Baader MKII CC + ZWO OAG v2 + ZWO2600MC-PRO - Back spacing & other problems...

[Aramcheck]

I've recently bought a ZWO Off-Axis Guider v2 & carried out the Primary Mirror sealant/spring modification  to my SW 200pds & fitted a Baader Steeltrack focuser.

Subsequent results over the last couple of months haven't however been good & last Thursday's session was absolutely dire... Part of the problem has been in getting the 57.5mm back focus which Baader specify with the MKIII M48 configuration. Initially I made the mistake of thinking I had to get 55mm & had been trying various spacers without much success. The best result over recent weeks was a go at M27, where I'd used the 21mm spacer that came with the camera & a (notionally) 1mm Baader spacer ring. This still had bad coma, but at least the central stars didn't look too bad...

I then realised that I needed a 23.5mm extension (The ZWO 2600mm has 17.5mm from face-plate to sensor & the OAG spec says it's 16.5mm, so that should mean I need to add a 23.5mm). On Thursday I tried using a Baader varilock set to 23.5mm, but even with 10 / 30 second exposures, the stars are distorted throughout the field.

For polar alignment I use an iPolar to align the AZ-EQ6 GT Pro mount, and that all seemed fine... My initial pre-focus 5 sec shot shows I think reasonable out of focus stars in the centre of the frame & what I presume is the effect of coma evident at the top right/left. On the 10 sec, 30 sec and 180 sec exposures the star eccentricity doesn't seem dependant on exposure time.

Also on Thursday, the satellite trails also have a high frequency oscillation, which follows the RA/DEC axes. I read that this could be caused by vibration from the camera fan or the mount gears. I did think the fan was a bit noisy, but I'm not 100% sure. Checking previous datasets, I've not had this problem before except for a couple of nights earlier in August, which had the same issue but much less severe. Last week's M27 session doesn't exhibit this issue & aside from re-calibration I don't think I've changed any of the PSD2 settings. (Earlier in the month I had tried reducing the guide period from 3s to 2s).

The only difference between the M27 image and last Thursday is the swapping the 21mm Ext + 1mm spacer with the Baader varilock.

Sorry this is a bit incoherent - I'm pretty much at my wits end trying to figure things out. Tonight looks like it may be clear, so I don't know whether to ditch the OAG & revert to using the finderscope for guiding, or try again with the Varilock.

Any ideas much appreciated!

Ivor

PS: Prior to trying the Varilock, I'd also bought a set of Astro Essentials spacers. These gave bad coma too, but I realised yesterday that one I used had a hole in the side and so was tilting the assembled tube...

PHD2_GuideLog_2022-08-25_221319.txt PHD2_GuideLog_2022-08-19_225352.txt

[alacant]

Hi

m27 shows triangular stars and there is constant elongation in your star tests. Our bet is that the 1º mirror is distorted. Either the sealant isn't thick enough or it has hardened, preventing the movement of the glass as the temperature changes. You probably pushed the mirror against the cork when you refitted it.

If you're using an OAG, you don't need to modify the cell. It's only needed if you're using a separate guide telescope or if you've unacceptable diffraction patterns around stars. As it is, your stars look to have the potential to be superb, so stick at it.

Strange diffraction from the secondary spider too, as if you've used an altitude-azimuth mount. Any ideas?

In any case, you're going to need more space between the sensor and the shoulder of the m48 connection. We find that most of the 2 element Baader ccs need closer to 60mm.

Our first call would be:

1. to set the mpcc at 60mm and go with the OAG. Or...
2. stiffen the tube with a longer Losmnady dovetail with the guide telescope mounted on a stiff top rail.

Both will need the mirror removed, silicone cleaned and replaced with the correct type and amount. Guidelines which work for us, here, the main part being [1]

 Seal the mirror to the cell using neutral silicone sealant, generous blobs to coincide with the gso cork. Leave the mirror to settle under its own weight on a horizontal surface, that a blob of the same sealant under the same ambience returns to its original form when distorted. Only then refit.

