Sunday, September 05, 2010


Uncle Rod’s Telescope Academy: The Autoguiding Auto de fé

What say I lay off the nostalgia for awhile? Oh, rest assured, you’ll eventually get more of the exciting adventures of Li’l Rod, Wayne Lee, and Miss Jitter, but not right now. Go to the old memory well too many times, and you start dredging up more bad than good. Most of my memories of my early days as an amateur are good ones, but, like just about everybody else’s, my early adolescent days had plenty of downs as well as ups. Time to leave-off and let some more good memories rise to the surface.

So, then, what’s the topic for this morning? One I’ve covered before. But also one I keep getting questions about. Lots of questions. And it is a confusing subject. What I’m a-talking about, brothers and sisters, is autoguiding. What hardware do you need to buy? How about software? How do you hook everything up? How in the H-E-double-L do you get it all work?

You are a sopping wet behind the ears newbie and ain’t got a clue what I am going on about? Well, listen up, and your old Uncle will edumacate you. Let’s say you want to take a picture of the night sky, of the deep sky. You want a detailed close-up of a nebula, a galaxy, or a star cluster. In other words, you want to take pictures through your telescope, not just through the lens of a camera riding “piggyback” on your telescope. What does that require?

Back when I was a youngun, I used to see advertisements for stuff like guide scopes and drive correctors in Sky and Scope, but for the life of me I couldn’t figure out what you did with the stuff. If you wanted to take a picture of, say, M42, you just mounted a camera in prime focus position on your telescope, plugged in the mount’s drive (no battery clock drives in them days) and let her rip. Open the shutter and come back in a half-hour or whatever. Sure, I knew you did need a good polar alignment, but that was it, right?

That’s what I thought till I got my first C8, a Celestron Orange Tube, and decided to get serious about deep sky picture taking. With careful polar alignment and balancing, the C8's drive could maybe have delivered OK 30-second images. I let it go for 30-minutes while I watched the latest episode of Mork and Mindy. When I developed my picture, I got, yep, STAR TRAILS. Bad ones. What the—?!

A little asking around at the club and a little reading of the few books on astrophotography I could find ‘splained what was going on. While I knew poor polar alignment could make the stars trail due to “declination drift,” I hadn’t heard about something called “periodic error,” which, it appeared, was the cause of my star trails.

The thing is, campers, most drive gears are not perfect, not even close. Take a gear, size it to where it’s appropriate for a normal, portable amateur mounting, drop the price to something acceptable to most of us, and you darn sure are gonna have imperfections. The gears will do a wonderful job for visual observing, but will stumble when it comes to imaging. Nothing is more demanding of a mount and its drive train than deep sky imaging. The smallest bump, the smallest irregularity in the gears, will cause the mount to speed up or slow down a small amount.

That is periodic error; it’s “periodic” because it repeats every time the bad spot on the gear rotates back around. The effect in the camera is a slow east-west back and forth of the target. In modern drives this deviation from proper tracking is usually less than 30 arc seconds, less than the diameter of Jupiter.

But that is enough. Even that small amount of periodic error will cause stars to trail in a long exposure. They will at best be oblong rather than perfectly round, and there’s not much tolerance for that on the part of most astrophotographers. Even just slightly off-round stars look horrible to most of us. What to do? You have three choices: live with it, throw money at it, or guide.

The simplest solution, and one that’s maybe recommended for beginning imagers is “live with it.” Almost any current mount, even an humble CG5, can track for a fair length of time without error. If the mount can expose for 30-seconds to 1-minute at reasonable focal lengths (say, less than 1000mm) and produce good stars, you can get started.

I found my CG5 was able to do good 30-second exposures at 800mm of telescope focal length most of the time. Yes, I’d get occasional frames where the stars would be oval or worse, but I just threw those pictures out. The rest I “stacked” in the computer to produce the equivalent of one long exposure. I could capture just about any deep sky object I wanted in this fashion with my C8, CG5, and Meade DSI camera.

