Is this PHD2 graph normal?

[abhoriel]

I've been having frustrating problems with guiding and to be honest, I've never managed to get it to work properly. I'm using an ST80 guide scope with a QHY5-ii mono guide camera with PHD2. I have an HEQ5 mount controlled using EQMOD and pulse guiding. I use a DSLR mounted on a ball head mount above the ST80 with a 500mm mirror lens for imaging. I get deviations usually of around 2" on the PHD2 graph, but sometimes these go up to 4". I've also attached some images I took last night which were made from 240 second exposures using PHD2 guiding, the 500mm mirror lens on a full-frame Nikon DSLR. There's a lot of trailing on the stars and I'm certain that it should work better than this at this exposure time and focal length. Is the problem evident in the graphs (see attached)? Or are the graphs ok and the issue down to flex (there could be a fair amount in my set up). I've been thinking it's the mount for a while (it's fairly old, been through many hands) and iI'm seriously thinking of doing Astrobaby's tune up and belt modding it..

Just to summarise, my questions are:

• Is this PHD2 graph normal or pathological? I don't really have anything to compare it to
• Could the problem be the mount or flex? (or both..!)
• Could it be balance? It's a little hard to tell if the balance is slightly off, as both ra and dec are a little stiff even when completely unlocked
• Is following Astrobaby's mount tuning and/or belt modding the mount likely to help?
• I've also recently ordered an ED80 to use instead of the 500mm mirror lens, maybe I can avoid flex without having to mount the camera on a ball head?

 

Any suggestions would be greatly appreciated

 

Ignore the last quarter of the PHD2 guiding graph, the guide star started to go behind some trees..!

[abhoriel]

here are the full sized images. Don't open full size on a device with metered data!

[Andyk93]

How close is your polar alignment? The closer I get mine the better my guiding is. Have you run the guiding assistant to see what that recommends? My neq6 guides better when you send Dec movements in one direction

[abhoriel]

 

2 minutes ago, Andyk93 said:

How close is your polar alignment? The closer I get mine the better my guiding is. Have you run the guiding assistant to see what that recommends? My neq6 guides better when you send Dec movements in one direction

 

Thanks for the suggestion. PHD2's Guiding Assistant said that the polar alignment error was around 8 arcminutes. Again, I'm not really sure if this is good enough or not. I usually just do a rough polar alignment with the polar scope. I hoped that it would be "good enough" for guiding to compensate for any issues. I did drift align with PHD once and got it down to a couple of arcminutes, but I don't recall it helping a huge amount. 

 

If it helps, the guiding was a little better when earlier in the evening before I had to do a meridian flip (DEC barely needed adjusting, it was mostly just RA)! I wasn't imaging then, just testing!

[abhoriel]

Does anyone know if this PHD2 graph would account for the star trails?

[Pompey Monkey]

53 minutes ago, abhoriel said:

Does anyone know if this PHD2 graph would account for the star trails?

I uploaded the jpeg of M42 to Astrometry.net and the results said that the image scale is approx 10" per pixel. The elongation in your stars is rather more than a pixel.

As your Phd statistics show an RMS error of less than 2", this suggests to me that the guiding is OK (although it could be a bit better for an HEQ5) and that the error is coming from somewhere else.

Your guess at differential flexure sounds plausible as you say that the camera an 500 mm lens, which probably weigh a bit, are mounted on a ball joint. Do you have a picture of your set up?

Other than the trails, nice work!

[abhoriel]

I don't have a picture, but I will see if I can get round to taking one tomorrow. Thanks for your help. Thats pretty much exactly what I needed to know. I didn't know how to translate the angle units into pixels!

There could be several sources of flexure. I use the st80 barlow (with the lens removed) as an extender to bring the qhy5-ii camera into focus. I guess the qhy5-ii camera's lead could be pulling on it. Then theres the ball head too I guess. The 500mm mirror lens isn't actually very heavy, because its a mirror lens. A 500mm refractive lens would be very large and heavy though.

Are there any other causes besides differential flexure which could lead to these results?

