Username:

Password:

UKAI is moving to Facebook

Author Topic: Innovations Foresight ONAG Mini Review  (Read 3429 times)

0 Members and 1 Guest are viewing this topic.

Offline chris.bailey

  • Administrator
  • Galactic Poster
  • *****
  • Posts: 10012
  • If at last you do succeed, don't try again
    • Rosemount Observatory
Innovations Foresight ONAG Mini Review
« on: 15:24:36, 14 April, 2016 »
Innovations Foresight ONAG Mini-Review

Background

With improvements in amateur astrophotography equipment over recent years, there is a tendency to use longer and longer sub-exposures. Fifteen minutes exposures that were pretty much unheard of even 5 years ago are now quite normal practice. Even thirty minute (and longer) exposures are not uncommon, particularly for narrowband imaging.

Long exposures, however,  are very good at revealing mechanical shortfalls, particularly flexure. Whatever the guide scope mounting system, some degree of differential flexure (DF) between the guide and imaging system is pretty much inevitable. Whether this impacts on the resulting images is largely down to the imager scale and how much out of roundness the user accepts. A rule of thumb is that absolute roundness values of greater than 90% are considered to be visually round. That is quite a tall order to achieve at higher resolutions, say below 1.5”/pxl, and/or long focal lengths.

DF is quite easy to measure (assuming there is access to plate solving software). With guiding running, take a series short un-dithered exposures (low in the east or west will tend to maximise DF) over a period of say 30 minutes. Plate solve the first and last and calculate the difference in co-ordinates between the two. As the guider has maintained position of the star in the guide scope, any movement of the image over the imaging period is predominantly DF. Compare the level of DF to the image scale.

For me, using a separate guide scope for guiding has never resulted in sub-exposures longer than  ten minutes with acceptable roundness and measured DF has been in the region of 10-15” per hour. Long (20 minute) narrowband imagers rarely had roundness values greater than about 80% and many were closer to 75%. Whilst this was just about acceptable at a wide-field image scale close to 3”/pxl, it really started to be obvious at 1.5”/pxl and below. Changing to an Off Axis Guider (OAG) gave immediate improvement but some small degree of DF remained. Robustly built Off Axis Guiders are hard to find. Add in the difficulty with using an OAG on larger chip cameras (limited room for the pick off prism on the periphery) and sub-optimal guiding on some systems due to poor guide star ‘shape’ at the edge of the light cone and it has never been a 100% satisfactory solution. On my long focal length RC (2000mm) guiding has been particularly problematic over the long exposure imaging at f8 requires. The Innovations Foresight ONAG unit seemed worth a try.

The ONAG only appears to be available from Innovations Foresight themselves. Ordering was simple and confirmed by email, with a further emails being sent upon shipping. The unit arrived in the UK within five days with the inevitable import duty, import administration fees and VAT being payable on delivery. These together add close on 30% to the costs so making it a similar cost in sterling to that quoted in dollars!

The Innovations Foresight  ONAG

There are two models of ONAG, the SC for imaging devices up to APS-C  (28mm diagonal) and an XM for large format CCD’s up to 50mm diagonal. This review covers the smaller SC unit though the larger unit is very similar in operation. The XM is considerably larger and heavier (850g compared to 770g). It is also more costly. The SC ports are fixed T-Thread with the XM having interchangeable ports to suit a variety of cameras, with custom adaptors being available. Both models have a non-rotating helical focuser (with locking mechanism) on the guider port to enable accurate common focusing of the guide CCD with the imager. Older models were not fitted with a helical focuser and were reportedly difficult to co focus. The are limits on the length of male T-thread that can be used on the scope port to avoid interference with the cold mirror mountings so some care is needed during assembly.

The basic theory of the unit is quite simple. A Dichroic Beam Splitter (Cold Mirror) reflects light in the visible spectrum upwards to the imaging system, with any light over 750nm wavelength  passing straight through to the guide camera. The Cold Mirror is made of optical grade quartz and is claimed to reflect >98% of the light up to 750nm. At the rear of the unit (guide port) there is an X-Y stage to allow the guider to be moved around, to find optimal guide stars, with an exploration circle of 46mm.

