Elongated stars

[paulf]

Whoa, your guiding graph should look a lot flatter than that with the short focal length of a finder guider. How good is your polar alignment?

I think it's good, but I haven't done drift alignment or anything. The wobbly guider bracket is effecting the RA graph at the moment. Once I've stiffened all the kit up then I'll certainly come back to the graph. 

[libraryman]

Take some of your recent images if you have a few and put them into deepsky stacker, there is an artical about using this program to measure your differential flexure. its good and will confirm that the issue is as others have suggested  differential flexure and most likely  coming from your finder guider mount.

Ray

[paulf]

Latest improvements now ready for testing when clear skies return.

1) I've changed the Skywatcher dovetail bar for a William Optics Vixen Long Dovetail Plate.  Would have liked to move to Losmandy but that is out of budget for now. The new plate gives a nice flat surface though, and feels a lot sturdier than the old dovetail bar.

2) I've added a pair of William Optics CNC tube rings. Again they feel a LOT more solid than the Skywatcher ones I had. They also have nice flat surfaces top and bottom and I was able to screw them down to the dovetail plate with three screws each rather than the single screw in the Skywatcher rings.

3) Have bought a DJ equipment 2" clamp as suggested to attach the guidescope to the front tube rings. Again the flat surface on the WO rings helps here.

4) A 1kg Baader counterweight clamps nicely to the underside of the dovetail plate. Have managed to get the rear ring over the mount centre. Think the focus drawtube is 1cm away from usual focus point though so when I've had a chance to check focus then might be able to move the dovetail plate back a little bit further.

Looking forward to testing it out. Then will concentrate on the PHD2 side of things, try to get that optimised.

[perfrej]

Well, that looks SO much better! Never under estimate the power of flexure - even in small amounts

/per

[paulf]

Stars are looking a lot better now, 200% centre crop. 1200s Ha:

The RA guiding is still very jumpy though.  As one line of enquiry I turned off guide output and here is the guiding graph covering around 20 minutes:

Does this look 'normal'? Period Error wouldn't look like this would it?  Was hoping to run for an hour or more to look for PE but clouds stopped play as you can see

[perfrej]

I don't really get the fuzz about PE. If you guide and your PE is guidable - which it almost always is - what's the point? You are going to guide anyway

/per

[paulf]

Does the graph look normal then do you think? Others have commented that the RA looked jumpy previously so although I'm trying different PHD settings (and I was over-correcting originally) I'd like to eliminate or identify any other sources of jumpiness.

If the above graph is the result of PE then I was surprised that the cycle took 20 minutes, I thought a PE cycle for an HEQ5 Pro was around 10 minutes?

[kalasinman]

Looking at the geometry of the main vs guide scope. The main center of FL is fairly near the center of the  mount RA axis . The guide scope is WAY forward of the mount center of RA. So a fairly small motion of the main scope would be magnified considerably in motion of the guide scope. A mechanical amplifier if you will. So the main moves a little, and the guide scope sees what it thinks is a big motion , and sends what it believes to be an appropriate large correction. That moves the main too far and then the guider sees that amplified. Balance at this point is secondary. Try to get the center of FL of the guide scope at the center of the mount RA axis, like mounted to the rear ring instead of the front one. Then balance to that.

[paulf]

Thanks kalasinman, that makes sense. Might need another counterweight though!

[Zakalwe]

Looking at the geometry of the main vs guide scope. The main center of FL is fairly near the center of the  mount RA axis . The guide scope is WAY forward of the mount center of RA. So a fairly small motion of the main scope would be magnified considerably in motion of the guide scope. A mechanical amplifier if you will. So the main moves a little, and the guide scope sees what it thinks is a big motion , and sends what it believes to be an appropriate large correction. That moves the main too far and then the guider sees that amplified. Balance at this point is secondary. Try to get the center of FL of the guide scope at the center of the mount RA axis, like mounted to the rear ring instead of the front one. Then balance to that.

That doesn't make any sense.

The process of calibration by PHD will work out exactly how the mount responds to a guiding input. So the guiding system will do (backlash excluded) whats needed to keep the star centred in the guiding system. So it doesn't matter where the guider is, PHD will issue the correct command to move the mount the correct amount. Yes, it isn't really guiding the main imaging rig, but the difference in focal lengths,pixel size and so on means that it will work well, regardless of where the guidescope is mounted.

Personally, I would look again at the mount balance and stiffness. I would feel happier if the guider was mounted with two clamps, not one. There's still room for flexure in that setup.