I was happy with this “unguided imaging” solution while I was learning (finally) the difficult art of CCDing. Before long, though, as amateur astronomers always do, I wanted More Better Gooder. What’s the problem with stacking unguided short exposures? For one thing, I got tired of throwing out 1/3rd to 1/4th of my images. I had to spend more time with each target than I should have. There was also the noise problem.

Every CCD camera produces a certain amount of noise due to various electronic gremlins. Some types of noise are more prevalent in shorter exposures than in longer ones. While stacking multiple frames together tends to reduce this electronic noise, the result is never quite as noise-free as longer exposures . This was especially problematical with my DSI. As much as I liked and still like the little one-shot color camera, it was by nature noisier than its more expensive and (especially) monochrome cousins.

What then? I could have gone the “throw money at it” route. If you are willing to shed 10k dollars for your mount, you can begin to do longer unguided exposures. Maybe several minutes. My problem with this approach? In addition to my naturally stingy—err… "thrifty"—nature, living where I do, it’s impossible for me to have an observatory and permanently mounted telescope. If you are going to spend a lot on a telescope mounting for unguided imaging, you really do need to have the polar alignment dialed in as well as possible and the telescope precisely balanced in order to maximize the mount’s unguided potential. Having to set up at the club dark site every time I wanted to take pictures does not lend itself to either of these things.

That left autoguiding. I was no stranger to guiding per se. In my second go-round with deep sky astrophotography back in the late 80s and early 90s, I guided plenty of pictures. That was well before the advent of the autoguider—for amateurs anyway. Which meant that for the duration of an exposure, which with film was usually ½ to one hour, I had to squint through a crosshair eyepiece and keep a dim star centered. That blamed star would wander, and I’d push a button on my hand paddle to put it back in the crosshairs. Not fun.

Things are different now. Theoretically, at least, you should be able to guide without tearing your hair out and assuming the old astrophotographer’s 1000-yard stare. Guiding has been automated. Computerized, actually, like everything else. A CCD camera watches the guide star instead of you. It pipes its images to a computer program that can tell when the star moves and issues guiding commands to the telescope drive to move it back where it oughta be.

That is way-oversimplifying autoguiding, of course. In the real world there are just so many things to go wrong: camera, guide scope, cables, computer programs, etc., etc., etc. It is possible to conquer the various demons that will most assuredly bedevil you, but before you can begin troubleshooting/tuning your system, you have to have a system, and you have to make some decisions before you get one.

Guide Scope or Off Axis Guider?

The first of which is “Guide scope or off axis guider?” There are a few cameras that can guide and image at the same time through the same telescope. But in most cases you are going to have to either provide a separate telescope for a separate “guide camera,” or a means of picking off some of the main telescope’s light and sending it to that guide camera.

The simplest solution is a separate guide scope mounted to, riding piggyback on, the big scope. The usual choice is a small refractor, a 60mm to 80mm rig, maybe. Back in the days of manual, visual guiding, it was necessary to use as high a magnification as possible to ensure accuracy. You had to be able to notice the tiniest “excursions” of the guide star. That is no longer necessary. A CCD camera can do perfectly well with a short focal length guide scope like a Short Tube 80 refractor or one of the ubiquitous 66mm ED lens-scopes.

Naturally, you’ll have to mount the guide scope to the main scope, and how you do that is critical. You want “as sturdy as possible.” If there is any potential for flexure in its piggyback mounting, the guide scope may move a little over the course of an exposure. If the guide scope moves independently of the imaging telescope, that will cause stars to trail in the image no matter how well the guiding actually worked. “As sturdy as possible” is the answer.

Sometimes the fault isn’t with the guide scope, but with the main scope. If you are an SCT fan, you have probably heard of “mirror flop.” Since the telescope’s mirror moves back and forth to focus and is not securely mounted, there is the possibility it may move a little during an exposure. That will, like guide scope mounting flexure, cause the stars to trail, since, in essence, the main scope has moved independently of the guide scope—its image has, anyway.

Solution? Some modern SCTs, like the Meade ACFs and the Celestron Edge HDs, have mirror locks to fasten down the main mirror, preventing flop. Your CAT ain’t got ‘em? Some mechanically inclined imagers have fabricated their own locks, usually with bolts threaded through the rear cell. That’s not an option for the all-thumbs brigade headed by your old Uncle, though.

