Last updated: 6 July 2005

Subject:	Optimising the ETX for CCD Photography
Sent:	Sunday, July 3, 2005 12:36:38
From:	Stephen Bird (stephen.bird@btconnect.com)
Optimising the ETX for CCD Photography
The ETX is great little scope for visual astronomy, but it is also good
for astrophotograpy as well, but it takes a little work to get good
results. In my quest to get good pictures from the ETX, I decided to try
the Meade LPI and DSI (original and Pro versions), I guess the price was
right and they ought to compliment the scope being made by Meade. Having
got some average snaps I wanted to move into the next league, so my mind
turned to the issues of "autoguiding", and I had an exchange of e-mail
with Dick Seymour, resulting in the following first thoughts. some of
this needs refinement, and may be a little wide of the mark, but hidden
in here are a few pointers for would be CCD astronomers.

Well I guess where I am coming from is I just wanted to know the
limitations of the "autoguiding" using an ETX. Because effectively the
ETX via Autostar Suite and the Autostar can autoguide using a single LPI
or DSI. I think it achieves this be interleaving your image capture
frames, with an alignment frame which you don't see. This extra exposure
between frames, uses the centroid box on screen to check for a change of
position of the chosen object, and send some guidance commands via the
Autostar to the ETX, to correct any drift in the image. I have at the
moment an LPI and a DSI Pro, I returned the DSI standard as it was a
"lemon" in my opinion but judging from the DSI Pro that I now have, I
was probably unlucky, so don't dispair if you have the DSI original
(standard model).

So here is my weighty tome on how I see it all. If any of it is wrong,
it is because it is not in the manual and I had to deduce it myself, so
I deny any responsibility, blah, blah, blah!

As an aside, the manuals being on the CD or on Meade's website is not an
issue. It is certainly a simple matter, especially in Windows XP, to
copy them to the LPI / DSI Documents folder, then create a shortcut for
each manual, and add the shortcuts to the start group (or create a start
group and add the manuals to it) for the manuals. Then they are easily
accessible via Start  Programs - Meade - Manuals. I have added the LPI,
DSI Quickstart, ETX AT, Autostar 497, and ETX Premier manuals to that
start group, so the information in each is instantly accessible. CD's
are not convenient in the dark!

What I have found for the Autostar Suite Autoguiding using the ETX is
that first of all you need to reduce all sources of drive error to an
absolute minimum. Here are the 10 worst error sources I have
indentified, I'm sure there are more:

1. Reset / training / calibration of the scope - you need to redo this
minimum every 3 months (as things just change for the hell of it

2. Change of power source - (new batteries, but don't use batteries, as
the voltage drops, old batteries cause errors to build in my opinion),
best thing is a big AC mains power pack with a good regulator to produce
a stable constant voltage supply or second best, one of those hefty 12
volt power packs, I use a 17 Amp Hour supply fully charged each evening,
and fed via a regulator (one of those things from the car accessory
store that has options on the voltage, it gives a good regulated 12 volt
output). Any unusual voltage spikes will be a source of drive errors.

3. Alignment - Accurate polar alignment is essential. If you are using a
permanent pier and wedge set acccurately, then you only have the RA
drive causing drive errors. If you use Alt / Az alignment, then you
multiply your errors many fold, as every time you "Go To", the DEC
errors accumulate. (You can see this by going back to Polaris over the
course of a nights viewing and watching the orbital distance from dead
centre (celestial pole) grow, it is larger for Alt / Az than Polar

4. Field Tripod - Using the field tripod also adds to errors, but they
are small if it is set level, fastened firm and not kicked or tripped
over! Also don't add a heavy piggy back camera, or significantly change
the set up of the scope over the night, or what happens is as the scope
moves and the centre of gravity shifts, the tripod "settles" and the
alignment goes out. Using the photo port at the back or putting the DSI
up on a Barlow lens can make a difference if the tripod is not rock

Mother-in-laws blinkies - Buy 3 Duracell 9 volt batteries, 3 flashing
red LED's from Radioshack / Maplins, 3 reusable tywraps and 3 lengths of
velcro. Use the tywraps to hold the wires of the LED's to the battery
connectors and the velcro to hold the battery to the legs of the tripod.
that way when the mother-in-law comes to visit, the old duchess won't
trip over the tripod legs in the dark!

