5/3/2010. Clouds and rain. I used this star-free interlude to take care of some housekeeping. Some things got done, others not yet. Here are a couple of Have Done and To Do lists:
Remaining on the To-Do List:
All in all, as they say, we could use some rain.
(1) needs something at infinity for a focus target, and (7) needs to be done under realistic loads and weather. (2) was a bit confusing while the Canon drivers (which must be present) and Nebulosity vied for control of the camera. Eventually, I convinced the EOS drivers to load on detection of the camera and the EOS utility not to run. That left Nebulosity in control. Looks good. (3) turned out to be a piece of cake. I tried Maxim DL5 first because it is familiar, and it worked at least as well on the Netbook as on the tiny Vaio with the compromised keyboard. Then I tried Nebulosity 2.0 which is a simpler, less ambitious program tailored for use in the field rather than in the imaging lab. All the critical bits I could test on the kitchen table worked well and intuitively for both the SBIG CCD and the Canon EOS. Purchase a licence for that puppy! Done. (4) Astrodon 1st Generation, E-series filters are in place in the CFW-8 on the ST2000XM CCD, and as far as I can tell they really are parfocal. "As far as I can tell" is about 15 feet: an image of the Christmas cactus at the north window of the galleria. Yes, I know that they should be the I-series for my camera, but nobody was offering those used at bargain basement prices when I was shopping for parfocal filters; I expect the E-series to be virtuous enough. (5) took shape from an acrylic sheet cut to cover the 5-inch dewcap, painted black, with two large holes, four bolts, and some electrical tape suitably arranged. Nevermind the handwaving; there'll be a picture by and by. I again followed Ron Wodaski's advice about arranging diffraction spikes to indicate critical focus. (6) was dispatched using speaker wire, a soldering iron, and more electrical tape. Four, 12v sockets are now available from the battery in a box and a dedicated charging cable hangs out one end of the case. It's a good start on portable power.
5/4/2010: As for item (1), use the same arrangement as for native FL imaging: FeatherTouch in A-P focusser, then Maxbright in FT, then Barlow in Maxbright, and rack the A-P focusser out about 1.2 inches. The 0.75x telecompressor is known to work with the custom-made spacer to get minimum FL for he SBIG CCD. Nail down options for the EOS bodies: according to notes from last year, only a straight-through arrangement works, without the FT, without the Maxbright, just coupled straight to the A-P focusser.
After I rolled the mount out today, I added a plastic milk crate to the trailer (stengthened with plywood on the bottom) to carry the chock-blocks and the counterweights. I really want the mount to come out in one trip. I don't want to be tempted to leave the counterweights attached while rolling it in or out of the basement because I think that small bumps, smoothed as they are by the pneumatic tires, may stress the gears (on AP-GTO, Roland advises that fully loosening the clutches will isolate the gears from stresses of this sort). I'll need to think about how to group and carry the electronics for line and 12VDC operation. I'm thinking the LVI and A-P controllers should travel together in a soft pack. 12VDC and AC power accessories should be packed in clear plastic and carried out as needed.
After mounting the OTA, I synched on the Sun, slewed to Venus, then centered and resynched on Venus. I used Venus to find a focal point for the barlow. If the weather holds, I'll try M3 with the TeleVue Big Barlow and see how that goes.
Under the stars, the computer aided focus was a mess; experiment with that some more. A LiveView session on Arcturus seemed to nail focus, such as it is with the stars boiling tonight. WATCH OUT: if you use Nebulosity to focus the Canon, the camera is left with the last settings you sent it. You might mean to be on bulb, but the shutter may be set to 5 seconds. You might mean to be shooting raw but Nebulosity may leave the camera in JPEG mode. Likewise, ISO may be reset. Etc. Double check all settings before starting a series of exposures. The first six or seven 10 minute exposures of M3 were JPEGs and were trailed in dec. I realigned on the pole, refocussed, and reacquired the 6.3 guide star before making a series of 5 minute RAW exposures.
Alas: there is no joy in Mudville (see the cropped image at right, the red channel of a 300s frame at 1,500mm focal length). All the RAW frames are trailed in dec by similar amounts. There is no systematic and increasing excursion from first frame to last, so it's not as if flexure or slippage is allowing the sensor to creep out of alignment over the series of exposures. Close examination shows that the excursions are not constant: the stars occasionally move N or S but there is usually a sharp "core" where they stay put for 30-40-50% of the exposure. Tomorrow I'll verify that backlash settings are zero, try guiding at 1x (per Roland and Howard's suggestions in the AP-GTO group), and get systematic with the LVI aggressiveness trials. Look at it like this: RA is perfect, so you're halfway home.
May 5, 2010. The backlash settings in the keypad are all zeroes. I lined up on a small sunspot group at 762/3.2 x 2 or about 475x and checked for dec backlash at 0.25x, 0.5x, and 1x sidereal. There was, at most, a couple of seconds hesitation when reversing direction at 1x and no sudden movements. I checked again with a lower power crosshair eyepiece and saw the same thing. According to various messages from Roland and Howard in the AP-GTO forum, this is as it should be. So the Dec gear mesh is fine and there are no other mechanical issues indicated. Software in the keypad and GTOCP3 box seems appropriately configured. So the declination response problem is almost certainly in the aggressiveness setting(s) for the SmartGuider. Let's start with the A-P's guide rates to 1x sidereal and work out guider settings tonight.
I left the telescope with crosshairs centered on the Sun. An hour later, I tweaked azimuth based on slow dec drift. Two hours later, dec drift was zip but now RA was off! What the hell?!? Then I remembered: it's the Sun, duffous. Owing to the Earth's orbital motion, it drifts a degree a day relative to the stars (you know, the "sid" in "sidereal rate"). Turned it all off, covered it up. Looking good for tonight.
Later that same night, on the same field as above, these parameters look pretty good:
All attempts to calibrate at 1x failed when the star left the chip. Maybe if it was perfectly centered to start with, but who wants to have to futz with placing the guide star just so when a clear night is wasting? So I calibrated at half speed. Then I systematically adjusted aggressiveness up from low levels and changed the guider speed from 0.25x to 0.5x to 1x as I went. Changing guider speed after calibration is specifically endorsed as an adjustment strategy in the LVI manual, so I hereby revise my previous description of this as a kludge. I didn't put a camera behind the telescope tonight in order to keep things simple and "focussed." The point of tonight's efforts is to nail down the guiding parameters, not to make a nice photo. Here's the best plot of the night (keep in mind that polar alignment is deliberately imperfect to give the autoguider something to do):
The first recorded settings above are good; the set marked "better" is, well, better. With the lower aggressiveness, excursions are small and do not increase. They're eventually reduced. With the second set, excursions damp out much more quickly.
:: top ::