Actually, I’ve rarely had problems with mirror flop over the years, since I am careful to do one thing and avoid another. I always finish focus “uphill,” counterclockwise, which leaves the mirror in a stable position. I am also careful not to image anything crossing the meridian, since the change in the telescope’s attitude when tracking across the Local Meridian is when flop often happens. If you want to be sure of eliminating mirror flop, no ifs ands or buts, there is only one real fix, however.

The off-axis guider, the “OAG,” doesn’t just totally eliminate mirror flop in SCTs, it eliminates the need for a separate guide scope. How? It does so by stealing a little light from the edge of the CAT’s field. An off axis guider is a very specialized prime focus camera adapter. In addition to allowing you to attach your camera to the scope, it includes a star diagonal-like setup. There’s a focuser tube into which either an eyepiece or a guide camera can be inserted, and there is a diagonal mirror, just like a star diagonal. The difference is that it’s a very small mirror, and extends only into the very edge of the light cone.

This mirror, the “pickoff” mirror, grabs a little of the main scope’s light and allows the stars around the very edge of the field to be delivered to an eyepiece or camera. Since it’s very small, the pickoff mirror only shows a few stars. In order to locate a good guide star, the OAG is constructed so the mirror can be rotated around the edge of the field. Usually, but not always, a suitable star can be found this way. Some pickoff mirrors can be extended a little farther into the field in the quest for stars, as well. Since it’s small and at the edge, the mirror’s shadow isn’t likely to show up in the field of the imaging camera.

How does the OAG eliminate mirror flop? Actually, it doesn’t. But it does allow the guide camera to see star movement caused by mirror movement. If the guider sees the star move, whether due to periodic error or flop, it will issue the appropriate guide command to follow it.

I used an OAG for years, and got some good pictures, but it wasn’t pleasant. The OAG’s irritants are twofold. First, it is a pain to locate a good guide star when you are restricted to the small choice offered by the small mirror. Often, I’d have to compromise framing of my target in the imaging camera by moving the scope to bring a good star in. Today’s guide cameras are far more sensitive than my eyes, though, so that is not as much of a problem as it used to be. Nevertheless, the guide scope is still the winner here, since it can be moved in its rings to search for more stars, and the guide camera is, of course, taking in the full field of the guide scope, not just the edge.

One thing that is still a problem with OAGs is the shape of the field edge stars. Many scopes, particularly SCTs and faster focal ratio refractors and Newtonians, deliver misshapen stars at the field edge due to coma and/or field curvature. A star that looks like a blob or seagull may be impossible to guide on. Field-flattener lenses or coma correctors can help, and with some telescopes may be a necessity.

One last potential down-check? The OAG comes between camera and telescope. While an OAG can potentially be used on Newtonians and refractors, those scopes may not have enough focus range to accommodate one; especially if a focal reducer is being used. The camera will be placed too far out to come to focus.

All in all? I prefer the guide scope. It’s less painful to get going. On the other hand, once you have located a good guide star, the off-axis guider is capable of delivering better results. The Celestron f/6.3 reducer corrector I use with my DSLR for imaging delivers pretty good edge-of-field stars, so I’m thinking I might dig out my old OAG (if I can find it), the next time I have a hankering for some deep sky snapshots.

There is a third option, too: “OAG without the OAG.” Santa Barbara Instrument Group, SBIG, the famous CCD maker, offers cameras with built-in guide cameras. There is a guide chip arranged right at the edge of the imaging chip, and this guide chip, like the OAG, picks up stars at the edge of the field. No OAG is needed, and the very sensitive nature of the guider in these cameras means finding a star is rarely a problem.

The main limitation with Santa Barbara’s “self-guiding” cameras is edge of field star shape. I’ve had good luck imaging with my ST-2000 CCD camera with the f/6.3 reducer; not so good, variable anyway, when using Meade’s f/3.3 reducer with the C8. If you are in the market for a new imaging camera, the SBIGs with built-in guiders most assuredly deserve your serious consideration. No guiding setup is closer to plug and play.