[kens]

You should post your PHD2 guide log for analysis. That said, I can imagine a DSLR and 500mm lens atop a ST80 could cause quite a bit of flexing unless you are using very solid rings. The ball joint would not help either. If it were me, I'd be going with a side by side setup. 

You've got some backlash on the dec axis so should probably guide in one direction only in dec.

You could try removing the camera and just guide with the ST80 to see how well it performs on its own.

[swag72]

Having battled with flexure in a dual system since August all I can say is that the SLIGHTEST movement that you can't even discern can cause issues. I'm in the process now of getting everything VERY well bolted down with scope rings and big dovetails (I thought it was all well bolted down, but clearly not  ) and I hope to try it later in Jan again. If you post a pic then I'm sure that the good folks here will be able to give you some potential issues.

As an additional point - There's something that a friend of mine ( @AndyUK ) told me about in DSS that can point to if the issue is flexure. You put all of your subs in DSS and then there's something in the output that will point you in that direction if applicable. I've not used DSS for years so I don't know what it is, but perhaps this will jog someones memory on here to tell you what  I'm rambling about  

[alacant]

9 hours ago, abhoriel said:

I use the st80 barlow (with the lens removed) as an extender to bring the qhy5-ii camera into focus

Hi. Did you enter the correct focal length in PHD2? With a 2x barlow this should be 800mm. I think it better to use an extension tube to get the camera to focus. Can you post the guide log with a calibration run? Cheers and good luck.

[abhoriel]

Thanks a lot for everyone's input. It's actually been so helpful.

I've attached the PHD2 logs and also a picture of how everything fits together. Its all mounted on a HEQ5 mount, and there are normally wires hanging off both cameras, which could potentially be another source of flex.

If you push hard on the DSLR, there is a noticeable amount of give where the ballhead connects to the scope rings. Given that micrometers of movement is enough to cause major differential flexure issues, I guess this could be a place to start! I'm going to get an imaging scope soon, which will replace the ballhead mount and mirror lens set up. But I intend to use lower focal length SLR lenses for wide fields, so how do people usually attach a guide scope to a DSLR + lens?

 

 

PHD2_GuideLog_2016-12-26_214651.txt

[abhoriel]

3 hours ago, alacant said:

Hi. Did you enter the correct focal length in PHD2? With a 2x barlow this should be 800mm. I think it better to use an extension tube to get the camera to focus. Can you post the guide log with a calibration run? Cheers and good luck.

 

Sorry I don't think I made it clear. I use the 2x barlow with the glass lens removed. Basically I didn't have a 1.25" extension tube, so I repurposed the barlow! You are right though, this certainly isn't optimal and actually doesn't add a lot of length, so the ST80's focuser has to be racked out close to its maximum.

[Pompey Monkey]

That ball-joint does look very suspect!

You could try attaching the camera directly to the tube rings using one of these: 

It would certainly be a cheap test

Also, you want the imaging train to be at least approximately lined up with the guiding train as this can also introduce another potential error into the system.

FWIW, I have a very second-hand HEQ5 and it works very well* at a pixel scale of 2.8" pp, with a load that I estimate to be between 8-10 kg.

*Except when imaging near the zenith!

[Thalestris24]

Hi abhoriel

Can I ask what make of lens you're using? I have a Samyang f6.3 500mm but I've never been able to get a decent image from it - there's always some kind of distortion present. Not to mention the difficulty with focus and maintaining it. I wouldn't say your guide graph was too bad - it's probably better than I usually get using a 180mm guidescope. It's worth running PHD2 with guiding outputs off - what you then see is the effects of seeing, plus any flexure. You may well have some dynamic balance issues with your current setup. As others have suggested, it's probably a good idea to ditch the ballhead. Also get an extension tube (~40mm) for the ST80. You want everything as rigid as possible!

Louise

[abhoriel]

38 minutes ago, Thalestris24 said:

Hi abhoriel

Can I ask what make of lens you're using? I have a Samyang f6.3 500mm but I've never been able to get a decent image from it - there's always some kind of distortion present. Not to mention the difficulty with focus and maintaining it. I wouldn't say your guide graph was too bad - it's probably better than I usually get using a 180mm guidescope. It's worth running PHD2 with guiding outputs off - what you then see is the effects of seeing, plus any flexure. You may well have some dynamic balance issues with your current setup. As others have suggested, it's probably a good idea to ditch the ballhead. Also get an extension tube (~40mm) for the ST80. You want everything as rigid as possible!

