Thursday, July 13, 2017

A Guide for Guiding

Yes, here it is, at last, a guide on guiding!

Background

Guiding is a technique used to firm up the way that a telescope mount tracks across the sky.  I'll speak specifically to German equatorial mounts, shorthanded as CGEMs. (although I believe it is possible to also guide an alt-az mount - my NexStar SE has guide ports).  CGEMs move on two axes - right ascension (RA), and declination (dec).  These are the coordinates with which the celestial sphere is mapped from our earthly perspective.  When properly polar aligned (meaning, the telescope is close to looking at Polaris at its home position), a mount mainly moves in RA as time passes through the night to track the object.

The mount moves by moving gears.  However, gears cannot be made perfectly, or else they would never actually be able to turn.  So there are small imperfections in the motion of the mount - too small for your eye to notice, but you will see it in the camera as your stars streaking.  This is called periodic tracking error, since it takes several minutes for the gear to make a complete turn, and you will see the same motions at each point along the gear.  Some mounts have Periodic Error Correction, or PEC, which counteracts this periodic tracking error.  This is one option for getting those long-exposure streak-free images, but requires you to manually guide the telescope for several minutes (8 minutes on my Celestron CGE) by watching a star and keeping it precisely in the crosshairs of a lit-reticle eyepiece using the hand controller while the mount records your corrections, and then plays them back.  To further complicate the matter, most mounts don't have absolutely-encoded gear positions, but only relatively-encoded - which basically boils down to the fact that you would have to do PEC every time you brought your telescope outside.  (My Celestron CGE Pro doesn't have absolute encoding, but it does have a well-defined switch position, which allows you to record PEC permanently).  There are fans of each method, but when I first heard about each of these options, guiding sounded better to me.

The gist of guiding is that you have a second camera locked onto a star that watches for when that star moves, and then a command is sent to the mount to the mount to move a tiny bit back the other way to compensate for it and prevent the motion from showing up in your picture.  There are several options for the placement of that camera - many people (myself included) mount a smaller telescope on top of their imaging scope, or beside it on a dual plate, and attach the guide camera to that.  Others do what is called off-axis guiding, which is where you attach a device to the back of your telescope before the camera & filters that picks off a small fraction of the light and sends it to the guide camera.  There are also expensive CCD cameras that have two chips - one for guiding, and one for imaging.
An 80-mm f/5 refractor (the Orion ST-80) sits atop my Celestron 11-inch SCT on my Celestron CGE mount.

The Guide Scope

In an ideal world, you would want your guidescope to be no less than half of the focal length of your main telescope.  If you have a 650mm Newtonian, then you would want at least a 325mm refractor.  That's easy enough - my Orion ST-80 only weighs a couple pounds and has a 400mm focal length.  Unfortunately for me, my C11 has a 2800mm focal length (yes that's right folks, there are 9 feet of optical path length folded up into that tube), which would need a 1400mm guidescope!  Luckily for me, I have an f/6.3 focal reducer that reduces the effective focal length to 1764 mm (2800 * 0.63 = 1764), and I can't take images much longer than 5 minutes around here anyway because of light pollution, so yes, I make a 400mm focal length telescope work.  You can really get away with less than that half marker.  But don't go too crazy!

The Guide Camera

Really, honestly, a cheap monochrome CCD camera will do.  I have an old QHY5 (yes, the original monochrome, the little red hockey puck) that works great.  Other examples include the ZWO ASI120MM (and probably the lower-numbered models would work too), really any of the QHY5's, Meade Deep Sky Images, Orion StarShoots (I would not recommend the Color Imager IV, however - I have one, and I have thus far gotten 0 stars to show up even with 10-second integration times), and a wide variety of other little CCDs, like $100-$250.  Unless you are doing off-axis guiding where you need high sensitivity, one of these will do you just fine.  Bonus, they usually also make great planetary imaging cameras if you add in some filters!

You will want to get monochrome for a guide camera - color will work, but you will get higher sensitivity from monochrome, and guiding software throws out the color data anyway.

Software

There are many great people out there who love to code great programs for fun.  And thank goodness for them, because there is some wonderful astronomy freeware out there!  The name of the game for guiding is PHD (Push Here, Dummy), and can be found here.  Good news, Unix-ers: it's available for Windows, Mac, and Linux.

There's a reason it's called Push Here, Dummy.  It's because you just hit a series of buttons along the bottom of the screen to get going, for the most part.  I'll do a walkthrough below, after a few more sections on drivers and connecting stuff together.

