Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
11-26-2014, 11:26 PM   #1
Veteran Member




Join Date: Nov 2013
Posts: 385
AstroTracer Quirks: Does anybody know when it works/does not work?

I can't figure out how to get reliable tracing from my AstroTracer unit. Sometimes it works perfectly but then all of a sudden I get a lot of star trails. I have been trying to see a pattern but have found nothing so far. Two things I have noticed:

- It seems to work better if pointed towards north at flat angle. But even then suddenly the tracking gets worse.
- After calibration it works best with the first object I am pointing it to. If I point it to another object the tracking gets worse.

Has anybody found a pattern what works/doesn't work with the AstroTracer?

11-27-2014, 12:26 AM   #2
Pentaxian
Sagitta's Avatar

Join Date: Feb 2011
Location: Maine
Photos: Gallery
Posts: 2,949
QuoteOriginally posted by maxxxx Quote
I can't figure out how to get reliable tracing from my AstroTracer unit. Sometimes it works perfectly but then all of a sudden I get a lot of star trails. I have been trying to see a pattern but have found nothing so far. Two things I have noticed:

- It seems to work better if pointed towards north at flat angle. But even then suddenly the tracking gets worse.
- After calibration it works best with the first object I am pointing it to. If I point it to another object the tracking gets worse.

Has anybody found a pattern what works/doesn't work with the AstroTracer?

From my own experiences...


Be aware of the topography you're shooting at. I've notice that when I shoot out in the boonies (or from my bedroom window of all places) I can track almost flawlessly. Out on the flat by the river? Forget it. Turns out that theres a lot of iron slag under the ground out there, which is probably causing the tracer to go haywire.


Ditto for your tripod. I had my best results by far tracking one night when I forgot my tripod head and was forced to improvise using a rock and a hat as a prop to hold the camera steady. (in other words, go old school and get a wooden or modern and get a poly tripod - avoid metal)

Did I mention avoiding metal? 'cause it seems to cause the tracer to go goofy.

Then again, that's just my experiences, YMMV of course.

These were my two 'hat' shots, just to show what I pulled off. First was UWA, so was easy-ish


90 seconds @ 10mm




Second was much less so.


This was a 60 second exposure at 300mm.

11-27-2014, 04:00 AM   #3
Forum Member




Join Date: May 2012
Posts: 98
I have used the Astrotracer unit quite a lot and have had a few occasions similar to yours.The two step calibration is not explained properly in the manual and I missed a few shots until I figured it out. When you first set up the GPS it requires a basic calibration process. However, when you turn on the Astrotracer function you must also do a "Precise Calibration" as well before taking pictures. The Astrotracer unit assumes that the precise calibration has been done even if it has not been done. The difference is the stars will have a slight, but noticeable trail. The length will increase with the time of the exposure.

Other suggestions:

I also always use the 2 second delay to ensure no vibration from the mirror movement. If your trails are smooth, it is a calibration problem. If they are jagged, it is a vibration problem. You need to have a solid tripod on solid ground. I have found a wooden deck transmits too much vibration and exposure to wind is also a culprit. I remove the strap from the camera as well.

When setting up the GPS in the menu, turn OFF the GPS Led indicator light in the menu. It is bright enough to affect your night vision if you look through the viewfinder.

Hope this helps.
11-27-2014, 06:20 AM   #4
Veteran Member
VisualDarkness's Avatar

Join Date: Nov 2011
Location: Uppsala, Sweden
Photos: Gallery
Posts: 4,441
So maybe my magnesium alloy tripod with a carbonfibre-metalalloy head is responsible for my lack of success with the tracer? I've done probably a hundred pricise calibration with the same bad result with the DA 300/4.