Louise

Hi Louise, I use an old Tamron adaptall f8 500mm mirror lens. (This one http://www.pentaxforums.com/userreviews/tamron-adaptall-2-sp-500mm-f-8-mirror-55bb.html ). I got an old one on eBay and then got a Nikon adaptall adapter. It is indeed very difficult to focus and very easy to accidentally defocus.. I just set it to infinity and take a few pictures to check (often getting it wrong anyway!). Defocused stars look like donuts with this lens! I have never managed take any decent day time pictures with it, but it works somewhat adequately for my level of astrophotography. As for distortion, it doesn't seem too bad, but if you're interested I'll upload some uncropped raw images if you want to see. You mention maintenance though.. As far as I can tell, the lens is maintenance free. Despite being a mirror based design, it doesn't need to be (and cannot be) collimated. You just need to keep it somewhere dry, or it'll end up with fungus, the bane of old lenses.

 

I could attach the camera directly to the scope rings and ditch the ball head, but it will be difficult to get it pointing forwards! I'll get a proper extension tube too for the st80. Very good advice. Many thanks

[alacant]

Hi. Definitely need to lose the ball joint. Have you considered side by side mounting? It would be good for when you get your new telescope too. Your guide_log: the DEC calibration is poor. It's best to knock the mount north at guide speed for a few seconds before you hit calibrate. It's regularly knocking DEC from one side to the other for over 5". You may be able to improve the backlash adjustment to improve this.  Otherwise it looks good. Remember to look at the end result snaps, not the graphs: your shots look great. HTH

[Thalestris24]

28 minutes ago, abhoriel said:

Hi Louise, I use an old Tamron adaptall f8 500mm mirror lens. (This one http://www.pentaxforums.com/userreviews/tamron-adaptall-2-sp-500mm-f-8-mirror-55bb.html ). I got an old one on eBay and then got a Nikon adaptall adapter. It is indeed very difficult to focus and very easy to accidentally defocus.. I just set it to infinity and take a few pictures to check (often getting it wrong anyway!). Defocused stars look like donuts with this lens! I have never managed take any decent day time pictures with it, but it works somewhat adequately for my level of astrophotography. As for distortion, it doesn't seem too bad, but if you're interested I'll upload some uncropped raw images if you want to see. You mention maintenance though.. As far as I can tell, the lens is maintenance free. Despite being a mirror based design, it doesn't need to be (and cannot be) collimated. You just need to keep it somewhere dry, or it'll end up with fungus, the bane of old lenses.

 

I could attach the camera directly to the scope rings and ditch the ball head, but it will be difficult to get it pointing forwards! I'll get a proper extension tube too for the st80. Very good advice. Many thanks

Hi

Sorry, I meant maintaining focus - my one, at least, changes a lot with temperature. I don't think many people use mirror lenses for AP? You could use an 80mm Bahtinov mask to help aid focussing. I'd be interested to see a full-size jpg if you can post one. It may be that the f8 lenses give a better image than the f6.3 ones.

Louise

[abhoriel]

24 minutes ago, Thalestris24 said:

Hi

Sorry, I meant maintaining focus - my one, at least, changes a lot with temperature. I don't think many people use mirror lenses for AP? You could use an 80mm Bahtinov mask to help aid focussing. I'd be interested to see a full-size jpg if you can post one. It may be that the f8 lenses give a better image than the f6.3 ones.

Louise

 

 

Possibly, but an f6.3 should collect 61% more light, swings and roundabouts.. I leave the lens outside for a bit whilst I set up the rest of my gear to let it cool. I've attached a couple of full sized unstacked "raw" jpegs. Looking at them now, there's clearly some coma going on near the edges. Also the donut shaped darker areas will be out of focus dust particles! I'm not quite sure what the oddly shaped darker area at the lower left part of the images is though! The m42 images are darker as I reduced the iso to avoid blowing out the core of the nebula and also because it was taken after 01:00 when all the street lights turn off

 

[cjdawson]

I've just taken a quick look at one of the first full size images.  My money is on the polar alignment first.  Looking at the image, I'm suspecting that image rotation is the main cause at the moment.  This will happen when your PA is off, the further off it is, the worse the problem.  Taking shorter exposers will help a bit, but it's still better to sort the PA out and not have the problem in the first place.