Guiding, therefore, is carried out in the Near Infrared (NIR) spectrum. Most cameras are sensitive to NIR though it is important that there is no UV/IR cut filter in the chain, for obvious reasons! The brightness of the guide star in NIR will be dependent on its spectral class but expect to need a guide star of between 1 and 3 magnitudes brighter than if guiding in full spectrum. Guiding in NIR is claimed to reduce the affects of seeing as the longer wavelength of light is less affected by atmospheric turbulence. The angle of NIR light passing through the cold mirror leads to quite severe astigmatism. This results in out of focus stars being a long streak which change orientation by 90 degrees as critical focus is passed. The in focus guide star is therefore a small cross in good seeing or more typically square like.

The ONAG SC unit takes 68mm of back focus to the imager port and 90mm to the guide port. This 22mm difference meant, for me, that with a C-T converter on the guider port (8mm) my SX H16 and SX Filter Wheel would be in common focus with the Lodestar without the need for spacers and this proved to be the case with only minor adjustment of the helical focuser. Initially the complete unit (ONAG, Filterwheel, Lodestar, H16) was fitted (rather insecurely) to a 3” Feathertouch focuser using a compression ring only. This has since been exchanged for a custom adapter made by Precise Parts of Miami (who have all the key dimensions of the ONAG units in their database) and, based on the cost, is made from high grade Unobtanium!

The ONAG unit is very robust, well machined and finished. T Threaded locking collars are provided to allow the guider and imaging train to be locked out in a certain orientation. I like to run the imaging camera with a rotation angle close to zero (imaging north up-down) but with the guider at 45 degrees (CCD Autopilot does not like the guider set with a zero rotation angle). Setting this all took a little time and it would be nice if the locking collars had capstans to help secure them as the one under the filter wheel is all but inaccessible, leading to a lot of trail and error. It does look very strange having the imaging camera riding above the unit and the guide camera at the rear but the centre of mass is shifted forwards compared to having the imaging camera and filter wheel at the rear reducing the net lever arm on the focuser. I often saw hanging a heavy camera off the back of a filter wheel to be a potential source of flexure, this arrangement does reduce that to a degree. I did need to add a small spacer between the focuser and ONAG to prevent the filter wheel fouling the housing at the focusers home zero position and this may be an issue with even larger filter wheels.

If a field flattener or focal reducer is required in the imaging train, things get a little more complex. If fitted to the scope port the total back focus may be too great (68mm + 25mm (filterwheel) + (17mm camera = 110mm) whereas putting it on the imager port is likely to require adding spacers to both imager and guider. This would all need great care to avoid the very flex you are trying hard to avoid.

Optec FocusLock

A by-product of the astigmatism caused by the cold mirror is that the guide stars can be used to measure and adjust focus using an electronic focuser. Through critical focus the guide star will follow a -- + | sequence and therefore has direction and magnitude information. FocusLock software continuously monitors the guide star and adjusts a connected electronic focuser accordingly. The software runs with Maxim or SkyX natively though can also interface with Ascom systems. A facility for setting filter offsets is included. Once calibrated and running, FocusLock should maintain focus over long imaging sessions and negate the need for temperature compensation or refocusing. My Williams Optics FLT110 is particularly sensitive to temperature changes and ideally requires refocusing at every 1 degree temperature change.

Note that the FocusLock software is an additional cost add-on downloadable from Optec.

ONAG First Light


Having got everything bolted together I managed to run a short luminance session on M51 (Whirlpool Galaxy) with a Williams Optics FLT110/3” Feathertouch Focuser on a Paramount MX. Mount modelling (TPoint) was turned off, though PEC was applied to give a residual PE of around 0.8”. I would normally run the MX with a large Point model so that guiding is merely providing an occasional nudge, but wanted to test the ONAG unassisted in this respect.

Getting the imaging scope and guide par-focal was much easier than expected. Though the helical focuser does have a degree of backlash, the fine thread allows very accurate adjustment. I used SkyX and @focus to get focus on the imager and then continuous imaging on the guider to get the best Half Flux Diameter (HFD) reading I could on the guider (Lodestar X2). Once in common focus the locking screw is used to securely lock off further adjustment on the helical focuser. The X-Y stage on the guide port is smooth and locks off securely. Moving it around has little effect on the focus of the guide star.

