[Gimbal]

In order to use the astro tracer it's necessary to have the electronic compass calibrated. And even though the calibration process says that the calibration procedure is "OK" the result can differ quite a lot from time to time.

With the ogps-1 on my K-5 i could after a couple of trials get a pretty good accuracy. Usually less then 3 degrees error for most of the directions but often there were one direction that could be up to 5 degrees in error. But all in all quite good.

With the K-1 I have yet to get under 15 degrees in error in the "bad" directions. Usually about 20 degrees in error in some directions and less then 3 in others. Very inconsistent.

But during my tests I also found out that the movable LCD affects the compass with up to 10-15 degrees in some directions depending on the position of the LCD. So moving the LCD in or out turns the compass. This does not happen in all directions though.

So I figured that since I will have the LCD out and tilted when doing astro work I will also have to have the LCD out when calibrating. But while that does give a different result compared to having the LCD tucked in the accuracy didn't improve much.


The procedure I use to check the compass is to first align the tripod head (which have a 360 degrees scale) so that 0 degrees on the scale has the camera pointing to the north. Then I attach the camera and watch the compass heading on the LCD and compares with the scale on the tripod. Turn into different directions and compare. Do not have the camera mounted on the tripod when aligning to the north since the camera body is full of magnets and will affect any compass used to align the tripod.


So the question is, has anyone else checked the accuracy of the built in compass, and did you get at better result?

[MJKoski]

I have now used AT during 5 night shooting sessions and in one place the calibration was badly off even though it said complete. Other than that I think it worked nicely. Compass reading was comparable with my analog compass used for orienteering. I keep my cellphone away at all times when doing the calibration.

[biz-engineer]

Compass reading changes when moving LCD in and out of its compartment +-5 degree. Not surprising since magnetic material changes the course of terrestrial magnetic field which is in the range of +-35 uT. But if you keep the LCD display in the same position during calibration and use, it should work. Did you see any effect on tracking of stars?

[monochrome]

I taught a course in Map and Compass to a group of Scouts. The first time I laid out the short orienteering course to complete the workshop I was surprised to see my compass bearings not lining up with known and visible map landmarks. Then I saw the yellow 'Underground Pipeline' warning sign.

You never know what is around you.

[Adam]

You basically need to keep re-calibrating until you get sharp (i.e. trail-free) shots at 30s or so. There certainly is some trial-and-error involved.

I recommend revolving the camera along one axis at a time, rotating 180 degrees and back in each of the 6 directions.

[Gimbal]

But if you keep the LCD display in the same position during calibration and use, it should work. Did you see any effect on tracking of stars?

That's what I figured as well, and even though I got a slightly different result it wasn't that much better. I still have quite a large variance in precision during a complete 360 degree turn. At some directions it is spot on, at other directions it is slightly behind and yet other directions the compass is way ahead. A little variance is expected but when the error reaches 20 degrees it is way to much. This could happen with the ogps-1 as well but a couple of re-calibrations later one usually could get a really good accuracy.

---------- Post added 08-28-2016 at 09:01 PM ----------

I recommend revolving the camera along one axis at a time, rotating 180 degrees and back in each of the 6 directions.

Yes, that is what I do. I usually rotate the first axis while still mounted on the tripod, then hook it off and do the other two "by hand".

[Adam]

That's what I figured as well, and even though I got a slightly different result it wasn't that much better. I still have quite a large variance in precision during a complete 360 degree turn. At some directions it is spot on, at other directions it is slightly behind and yet other directions the compass is way ahead. A little variance is expected but when the error reaches 20 degrees it is way to much. This could happen with the ogps-1 as well but a couple of re-calibrations later one usually could get a really good accuracy.

---------- Post added 08-28-2016 at 09:01 PM ----------



Yes, that is what I do. I usually rotate the first axis while still mounted on the tripod, then hook it off and do the other two "by hand".

Yes, that is what I do. I usually rotate the first axis while still mounted on the tripod, then hook it off and do the other two "by hand".

I'd just take it off for all 3, since you want to fix the position of the camera, ideally.

Adam
PentaxForums.com Webmaster (Site Usage Guide | Site Help | My Photography)



PentaxForums.com server and development costs are user-supported. You can help cover these costs by donating or purchasing one of our Pentax eBooks. Or, buy your photo gear from our affiliates, Adorama, B&H Photo, KEH, or Topaz Labs, and get FREE Marketplace access - click here to see how! Trusted Pentax retailers:

[stevebrot]

You basically need to keep re-calibrating until you get sharp (i.e. trail-free) shots at 30s or so. There certainly is some trial-and-error involved.

I recommend revolving the camera along one axis at a time, rotating 180 degrees and back in each of the 6 directions.

Too bad there is no option for manual override if you have a known bearing.


Steve

[Gimbal]

I'd just take it off for all 3, since you want to fix the position of the camera, ideally.

I've tried that as well. But this rotating of the body is most probably only for calibrating the 3-axis electronic compass. The position fix is made by the GPS, and a GPS does not need any manual calibration, but an electronic compass do. In fact you can do the calibration process while still being in house without the GPS ever getting a lock. (But of course is should also be done at the photo site, but for practicing purpose you can do it at home in the house.)

---------- Post added 08-28-2016 at 09:51 PM ----------

Too bad there is no option for manual override if you have a known bearing.


Steve

I fully agree. Or even better a manual deviation table for the complete 360 degree turn.

[MJKoski]

Just FYI: It is possible to "complete" the calibration by rotating around one axis only if done long enough. Thay will obviously result in bad output.

[leekil]

I'd just take it off for all 3, since you want to fix the position of the camera, ideally.

Maybe the camera should be calibrated completely while attached to the tripod? I assume the tripod is metal or has metal in it, and since the movement of metal (like the screen) affects the magnetic field, therefore the tripod might as well.

[repaap]

I did read in manual that every time when you change lens or focal lenght you need to calibrate, IMO even when you Change your camera postion, calibrate. I calibrated my camera by trying to keep sensor in one place and rest turning around it. It is quite difficult, but I did get rid of trailing up to sertain point. wideangles(20 mm) are more difficult than for example 43 or 77 mm... because edges smeares in some point no matter what.

[Adam]

Maybe the camera should be calibrated completely while attached to the tripod? I assume the tripod is metal or has metal in it, and since the movement of metal (like the screen) affects the magnetic field, therefore the tripod might as well.

I've gotten perfectly fine results while hand-holding and calibrating. But you have to be careful to keep the camera still and your motions precise.

[MJKoski]

I have noticed no difference considering the stability of the calibration procedure. My first calibration attempt was quite "unstable" I would say But still, tracking worked fine. During the winter it WILL be PITA to do in cold with stiff fingers and all that. Be Prepared to dig something from the snow...

[PhilRich]

I did read in manual that every time when you change lens or focal lenght you need to calibrate, IMO even when you Change your camera postion, calibrate. I calibrated my camera by trying to keep sensor in one place and rest turning around it. It is quite difficult, but I did get rid of trailing up to sertain point. wideangles(20 mm) are more difficult than for example 43 or 77 mm... because edges smeares in some point no matter what.

So, if using a zoom lens, the camera doesn't calculate compensation for focal length changes?