Has anyone tried making these DIY OSC CCDs?

[abhoriel]

Our Ukrainian counterparts have designed some DIY CCD cameras based on the 6MP colour Sony ICX453AQ CCD (also used in QHY8Pro). The CCD can be obtained from old defective Nikon SLRs on ebay very cheaply. It should presumably be well known about here, but I don't really see any topics on it! They have been progressive improving the design cam10, cam81, cam83, cam84, cam86 etc 

They provide instructions, components lists, gerber format PCB diagrams.

Here is the info on the older cam84 (which is in English)

http://astroccd.org/2015/04/cam84/

Here is the newer cam86 which uses a microcontroller and can handle TEC cooling etc. Its in Russian, but google translate does a good job:

http://astroccd.org/2016/10/cam86/

 

Theres even a cam90 which uses a newer, higher resolution CCD.. the schematics/PCBs are on the Ukrainian forums but I think its kind of under testing.

 

Here's the Ukrainian language forum topic (full of useful info, but use google translate): 

http://www.astroclub.kiev.ua/forum/index.php?topic=28929.0

There's also a cloudynights topic, which seems to focus on the older cam84 much more:

https://www.cloudynights.com/topic/497530-diy-astro-ccd-16-bit-color-6mpx-camera/

And a French topic on it (focusing much more on the cam86), in which pagpatrice have done a great job of organising everything (again, google translate!):

http://www.webastro.net/forum/showthread.php?t=141764

https://www.diycam.fr/index.php/fr/9-all/construction/5-cam86-fournitures-part-i-la-carte

 

 

I don't know if anyone is interested in this here, just wanted to see. I'm considering giving it a shot for the learning experience primarily I suppose!

[Gina]

Several years ago I carried out experiments with Canon cameras with APS-C sensors, cooling the sensor to various temperatures and at various gain settings but little improvement was gained in signal to noise performance below 0C and nothing at all below -5C which was disappointing.  I don't know what modern DSLR sensors are like or Nikon v Canon, they may be different.  One thing I do know is that the sensors used in ZWO cameras do benefit from cooling to -30C so there have been developments in CMOS sensors.

Edited October 27, 2017 by Gina

[abhoriel]

8 minutes ago, Gina said:

Several years ago I carried out experiments with Canon cameras with APS-C sensors, cooling the sensor to various temperatures and at various gain settings but little improvement was gained in signal to noise performance below 0C and nothing at all below -5C which was disappointing.  I don't know what modern DSLR sensors are like or Nikon v Canon, they may be different.  One thing I do know is that the sensors used in ZWO cameras do benefit from cooling to -30C so there have been developments in CMOS sensors.

Sure. But nikon sensors are all sony made and are different from the canon ones. Also, this project uses a sensor from an older SLR which uses a CCD rather than a CMOS sensor. The same CCD is used in the qhy8pro, in which the cooler presumably helps!

[Gina]

Ah yes - I see.  Yes, I gather CCD are eminently suited to cooling.

[Mick J]

I have glanced (in the past) at a thread on IceInSpace started by Brendan, covers the same camera's I believe.  A long read at 42 pages and it's part 2 !!

Looks good work from everybody.  (as a btb - always good diy stuff on iceinspace)

Edited October 27, 2017 by Mick J

[abhoriel]

6 minutes ago, Mick J said:

I have glanced (in the past) at a thread on IceInSpace started by Brendan, covers the same camera's I believe.  A long read at 42 pages and it's part 2 !!

Looks good work from everybody.  (as a btb - always good diy stuff on iceinspace)

Ah yes, I should have included that topic here!

[GVidinski]

Hello stargazers!

Let's resurrect this thread. How many of you have already done one of those CAM86/CAM87/CAM90/CAM10? I'm interesting if you can share some feedback regarding assembling, parts, enclosures, mounts etc. as I'm going to build one for myself.

 

Happy holidays!

Georgi

 

[skybadger]

I have one PCB mostly assembled on the bench but had to unsolder the surface mount chips to try to find a power bug which caused the 15v line to disappear. Still Haven't found it. My approach is to make up a second board and see if the same thing happens now that I'm better at soldering. 

Have a handful of d70 chips to desolder and separate. Mostly lacking time to do all this as well as other projects. 

 

[Cooking Blight]

I have actually built two of the CAM86 version. One for myself and another one for a colleague. Because I was going to use it with a hyperstar on my CPC800 I had to design my own round version to keep it from blocking too much light. I also implemented a few improvements in the circuit that had been discussed on the main topic. I also chose a different cooling solution with PTC directly underneath the CCD. The weak point in this design is the custom heatsing which also acts as the front housing. A fan at the back pulls air in from the front which gives it a temeprature delta of around 14C below ambient which is not that great. If I could pull away heat faster it could probably get to around 30-40C below ambient. I have been toying with the idea of a liquid cooling loop to the camera, but I have seen little to no benefit with temperatures below 0C. Because I live north of the polar circle high temperatures are not an issue.

The housing is made from aluminium and has mostly been anodized black. A few structural components have been left without any surface treatment. The front is sealed off with a 2" filter (Not shown).

One other thing I decided was to re-write both parts of the firmware and the complete low level driver. The firmware was compiled using a software that required registration if I remember correctly, and the driver was written in pascal. The low level driver I have been doing some testing with is now written in C++ and the firmware is compiled with XC8. The driver required a slight re-write of the ASCOM driver due to some of the datatypes used. I have also briefly tested a 64bit driver with NINA, but haven't gotten that far with it.

[markse68]

Wow- really amazing work @Cooking Blight! Quite beautiful too! Can’t tell from your pics how long the fins are on the heat sink/front housing- do they continue down inside the housing? Would increasing their surface area help with the cooling?

Mark

[Cooking Blight]

Thank you!

The fins don't extend into the housing as there is no space left. Increasing the surface area would improve cooling significantly. I can't remember exactly what the temperature difference is between the hot and cold side of the PTC at its current "setting", but it is probably around 25-35C. With increased heat dissipation you could get the temperature on the sensor a lot lower. I was thinking about machining a new heatsink for the camera with fins extending on the outside. The problem is the time it takes to machine this part as it is rather complex. I made enough parts for a total of 6 cameras while I had the setup for it (CNC lathe with live tooling), but couldnt get a high enough spindle speed to run smaller end-mills and thereby increase the number of fins and surface area as it would take too much time.

A liquid cooling jacket instead of the cooling fins would significanly increase the cooling capabilites, but would also add some complexity to the camera. This is something I will do with one of the space camera housings at some point, but I'm currently working on a lot of other projects for the telescope and other things.

[Ratlet]

Might have a look at this.  I've got a Nikon d70s sitting in a box waiting to go to the charity shop.

[skybadger]

That's a lot of hard work gone into that and it comes out looking a lot like my atik11000. Very nice. 

Water cooling is a risky business from my experience, again with the 11000. 

There's no real reason you can't get good low temperatures but you'll need to size the tec unit appropriately and raise the ccd on a socket to get sufficient clearance for the tec and good conductivity via a heat pipe to the case for cooling. Getting a good thermal sink to the hotside block is the key. 

I have some spare PCBs for the original format cam86 if anyone wants any but they are cheap enough for a set of 10 from the web.