Azimuth / inclination for astrophoto


The idea: to have precise azimuth / inclination angle displayed on a LCD screen

Hardware : many inclination sensors are available - any suggestions? Would be controlled by Arduino (or one of its smaller siblings) and powered by batteries.

Use: precise alignment of the scope mount for astrophotography (that means the operation temperature range is -25 to + 35 Celsius). Two sensors will probably be required as I ned to measure the position of two axis of rotation, so one sensor would be the compass, the other would be for altitude angle.

Precision given the fact that distances are, well, astronomical, tiny errors in angle are much amplified so the precision I'm looking for is at least 0.1 degree.

Would you think this is feasable?

If you can find an accurate enough sensor, Arduino can certainly receive data from it and display on a LCD.
For example:

I think so too - Arduino should be able to talk to these kinds of sensors.

Was wondering about calbration - while easy enough for inclination - I think it may prove problematic for azimuth. Do sensors know which way north is based on the magnetic field of the Earth? Because if I need to first show it the 'zero' direction then it's quite useless.

And if indeed they are sensitive enough to see the magnetic field, wouldn't they be confused with all the interference from power lines, motors running, cell phones, etc. ?

Most consumer grade orientation sensors are accurate to about 1 degree, and to get that level of accuracy they must be carefully calibrated. This site describes the best approach (works for both accelerometers and magnetometers).

Of course, magnetometers are confused by nearby metallic objects, power lines, etc.

It's true that +/-1 degree seems to be the norm. That calibration technique looks very pro, would be worth trying it with a more precise sensor. A tenth of a degree would be enough for my scope - I'm sure I'll find something in that range.

Many thanks for the replys - happy to see this seems to be feasable.

You could always zero-out your mechanical apparatus to True North, which doesn't change, vs magnetic north, which does vary, and then the mechanical turn away from True North would always be the same. Your building is not moving, correct?

If you use a compass for azimuth, then you will need to know and factor in the local inclination to compensate for the difference between mag and geo north.

Also as already stated the compass reading will not be stable due to surrounding factors, such as human beings with mobile phones with high gauss magnets in them.

If you are a fixed mounting, then use a rotary encoder and have the base of the mount already calibrated to geo north.
I don't think any az/el astro systems use a compass, they just plant the base in the ground aligned to geo north, I believe you can use the stars to do this.

Tom..... :slight_smile:

(that means the operation temperature range is -25 to + 35 Celsius).

That rules out an Arduino then.

I used optical shaft encoders on my two axis and fed the information to a computer sat indoors. But that was a lot of years ago.

CrossRoads - it does, it's my portable scope on a tripod. The mechanics are nice and precise, but no markings to be able to turn a given number of degrees. Operation also includes moving the scope around on both axis to point it where I need after the initial alignment.

TomGeorge - yes, it's about a degree and a half due east. And I do use the polar star and an optical alignment tool, but believe me it's a royal pain in pitch dark, snow and negative temperatures.

Grumpy_Mike - yes, that's the luxury high-tech solution but a computer on top of all the other stuff would surely have the better of me and I'd become clinically insane. As for the temperatures, that's the extreme backet. More commonly it would be +/- 10. And I've found the Teensy to be quite smaller and equally suited for this.

Ok, on a tripod you're going to need to calibrate after powering up then.
'328P chip itself is good for -40C to +85C, you can easily build up a standalone design with parts selected to meet that range. I do a board for a customer in Africa where all parts are rated -40C to +125C, including the processor, Atmega328P-15AZ.


gives you all you need plus temperature and barometric pressure.
All the drivers do exist for arduino/clones.

What may not work properly at -25degC is an LCD display and a battery :slight_smile:

The accelerometer gives you 3axis inclination, magnetic sensor the azimuth (3axis).
The magn. sensor there is quite sensitive, so be careful with nearby metallic objects.

Excellent - thank you all. I shall return for your kind advice as I'm slowly working through this. Was already considering what other data I'd like to have :slight_smile:

I'm quite amazed that Celestron or the like would not offer this as standard equipment on their higher end gear. I mean, if I could theoretically build it with cheap components, it should cost nothing for their factories in China, right?

Maybe they just don't want to support it.

If a relative measurement is fine, you could use a patterned wheel with an optical encoder and calibrate with the north star.

I'm quite amazed that Celestron or the like would not offer this as standard equipment on their higher end gear.

Taking some time to wonder exactly why might be better than simply being quite amazed. It may be because they know what they are doing, and what you propose is not a very good idea. You are working on an alt/az mount while a serious astrophotographer with higher end gear from Celestron would use an equatorial mount and wouldn't give what you propose a moment's consideration.

I imagine Celestron would have a "where you are now" recording facility included in all their stuff, but it would be in RA/Dec, which is useful information, rather than azimuth/altitude, which is what you are doing and is useless information. I was going to point this out last night but I deleted my post when I understood that you only wanted a display, not a proper guidance system.

I now realise that even that is pointless because the information you intend to display is ephemeral. Don't worry about the 0.1 degree accuracy. That's the least of your problems, and all that stuff about compass accuracy, accelerometers, and low temperaures is all well and good, but I submit it's also baloney and a fine example of using an Arduino to build a square wheel. But, if you insist on doing this, you might include a clock, perhaps one with sidereal time.

with a GPS, you can ascertain your longitude and latitude. the first numbers needed for the formula.

time of day is key and that too can come from the GPS.

with our global position, you can put in your correction for true north and not worry too much about a cell phone in the initial stages.

line up on VEGA. press the vega key
line up on polaris, press the polaris key
line up on Rigel or maybe Vulkan and once you have each of these marked, your software can orient itself to your zero and you are ready for hours of freezing in the night air.

Nick_Pyner - the mount is equatorial, and while I take pride in not being too serious about it, it's been 10 years or so that I take photos of the sky, so I do have a vague idea of what is needed, or at least useful.

You should also know that sky coordinates can be equatorial or alt/az, and both systems are capable of pinpointing an object in the sky with sufficient accuracy for astrophoto. I am also very familiar with what is offered by astro gear manufacturers, and I do have some fine glass and sexy mechanics from several of them.

Now as to the usefullness of the ephemeral information - this is my mobile rig, and I often change both latitude and longitude when I use it, so the inclination of my RA axis has to be properly set to the latitude of my observing spot. As such, I do believe a sensor that shows me that angle can be useful.

And finally, please know that this is an add-on and not a replacement of my drive motors.

You seem pretty pissed-off about something, and I fail to see why. I did not ask for your opiniin whether it's useful, rather I asked whether it's doable. Let me worry about the square wheel, cause you seem a bit clueless about astrophoto.

line up on Rigel or maybe Vulkan and once you have each of these marked, your software can orient itself to your zero and you are ready for hours of freezing in the night air.

Never been able to find Vulkan, it's hidden behind Nibiru or something :slight_smile:

I'm not trying to build a GOTO, nor a pushTo. I want to be told what azimuth I have and what angle the RA axis is at.