Re: It's Saturday and Clear

[Rocket Pooch]

Up and running for a test image with the AOG at 1.4 meter FL getting .24 and .27 pixel guiding, Polar alignment is a little out, I did not want to re-align it, I just wanted to get the AOG going after 22 month's since I bought it.

[MarkS]

If you're getting .24 pixel guiding at 1.4 meters that's really excellent.
I'm getting .16 pixels RMS at 300mm

I've nevertaken my EQ6 apart for regreasing or bearing replacement.  I need to do that.

[Les R]

All clouded over.

Managed to polar align again ok. EQMod played up though and I eventually changed the port. Got it working then unable to slew with ciel.... Until I realised I need to "connect telescope"... All good though... Something I will remember next time!

Aligning seemed ok, so decided to give live view and the canon a try.

That's when I realised it wasn't gonna work. Using the diagonal or just straight connection, I needed to extend the barrel by about another 1/2 inch to get a focus. God knows how the guy I bought it from got it to work. I'm guessing he used extension tubes and forget them! Ho hum.... At least that's something I can sort.

But I couldn't get my head around APT once that was loaded. It connected to the camera ok, but I couldn't work out how to set exposure details. (I was going to try to do M31 ... With just 60 second exposures. Something else I need to read up on before next time..... But even though it ended with failure, a little bit more has been learnt.

And one thing I've worked out from the future (and probably the reason for problems last night.) There is no spirit on my heq5 and last need the lens tray/tripod support was used. In setting up tonight, I made sure everything was level before putting the mount on and tightening the frame. That tray was well off level..... But..... I realised I could use the underside of where the mount sits and get perfect levels, which regardless off any adjustments, I'm going to be able to make sure of its level before I get going.

[Rocket Pooch]

My guiding settles down to .21 dec and .18 ra once three work periods went through.  Mind you I just breezed by the mount and it has a wobbler when I bumped a cable 

[JohnP]

Chris/ Mark - How do you calculate your pixel guiding number please? John

[MarkS]

John, if you're using PHD go to Tools|EnableGraph and this displays RMS pixel error on left hand side of graph.

[Rocket Pooch]

Same in AA5 John it tells on the guiding graphic.

[JohnP]

OK - Not using PHD or AA5.....

[The Thing]

John, if you're using PHD go to Tools|EnableGraph and this displays RMS pixel error on left hand side of graph.

If your using PHD2 you get RMS for each axis and a composite figure. If you enable Trendlines (checkbox on the Graph) it will also tell you how good your polar alignment is based on drift being corrected. Mine was around ~0.5 arcmins out which is probably a bit improbable but maybe just luck.

My RMS's were ~1.1 (RA) and ~0.87 (DEC) which is about a pixel on both the guide camera (QHY5/OAG) and the 1000D at 1250mm. I have just  bought a new SKF bearing set for both axes of my HEQ5Pro as they are a bit sticky so I hope to improve this. FYI £53 delivered from leemancity7 on Ebay. He does bearing sets for bikes and scooters so message him with your needs.

[Rocket Pooch]

John.

My guide camera at 1.4m binned 2x2 gives 1.96 a/s per pixel.  So I think the guiding was quite good last night.

Chris

[MarkS]

My RMS's were ~1.1 (RA) and ~0.87 (DEC) which is about a pixel on both the guide camera (QHY5/OAG) and the 1000D at 1250mm.

That's interesting.  Tonight I was getting 0.2pixel RMS on one mount and 0.22pixel RMS on the other mount (I'm using PHD without separate figures for RA and Dec).  Guidecam was QHY5 at 300mm.  Scaling up to your 1200mm that translates into 0.8 and 0.88 pixel RMS which is very similar to your figures.

Trouble is, the mount giving me 0.22 RMS is the mount on which I was using the C11 for imaging at 2200mm.  The stars looked really blurry on the final image so I'm not very happy.  I'm going to need much better guiding accuracy.