Cheers and HTH

[1] gso used but also tested with a 200pds, 130pds, bresser etc; the same applies to any low end Newtonian.

Edited August 27, 2022 by alacant

[Aramcheck]

Thanks @alacant I'll try increasing the back focus to 60mm.
On the primary mirror I just let gravity do the work & left it on a flat surface for a couple of days, until the separate test blob had cured ok. (i.e. no pushing was involved!) I've attached a couple of pictures showing the amount of sealant used & one of the rear 'after' shots.

I must apologise too, as I'd forgotten that the M27 I posted was a combination of some 2021 data and from the 19th Aug 2022. The 2021 data set was at a vastly different angle & only overlaped a central band, so that's why the diffraction spikes look odd.

 I've attached an unprocessed M27 image with just the 19th Aug 2022 data & a STF applied.

Cheers
Ivor

[alacant]

20 hours ago, Aramcheck said:

I'll try increasing the back focus to 60mm

You'll always get some astigmatism in the corners over aps-c with the Baader mpcc anyway and the corner elongation should respond in going to ~60mm spacing.

Apart from the points above, the new m27 stars look fine to me.

Guiding at 1000mm is never going to be easy and you'll need nights of very good seeing to be able to do it justice. For other nights, maybe invest in something like a 130pds? Oh and a GPU cc;)

Cheers and HTH

Edited August 28, 2022 by alacant

[michael8554]

Looking at the star shapes of the 10s exposures, I'd agree that more backspacing is required.

The 180sec exposures will have guide errors, so don't judge the backspacing using those.

Looking at the GuideLog for the 19th:

The Calibrations had warning messages after the Cals,  and again in the Guide Assistant results:

"Last Cal Issue = Rates,"

"Norm rates RA = 17.6"/s , Dec = 14.3"/s; ortho.err. = 7.6 deg"

"GA result- Recommendation: Consider re-doing your Calibration (Prior alert)"

Which you seem to be ignoring ?

You would get better Cals if you pulsed north until the guide star moves consistently, showing that Dec Backlash had been taken up.

THEN Cal.

Near south and near Dec = 0 if possible, but Dec = 28 isn't too far out.

It would be worth Binning the guide camera, to improve sensitivity and pixel scale.

Your Dithers are only a maximum of about 6 pixels on the imaging camera.

Pretty much the same errors on the 25th.

Michael

 

[Aramcheck]

Thanks @alacant & @michael8554 - much appreciated.

@alacant :- I do have a 130pds, but only tend to use that if it's too windy for the 200pds. A  better coma corrector is something I probably will look into when finances permit. Not sure what the advantage of a Losmandy plate are, but I do intend to put a 2nd bar on the top of the OTA, as you've suggested in other posts.

@michael8554 :- Thanks - the tip about moving North before calibration is a great suggestion. I'll also try guiding with binning x 2. Since moving over to the OAG I haven't managed to get PHD2's guide assistant to run successfully, due to Dec backlash not giving enough movement South. Normal calibration has also been hit & miss, but even using the 50mm finderscope for guiding it wasn't uncommon to get calibration warnings about large Dec backlash. On the Dithering I was working on the assumption that with the OAG the guide camera & imaging camera had the same pixel size then I should use a setting of 1 in Astro Photography Tool. I'll try increasing that, although I haven't noticed much fixed noise.

Another suggestion I read about, was to increase the size of the box around the guide star in PHD2. When the scope is pointing at lower altitudes I sometimes get large deviations which make subs unusable... I'd put these down to wind/glitches/cable snags, but I think now that it's occuring when the guide star randomly moves outside of the target box in PHD2, presumably due to the larger atmospheric thickness at lower altitudes. Something else I'll need to look into.

Cheers
Ivor

 

[alacant]

10 hours ago, Aramcheck said:

Losmandy plate are, but I do intend to put a 2nd bar on the top of the OTA,

If you're going with the OAG, neither are needed.

