up that tutorial, I'd seen it before but forgotten about it.

[dugpatrick]

With 10 minute subs you might be looking at differential flexure. You should examine the set and see if the flexure continues throughout the session. If the movement is slow and in one direction then probably it's flexure. FYI, PHD will show the relative offsets for the frames.

One obvious source of flexure are the rubber rings sometimes found in 9x50 finder/guiders. You absolutely need to remove the rubber ring and replace it with something solid. I drilled new holes and added thumb screws to my finder-guider. It made a huge difference in flexure.

Doug

[badgers]

I'm with you Doug,

I've been looking in great detail at last night's subs and although the tracking graphs were superb, there is still some irregularity in star shapes that is constant across the frame.

They are much better than the previous attempts, but still not perfect.

This residual issue does indeed seem to be Differential flexure....

1) The remaining egginess is uniform across the frame

2) The first sub taken after calibration looked better than later frames

3) The image drift between frames is all in one direction and creeping up each sub by about 1.6 pixels per sub for dX in 10 minute subs.

4) The image drift on 5 minute subs was 0.84 pixels per sub and they definitely appeared rounder.

I've not considered DF a problem before as my finder-guider is bolted into very solid guide rings and I can't feel any movement.

But all the evidence is pointing this way now.

These are the guidescope rings I'm using:

They are bolted onto a dovetail on top of my OTA tube rings.

I guess the weak spot is that the Rings clamp onto the metal at one end of the 9x50, but onto the plastic objective housing at the other end.

[badgers]

I sense an OAG in my future

These are animations of 5 consecutive subs each, cropped to the same position of each frame.

Each was guided from a single PHD calibration on a reasonably bright star as close to the centre of the frame as possible.

First the subs from the initial attempts on M57 with poor balance and poor PHD graphs.

This is 5 subs from last night, much less eggy but still not round. PHD graphs looked extremely good.

There's definitely shift for a guided image. I did not ask for dithering so I guess it must be flexure.

There is a little field rotation, but nothing to write home about.

In the first picture the drift coincides almost perfectly with the egginess.

But in the second animation the image drift is orthogonal to the egginess.....

i.e. the shift is South East to Northwest, while the egginess appears to run along South West to North East.

Any thoughts ?

[dugpatrick]

You could also perform a flexure test on the focuser. For example, put a laser collimator in the focuser and aim the laser at the center dot in the mirror. Then add a weight to the focuser and see how much the laser dot moves. Unfortunately I can't say how much movement would be normal or acceptable for 10 minute subs. With my 8" newt I can go up to about 8 minute subs without flexure but maybe my newt has a stronger tube or stronger focuser.

Jim Solomon's web page has a list of areas for possible flexure:

http://www.saratogaskies.com/articles/gs-vs-oag/index.html

Good luck! I spent months working on flexure before I found the source. For me it was a stiff rubber ring in the guidescope, but I think your guidescope should be solid. Some people report problems where the focuser attaches to the tube; reinforcement might help if adding weight to the focuser causes drift.

Doug

[badgers]

I do worry about the way the focuser is attached on skywatcher newts. A reinforcement plate wouldn't hurt if I can get one made up.

I have a moonlite focuser coming soon that I bought 2nd hand and it would be worth reinforcing it.

Probably worth doing this before I go down the OAG route.

At least I'm not loosing any imaging time at the moment, I've seen stars twice in the last 6 weeks.

[swag72]

Here's my idea which I've not seen mentioned yet.

I suffered from eggy stars during long exposures. They were all OK on short subs, but longer ones, forget it. I was looking at flexure as I really couldn't believe it was anything else. I was getting totally flat guiding graphs and had some drift in the subs.

My cure was polar alignment. I thought it was good, I'd used EQMOD polar align and so never for one minute thought that PA would be my issue. I wish I'd checked it sooner.

I downloaded PEMPro (free 60 day trial) and nailed the PA, my stars are now fine. I made sure that everything was tight, it has become something of a fixation with me now. I have learnt now to never think that PA is good enough - It needs to be spot on.

I reckon before you start trying to go down the OAG road and the hair pulling issue of flexure, just try a good programme to check your polar alignment. It may be an easier fix that you thought.

If you think this is all improbable then that's cool, it's just my experience and opinion but I reckon it's got to be worth a try before anything else. It's the easiest fix of the other potential problems you mention. Sometimes I think we get so certain of what it can and can not be that we miss the easier stuff and can be dragged down a very different path.

I hope that this helps and gives you another possibility - But this time an easy one to test. Download PEMPro for the trial, you have nothing to lose and it is very accurate.

[badgers]

Thanks Sara,

I've read all your posts on your eggy stars with interest. Really glad you got it sorted.

Nailing PA would be great as it would help me narrow down issues elsewhere.

PEMPro looks great, but I've avoided it so far as it looks very complicated. I also need to know what requirements it has for the stars it uses.

I only have uninterrupted views from West to North at my usual site. Seeing stars low in the South or East is impossible for me.

I've also been worried by the view that PA can be too good for PHD guiding, so I've not obsessed on PA yet.

Currently, I've been doing the Synscan Polar Realign, followed by a short drift alignment.

Anton

[badgers]

I bought a cheap aluminium plate from B&Q that I'll cut and mount on the inside of the OTA after bending it to the right curvature.

When the moonlite arrives, hopefully this will provide plenty of stability as it will hold the focuser to the tube much more solidly and across a wide area on both surfaces.

If I take Sara's advice and try PEMPro and also have the reinforced mounted moonlite, I should have eliminated 2 more sources of error.

I'll get round stars if it kills me!

Anton

[swag72]

I'm not sure how you are running your scope and mount, but PEMPro does require Maxim or CCDsoft to control the imaging camera and also ASCOM.

It is actually very easy indeed and is all click here to slew there etc, you actually need very little user information, it will do it all itself. Interesting, I've never heard that PA can be too good for PHD, I always believed you wanted the best PA possible.

[badgers]

I've got Maxim and I can hookup the mount if needed to PEMPro. But the thing that worries me is where it asks you to choose a star low in the South.

I'm stuffed in this regard

I guess the idea is that if the PA is 'too good' then DEC corrections are small and probably changing direction frequently, so backlash may come into play. If its slightly off then all corrections are usually in the same direction. I don't know though, there seem to be 4 schools of thought on every issue I encounter

You've solved your eggy stars so in your case nailing PA seems like a winner.

In my case with a 6" Newtonian my concerns are with the focuser, they really are flimsily attached compared to a refractor.

But I'd still like to find a good PA strategy where I don't need perfect views of the south.

Actually what I'd really like is a pier and in an obsy on a deserted island near Hawaii, but I can't have everything

Anton

[swag72]

I've got Maxim and I can hookup the mount if needed to PEMPro. But the thing that worries me is where it asks you to choose a star low in the South.

I'm stuffed in this regard

It doesn't ask you to pick a star low in the south. It does it all either 30 degrees east or west of meridian - You decide which side. And then a little later it slews to 5 degrees east of west. I did mine last night and the scope really was not low in the South at all.