This LRGB image of the Cocoon Nebula is the first image that I’ve processed from data taken with my “new” Astro-Physics Starfire GTX130 refractor.  I took this image mostly just to test out the scope—not to produce a world class image.  The Cocoon lies among some of the densest star fields in the Milky Way so it’s a great choice for looking closely at field performance and for computing contour plots of FWHM and eccentricity.  So, the story here is mostly about the scope and only a little about the image.  Of course, I hope that you enjoy the image but this is mostly a gear post.

Early in 2019, the opportunity to acquire an Astro-Physics Starfire 130GTX refractor fell into my lap completely out of the blue.  Rowland doesn’t make very many of these scopes so they can be hard to get. Furthermore, the folks at Astro-Physics sometimes get grumpy when these scopes change hands so I won’t go into the details of how this went down.  It took quite a while before I had my hands on the scope and this was during the period when I was working pretty hard on my 20” so didn’t give much thought about what I wanted to do with it.  My first notion was to mount it alongside my 20” on its L500 mount, which I was preparing to send to Chile.  After thinking about that idea for a while, I concluded that it just wasn't a good idea to run it in parallel with the 20".  Wiring it was complicated and my imaging goals for the 20" were going to be completely different than what I wanted to do with a wide-field system.  Furthermore, the 20" system was complicated enough without adding more complexity and I didn't want to delay getting it up and running with a side project that just didn't make sense.  I gave up on the idea of sending the refractor to Chile, thinking that it might be good to have a small scope that I could use locally in Oregon when I felt like imaging.  I had a goofy idea that it was small enough to carry to some dark remote sites in Arizona as well.  However, as it came together, I realized what a stupid idea that was!  This thing turned out to be pretty big and a lot heavier than I’d like to lug around!

Later in 2019, I had been at the NEAIC meeting where Astro-Physics introduced the Mach2 mount and on a whim, I put my name on the list--and forgot about it.  By the time I was starting to wonder how I was going to mount the refractor, I got a note from Astro-Physics saying that my name had come up on the list and that they had a new Mach2 ready to ship to me.  Well, bingo...that solved the mount problem!  I had a FLI-ML16200 camera on the shelf that I planned to use but my overall plan for the configuration was really scattered.  It's funny, I am very energized by working on large aperture scopes but I had a lot of trouble focusing on this wide-field system, so pulling this project together turned into a very drawn-out process.

Configuring, buying, and assembling all of the parts took quite a while—in a large part due to long post-Covid delivery times and supply problems.  I've had good luck with FLI CFW5-7 filter wheels so I ordered a CFW2-7 for this system.  I populated it with 50 mm round Chroma Filters.  I briefly considered running the scope unguided and while that approach can work, that’s a path lined with potential problems.  So, to mitigate risk, I installed a new Optek Sagitta OAG behind the f/4.5 reducer.  I pulled an old Starlight Xpress Lodestar off the shelf for guiding.  An Optec Flip-Flat keeps dust out and makes taking flats easy.

The real challenge was the focuser.  The 3.5" feather-touch focuser that comes with the GTX130 is a work of art and it operates incredibly smoothly.  So, my first thought was to simply buy a Posi-Drive Motor System (PDMS) from Starlight Instruments to motorize focus.  If that didn't work, I figured that I could always step up to something like a Moonlite focuser to add a rotator and a focuser.  Almost as soon as I ordered the PDMS, I heard that Jon at Starlight Instruments had passed away.  His right-hand man, Wayne had left the company and things were left in total disarray.  Regardless, Jon's wife eventually answered emails and she assured me that they would ship the part in 6-weeks.  After 6 months, she was telling me the same thing.  Astro-Physics sells the same part and they told me that they could get the part for me in 6-weeks so I gave them the order.  9 months later, the delivery time was still listed as 6-weeks!  Along the way, I ordered a QuickSync FT-40 focuser from Optec.  It's a nicely made focuser but unfortunately, it connects to the fine focus knob on the focuser, which is no good.  Fine-focus knobs are friction coupled and they slip against heavy loads so this thing was a complete fail.  I had other things going on and I became discouraged with the whole project so I put it all aside for a while.  Then, well over a year after I submitted the order, a Starlight PDMS from Astro-Physics showed up on my doorstep!  It couples directly to the course focus, rack and pinion drive and it works beautifully!  That was all it took to kick my interest in the project back into gear.

I constructed a dedicated control box to hold the PC, IP power switch, power supplies and a substantial UPS.  I chose a Gator rack case designed for holding amplifiers and mixers for musicians.  It has covers for both ends along with wheels, and a retractable handle to make it easy to roll around.  I had a rear panel laser cut by SendCutSend so that I could mount pass-through connectors for all of the connections to the PC and IP power switch. I added a couple of cooling fans to keep the inside cool when all the covers are on. The rack mounted Tripplite UPS that I selected is pretty heavy (~30 lbs) so the entire control box might weigh around 70 lbs. Regardless, I can roll this box out and have the scope connected and running in less than about 2 minutes.   