10 hours ago, Aramcheck said:

large Dec backlash.

To help diagnose issues such as these and to give your software a fighting chance,  the mount has to be better than perfect. Out of the box, we've yet to see a correctly adjusted, internally clean and correctly lubricated sw mount. An hour or so spent dismantling and adjusting would serve to lower any frustration to a minimum. There are many guides on disassembly and adjustment.

Cheers and HTH

[michael8554]

"Since moving over to the OAG I haven't managed to get PHD2's guide assistant to run successfully, due to Dec backlash not giving enough movement South."

The GA run was successful. The optional final measurement, you didn't have "Measure Dec Backlash" ticked.

Because I guess you already knew that previous GA runs had advised the backlash was too large to measure.

Need to adjust that.

"Dithering I was working on the assumption that with the OAG the guide camera & imaging camera had the same pixel size then I should use a setting of 1 in Astro Photography Tool"

Well the figure you set that to depends on how much dithering you need, 12 pixels is a commonly quoted figure.

"Another suggestion I read about, was to increase the size of the box around the guide star in PHD2. When the scope is pointing at lower altitudes I sometimes get large deviations which make subs unusable."

That will only work if the "large errors" are moving the star out of the box - shouldn't happen.

Michael

[alacant]

11 hours ago, Aramcheck said:

haven't noticed much fixed noise

I don't think you'll need dither with a 571.

[Aramcheck]

I finally got a couple of spells of clear-ish nights. With the back focus set to 60mm the coma looks much reduced. I've also adjusted the Dec backlash after watching a few youtube videos (Martin's AP, Cuiv The Lazy Geek and Astrobloke). This helped PHD2 to complete the calibration / guide assistant process, although I haven't yet tried binning... On the 12th I thought everything was going ok , but the stars still had a triangular shape. On the 13th I made sure that the primary locking nuts were not touching the mirror cell & took a few shots of Vega whilst varying the focus. (The full sequence can be seen here: https://youtu.be/KuIURZ9u1p8) I also took a set of 10s exposures in addition to my regular 180s ones for a couple of targets - M27 & another near zenith. These also show triangular stars. I didn't bother to dither on either night, to maximise the amount of data I could collect.

I presume the next step will be to cut the primary mirror free & to try again with the sealant. Am I missing anything else?

Photo of the stacked/unprocessed 10s M27 subs with a STF applied.

Thanks again for all the help!
Ivor

PHD2_GuideLog_2022-09-13_220027.txt PHD2_GuideLog_2022-09-12_213448.txt

[edarter]

I could be wrong, but with regard to the sealant I think best practice is to pack between the mirror and the base while the sealant cures, in order that it doesn't just squidge and settle with the mirror touching the base? Thats what I had noted anyway for when I get round to doing this mod on my 130PDS.

Thanks
Ed

[Aramcheck]

Found an interesting reply last night on the SGL on the thread linked below from @emyliano2000- I will have to check the alignment of the focuser I think before trying to cut the mirror free.

Cheers
Ivor

 

[michael8554]

GuideLog for 13th looked good.

Similar RA and Dec errors should mean round stars.

You could try Binning the guidecam, which will up the pixel scale and increase sensitivity.

Which would allow you to reduce the 3 sec exposure time, as MultiStar compensates for "Chasing the Seeing" to a degree.

Michael

[Jerry Barnes]

I wish I'd read this thread earlier, very helpfull and I realise I need to increase my back focus having essentially the same setup.  Thanks for starting it.

[Aramcheck]

On 17/09/2022 at 13:52, michael8554 said:

Similar RA and Dec errors should mean round stars.

You could try Binning the guidecam, which will up the pixel scale and increase sensitivity.

Thanks @michael8554 . I had a go at binning the guidecam x2 last night. Results were as good as x1 but it made it easier to find a star with decent SNR. In the middle of the session I tried switching back to the x1 binning, to see if I could could get a comparison, but the SNR was too low to bother with, so I switched back to the x2. I then had to recalibrate & continue without Dark frames as I messed up 1st time around (cancelling building the dark library after 3.5 secs). Both guide sessions seemed good.