11-27-2014, 06:27 AM   #5
Loyal Site Supporter
Loyal Site Supporter




Join Date: Jun 2009
Location: Tumbleweed, Arizona
Photos: Gallery | Albums
Posts: 5,387
Well, here are a few more ideas....
  • Fresh battery - the satellite signals are very weak. It is incumbent on the receiver to acquire and lock on them. A marginal battery will not help. I have started carrying a spare with me in my camera backpack. I actually had a battery die on me the last time out - the spare came in handy.
  • Surrounding Metal - Trying to calibrate for astro (the three axis exercise) needs to be done away from lots of metal as was suggested before. Trying to cal next to my truck was a mistake. Also, a large old metal trestle bridge was not too helpful. See the multipathing explanation below.
  • Large flat vertical structures - I was down in Oak Creek Canyon and had a large rock face (base of a cliff ~ a couple hundred feet high) across the river (creek) from me. I was not trying to astro traking, but just geo locate. When I got home, I looked on the map function within LightRoom and my position varied 100 meters up and down the creek - really all over the place. This is a function of multipathing. Multipathing is where the satellite signal bounces and reflects off of things on its way to the receiver. The time delay from the bounce throws the position off. If the position lock is lost, the astro tracking is going to get lost, which would probably appear to be star trails.
  • Large flat surfaces - I have found that sometimes trying to geo locate when standing in a large field of broken shale rock, lots of flat surfaces reflecting the signal does not help. I think that the GPS unit would have an antenna mostly pointing up, but for good signal geometry, you want a lot of satellites located around you just up off the horizon for best tracking. The signal geometry will vary with the time of day (the satellites are always moving) - at most, just wait 5 minutes and you will have another geometry to work with.
  • Under a tree canopy - Probably not a particular problem with astro tracking, but if you are under a canopy of trees, they tend to absorb the satellite signal, which would result in either loss of lock or continuous re-locking, which would cause bad location.
  • Signal Jamming - Close to cell phone towers, radio broadcast towers, etc. These will just jam or at least interfere with the GPS receiver and its ability to lock location.
  • Signal acquisition and locking - The little blue light will indicate lock (on steady) or attempting to re-lock (flashing). Signal locking is separate from calibration. You still need to calibrate in order to be able to astro track.
  • Enable / Start Astro Tracking - You need to go into the GPS menu, calibrate (the three axis exercise), setup the astro tracker - how long you want to track etc. And then enter the astro tracking sub menu and start the exposure within that submenu.
  • Focal Length - Its a bad habit of mine, but I change lenses without turning off the camera. So, if I go to an old K or M or M42 lens, the focal length may not be known to the camera. Astro tracking needs the focal length in order to track. If its wrong, the tracking is going to be off.
  • Elevation Angle - It seems if the camera is pointed up at an angle greater than 45 degrees, confusion sets in. So, straight up does not work either.
QuoteOriginally posted by VisualDarkness Quote
So maybe my magnesium alloy tripod with a carbonfibre-metalalloy head is responsible for my lack of success with the tracer? I've done probably a hundred pricise calibration with the same bad result with the DA 300/4.
Then there is your actual location on the earth. The satellite orbits, tend to be some what polar (see diagram). The farther north or south (nearer the poles) you will be able to track, but your satellite geometry is going to not be the best. You need to acquire and lock in order to calibrate. With a long focal length (300mm) your astro tracking time is going to well less than a minute as the stars will be moving quickly through your field of view, and the satellites will be very rapidly changing their geometry relative to your position - which will make for a less than ideal situation to try to track in. You need 4 satellites to astro track - 1 for each axis (latitude, longitude, altitude and time) with good angular separation (they all can not be clustered to the south of you). At best if you have lousy tracking, wait 5 minutes and try again. If you are up in the norther part of Sweden (I would say 55 degrees north and above), are you above the satellite's northern part of their orbits? If so, then you might have the bad angular separation problem. Waiting a few minutes to let them fly to a better separation geometry for locking and tracking would be a potential fix. I use a metal tripod without any problem.