Drivers

In addition to PHD, you're going to need a few drivers.
First, the ASCOM platform.  ASCOM is the standard for controlling a wide variety of astronomy devices, from cameras and telescopes to focusers, filter wheels and even motorized dome roofs.  The platform is the basis for all of your other ASCOM drivers.

Second, both the camera driver and the ASCOM driver for your camera.  Get your camera driver from your manufacturer, and you should find the ASCOM driver either also on your manufacturer's website, or on the ASCOM camera driver page.

Third, if you are going to use a program like digiCamControl or DSLRShutter or something to control your DSLR now that you will be able to do images longer than the built-in 30 seconds, be sure to download your camera's driver too.  (BackyardNikon/BackyardEOS comes with all the drivers already).

Connecting Stuff Together

All right, so this part takes a little attention on your part so that you don't fry your mount.  Hopefully, your mount came with a cable that plugs into its RS-232 port - it will look like a phone jack port. (Don't use the "Auto Guide" port.)  On Celestron mounts, this port is located on the hand controller itself - the same one you would use to update the firmware.  Unfortunately I am not at all familiar with basically anything besides Celestron, so consult your manual to find out where your RS-232 port is.  But the cable will have what looks like a phone jack on one end, and a serial cable connector on the other.  If you don't have one, it is very important that you get the one that is designed for your mount.  If I plugged my Celestron mount cable into the Losmandy Gemini mount out at the observatory, I would fry the mount.  If in doubt, consult the manual - it will have the pinout (which pin is signal, which is ground, etc.)

Once you have the RS-232 cable in hand, pick up a serial-to-USB converter on Amazon on the cheap.  There's not really anything special you need to get here.

Plug the serial-to-USB converter and your guide camera into your computer.  Since I use a tablet with exactly one USB port, I got a USB hub.  I got one that can be powered (by plugging in a micro-USB cable to a standard cell phone charging plug, spare battery, or telescope power tank with USB ports, which delivers 5V/1A) since I discovered that my guide camera uses a lot of my tablet's battery power.

Use the USB port on your guide camera, not the ST4 port (the one that looks a bit like an ethernet or phone jack).

Getting Started With PHD

There are a lot of tutorials out there for how to use PHD.  There is probably stuff that I am not doing that would bring me greater success.  But I'll walk you through how I do guiding that gets me the kinds of results I'm currently getting - which, let's just say, are pretty sweet.

When you start it up, click on the first icon on the bottom of the screen - the picture of the plug.

Click Manage Profiles, and then New Using Wizard.
Select your guide camera from the list.  (Note that my Orion StarShoot Color Imager IV shows up under "Windows WDM-style wecam camera" instead of Orion StarShoot, and my QHY5 shows up under COMS QHY5 (ASCOM) so it may not be the one you think it should be - try a few).  Once you've selected your camera, you can click "Detect" and have it automatically detect the pixel size.  On that same screen, enter your guide scope's focal length (in mm).  Leave the guide speed at 0.50 (although make sure your mount is also set to this value).  Click Next.

Select your mount from the list.  This will show any ASCOM mount drivers you have installed, along with some other options.  Click Next.  Click Next again, unless you are super cool and have adaptive optics somehow on your stuff.  Name your profile ("CGE Pro with QHY5 & Orion ST-80," for example, or if you cleverly name each of your rigs), and then hit Finish.  Go ahead and leave that "Build Dark Library" box checked - you will want to do that.  Finally, hit Connect All, and your devices should connect.  

If you get connection errors with the mount, a) check all your cables, and then b) click on the little icon between the box with the mount name and the "connect" button and make sure you have the right COM port chosen.  You can try each of the options there, or you can find out which one it is by going to your Start Menu, searching for Device Manager, scrolling down to "Ports (COM & LPT)" and looking for which one has your serial-to-USB converter on it.  I always plug my mount into the same USB port on the hub to avoid having to deal with this.

Now you should see a "Build Dark Library" window.  Follow its instructions, and the defaults are all fine.  Basically what a dark library does is it eliminates the hot pixels that your CCD camera is bound to have so that PHD doesn't think they are stars.  This process will take a few minutes.  Make sure that you are not shining any light anywhere near your guide scope during the process, and don't do this during the day (unless you do it in a dark closet).  

Now you are ready to start acquiring from your guide camera.