Which Guide Camera?

Beginners tend to shy away from SBIG’s self-guided cameras for a number of reasons: cost, perceived complexity, they just want to use the family DSLR for astrophotography, etc. If, for whatever reason, you don’t want an SBIG ST, you will need a separate guide camera in addition to your imaging cam (self-guiding is proprietary to SBIG).

One thing I urge is that you not scrimp. Some new imagers waste precious time and reduce their hairlines by playing around with webcams. One can work with the appropriate software, like the freeware Metaguide, but only if you are lucky enough to have a nice, bright guide star in the field. Which is rarely the case. Unless modified, a webcam cannot expose for longer than 1/30 second. Even semi-webcams like the Meade LPI or Celestron NexImage that can expose for longer are not sensitive enough for easy guide star acquisition and good guiding.

If you need to save bucks, almost any CCD camera can work. Many last-generation cameras can be had for a song on Astromart, and if you can find guiding software that supports your camera of choice, you are in like Flynn. One caveat: eschew really old cameras that use a serial or parallel interface instead of USB. You will have a very hard time getting them going for a number of reasons, including that modern PCs have neither parallel nor serial ports. A good choice of used camera in my opinion is one of the (no longer produced) Meade DSIs. I, II, III, it really don’t matter. All are sensitive enough to make good guide cams.

Don’t have an old DSI lying around and don’t want to buy used? Today, there are numerous cameras sold specifically for use as guiders. Many of ‘em come from the same factories in China, and all are basically similar. A lot of us are using the Starshoot guide cams from Orion. Their attractions are that they are quite inexpensive—the base model is less than 300 bucks—and are equipped with ST-4 guide outputs (more on that later). The Orion guide cameras do have the deficiency of using CMOS instead of CCD chips, and are less sensitive than the Meade DSIs and similar cameras. Nevertheless, I have always been able to locate a suitable guide star in any field with my Starshoot.

There’s another type of guide camera you may be interested in; especially if you don’t like toting a computer in the field, the self-contained guiders. They don’t require a PC to operate; they have a small hand unit that takes care of all the computer work. The old SBIG ST-4 and STV cameras, long since discontinued, used this all-in-one concept, but for the longest time, there didn’t seem much call for it. If you were using a CCD camera, you’d need a computer in the field anyway, so why not just let it handle the guiding, too?

Self contained “solitaire” guiders are back in a big way. Things changed with the advent of the DSLR for astrophotography. Some of us run our DSLRs with laptop computers, just like we do CCDs. But some DSLR imagers eschew PCs and use a simple remote shutter release, relying on (increasingly good) DSLR video displays for framing and focusing. These folks don’t need a PC for imaging, so it’s wonderful for them to be able to dispense with one for guiding, too. If this sounds good, look at the Orion Starshoot Solitaire, the LVI Smartguider II, the SBIG SG-4, and the Celestron NexGuide.

If you are not using a solitaire guider, you will operate the guide camera with software running on a computer. There are numerous CCD software packages that will autoguide. Most of the top-of-the-line imaging programs, like MaximDL and CCDsoft guide as well as image. Be sure these programs will interface with your guide cam before you pull out your wallet, though. And be sure you need this expensive software. If you are using a DSLR, you probably don’t.

Whether you run a DSLR or a CCD, one autoguiding program is head and shoulders above everything else, and, believe it or not, it is free: Stark Labs’ PHD Guiding. “PHD” in this context don’t mean “Doctor of Philosophy,” it means “Push Here Dummy.” And that’s the truth. Hook everything up, push a couple of buttons, and the puppy just LOCKS ON. Hell, my first night of imaging with PHD, I had set up EQMOD wrong. My Atlas mount was tracking at the wrong speed as a result. PHD said “so what?” and kept the guide star in the crosshairs anyway.

PHD is not for imaging, only guiding, so you’ll need a separate program for your imaging camera. While PHD works great with Stark Labs inexpensive (and great) imaging program, Nebulosity, it is just as happy alongside any other imaging program.