5. Cords and cables - My favourite bugbear! Hanging off the scope,
dragging around every time you need RA movement (all night!) and DEC
movement are cables. They drag, they snag, they tangle, and each time a
clutch slips a bit. I have one USB cable for the DSI or LPI, one
focusser cable for the electric focusser and one power cord for dew
heaters. I put all my "eggs in one basket". Strip the insulation from
the USB cable where it will drag against the scope in any way, (when the
plastic cools, it gets less flexible, increasing drag and snag), thread
the USB cable through the coils of the focusser cable. For the length of
power cord for the dew heaters that will drag against the scope, use
single insulated wire of the minimum power rating (one red, one black
wire rated 240v 0.5A is fine for 35W (at max setting) of heaters running
at 12v). thread them inside the coils of the focusser cable. Now you
only have one cable route to contend with.

To keep my cables in check, I use a cable tray between the ETX base and
the tripod wedge. It is a black large size plant pot tray. The scope
power cord and focusser cable from the ports on the ETX switch / control
panel exit through a hole in the cable tray immediately below the switch
/ control panel. This presents as few snagging opportunities as possible
in this area. The focusser, dew heater and USB cable bundle pass up the
right hand fork to the OTA, fastened to it with the same velcro strap
that holds my 2 channel dew heater controller to the fork, at which
point they split into the component cables. Each of which is routed to
avoid snagging or dragging as the scope makes DEC movements. At the
other end the cable bundle exits the cable tray at the side of the tray
approximately in the "due South position", because in Polar mount this
is where the cables naturally fall. The trick is to only have sufficient
loop of this cable bundle laying in the tray to allow the scope to go
from RA lock to RA lock. On an ETX, this is plus and minus 0.875
rotations from the centre of the run lock to lock, this figure will make
sense when you realise, it is the minimum cable length needed in the
tray. Not yet added to my tray is a PTFE lining (sad, yes I know, but I
don't want snagging and dragging!).

6. Alignment Method - Forget One Star, and any variant of Easy
alignment. Waste of time! Get out your planisphere and choose two stars
that are not at the zenith and not at the horizon, and are at least 90
degrees of arc apart. I have found, and my reasoning is too lengthy even
here to go into, that one due East and one due West at an Alt of 45 deg,
or as near as possible is best.

7. Drive Percentages - Spend some time getting the Alt / Az percentages
optimum. If your scope doesn't move for 30 seconds on speed 1 or 2 when
you press the button, then you are never going to get good autoguide

8. Drive Training Optomisation - I use the spreadsheet tool and patches
from the ETX site to optomise drive training. Every little bit of
accuracy helps!

9. RA drive mechanics - The RA drive is without doubt the weakest link
on the ETX.It suffers from either too much lube or too little. I open it
up once a year and check the lube situation. The wires to the DEC drive
pass through the key way and RA clutch / drive, and are prone to
wrapping and snagging. At the least sign of trouble, they need checking,
and again once a year servicing checks. The mountings and settings for
the whole RA drive mechanism need checking, though with a little
Locktite, once set, even after a year mine seems fine. Amazingly I have
never had DEC problems with this scope, yet!

10. Too hot / too cold - Well in the UK, these are the two temperature
extremes that dictate our whole weather pattern! But due to lubrication
and I suspect tolerances, (and probably shivering / sweaty hands), drive
errors are larger below -5 deg C and above 25 deg C. They just are!!

Now having reduced the above 10 items, you can make a start on the
problems of autoguiding.

First off, connections. If you don't have the RS232 port on the Autostar
handbox connected to the serial port on the PC, and the Autostar via
comm port or Autostar via network setting on, nothing is going to
happen, yes been there, seen, done it!

Next the tracking centroid. Choice of object is crucial. If you choose
something bright, make sure it is the brightest object in a big area
surrounded by either a complete dark area or objects that are far less
bright and / or prominent. The size of this area is related to the drive
error and exposure time per frame you will use. My reasoning for this is
as follows:

When you autoguide, you shoot one alignment frame, followed by one image
capture frame. The alignment frame is short. So if you choose a dim or
low contrast object, it might be visible in the live image exposure and
the image capture exposure, but not in the alignment frame exposure,
which is based upon a longer exposure than alignment frames. So what I
am saying is Meade runs an algorithm, that makes the alignment frame a
fraction of the exposure time that you set for the image capture frame.
Now I'm not sure, but I suspect that what we actually have here are
three frame timings, the live image refresh rate, the alignment frame
rate and the exposure frame rate.

