Barn-Door tracker test

I've been thinking for quite awhile about building a small barn-door tracker to improve my Milky Way shots. I got it built a couple days ago, and was able to get just a couple shots in last night. We had gone on a night float, and I had two other people with me who didn't want to spend much time out at 2:00 in the morning! I set up an extremely rough orientation with the North Star (as in, I couldn't even find it and took a wild guess) and fired away.

Here is the first shot I got.
Canon 5d, Samyang 14 2.8 @ 2.8, ISO 800, 191 seconds.

IMAGE LINK: https://flic.kr/p/ocCk​co

IMAGE LINK: https://flic.kr/p/ocCk​5Q

IMAGE LINK: https://flic.kr/p/otRc​yB

IMAGE LINK: https://flic.kr/p/ocCu​v2

davidmtml wrote in post #17059169
This is really the piece that does the magic. Minds smarter than my own figured out that if you have a screw with 20 threads per inch, place it exactly 11.42 inches from the hinge, and make 1 rotation per minute (in this case, 1/2 rotation every 30 seconds, and you must turn more frequently the longer your focal length), the camera will track the movement of the stars

Good show.. even a little crude tracking makes a big difference.

But focal length doesn't change the speed the earth rotates. What you're trying to do is:

Camera rotation = earth rotation - mount rotation.

You want to make mount rotation such that camera rotation = 0.. so focal length doesn't change anything.

Well done anyway.

rfdesigner wrote in post #17060844
Good show.. even a little crude tracking makes a big difference.

But focal length doesn't change the speed the earth rotates. What you're trying to do is:

Camera rotation = earth rotation - mount rotation.

You want to make mount rotation such that camera rotation = 0.. so focal length doesn't change anything.

Well done anyway.

Focal length affects magnification. More magnification means a smaller rotation of the earth will be visible. So more magnification - longer focal length - requires that his barn door needs to have the screw tweaked more often. Not rotated faster, but more often.

So instead of one turn every minute, he needs to switch to a half-turn every 30 seconds or a quarter turn every 15 seconds. Remember that he hasn't a motor that keeps the screw constantly rotating.

pwm2 wrote in post #17060877
Focal length affects magnification. More magnification means a smaller rotation of the earth will be visible. So more magnification - longer focal length - requires that his barn door needs to have the screw tweaked more often. Not rotated faster, but more often.

So instead of one turn every minute, he needs to switch to a half-turn every 30 seconds or a quarter turn every 15 seconds. Remember that he hasn't a motor that keeps the screw constantly rotating.

Ah.. that wasn't what I understood him to mean. Yes, in that case absolutely correct... re-reading I see this now.

I always think in terms of constant rotation, just trying to keep the rate approximately correct. For some reason I'd never considered stop and go.

Yep, you are both right haha. I see now how my original wording was confusing. 1rpm could be half a turn ever 30 seconds, a quarter turn every 15 seconds, 1/6 turn every 10 seconds etc...The ideal setup is to use a motor at 1rpm so that it IS a constant rotation. But luckily at wider angles, it is pretty easy to just make half a rotation every 30 seconds. As you get to longer focal lengths, it would be much easier to have a constant rotation.

Obviously at some point I would love to get it set up with a motor, but for now I was just going for cheap and simple!!

What about this setup prevents "camera shake" or vibration reaching the camera when turning the screw?

Pretty cool idea and implementation!

dlos wrote in post #17062367
What about this setup prevents "camera shake" or vibration reaching the camera when turning the screw?

Pretty cool idea and implementation!

Only the long exposure time. Same result as when you walk infront of the camera while using a 10-stop ND - the camera doesn't pick up walkers but only people who have stopped for a while.

pwm2 wrote in post #17062390
Only the long exposure time. Same result as when you walk infront of the camera while using a 10-stop ND - the camera doesn't pick up walkers but only people who have stopped for a while.

Yep, basically this. However, I do my best to be as gentle as possible while turning the screw.

davidmtml wrote in post #17063029
Yep, basically this. However, I do my best to be as gentle as possible while turning the screw.

The most critical part is that the tripod stands on stable ground, and the head doesn't slowly sag - the long exposure times can handle quite a bit of vibration but not a changed position/orientation of the camera.

why do you need a rubber band there?
do you have the link to the instructions?

pkim1230 wrote in post #17065034
why do you need a rubber band there?
do you have the link to the instructions?

The rubber band shouldn't be needed, but reduces risk of bumping the upper part. And extra load might also reduce issues with play in the hinges.

But depending on where you live (how aggresive angle you need to use) and how long the screw is, you might need the rubber band to make sure the camera weight doesn't overhang the hinges too much.

Nice example of the construction. Thanks!

Wonder if this could help automate things. https://www.sparkfun.c​om/products/12219

losangelino wrote in post #17065241
Wonder if this could help automate things. https://www.sparkfun.c​om/products/12219

A problem with it is that it is a "stupid" motor that gains more torque as you apply more voltage - so it also increases speed with the voltage while losing speed with the weight of the work it has to perform.

This means that you need to add some feedback sensor detecting how far it has rotated to decide if the voltage needs to be increased or decreased to actually maintain that speed of 1 turn/minute.

It's way better to use a stepper motor solution where a microcontroller issues step pulses at the required frequency. As long as they have enough hold torque and the step frequency isn't too high or the step acceleration too high, they will guarantee the requested turn rate.

I got inspired yesterday so had to make a quick check at the local "toy store" but they had no stepper motors home. So I have to decide if I have enough time to mail-order some gear and build. Or look at what older projects I can scrap to reclaim a suitable stepper motor.