The first thing you notice is how much darker the guider background is in NIR. Using a dark frame on the guider (to negate hot pixels) the background values during guiding are near enough zero! M51 is a tough part of the sky for guide stars and the X-Y stage was needed to find one of sufficient brightness in NIR, this worked well but raises some concerns for automated imaging. Even so expect to be guiding at much lower brightness values than with full spectrum and longer than normal (10 second) binned 2x2 guide exposures seem to work best. SkyX calibrated easily, however, and guiding went smoothly for a dozen 10 minute dithered sub-frames. Star roundness values of 87-93% in the frame centres was typical and consistent despite seeing conditions being in the 3” region, so sub-optimal.

One thing to be aware of is that images from the ONAG are mirrored, this is easily corrected in post processing but easily forgotten.

Plate solving images over the session gave effectively zero image displacement over the period and no star alignment was really needed to stack images. Given the micron level consistency needed to achieve that it is quite an achievement. CCDInspector results also suggested that the previous tilt in my imaging system is much reduced.

ONAG Second Light

A greater test was mounting the same FLT110 on an EQ6 Pro at Kelling Heath Spring Star Party. Though most nights were too windy for decent imaging, the middle night was perfect with no wind, decent seeing and good transparency. Again the ONAG performed very well over 23 six minute luminance sub-frames of M101 with star roundness values typically in the high 80%’s from a mount with no PE correction applied. The EQ6 does not like long guide exposures so finding a guide star with short 3 seconds exposures was a bit of a struggle but not really any more than it would be with an OAG (which often necessitates a small slew displacement to find a guide star)

Having the ONAG, filter wheel/imager and guider all bolted securely together meant transporting everything as a single unit, which made setup at the star party and again back at home very quick and easy. The guider and imager remained in co-focus throughout.

Even on the EQ6 Pro, sequences of exposures show differential flexure on the verge of unmeasurable

FocusLock First Light

I have yet to use FocusLock in anger on an imaging session but have tested that it calibrates and responds to manually imposed focus moves. Manually moving the focus by 50 steps initiates FocusLock adjusting focus back with it arriving back within a few steps of the starting position within a matter of a couple of guide cycles. This was repeatable over several cycles in either direction.

Unfortunately, FocusLock has not been picked up by the automation software writers such as CCD Autopilot though it could possibly be scripted in. This limits its use a little as it the all night unattended sessions where it would prove of most value.

Early Conclusions

Pros -

  • Well made robust unit built for years of use
  • Zero differential flexure measured in tests so far
  • Use with FocusLock promises no focus drift over long imaging sessions
  • The ONAG, guide camera, filter wheel and imaging CCD essentially become a single unit making scope to scope transfer simple


Cons -

  • Using a focal reducer/flattener could be frustrating to get correctly spaced
  • Getting all elements into the required orientation is fiddly enough to add significant funds to the swear box. Capstans on the locking rings would help enormously.
  • Additional weight of ONAG, may over-stress small focusers
  • Cost (import from USA with attendant VAT and Import Duties) helps make it a very expensive guiding solution
  • Need for higher magnitude guide stars for effective guiding. Guide stars may be difficult to find in automated sessions
  • FocusLock not currently adopted by automation software requiring custom scripting
LX200|ZS70|FSQ85|FLT110|Altair DF250RC|EQ6 Pro(Rowan Belt Mod)|ParamountMX
ATIK383L+/EFW2/OAG|Lodestar|Baader 36mm LRGBHaSIIOIII
Starlight Express SXVRH16/ONAG/FW|Lodestar X2|Baader 2" Filters
Starlight Express SXVRH814/ONAG/FW|LodestarX2|Baader 2" Filters
Lunt LS60PTBF1200|DMK41|Quark Chromo
Samyang 135mm f1.8

Offline chris.bailey

  • Administrator
  • Galactic Poster
  • *****
  • Posts: 10012
  • If at last you do succeed, don't try again
    • Rosemount Observatory
Re: Innovations Foresight ONAG Mini Review
« Reply #1 on: 10:03:15, 21 April, 2016 »
Given the full moon clear sky last night I tried FocusLock on a live session.