I'm planning on trying Alignmaster myself, it looks ideal helping to get things aligned faster than a drift alignment would.

[Thalestris24]

14 minutes ago, abhoriel said:

Possibly, but an f6.3 should collect 61% more light, swings and roundabouts.. I leave the lens outside for a bit whilst I set up the rest of my gear to let it cool. I've attached a couple of full sized unstacked "raw" jpegs. Looking at them now, there's clearly some coma going on near the edges. Also the donut shaped darker areas will be out of focus dust particles! I'm not quite sure what the oddly shaped darker area at the lower left part of the images is though! The m42 images are darker as I reduced the iso to avoid blowing out the core of the nebula and also because it was taken after 01:00 when all the street lights turn off

 

Hi

How long were the exposure times on these? I think the star distortions are unlikely to be caused by PA error. Some are similar to what I get with the Samyang. But, they could be caused by flexure/balance issues. Taking some short exposures of bright stars should clarify things.

The dark things are probably dust motes - possibly on the back of the lens, and stacking with flats will make them disappear and will also deal with vignetting.

Louise

 

[abhoriel]

Both of those images were 240s, or maybe 238s because of mirror delay!

Yeah there is a lot of vignetting, which along with the dust motes are generally corrected with flats (which I do take!).

[abhoriel]

My issues now seem to be largely resolved, so I thought I'd leave a follow up post here for anyone who was interested.

I have since acquired an Orion ED80 scope and as a result, my set up has changed quite a lot. I no longer use the ballhead mount or lens. I have the DSLR connected to the ED80 focuser using a T ring and extension. The ST80 is mounted on top. Both scopes are mounted on dove tail bars with scope rings. I've also purchased longer wires for various things like the DSLR and guide camera, as they weren't long enough before and were a potential source of tension. I'm not certain exactly what fixed the issue, but I strongly suspect (as everyone here suggested!) that it was the way that the DSLR was being mounted before using the ballhead. 

Also, interestingly, the PHD2 graph is generally flatter now, with everything staying within +/-2'. Not really sure why this improved either. Maybe I got luckier with the balance this time or maybe the mount prefers aiming close to the zenith. The set up is significantly heavier now, and I use a second weight (before I used one weight which was set pretty high up the bar to achieve balance); perhaps the extra weight helps stability somehow.

I've attached a random 8minute exposure of Andromeda, and the stars are vastly more circular than my previous shots despite a doubling in exposure time. I didn't get any useful data though, the almost full moon really didn't help and I've only had time to construct one dew heater so the guide scope lens got covered in dew bringing the imaging session to an early end! This image also seems to suffer from severe coma, I guess thats why everyone buys a field flattener..

[Herra Kuulapaa]

Haven't read every message on this thread, but in generally a vixen plate would help you with this and many other mounting challenges - one can mount things parallel.

For example this

[newbie alert]

On 28/12/2016 at 14:17, Pompey Monkey said:

That ball-joint does look very suspect!

You could try attaching the camera directly to the tube rings using one of these: 

It would certainly be a cheap test

Also, you want the imaging train to be at least approximately lined up with the guiding train as this can also introduce another potential error into the system.

FWIW, I have a very second-hand HEQ5 and it works very well* at a pixel scale of 2.8" pp, with a load that I estimate to be between 8-10 kg.

*Except when imaging near the zenith!

Where did you get one of those and do you know whst its called..i got a job for one of those..it be perfect..

[abhoriel]

36 minutes ago, newbie alert said:

Where did you get one of those and do you know whst its called..i got a job for one of those..it be perfect..

I dont know what that thing is called (so I shall refer to it as 'thing'!), but I actually found one attached to the top of the scope rings that came with my ST80. If you have rings for a skywatcher scope, you may have this thing already.