When you lock that centroid to your object of choice, Autostar Suite is
going to take an alignment picture frame, register and caluclate on it
and follow it, if it can! So if the object is one that is the same or
very similar to a number of objects around it, how can Autostar Suite
decide which one was the original if it has to make a choice, next
alignment frame? This gets more complex with the DSI Pro software, as
you now can use two centroids to take out rotational misalignment (or is
that easier as Autostar Suite can perform a checksum to see if the two
followed objects match the original distance between centroids in the
first alignment frame??).

With me so far?

Now if your drive errors are small, and atmospheric turbulence low, then
between alignment frames, the centroid will still be on the alignment
object, and Autostar Suite only has to apply a couple of guide pulses to
the RA and DEC drives. But if your drive errors are larger, then this
takes more work, all of which has to happen before the next image
capture frame is shot. And, you lose alignment if the drive error moves
the alignment object off screen or Autostar Suite cannot decide which of
two or more objects (three or more with DSI Pro!) is the alignment

So, with the LPI, when you are shooting image capture frames of a 1/10
of a second, and locking the centroid to a planet which has real angular
size and is unarguably the brightest object within a country mile, as
long as there is not a 30 second lag before your RA and DEC motors take
up the drive slack, then you would be hard pushed to lose alignment.

Move to the standard DSI and things become a little more complex, as you
start to shoot open clusters and globular clusters, you are aligning on
stars, and there are a lot more of them in the field of view. With
exposures up to 10 seconds, you could get object confusion errors,
whereby the centroid doesn't know which object to lock on to if you have
drifted outside the box, again my reasoning is this, in laymans terms;

When the box is drawn on screen, your object is inside it. The Autostar
Suite software checks the object size and brightness and position in the
box using a bit of math (checksum values) on the pixel values. Next
alignment frame, if the same math is roughly in the box, Autostar Suite
can send a few drive pulses to the ETX to centre the object in the box.
If the math is way out, Autostar Suite looks for the group of pixels
close by with matching math, and sends more drive pulses to the ETX to
catch up. If you choose an object too close to another similar object,
confusion could arise and you lose alignment!

OK now move to the DSI Pro, and shoot Deep Sky Objects, with exposures
of minutes, and L,R, G and B filter frames to take. Alignment can be
lost unless you have a well prepared scope, with few drive errors. I'm
sure that the two centroid solution is going to help (when I get that
far). But my recommendation for the ETX has to be to stick to the
shortest image capture exposure time possible and stack the images.
Unfortunately Autostar IP is not great for checking pixel saturation
levels quickly to adjust exposure times for DSO's, Maxim DL is better
for that. Add in the time taken for Bias Frames (if you bother!), Flat
field frames (do you bother?) and Dark Frames at regular intervals, and
you have a long night ahead! And of course if you want more than pretty
pictures, e.g., spectra, measuring asteroid light curves, calculating
orbital periods of eclipsing binaries, then you need to shoot single
exposure frames.

Now some late thoughts:

If the ETX drifts due to drive errors during the image capture frame,
you get smeared images, i.e., star trails. So maybe all the autoguiding
is doing primarily is providing a frame of reference for image stacking.
Any significant drift due to drive error and you just stack smeared
images to get a high resolution smeared image.

The DSI Pro software with the dual trackboxes would help to remove
rotational smearing to a degree, which is the point behind it if you
read the small print, but in the end it has to be all down to reducing
drive errors to a minimum and the use of edge enhancement and sharpening

Certainly by using the DSI Pro with its IR, Red, Green and Blue filters
there is a chance of sharper focus per image component as with less
bandwidth by definition you can achieve sharper focus. Again precise
focus is hard to achieve using methods like FWHM with Autostar IP.

Seems the only way to get decent long exposure guiding is with a second
camera, perhaps a low weight guide scope and the LPI has some merit.


Stephen Bird

Return to the top of this page.

Go back to the Astrophotography Page.

Go back to my ETX Home Page.

Copyright ©2005 Michael L. Weasner / etx@me.com
Submittal Copyright © 2005 by the Submitter
URL = http://www.weasner.com/etx/astrophotography/2005/ccd_optimizing.html