EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: adhdengineer on May 12, 2013, 09:21:46 pm
-
So I took up astronomy as a hobby but living in Northern Ireland means i get very few cloud-free nights. This lead me to revisit my old teenage hobby of electronics.
Basically i want to build my own long exposure camera. I found a CMOS image sensor on farnell and bought myself a couple (at £15 per chip), now i need to try to use them.
Problem is that the footprint is unique to the IC (at least it is as far as i can tell) so i can't use an off the shelf breakout board.
Therefore i face the dilemma of making a straight pin breakout to turn my 48pin into a 48 pin DIP or making a basic board that has the power rails and IOs broken out and bypass caps in place.
I've decided to at least pretend to do it right so here is my breakout with simple power arrangements.
There are a few things i'm doing i know i probably shouldnt. For example the IC has an analog VDD/GND but i've tied that into one of the digital VDD and the main GND, mainly since i don't know how to join digital and analog ground.
I'm also tying VDDx pins that are next to each other together and having them share a bypass cap which is probably not the best.
http://www.aptina.com/products/image_sensors/mt9p031i12stc/ (http://www.aptina.com/products/image_sensors/mt9p031i12stc/)
that's my image sensor btw.
So my question is: how bad is my schematic/board
Ok so it wont let me attach the eagle files so i've taken a screenshot of them.
-
I think you can definitely learn a lot from layout out your own board, and it does allow you to place any decoupling parts close to the IC itself as needed, but you should know that there are ILCC breakout boards available (http://www.epboard.com/eproducts/doc/E48-0290_p.pdf (http://www.epboard.com/eproducts/doc/E48-0290_p.pdf))
-
I think you can definitely learn a lot from layout out your own board, and it does allow you to place any decoupling parts close to the IC itself as needed, but you should know that there are ILCC breakout boards available (http://www.epboard.com/eproducts/doc/E48-0290_p.pdf (http://www.epboard.com/eproducts/doc/E48-0290_p.pdf))
looks like they sell the one i want. $36 per board tho, seems a bit... steep. I should be able to make a straight DIP breakout board myself that would do the trick.
-
I have two questions:
1. Why do you want to build your own camera? Is it the bulb mode in normal cameras not enough for you?
2. How to do in Eagle the ground coper fields like you have. Is it some standard function?
-
I have two questions:
1. Why do you want to build your own camera? Is it the bulb mode in normal cameras not enough for you?
2. How to do in Eagle the ground coper fields like you have. Is it some standard function?
1) to see if i can, for the challenge, it's something to do. I do have a DSLR and i do use that but it would be nice to be able to do it myself so to speak.
2) you create a polygon covering the area you want as the ground plane then hit the rats-nest button and it fills it in. You have to use the Name tool to set to the ground plane or else it's an unnamed net, not connected to anything else.
-
I see.
I made my own timer/shutter for the DSLR camera. Also for counting the BULB mode. But astrophotography still in front of me.
Good luck with your project
-
Those tiny labels are going to be nothing but little white smudges, but it sounds like you're making your own board so no biggie. How confident are you in your ability to solder those caps and resistors?
-
Those tiny labels are going to be nothing but little white smudges, but it sounds like you're making your own board so no biggie. How confident are you in your ability to solder those caps and resistors?
I'm making my own reflow oven since the CMOS sensor is basically a QFN package so there's no way to solder it without reflow. Basically just an excuse to make a reflow oven.
I'll play around with some other boards to experiment with the reflowing since the caps are dead cheap i don't mind wasting them, but the cmos is expensive so i don't want to make a mess with that.
-
A couple of points:
Did you think to attach the board to anything, maybe? Mounting holes might be a good idea in that case.
Another thing since this is supposed to be an astro camera. How about cooling? I could find no noise data in the datasheet you attached but maybe i didn't look hard enough. However it used to be the case that astro imaging elements were/are cooled to the max to reduce thermal noise so you might want to spend a thought to that. The usual solution is a) place all hot items as distant from the sensor as possible and b) set up a Peltier element to suck the heat out of the sensor. Point b) can be a mechanical challenge.
Regarding the sensor, are you sure this particular one is suited to purpose? A quick look at the datasheet leads one to think it might be more oriented towards video capture than long exposure low light level photography... To make any proper sense it should have detailed bin sensitivity data otherwise astrophotometry at least is out. But then maybe you are not aiming for that.
-
A couple of points:
Did you think to attach the board to anything, maybe? Mounting holes might be a good idea in that case.
Another thing since this is supposed to be an astro camera. How about cooling? I could find no noise data in the datasheet you attached but maybe i didn't look hard enough. However it used to be the case that astro imaging elements were/are cooled to the max to reduce thermal noise so you might want to spend a thought to that. The usual solution is a) place all hot items as distant from the sensor as possible and b) set up a Peltier element to suck the heat out of the sensor. Point b) can be a mechanical challenge.
Regarding the sensor, are you sure this particular one is suited to purpose? A quick look at the datasheet leads one to think it might be more oriented towards video capture than long exposure low light level photography... To make any proper sense it should have detailed bin sensitivity data otherwise astrophotometry at least is out. But then maybe you are not aiming for that.
This is just a dev board so i'm just going to be mounting on a breadboard. Any final design would have mounting holes.
I do plan to add a TEC to the final design to try to improve low light long exposure noise performance so i'm planning on leaving the bottom of the board clear with no components.
The sensor probably isnt that suited to the purpose if i'm honest but it was pretty much the only one i could find. Ideally i'd like an APS-C or similar but i've no idea where you can source them in one off quantities. The project is just a bit of fun for cloudy nights, tho i might modify it to become an auto-guide camera for use with the telescope mount i'm saving up for.
-
how do you plan to do with the lens? any mount?
For any long sky exposure you need tracking. I think even 1-2 minutes will show the sky spin.
-
how do you plan to do with the lens? any mount?
For any long sky exposure you need tracking. I think even 1-2 minutes will show the sky spin.
It's not going to use a lens, just going directly into the telescope in place of the eyepiece.
Although i did have an idea to use a macro extension tube (the more expensive kind with the electronics pass through) to mount a canon lens and control it with a micro, just for giggles you understand, but i don't want to get distracted from the main project. That would be phase 2.
I do have a very basic equatorial telescope mount i can use, it has a simple motor to counter the earth's rotation. It's not great but at shorter focal lengths it's ok.
I'm saving up for the EQ-6 mount, but it's about 950 quid so it may take me a while. I want to get a new DSLR first (for shooting stills and videos of my 7 month old girl)