I'd also look at the focuser. The Equinox focuser is good, but it can be improved with a bit of tweaking. The silver lock ring has a habit of loosening up and some users have complained that the focuser isn't perfectly orthogonal. I know that mine used to shift the drawtube slightly when using the locking screw...replacing it with a Moonlite made a considerable difference.

If everything is as solid as it can be, then I'd look at the PHD settings. The Min motion looks low and and I'd increase the RA aggressiveness.

[kalasinman]

That doesn't make any sense.

The process of calibration by PHD will work out exactly how the mount responds to a guiding input. So the guiding system will do (backlash excluded) whats needed to keep the star centred in the guiding system. So it doesn't matter where the guider is, PHD will issue the correct command to move the mount the correct amount. Yes, it isn't really guiding the main imaging rig, but the difference in focal lengths,pixel size and so on means that it will work well, regardless of where the guidescope is mounted.

Personally, I would look again at the mount balance and stiffness. I would feel happier if the guider was mounted with two clamps, not one. There's still room for flexure in that setup.

I'd also look at the focuser. The Equinox focuser is good, but it can be improved with a bit of tweaking. The silver lock ring has a habit of loosening up and some users have complained that the focuser isn't perfectly orthogonal. I know that mine used to shift the drawtube slightly when using the locking screw...replacing it with a Moonlite made a considerable difference.

If everything is as solid as it can be, then I'd look at the PHD settings. The Min motion looks low and and I'd increase the RA aggressiveness.

yes, I don't know what I was thinking. Sorry.

[paulf]

Personally, I would look again at the mount balance and stiffness. I would feel happier if the guider was mounted with two clamps, not one. There's still room for flexure in that setup.

I did buy 2 clamps so will look at that. They don't sit well on a skywatcher vixen bar, so I guess it's off to FLO again to find some sort of plate to mount on the tube rings!

The Equinox focuser is good, but it can be improved with a bit of tweaking. The silver lock ring has a habit of loosening up and some users have complained that the focuser isn't perfectly orthogonal. I know that mine used to shift the drawtube slightly when using the locking screw...replacing it with a Moonlite made a considerable difference.

You did help me a couple of months back with loose focuser. It's all as tight as it can be now. I've adjusted focuser tension as stiff as feels sensible. Should I keep adjusting tension until the motor focuser can only just move the drawtube? Or just enough until there is minimal play? Would love a Moonlite

If everything is as solid as it can be, then I'd look at the PHD settings. The Min motion looks low and and I'd increase the RA aggressiveness.

Yes, I'll try a few things. Does my unguided PHD graph look 'normal' though do you think?

Paul

[Zakalwe]

I did buy 2 clamps so will look at that. They don't sit well on a skywatcher vixen bar, so I guess it's off to FLO again to find some sort of plate to mount on the tube rings!

You did help me a couple of months back with loose focuser. It's all as tight as it can be now. I've adjusted focuser tension as stiff as feels sensible. Should I keep adjusting tension until the motor focuser can only just move the drawtube? Or just enough until there is minimal play? Would love a Moonlite

Yes, I'll try a few things. Does my unguided PHD graph look 'normal' though do you think?

Paul

The locking ring wont affect the drawtube. I used to lock the ring up fairly tight as I never used it to rotate the focuser.

There shouldn't be any need to use the focuser locking screw when you have the motor on it. If the tension is adjusted correctly then the motor will hold the drawtube. The focuser lock screw just ends up shifting the focus slightly.

Keep an eye on the locking screw that clamps the field flattener...these can shift the flattener ever so slightly and can result in the camera sitting out of square. Those single locking screws have no place on an imaging scope and should be condemned to Room 101 IMHO. Three locking screws are better (but not much). The best is a screwed connection all the way- no chance of anything slipping or becoming tilted.

The graph looks a bit odd, almost as if the mount isn't responding to the guide commands. Again, I'd get everything as stiff and as tight as possible before messing with any settings.

The ultimate test, though, is how good the stars look in the image, not the graph. You can make a graph look lovely and flat and have horrible stars. If the graph is moving, then it's showing that it's doing its job, after all/ The stars should be tight and "sharp"....you can have round fuzzy stars from random guiding movements (or bad seeing!).

Your latest image shows a massive improvement from the initial one, so I think that you are on the right track with the improvements in the mechanics of the rig.

I did buy 2 clamps so will look at that. They don't sit well on a skywatcher vixen bar, so I guess it's off to FLO again to find some sort of plate to mount on the tube rings!