Focus & Calibration

Slew to a bright star - you'll need to focus your guide scope.  Click the next button along the bottom panel - the one with the green circled arrows.  This will start image acquisition.  For focusing, especially with a bright star, I'd choose 0.5s from the drop-down menu along the bottom.  At first, the image may appear extremely noisy with lots of banding.  Don't fret - PHD heavily stretches images in order to find stars.  This will calm down when you are near focus.  In fact, that's how I tell when I'm near focus - when the noise starts to drop.  Go slowly, since the image is only updating twice per second.  Bear in mind that your guide scope may not have enough in-travel or out-travel for your guide camera - this is where extension tubes and mirror diagonals come in handy (for when you need to go out further), or a Barlow (for when you need to come in closer, but keep in mind with a Barlow, your light intensity will drop and your field of view will shrink, so you may need to increase your gain or integration time.  This also effectively doubles your guide scope's focal length, so set settings accordingly).  

Once you are focused (you'll see stars, and they don't have weird fuzzy circles around them, unless the star is very bright, like Vega or Sirius), then you are ready to calibrate.  Slew to your target.  Since you entered your focal length in the setup wizard, PHD will already have calculated what step size it needs to take in order to see the star move.  Change your integration time to 1-4 seconds - 3-4 seconds is best so that you don't guide on the seeing (atmospheric motions can cause the star to appear to move when it has not - longer integration times smooth this out).  Click on a star in the image, or press Alt+S to have PHD pick one for you (it can oftentimes see stars that you cannot), and then click on the PHD icon in that bottom panel, the green circle with the crosshairs.  This will start the calibration process, which will take several minutes.  It will move the telescope north, south, east, and west, and watch how the star moves and how much it moves when it sends a certain-length pulse to your mount.  Once it's complete, it will begin to guide your mount, and it will say "Guiding" at the bottom of the screen.

There are two error messages I tend to get - something about non-orthogonality, and something about too northerly of a declination.  The non-orthogonality one, I finally discovered, usually simply means that your camera is at a cant with respect to the telescope.  It's best to attach it "upright" - meaning, the USB and ST4 ports are in line with each other and parallel to the ground when the telescope is in the home position.
Left - ports parallel with the ground when the guide scope is upright.
Right - camera is canted, and can throw weird PHD errors.

As far as the northerly declination goes - the advice I have been given with PHD is to have new calibration data for each of your targets, and to take that calibration at or near your target.  Sometimes, my targets are far to the north, like M51 or M81.  These can cause trouble for PHD because the telescope won't move hardly at all in RA when it's looking north, since that is where the celestial pole is.  But I usually have success guiding here anyway, and just ignore the message.

Speaking of that advice to get a new calibration at each target, here is how you do it: first, stop guiding before you move to your next target by clicking the stop sign button in the bottom panel.  Slew to your new target (don't forget to focus your main telescope again if you have an SCT!).  Then go to Tools -> Modify Calibration -> Clear Calibration Data.  At your new target, hit the loop button (green arrows), then the guide button (PHD icon - green circle with crosshairs), and PHD will re-do the calibration process.

Guiding

Now that you are calibrated & focused, you are ready to guide!  If you had any issues with calibration or other things, you can click the brain button to access settings.  Consult Dr. Google.

Now that guiding is going, you'll see the graph on the bottom start to populate.  This shows the corrections that PHD has had to send to your telescope, and how severe they were.  If you have tall spikes larger than 3 or 4 arcsecs that are all over the place, then your alignment or polar alignment probably isn't very good.  Keep in mind though that it takes a minute or two to "quiet down," so wait for things to shake out first.  (Or, try imaging anyway - sometimes my images turn out just fine even when the guide plot is crazy).
An example guide plot.  My stars came out perfectly round like this on an 800mm refractor using a 102mm refractor to guide.

If a cloud rolls through, or you forget to turn the observatory dome, or your guide scope gets dewy, or any number of things that cause PHD to lose the guide star, it will start making computer notification noises at you.  So keep your volume turned up!  It can save precious time during the night to know right away when something is amiss.  (There's a plugin for Sequence Generator Pro that will send you emails when PHD is having a problem).  

All right, so those are the basics, and that is my process.  Again, check out the variety of other sources on the internet that have tutorials and advice for guiding and PHD, especially when something isn't working.  There are a lot more settings to tweak than I've talked about here (or have even tweaked myself), but with this process, I've gotten at least 7 minutes of guiding on my CGE Pro with my C11 as the imaging scope and my Orion ST-80 as the guide scope, and at least 20 minutes of guiding out of the memorial scope, which is an 800mm refractor with a 102mm guide scope on a permanent pier.  

No comments:

Post a Comment