Last edited by interested_observer; 11-27-2014 at 10:00 AM.
11-27-2014, 07:05 AM   #6
Veteran Member
VisualDarkness's Avatar

Join Date: Nov 2011
Location: Uppsala, Sweden
Photos: Gallery
Posts: 4,441
Yes, I'm at about 59 degrees, so that is a unsolvable negative for me. Add to it that shooting low towards the horizon is most often impossible for me due to obstacles or light pollution and that makes me shoot pretty straight up most of the time. Maybe the tracer simply is a bad device for my situation.
11-27-2014, 09:31 AM   #7
Forum Member




Join Date: May 2012
Posts: 98
I think another point to make is that the long the focal length, the more precision is required for tracking. I tried a 300mm and I was not pleased with the results. This is where the O-GPS1 unit shows it's weakness. I find it very good up to 135mm, but on a deep sky object that is very faint you need more exposure time. Also, I noticed edge coma on long focal lengths at the maximum time settings. This may be caused be the tilting of the sensor relative to the lens.

I ended up getting an iOptron Skytracker and it solved all the problems. It does not rely on GPS and it moves the entire camera and lens. You just have to polar align the unit and you can use any focal length you want and image as long as you want. (5 minutes is about the maximum. You are also able to use the multiple exposure functions. That unit mounts on a tripod and costs about US$400.
11-27-2014, 09:35 AM   #8
Veteran Member




Join Date: Nov 2013
Posts: 385
Original Poster
QuoteOriginally posted by Sagitta Quote
From my own experiences...


Be aware of the topography you're shooting at. I've notice that when I shoot out in the boonies (or from my bedroom window of all places) I can track almost flawlessly. Out on the flat by the river? Forget it. Turns out that theres a lot of iron slag under the ground out there, which is probably causing the tracer to go haywire.


Ditto for your tripod. I had my best results by far tracking one night when I forgot my tripod head and was forced to improvise using a rock and a hat as a prop to hold the camera steady. (in other words, go old school and get a wooden or modern and get a poly tripod - avoid metal)

Did I mention avoiding metal? 'cause it seems to cause the tracer to go goofy.

Then again, that's just my experiences, YMMV of course.

These were my two 'hat' shots, just to show what I pulled off. First was UWA, so was easy-ish


90 seconds @ 10mm




Second was much less so.


This was a 60 second exposure at 300mm.

If the metal theory is right then a carbon fiber tripod should help too. Are there any carbon fiber ball heads?

---------- Post added 11-27-14 at 08:38 AM ----------

QuoteOriginally posted by Hamiltom Quote
I think another point to make is that the long the focal length, the more precision is required for tracking. I tried a 300mm and I was not pleased with the results. This is where the O-GPS1 unit shows it's weakness. I find it very good up to 135mm, but on a deep sky object that is very faint you need more exposure time. Also, I noticed edge coma on long focal lengths at the maximum time settings. This may be caused be the tilting of the sensor relative to the lens.

I ended up getting an iOptron Skytracker and it solved all the problems. It does not rely on GPS and it moves the entire camera and lens. You just have to polar align the unit and you can use any focal length you want and image as long as you want. (5 minutes is about the maximum. You are also able to use the multiple exposure functions. That unit mounts on a tripod and costs about US$400.
I have never used a tracker. Do you have to be able to see Polaris to align it?

11-27-2014, 09:40 AM   #9
Veteran Member
VisualDarkness's Avatar

Join Date: Nov 2011
Location: Uppsala, Sweden
Photos: Gallery
Posts: 4,441
QuoteOriginally posted by maxxxx Quote
If the metal theory is right then a carbon fiber tripod should help too. Are there any carbon fiber ball heads?
My ballhead is "carbon fiber" It means that the metal is a tad thinner and reinforced with carbon fiber to save a tiny tad of weight. FEISOL Ball Head CB-50DC Carbon Fiber with Release Plate QP-144750
11-27-2014, 09:55 AM   #10
Veteran Member




Join Date: Nov 2013
Posts: 385
Original Poster
QuoteOriginally posted by VisualDarkness Quote
My ballhead is "carbon fiber" It means that the metal is a tad thinner and reinforced with carbon fiber to save a tiny tad of weight. FEISOL Ball Head CB-50DC Carbon Fiber with Release Plate QP-144750
It would be curious if a Formula 1 team built one. It would probably be made of pure carbon fiber and weigh only 100g. And of course cost $50000.
11-27-2014, 10:01 AM   #11
Veteran Member
VisualDarkness's Avatar