No need for that. You just need a bit of flat aluminium stock like this:

http://www.amazon.co.uk/Aluminium-Flat-Bar-400-MM/dp/B00PMBBX76/ref=sr_1_16?s=diy&ie=UTF8&qid=1425748397&sr=1-16&keywords=aluminium+flat+bar+2%22

Just cut it to length and drill a few holes in it to mount it on top of the CNC rings. Then bolt the guidescope clamps to it. Job jobbed, solid as a rock and a heck of a lot cheaper than anything that ADM will have. Not as pretty, mind you!

[paulf]

There shouldn't be any need to use the focuser locking screw when you have the motor on it. If the tension is adjusted correctly then the motor will hold the drawtube.

I was referring to the tension adjustment rather than locking screw. e.g.: http://www.firstlightoptics.com/blog/adjusting-skywatcher-equinox-focuser-tension.html

I removed the locking screw as it go in the way of the motor bracket and as you say I assumed the motor would hold the drawtube.

Keep an eye on the locking screw that clamps the field flattener...these can shift the flattener ever so slightly and can result in the camera sitting out of square. Those single locking screws have no place on an imaging scope and should be condemned to Room 101 IMHO. Three locking screws are better (but not much). The best is a screwed connection all the way- no chance of anything slipping or becoming tilted.

The single flattener screw isn't in active use (only as a backup). The Hotech flattener uses expanding rubber O rings to lock into place so is 'self centering'. The Skywatcher flattener I had before certainly had some movement with the single locking screw though, which is why I changed to the Hotech!

The graph looks a bit odd, almost as if the mount isn't responding to the guide commands. Again, I'd get everything as stiff and as tight as possible before messing with any settings.

Sorry, maybe wasn't clear on this one - the graph shows 20 minutes UNGUIDED e.g. with 'guide output' unticked in PHD settings. So the swing in RA over that time period, does it show anything odd? Is it just Periodic Error - or some sort of random wandering?  I had hoped to do 1 hour to see if the 20 min cycle repeated, but clouds stopped that.

No need for that. You just need a bit of flat aluminium stock like this:

http://www.amazon.co.uk/Aluminium-Flat-Bar-400-MM/dp/B00PMBBX76/ref=sr_1_16?s=diy&ie=UTF8&qid=1425748397&sr=1-16&keywords=aluminium+flat+bar+2%22

Just cut it to length and drill a few holes in it to mount it on top of the CNC rings. Then bolt the guidescope clamps to it. Job jobbed, solid as a rock and a heck of a lot cheaper than anything that ADM will have. Not as pretty, mind you!

Ah great! The words 'cut' and 'drill' make me shiver but yes £10 sounds better than (another) £50+!

Thanks for help, much appreciated,

Paul

[Zakalwe]

Sorry, maybe wasn't clear on this one - the graph shows 20 minutes UNGUIDED e.g. with 'guide output' unticked in PHD settings. So the swing in RA over that time period, does it show anything odd? Is it just Periodic Error - or some sort of random wandering?  I had hoped to do 1 hour to see if the 20 min cycle repeated, but clouds stopped that.

Im not too sure Paul, I'm no expert in guiding...maybe chuck the log into PecPrep and see what it says. If nothing else it will produce some fancy graphs for you!

Don't worry about hacking the ali bar...aluminium is dead easy to cut and drill.

[paulf]

Have made the final (for now!) stiffening adjustments to my setup. New aluminium bar and double clamps.

Bolted to WO scope rings with M5 bolts. Clamps bolted to bar with M10 bolts.

Drilling the holes was a tale of two halves. The M5 holes were drilled handheld with the HSS drill bits supplied with drill - took nearly 1/2 hour for each hole.  The M10 holes were drilled using a cheap drill stand, and I bought a new mid-price drill bit set. Took about 15 seconds each!!

Looking clear for tomorrow night, so potential guiding issues aside, hoping to make a step forward.

[perfrej]

That looks seriously good!

/per

[paulf]

Clear skies last night and managed to squeeze 4 hours of imaging in. Processed this morning and all looking good (in terms of star shape anyway!), so I'd like to thank everyone who has contributed to this thread - you've been a great help

M101 and NGC 5422. 6x600s each for LRGB, stars RGB only.

[ollypenrice]

Looking good to go!

Olly

[terrym2002uk]

I think you have started a run on those truss clamps - out of stock at Phase One  

If anyone else is hunting them, then have a google for Rhino Narrow Truss Clamps.

Terry