I put the Mach2 on my old C14 roll-out pier, which is really stout.  Since the mount is prewired to only pass a single power cable and single USB3 cable, it was challenging to get all the wiring set up so that nothing would snag as the scope slews.  Last spring, I finally achieved first-light with the FLI camera and right away, I could tell that the ML16200 was a really poor choice for that scope--which is what I suspected all along.  At F/4.5, the 5.4-micron pixels are just too big and the chip is too small for the field coverage possible with that scope.  So, I ordered another QHY600M-ph.  After discussions with Chris White and looking at data from his GTX130 and from my scope, I ditched the idea of using the reducer.  I've never believed in reducers and this experience further confirmed my bias and added yet one more failed reducer to my pile of stuff that I'm not happy with.  The reducer for this scope was designed in the day of the 16803 but it's not up to the field and sampling provided by the IMX455 chip.  So, I ordered the F/6.7 field flattener from Astro-Physics.

My first images with the new camera revealed an obvious problem—there was a LOT of sensor tilt.  So, I closely examined the image train and the problem was easy to spot.  The FLI CFW2-7 filter wheel mounts the camera dovetail with two set-screws at 120 degrees.  With just two screws, the camera gets cocked at an angle as the screws are tightened.  So, I fooled around with some shims to reduce the problem.  I’ve posted the FWHM plots that I took before and after the fix.  I made it better but it’s hard to make it perfect and that’s why this image is cropped a bit from the full frame.  One corner is a real mess, which suggests residual tilt along with a spacing error—and maybe a bit of decenter.  I’ve ultimately decided that this problem is not easily fixable and I’m working on installing an Indigo filter wheel from Pegasus Instruments. Once I got the scope out under the sky, I dialed in the focusing system.  The belt drive on the PDMS works really well and after tensioning, there appears to be very little backlash.  This is the same focus motor that I used on my C14 and it had a really good performance and a perfect service history.  The thing that amazed me about the V-curve focusing system in SGP is just how slow it is.  I’m used to astigmatic focusing where the system checks and tweaks focus every guide cycle so V-curve focusing seems extremely primitive to me.  I haven’t looked carefully at the thermal sensitivity of the GTX130 so I guessed and set it up to refocus every 40 minutes or so.  Even at that frequency, the throughput of the system is significantly reduced by the need to stop to refocus.  That has me considering rebuilding the OAG to use a rear surface pick-off mirror so that I can use astigmatic focus sensing again.

I’ve attached a guiding plot to show how nice the system works under good seeing.  Frankly, this is better than what I’ve been seeing from my 20” lately.  The seeing can occasionally be really good during this time of year in central Oregon.  This image is simply a pure LRGB image. I took some Ha data but not enough to blend in.  One problem I need to investigate is that the flip-flat does not emit very much light in the red end of the spectrum.  It looks like it might require many minutes of exposure to take a flat through the Ha filter and even more through the SII filter.  This seems to be a problem that's more common with EL panels than LED panels.

I configured the system to run with SGP and that’s the software that I used to take this image.  However, I’ve purchased Voyager and I’m looking into switching to that code once I get a chance to test it out.

Last winter, the folks at Obstech told me that they were going to construct a new building dedicated to small scopes.  The base rental rate is about 2/3 of what I pay for the 20” scope so I’ve signed a contract to send this scope down there this fall.  That will solve my last major problem with this scope—namely that I just can’t tolerate running it in person any more.  This “getting old” thing really stinks and my health gets really screwed up when my sleep schedule revolves around my scope rather than my normal day-night cycle.  So, remote has become my only hope for imaging with my scopes.  (Editorial aside:  I've recently become aware that some folks around here consider remote imaging to be akin to "cheating", which I suppose would make James Webb the ultimate cheat. That's a completely absurd notion.  IMO, remote imaging is certainly NOT easy, nor is it "cheating" and I suspect that anyone who questions it has never tried it.)   Hopefully I can get all the little issues cleaned up and have the scope ready to ship by October…or so.

Whew!  This was a super long post and I’m impressed if you made it this far, but hopefully there’s something useful here for others working on similar projects.  Feel free to post suggestions, comments, or feedback.  In the end, I found that getting this scope configured and up and running wasn't any easier than it was to get my 20” going!

  John

PS. When I processed my image of M45, I discovered that my flat data for the green channel was incorrect.  I don't know what happened to it but it was very clearly wrong so I retook the flat data for the green filter.  Revision 'O' is the image reworked with the correct flat data.  This completely explains why I struggled so much with this data set, why the colors never looked quite right, and why the colors varied so much over the field.  All of you who gave me feedback that something looked off with the colors were right!  I don't know how I missed it but it sure cleaned up nicely.  I'm always amazed that I could miss such a glaring error...but sometimes that happens!