In the graph below I purposely moved focus off by 60 steps. FocusLock sees this change to the guide star roundness and adjusts focus back again. More testing/fine tuning is needed but first impressions are that this would work very well on long unattended imaging sessions
LX200|ZS70|FSQ85|FLT110|Altair DF250RC|EQ6 Pro(Rowan Belt Mod)|ParamountMX
ATIK383L+/EFW2/OAG|Lodestar|Baader 36mm LRGBHaSIIOIII
Starlight Express SXVRH16/ONAG/FW|Lodestar X2|Baader 2" Filters
Starlight Express SXVRH814/ONAG/FW|LodestarX2|Baader 2" Filters
Lunt LS60PTBF1200|DMK41|Quark Chromo
Samyang 135mm f1.8

Offline mcgillca

  • Full Member
  • ***
  • Posts: 136
Re: Innovations Foresight ONAG Mini Review
« Reply #2 on: 14:21:46, 22 April, 2016 »
Great review, Chris.

I also use the ONAG and have found the differential flexure to be essentially zero - see plot. This is generated from a series of 60 1 minute subs where I've calculated the displacement from the initlal frame. As you can see, there is a random scatter of about 0.5 pixels (about 0.3"), but essentially no flexure - the trends on the plot are close to random.

It was a little tricky to set up, but once I'd got the spacing right, has worked near flawlessly. As a test I combined three one hour subs together with no displacement and got great roundness (~ 83%) and FWHM 1.75" - see central cropped area of SH2 106 (there are lots of hot pixels since there is no dithering).

I use a cooled Atik 214L+ as my guide camera, binned 3x3. With that, if I set my minimum guide threshold at 1500 adu in CCDCommander, provided there is a magnitude 10 star somewhere in the field, I'm generally fine. Very occasionally I have to reduce this to 800 adu to find a guide star (M33 always gives ms problems), but with an 8" RC, I rarely struggle to find a star.

It certainly works well at a permanent observatory - not sure how fiddly it would be if you try and set up each day, but perhaps no more fiddly than a conventional OAG?

Also, I found that the support from Innovations Foresight is superb - Gaston built me a special version with a fixed stage since I was initially having problems eliminating flexure.

Colin

Main scope: Ikharos 8" RC
Camera: Atik 460ex with Atik EFW and Baader LRGB Ha, OIII and SII filters
Guide scope: On Axis Guider with Atik 314L+
Mount: Paramount MX located in Astrocamp, Nerpio, Spain

Offline mcgillca

  • Full Member
  • ***
  • Posts: 136
Re: Innovations Foresight ONAG Mini Review
« Reply #3 on: 14:23:15, 22 April, 2016 »
Ooops - forgot the images...
Main scope: Ikharos 8" RC
Camera: Atik 460ex with Atik EFW and Baader LRGB Ha, OIII and SII filters
Guide scope: On Axis Guider with Atik 314L+
Mount: Paramount MX located in Astrocamp, Nerpio, Spain

Offline mcgillca

  • Full Member
  • ***
  • Posts: 136
Re: Innovations Foresight ONAG Mini Review
« Reply #4 on: 14:26:57, 22 April, 2016 »
And not much help without stretching the image...
Main scope: Ikharos 8" RC
Camera: Atik 460ex with Atik EFW and Baader LRGB Ha, OIII and SII filters
Guide scope: On Axis Guider with Atik 314L+
Mount: Paramount MX located in Astrocamp, Nerpio, Spain

Offline chris.bailey

  • Administrator
  • Galactic Poster
  • *****
  • Posts: 10012
  • If at last you do succeed, don't try again
    • Rosemount Observatory
Re: Innovations Foresight ONAG Mini Review
« Reply #5 on: 15:16:27, 22 April, 2016 »
Colin

That is pretty much what I have found. Any flexure is effectively lost in the seeing. Your ADU count limit is pretty much what I have found, without the ONAG I would not have tried guiding on anything much below 9,000. With the ONAG 2,000 is pretty hard to achieve.