Join Date: Nov 2011
Location: Uppsala, Sweden
Photos: Gallery
Posts: 4,441
QuoteOriginally posted by maxxxx Quote
It would be curious if a Formula 1 team built one. It would probably be made of pure carbon fiber and weigh only 100g. And of course cost $50000.
And break as soon as you bumped it into something, making you go go home for a pit stop to replace broken parts.
11-27-2014, 10:12 AM   #12
Veteran Member




Join Date: Nov 2013
Posts: 385
Original Poster
QuoteOriginally posted by VisualDarkness Quote
And break as soon as you bumped it into something, making you go go home for a pit stop to replace broken parts.
But the pit stop would take only 2.5 seconds.
11-27-2014, 10:29 AM   #13
Veteran Member
VisualDarkness's Avatar

Join Date: Nov 2011
Location: Uppsala, Sweden
Photos: Gallery
Posts: 4,441
QuoteOriginally posted by maxxxx Quote
But the pit stop would take only 2.5 seconds.
Unless it suddenly burst into flames.
11-27-2014, 10:41 AM   #14
Loyal Site Supporter
Loyal Site Supporter




Join Date: Jun 2009
Location: Tumbleweed, Arizona
Photos: Gallery | Albums
Posts: 5,387
QuoteOriginally posted by VisualDarkness Quote
My ballhead is "carbon fiber" It means that the metal is a tad thinner and reinforced with carbon fiber to save a tiny tad of weight. FEISOL Ball Head CB-50DC Carbon Fiber with Release Plate QP-144750
There is not enough metal in a ballhead to really make a difference. A truck - yes, a metal bridge - yes, especially with multi-pathing where a real close object with lots of reflective (metal) surface is bouncing the signal around. I have an old metal Manfrotto 3001 and it has posed no problem so far. As to a head, I use a small Acratech GP, but a while ago picked up a Manfrotto 410 geared head. The 410 is one large massive hunk of metal. Its probably about 8 inches across. If that does not cause a problem, then a mere ballhead should not cause a ripple.

I have been shooting with a set of pretty wide angle lenses (10mm to 31mm). After about 4.5 minutes, you can start to see the stars in the corners especially start to trail, while the ones in the center are rock solid. As the focal length increases the tracking time will decrease. I think someone in another thread said that at 300mm 30seconds was the limit. I could believe that the stars would be starting to track there as well. Although I read that the stars were tracking during the entire frame, so I don't think the time frame was the problem - probably the GPS unit dropped out of lock and was out of calibration, or the frame was engaged while not in the astro tracking sub menu. That is just a guess on my part. I caught myself doing that a couple of times too.

11-27-2014, 12:50 PM   #15
Pentaxian
Sagitta's Avatar

Join Date: Feb 2011
Location: Maine
Photos: Gallery
Posts: 2,949
Also, when doing your precise calibration, its actually *two* movements per axis, not one. Rotate camera to the left just over 180 degrees, repeat to the right / rotate up just over 180 degrees, repeat downwards / rotate clockwise just over 180 degrees, repeat counterclockwise.

If you simply rotate 360 degrees on each axis it (usually) won't calibrate correctly.

This is the wrong way:

This is the correct way (as per Ricoh/Pentax, who posted this):

http://www.ricoh-imaging.co.jp/english/products/o-gps1/calibration.html

Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
astrotracer, carbon, exposure, fiber, flickr, length, lens, metal, object, pattern, time, tripod, unit, worse
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Does anybody know what to do with K-3 RAW HDR images? maxxxx Pentax K-3 13 11-13-2014 04:56 PM
Does the O-GPS1 Astrotracer work with manual mirror lenses? maxxxx Pentax Camera and Field Accessories 8 01-25-2014 09:42 AM
Does anybody know? Vantage-Point Pentax DSLR Discussion 7 07-07-2011 07:47 AM
Shutter does not work when autoflash is on. Daniellah Troubleshooting and Beginner Help 8 04-25-2011 06:49 AM
Does anybody know the music played? LeDave General Talk 2 02-15-2010 01:53 PM



All times are GMT -7. The time now is 05:38 PM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top