I also had to readjust the Baader varilock as it had come loose after the previous session a fortnight or so ago. Star elongation however last night was quite pronounced, but roughly in the same direction in all positons across the image. I tried taking 20 x 10s subs & they look a bit better, but there is still significant elongation. On the 180s subs only about half  could be successfully star-aligned in Pixinsight. The PHD2 log measured polar alignment as 0.9' & 0.6'. At the end of the evening I also tried reconnecting the iPolar & that hadn't changed much, so I don't think polar alignment was the cause. The elongation appears I guess about 30 deg from the Dec axis.

Early in the evening I also tried taking 20x0.1s subs with Vega de-focused & placed in different parts of the image. Also 1x0.5s sub with vega focused... I'm not sure what if anything these indicate...

Have just ordered a Hotech 2" SCA Laser collimator and an M48 to M42 adapter, so I can try next time without the varilock. Hoping also to get some time off work so I can go over the collimation & focus tube alignment.

Any advice much appreciated!!
Cheers
Ivor

PHD2_GuideLog_2022-09-24_221843.txt

[michael8554]

From the GuideLog, the final guide session was good  RA = 0.37arcsec, Dec = 0.34arcsec,

So stars should be round, but images show elongation predominately in Dec !

"but it made it easier to find a star with decent SNR."       " but the SNR was too low to bother with"

Does that mean that you are choosing stars instead of PHD2 ?

a) so long as the star that PHD2 autoselects shows "SNR" on the toolbar in green, then PHD2 is happy, no need to look for "better" stars.

b) manual selection disables MultiStar, not a problem if the OAG only gives you one star.

You don't seem to be Calibrating at +/-20 from Dec = 0 ?  Dec = 66.2 deg,    Dec = 31.8 deg,

Michael

[Aramcheck]

6 hours ago, michael8554 said:

Does that mean that you are choosing stars instead of PHD2 ?

a) so long as the star that PHD2 autoselects shows "SNR" on the toolbar in green, then PHD2 is happy, no need to look for "better" stars.

b) manual selection disables MultiStar, not a problem if the OAG only gives you one star.

You don't seem to be Calibrating at +/-20 from Dec = 0 ?  Dec = 66.2 deg,    Dec = 31.8 deg,

Michael

I usually just let PHD2 select the main star, but with the OAG I've also been adjusting the camera gain. eg if the only visible star in the OAG is saturated then I'll reduce the gain & then re-select it in PHD2. If the chosen star peaks less than 200 then I'll try increasing the gain a bit. Whenever possible I use the multistar.

From our back garden guide calibration is a compromise between sky brightness towards the city centre at lower altitude & the available small patch of sky to the south around neighbouring houses / trees.

BTW - I've attached a comparison with a stack of the 10s and 180s subs to show the variation in star elongation. I'm now wondering if vibration from the ZWO 2600 fan could be a problem. That'll be easy to check next time...

Thanks
Ivor

[alacant]

Hi

I see the triangular stars are noticeable by their absence:)

For an eq6 at 1000mm focal length, good. 

If you really can't live with the elongation, read on. Otherwise, just correct in software.

DEC:

sudden spikes:

This is caused by something mechanical/cable/wind/animal; the DEC motor is not running before the spike. PHD2 tries to correct, but the magnitude of the pulse causes DEC to overshoot:

You could try lowering aggression and DEC backlash, but in the end, perhaps the best cure is to dismantle and re-build the mount. If you've ever witnessed the inside view of a low end sw mount, new or otherwise, you'll see why.

RA

Before dismantling and rebuilding, try PPEC with a fixed peiod of 122s. Unfortunately though, unless DEC issues are addressed, this is likely to make the elongation narrower, so more noticeable.

A few last bits:

- Do you still have the rubber flex inducing washers (7 off)  in the focuser?