Chris
LX200|ZS70|FSQ85|FLT110|Altair DF250RC|EQ6 Pro(Rowan Belt Mod)|ParamountMX
ATIK383L+/EFW2/OAG|Lodestar|Baader 36mm LRGBHaSIIOIII
Starlight Express SXVRH16/ONAG/FW|Lodestar X2|Baader 2" Filters
Starlight Express SXVRH814/ONAG/FW|LodestarX2|Baader 2" Filters
Lunt LS60PTBF1200|DMK41|Quark Chromo
Samyang 135mm f1.8

Offline tomhow

  • Tom How
  • Galactic Poster
  • ******
  • Posts: 3137
  • Making equipment is far more fun than buying it.
    • The Curdridge Observatory
Re: Innovations Foresight ONAG Mini Review
« Reply #6 on: 18:31:00, 25 April, 2016 »
Thanks for the review Chris. Very interesting reading. I've done a little bit of scripting work with somebody in Australia who uses  FocusLock but now I understand a lot more about how it works!

I guess any residual flexure is more likely in the camera+filterwheel than the ONAG itself. Very good result.

If it is just a case of turning the system on and off, perhaps it can be implemented as an ASCOM switch interface to allow use with automated scripting?

Tak Sky 90, Atik 490, Homemade Mount, OAG, Lodestar

The Curdridge Observatory

Offline chris.bailey

  • Administrator
  • Galactic Poster
  • *****
  • Posts: 10012
  • If at last you do succeed, don't try again
    • Rosemount Observatory
Re: Innovations Foresight ONAG Mini Review
« Reply #7 on: 19:24:09, 25 April, 2016 »
Tom

There is a scripting Api so that should be possible. There is an inevitable offset between best focus on the guider and best focus on the imager. This is allowed for in the software. In short you do a best focus on the imager and then lock that into the guider i.e. it know that the guide image may not be at its own best focus to result in best focus on the imager. In the couple of times I have tried it so far, that offset value has varied a little. That may just be a result of conditions but I would like to do a few more tests before trusting it.

public ScriptFocusLock()
This initializes the scripting object and starts the FocusLock UI.

 public bool isLockAvailable()
This reports true when FocusLock is able to start focusing and engage the lock.

public bool isAtFocus()
       This reports true when the guide image is in the Critical Focus Zone.

public bool isConnected()
       This reports true if FocusLock is connected to the necessary devices.

public bool isLockRunning()
       This reports true if FocusLock is maintaining the focus. This is called Lock mode.

public bool IsRunning()
       This reports true if the scripted copy of FocusLock is running.

public void ConnectDevices()
       This will instruct FocusLock to connect to the preselected devices. Any errors will be displayed in FocusLock. A pop up Windows Notification will appear if all are connected.

public void DisconnectDevices()
       This will instruct FocusLock to disconnect from the preselected devices.

public void StartLock()
       This will instruct FocusLock to begin focusing and to engage the lock.

public void StopLock()
       This will instruct FocusLock to stop focusing and to disengage the lock.

public void CloseFocusLock()
       This will tell the scripted copy of FocusLock to close.

Chris
LX200|ZS70|FSQ85|FLT110|Altair DF250RC|EQ6 Pro(Rowan Belt Mod)|ParamountMX
ATIK383L+/EFW2/OAG|Lodestar|Baader 36mm LRGBHaSIIOIII
Starlight Express SXVRH16/ONAG/FW|Lodestar X2|Baader 2" Filters
Starlight Express SXVRH814/ONAG/FW|LodestarX2|Baader 2" Filters
Lunt LS60PTBF1200|DMK41|Quark Chromo
Samyang 135mm f1.8

 

ukbuysellRemote Imaging from AustraliaSharpSkyblank APTUKAI on Facebook
Powered by SMF 2.0.15 | SMF © 2006, Simple Machines LLC
DarkBreak by DzinerStudio. Theme modified by The UKAI Team

Page created in 0.406 seconds with 36 queries.
TinyPortal © 2005-2012