Hi all,
for my project I need to detect the exact inclination of my Telescope (Dobson) but I don't know which sensor is to be used to get this information.
Can you give me some advice?
Thanks all!!!
Hi,
Some info that has been written about it Tilt only sensor for telescope. - Sensors - Arduino Forum before.
It seams to me that you can roughly detect inclination, but that it will be quite hard to detect the exact inclination with a sensor.
With a fixed (or home) -point and shaft encoder you could read which direction the telescope is moving and how many steps are made/ how much degrees it moved. It works, but in case the power fails, you'll need to move the telescope back to it's home-point again to be able to detect where it is when you start using it again.
You could do something similar with steppermotor/controller as well. Some give you the possibility to rotate a motorshaft 1 single turn in up to 3200 steps. You'll need to start with a fixed point as well, but once you know how many steps are needed to move the scope completely up/down, it's easy to move it very accurately.
You won't really "sense" inclination in both cases, but you'll know what it is.
What about a device like the Digital Angle Sensor on this page http://www.rdgtools.co.uk/acatalog/Digital_Measuring_Equipment.html
...R
you'll need to move the telescope back to it's home-point again to be able to detect where it is when you start using it again
Absolute shaft encoders won't have this problem.
Rob
Could I use an accelerometer (MMA7361) or a gyroscope (L3G4200D) or both (MPU-6050)?
To be useful only a high resolution rotary encoder is practical (and common) for this kind of telescope application.
shedo:
I need to detect the exact inclination of my Telescope
What resolution do you need?
Are you also trying to detect the bearing? The accelerometer-based approaches would be no help with that, and it would be to your advantage to adopt a common solution for both measurements.
PeterH:
What resolution do you need?
For resolution you mean how many numbers after the decimal point I want?
PeterH:
Are you also trying to detect the bearing?
For bearing you mean the rotation axe of telescope, "the azimuth axe"?
shedo:
PeterH:
What resolution do you need?For resolution you mean how many numbers after the decimal point I want?
Yes. Telescope aiming units are given in degrees/mins/seconds (or just degrees with adequate number of decimal points) and at typical magnification ranges used with telescopes one requires very high resolution encoders to be able to actually move the telescope to a given point in the sky and hope to be able to find of object of interest in the eyepiece field of view.
PeterH:
Are you also trying to detect the bearing?For bearing you mean the rotation axe of telescope, "the azimuth axe"?
It is possible to read inclination with an accelero-meter, but with a relative low resolution. It may also give you different numbers at different temperatures and each probably has a small error-rate.
In a lot of projects those aren't real disadvantages but they certainly could be in yours, depending on the nr. of decimals you want.
I currently use this for the inclination and this for azimuth.
For now my goal is just to have the current coordinates of my telescope on the phone instead of having it on the telescope, I don't want to do a GoTo system. This is and example of the kind of accuracy that I need:
Azimuth: 172°14'
Height (inclination): 31°25'
Your protractor is accurate to 6 minutes with a repeatability of 6 minutes http://djtnancy.en.made-in-china.com/product/IMFQHiqjEzhP/China-Mini-Digital-Protractor.html. It's hard to tell what someone wants when talking accuracy and astronomics, but if your protractor is accurate enough, you probably can use an accelerometer as well.
This is the most accurate accelerometer/inclinometer module (0.1 degree accuracy) that I know of, but it is not cheap: ADIS16210 Datasheet and Product Info | Analog Devices
The cheap ones are only accurate to about 1 degree in any one measurement, but with careful calibration and lots of averaging you should be able to do better.
Is there the possibility to use more sensors together to have more accuracy? For example gyroscope and accelerometer. I read something like 6DOF but I don't understand what it mean exactly (I think the precision).
I don't need exceptional accuracy, but something similar or little better than what I have already.
Are there any tutorials for Accelerometer/gyroscope for my purpose?
It may be possible to increase accuracy a small bit by comparing 2 accelerometers. But it's probably not worth the money. With an analogue component inside it, reading it multiple times and averinging the outcome may increase resolution as well, but probably not much.
You can do very little with a gyro unless the object is moving faster/ keeps moving.
6 minutes may sometimes last hours by the way 
I've tried to learn more about your mini protractor. It may be interesting to look which sensor was used/ post a high resolution photo of its PCB. It doesn't seem to be "superexpensive" and items probably sell better when sold with a 0.1 degree accuracy. (why not if it displays up to 0.01 ?/selling stuff better as it is not only happens in china)
If it is in reality not as accurate as sold in the advertisement, but still accurate enough for your goal... other accelero meters may be as good as well.
Must say I've still got to learn a lot when comparing accelerometers at mousers http://www2.mouser.com/Sensors/Motion-Position-Sensors/Accelerometers/_/N-axgd7?Keyword=accelerometer&FS=True
You don't want/need one with a high g rating.
More bits of sensitivity are better - avoid analogue-only ones, unless you've got external ADCs with better than ten bit resolution.
Then the MPU-6050 (Integrated 16bit ADCs, page 10 of datasheet) is the best one for my project?
Or if you have another sensor to suggest me, I'm all ears! 