- Although unlikely due to use of an OAG, steel sw tubes are nortoriuos for flex. Maybe consider adequate support.

- It seems you now have the primary mirror cell performing. If not ...etc.

- are you certain you have the 120mm after an ir-cut filter?

Cheers and HTH

Edited September 29, 2022 by alacant

[Aramcheck]

Thanks @alacant. I had a bit of clear sky this week & had another go with some M48 spacers replacing the Baader varilock (total backfocus now 60.25mm).

I wanted to see how the stars changed with camera rotation & with the fan switched off. After calibration I took a few subs of M27 & then rotated the camera, recalibrated & took more subs. My polar alignment was poor (I must have knocked the scope), but I think the results eliminate the camera end of things as the source of the problem. Here's a section from near the middle of the frame:-

After M27 I then went to NGC7129 which was at a higher altitude and the source of the elongated stars on the previous session - this time the stars looked rounder but still with the distortion.

On the OTA I replaced the SW focuser with a Baader steeltrack at the same time as making the primary mirror cell mod. I'm currently awating delivery of a Hotech SCA laser. Once that arrives I'll check the mirror & focus tube alignment...

I'm certain on the IR-Cut filter fitted to the guidecam - it's a cheap Astro Essentials filter.

When you say try PPEC do you mean within PHD2, or should I use EQMOD & PecPrep to create / save a PEC curve to the mount firmware? I have to set-up the scope at each session, so I'm not confident that saving a PEC to the firmware would be repeatable.

Thanks for pointing out the large DEC movement occuring before the DEC pulse. There may be vibrations when I have to walk past the scope or poor lead management. I use some Suppression pads underneath the scope legs & an old rug to walk on as I have to pass within 1m of the scope when checking the laptop (the flat / dark bit of our back yard is quite small). I should be able to improve the cable management though...

I've looked at some of the youtube videos on a dismantle & rebuild of the mount & a pdf guide including this SGL thread. It's definitely something I'll attempt, but only after making sure I've got the mirrors / focus tube aligned as best I can.

Thanks again for all your help!

Cheers
Ivor

PHD2_GuideLog_2022-09-29_204409.txt

[michael8554]

Yes try the PHD2 PPEC algorithm at a fixed 122 secs.

From the log I see           Norm rates RA = 18.1"/s @ dec 0, Dec = 14.8"/s

At mount settings of      RA Guide Speed = 13.5 a-s/s, Dec Guide Speed = 13.5 a-s/s,

I'm really not sure of the effect of the calculated RA guide rate being more than sidereal rate.

But pretty aggressive guide rates, perhaps try a lower RA and Dec mount setting and see what happens ?

Guide figures on the last session were good  RA = 0.36arcsecs,  Dec =  0.34arcsecs.

Though reported PA error was 11.2arcmins at Dec = 66.2

The previous run was 3.7arcmins at Dec = 22.8

Tripod shifting ?

Michael

 

 

[alacant]

Hi

I see one frame with elongated stars and another with round(er) stars. Have you had a look at the guide log to see what, if anything, is different between the frames? Over 1000mm, you need only slight protuberances to cause elongation. If not, something is moving during the exposure.

On 01/10/2022 at 11:29, Aramcheck said:

PPEC do you mean within PHD2, or should I use EQMOD & PecPrep

EQASCOM already has a capable PEC algorithm, so no need to introduce anything else into the mix. PHD2's PPEC will do a good job of removing the 122s RA oscilation. If you want to get rid of it all, record VS-PEC for 40 minutes when the PPEC guiding has had chance to settle. 

On 01/10/2022 at 11:29, Aramcheck said:

definitely something I'll attempt, but only after making sure I've got the mirrors / focus tube aligned

FWIW, we work the other way around. Before anything else, the mount is stripped, cleaned, rebuilt and adjusted.  We then perform a guiding session under INDI to make certain we can eliminate mount mechanics. We find it quicker and with fewer lost frames that way.. We speak from bitter experience!

Cheers

Edited October 2, 2022 by alacant