QHY5 has 5.2 micron pixels.  So at 300mm focal length each pixel is 3.55 arcsec.  So 0.22 pixels translates into  0.78 arcsecs i.e. we are getting RMS 0.78 arcsec guiding.  You can pay a lot more money for a mount and still not improve much.  For instance the Celestron CGE Pro boasts RMS 0.58 arcsec with both PEC and autoguiding switched on:
http://www.celestron.com/astronomy/celestron-cge-pro-mount.html

Chris was getting 0.18 pixels in RA with 1.4m focal length where the pixels were binned 2x2 giving 1.96 a/s per pixel.  So that gives a guiding accuracy of 0.18 * 1.96 = 0.35 arcsec RMS

That's approx half of what we were getting i.e. twice as accurate.

If it's clear tomorrow night I'll try to do an EQMOD PECPrep analysis - this will help identify what's going on.

[Rocket Pooch]

Hi Mark,

The reason why I went the OAG route (this is my 2nd OAG) was due to the some issues I had with long focal length guiding. 

I remember early year before last I was getting .12 ish on both axis at 400mm guiding and the stars were not right at 2 meter's imaging.  I think there's are slight issues guiding with short focal length guide scopes, mainly due to alignment but also due to under-sampled guide stars. 

I really have only see an issue using a separate guide scope when imaging at very long full lengths (well long for an EQ6).

As I have said before wide field imaging really is easy compared to long focal length imaging with all the extra issues you get.

Chris

[RobertM]

Chris,

I agree but a lot depends on how the rig is set up.

There can be a few issues that a guide scope can cause, what I've seen is mostly to do with movement in the guide scope/camera.  Issues such as:

Camera not totally tight in the drawtube - those single screws are not good enough unless you pack the camera nosepiece with some tape.
Slop in the focuser assembly.
Scope not rigidly enough mounted to the main scope (guide errors+rotation)
Dangling leads (USB/guider/dew heater)

Basically my take is that if you can move it with a bit of pressure then it's too loose.

Another one is not calibrating star movement over enough pixels of the guide camera for accurate guide correction calculation.  I think a lot of people miss this one.

The longer the f/l of the main scope the more amplified the factors will be. I think an OAG is definitely the easy way to go for fl>1500mm but with the right attention certainly up to that (and possibly beyond) is doable via a guidescope.

Sorry a bit off topic this one - luckily the  haven't noticed yet

Robert

[The Thing]

If it's clear tomorrow night I'll try to do an EQMOD PECPrep analysis - this will help identify what's going on.

Mark, you can use EQMOD to gather the data while you guide with PHD. Use the top left control to change the screen to the Auto PEC function and set it recording (red circle button). It records 5 worm cycles by default and creates a PEC table and applies it - and you can then save the data and load it into PEC Prep for analysis.

[MarkS]

Well I played with PECPrep until the clouds came over.  Ultimately EQMOD's Periodic Error Correction is no different to any other PEC and so is a big disappointment.  PECPrep itself is very powerful - it does a Fast Fourier Transform analysis to break down the periodic error into its component frequencies.  This is very informative - it shows (in my case) the worm gear contributes about 30% of the error.  The rest is due to other gears in the chain and it helpfully lists the frequencies associated with each gear.  But EQMOD can't correct for the other gears in the chain because PECPrep can only generate a PEC file of exactly one worm rotation long.

Under such circumstances (i.e. where the majority of the periodic error is produced not by the worm gear), any approach that uses the information gleaned from a few worm cycles to produce a "one worm gear rotation" correction chart used for all future cycles, is doomed to failure.  In fact it will actually amplify the periodic error a few cycles further on (but they don't tell you that in the small print!).

The proper solution is to utilise the powerful analysis built into PECPrep to dynamically isolate the contribution to Periodic Error caused by each gear and then to correct for it dynamically.  The residual errors can be analysed for remaining frequencies to continaully update and improve the corrections.  But it's not easy to do!

The better solution is to buy a mount with periodic errors that are more easily correctable.