Another freebie that should be investigated is the aforementioned Metaguide. Not only will it do a superb job of autoguiding, it will help you precisely collimate your telescope with the aid of a webcam. The only problem is that it’s limited in the cameras it can use, being restricted to webcams and video cameras at this time.

Hooking It All Up

Hokay, you done got a guide camera, a telescope mount, and a PC (maybe). How do you hook it all together? This is where a whole lotta folks get awful confusticated. They plunk down mount, camera, and computer and start pluggin’ in cables where they think they should go and are dismayed when nothing at all (good) happens.

The usual first mistake is that they run a serial cable from the laptop’s serial port (more like a USB – serial converter these days) to the autoguide port on the mount. Sounds logical. You’re guiding with the computer, you run a computer cable to the mount and plug it in—where else?—to the jack labeled “autoguider.” That’s fine, except for the fact that it will never work.

The problem is that the autoguider port don’t speak computerese. More on what it does speak in a moment, but for now, let’s address “serial guiding.” In this “Way One” type of guiding, commands to move the scope are sent over the RS-232 serial interface. How do you make that work?

First thing you do is plug the camera’s USB cable into a USB jack on the computer. If this is the first time, follow the camera-manual’s instructions religiously when it comes to installing software drivers and suchlike. This connection will convey the images your guide camera is taking, and it will also carry data and commands to and from the computer.

Now the sticky part: that serial cable. If you intend to guide via a serial cable, you do NOT connect it to the autoguide port. You connect to the serial port on the telescope mount, usually on the base of the hand controller. You also do not use the guide cable that came with the guide camera, if one came with it. You connect using the serial cable designed for your mount. The same one you use to send the scope on go-tos via a planetarium program running on the laptop.

In serial guiding, when the computer needs to move the mount to follow the guide star, it sends a computer command not much different from what gets sent from a planetarium program to move the telescope. Don’t worry about congestion on the RS-232 freeway if you intend to use your planetarium software while imaging. Cartes du Ciel (for example) and PHD Guiding (for example) will coexist peacefully. When you are guiding, not too much is happening with the planetarium software, anyhow.

Naturally, before you can begin guiding through the serial port, you have to tell the software that’s what you intend to do. In PHD, that involves using the (free) ASCOM telescope interface software. Once ASCOM is installed, it’s not much of a hassle. Just select your telescope from the “chooser” that pops up when you begin the connection process.

Some imagers don’t like to guide serially. There are a variety of reasons for that. One is that some think the autoguide port on the mount is better, that there is less of a time-lag between issuing guide commands and telescope response. That may be, but I don’t think there’s enough difference to make much difference. I’ve used both serial and autoguide port (“ST-4”) guiding and have never noticed much difference. It is true, however, that if the mount and camera both both have autoguide ports, you can save a little cable tangle by eliminating one wire going to the PC.

What do you need for ST-4 “Way Two” guiding? In addition to an autoguide port on the mount, you need the appropriate cable. That is not a serial cable, but an ST-4 compatible cable. What’s an ST-4? That was SBIG’s famous early imager/guider. One of its breakthroughs was that it inaugurated a simple relay/switch closure guiding system. Yep. What the autoguide port on the mount understands is not fancy computer commands, but simple switch closures. In this day and age, the switches/relays are virtual, electronic switches, but what’s happening is still the simple opening and closing of connections. Just like pushing east/west/north/south buttons on the HC.

So, to get started you will need an ST-4 type cable, which is wired differently from a serial one. If the guide camera features an autoguide output, you were likely provided with an ST-4 cable. How do you hook up for ST-4 guiding? If the camera has an ST-4 output, it’s simple. Plug the ST-4 cable into that and into the autoguide port on the mount. You will still need to connect a USB cable between the camera and computer for image downloading and camera control, of course. What do you tell the guiding software? In PHD, select “on camera” for the guiding interface. You do not need and will not use ASCOM for autoguide port/ST-4 guiding.

There’s also a Way Three. Let’s say you want to use your mount’s autoguide port, but are using a camera, like a DSI, without an ST-4 guide output. Does that make ST-4 guiding impossible? No, it doesn’t, thanks to a little company called “Shoestring Astronomy.” They provide something called the “GPUSB.” This is a small module that plugs into USB ports. It allows the guide software to send ST-4 switch closure commands out a USB port and down an ST-4 cable that plugs into the mount’s autoguide port as per normal. Assuming the software will accommodate a GPUSB, you have to tell it about it. In PHD, just choose “GPUSB” on the “mount” menu.

Is there a Way Four? Yes there is if you have a solitaire style autoguider. The exact connections will be peculiar to the particular guide camera, so read the manual, but there’ll be a cable (or cables) between camera and the solitaire’s handbox. Most of the time there’ll also be a normal ST-4 output either on the camera or on the handbox to connect to the autoguide port on the mount. Read the instructions, though, since these systems tend to be more involved than simple PC-operated guidecams.

So that’s all there is to guiding, huh? Not hardly. We’ve Only Just Begun. We’ve selected a camera and got it all hooked up. But that’s when them dadgummed gremlins rear their heads, when you actually start trying to use the junk. Exterminating those buggers and getting everything working in optimum fashion is somewhat long story, though, and it will have to be in a “Part II,” which I’ll give you-all Real Soon Now.

Next Time: By the time you read this I will probably be back from the 2010 Almost Heaven Star Party in the hills of West Virginia. It’s one of the country’s great star parties, muchachos, and you will want to hear all about it and you can bet your bippy I will tell you about it next Sunday.

Rod, have you seen the Skywatcher (Synta?) Synguider?

It's a standalone all-in-one guider camera computer which connects straight from the guidescope eyepiece tube to your mount's ST-4 autoguider port.
Have not seen the Synta one...url?
Hi Rod,
Great post! Just the info I was looking for. So the Orion Starshoot basically acts as a camera/GPUSB combo, if I understand it correctly?
Yep. And please see last Sunday's blog on this same subjet. R.
My apologies if you already answered this, but I have the (somewhat new) canon 60Da (unique for astrophotography) can I use this camera to take pictures as well as autoguide with appropriate software (if it exists)?

this camera does have a "live-view" if it makes a difference
You can probably autoguide with it using the live video output (live view). But you can't image and guide at the same time. One or t'other. ;-)
I have a Ultima 2000 GOTO Computerized Fork Mount and the Orion star shoot all in one.the Ultima 2000 GOTO Computerized Fork Mount
Has a aux on the base Is that my ST4 that plugs into my all in one camera?
Thanks Steve
NO...the Aux port is identical to the hand control port. Plugging anything else into it an damage the scope. An ST-4/autoguiding port will always be labeled "autoguide" or similar.
Thanks for you answer. Have one more question,will the celestron sky portal work on the aux port work for me?
Thanks for you information, will the celestron sky portal work on the aux on the same mount?
Thanks Steve
Yes, the SkyPortal link can be plugged into the aux port. But remember, you can use either it and a smart device OR the hand control, NOT BOTH.
thank you for that. I purchased a $500.00 dell computer and a meade lpi-g camera that had ascom software. it could not load the supplied ascom platform or the ascom camera driver! spent another 127.00 on a dell tech and meade tech, over three hours trying to figure out what was wrong to no avail! will a zwo camera work with phd without ascom? I have a losmandy g-11 gem mount.

The problem with the LPI is that there are no current drivers for it. PHD2 should work without ASCOM.
I bought a Celestron SkyPortal and when I plug it into my aux port on the ultima 2000 I get no led light on the SkyPortal . Nothing, is there some setup on my hand controller I need to do?
When I do plug the SkyPortal in the motors on the fork mount start to hum or run very slowly.
Any information you have I would be glad to try.
Unfortunately, Steve, I do not believe the SkyPortal is compatible with the Ultima 2000. Sorry. :(
Many thanks for taking the time to write this really helpful blog. Ten years later and it's still helpful..even here in a cloud covered UK!
I guess the two things I was most interested in were telescopes and steam engines. My father was an engineer on a threshing rig steam engine and I loved the machinery.
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?

stats counter Website Hit Counters