Choosing a camera for near IR band

[Tech]

Hi,

I don't have much experience with cameras, but i need a camera for astrophotography in the near infrared band, and i was hoping you can help me with choosing it.


I need to photograph stars up to magnitude 2.5 (perhaps 3) – but this classification relates to magnitudes in the visual spectrum. Eventually I’d like to photograph stars up to magnitudes 5 or 6.



I need as high resolution as possible (but am currently not willing to put in $2000 yet), and high shutter speeds.
Honestly, i don't know what the shutter speeds will be, can someone tell me from experience?
The viewing angle will be 11° and i intend to use a 100 mm diameter lens.
The stars on the picture should be seen as at least 4 pixels wide and easily distinguishable from noise.
The picture will be (probably) filmed in, or post-processed to, grayscale.

I have chosen the following components:
· - Nikon d3200 - good price to resolution ratio, and as i read it has good sensitivity in near IR
· - 100 mm diameter manually adjustable - i don’t know which one to choose
· - an I or R band filter, perhaps a combination of both? - as far as i understand this should be chosen together with the lense

Are these suitable for my needs?
Is there a better set for what I need (for roughly the same or lower price)?
Is the lens unnecessarily large?


Thank you

[ZeroID (Brent)]

I assume you are talking about Ha signal ? (656 nm )
Most DSLR's require a modification to remove the factory installed IR filter on the sensor to see near Infrared spectrum. The filter is removed and an external IR filter which is open to Ha wavelength is used to extend the sensitivity.
I've modified a spare DSLR and added an external filter which gets it back to near normal performance but yet to use it in anger. From the little I understand about Ha high Resolution is not absolutely necessary or even good but I may be wrong. Most DSLR astroimagers are using Canon cameras for imaging as there is plenty of excellent software for them. EOS 1100D & 1200D appear to be cheap, low noise and work well IIRC.
High Shutter speeds ? Near IR is low luminance and requires longer exposures with higher ISO settings normally but maybe I am misunderstanding your project requirements.

[Tech]

Hi, thank you for the reply.

Actually I’m looking for 700 – 1000 nm, to avoid light scattering (to photograph stars during daylight).

Yes I read about the IR filter, and I just rechecked it – seems a bit complex, but if it has to be done, it will be done.

The pictures will be used for calculating distances between stars, thus high resolution is necessary. The leakage errors should not be a problem since a computer will do all the work.

Is there a way to calculate how low must the shutter speed be to register a star of a given magnitude – the star should register as at least 5 pixels in diameter (at this the constituting pixels should be at least 2 times brighter than noise)? If possible I’d like to calculate this before I buy the actual camera – I know speeds depend on the sensor, but I’d like to make at least an estimate.

[ZeroID (Brent)]

700-1000 is right up in the IR band. Depending on star brightness it should be quite short exposures, several seconds to maybe 10-20 at ISO1600.
It will also be affected by local 'seeing', skyquality and light pollution so an exact number would have to be experimented for.
Most modern DSLR cameras are highly IR sensitive once the IR cut filter is removed and have high ISO's and low noise. SONY chips appear to be used for a lot of dedicated astro cameras, very low noise and very sensitive. I think Nikon use SONY sensors, you would have to check the specs.
I occaisonally use my SONY SLT A77v @ 24 megapixel but not IR modded. Still gets some amazing detail.

When you say 'distances between stars' are you talking angular distances or actual LY distances spatially ? I am unsure as to how you would expect LY distance could be measured using a DSLR and computer.

[cometcatcher (Kevin)]

In daylight? The Sun still emits heaps of IR.

[sil (Steve)]

Originally Posted by ZeroID

SONY chips appear to be used for a lot of dedicated astro cameras, very low noise and very sensitive. I think Nikon use SONY sensors, you would have to check the specs.

Only some of the current Sony cameras are particularly good, they have made big advances but its not across all Sony cameras. Similarly only some of Nikons use good Sony sensors.

100mm diameter lens? Are you sure you don't mean focal length? The Nikon 105mm f2.8 Macro is the only choice. Fantastic lens for astro, been using it for years, Canon do make an equivalent one. BTW the f2.8 Nikon lenses are great, pin sharp, i use 14-24 and 70-200 as well.

[glend (Glen)]

Lifepixel in the US offers DSLR conversions to full IR only band operation. They also do other Narrow band mods. Check out their website. Prices seem reasonable and they deal international. FAQ pages covers evrrything they do, the home page is abit uninformative.

[bobbyf (Bob)]

Just in case anyone following this thread is interested, I'm just about to put my Nikon D7000 up for sale. It's been IR modded professionally by Camera Clinic in Melbourne and has the 720nm filter fitted.
It uses the Sony 16MP Exmor sensor which has an excellent reputation.
PM me if interested, and I'll send pics etc.
Regards

Bob

[Tech]

Thank you all for the input; i'll see what currently gives the best price/performance ratio and is available locally (I don’t know how sensitive the big lenses are thus I’d like to avoid any long and rough transport).

Originally Posted by ZeroID

When you say 'distances between stars' are you talking angular distances or actual LY distances spatially ? I am unsure as to how you would expect LY distance could be measured using a DSLR and computer.

I need it for measuring angular distances, which will later be used for additional operations.
I suppose LY distances could be measured with stereoscopy? Hmm… I wonder if that’s doable.

Originally Posted by cometcatcher

In daylight? The Sun still emits heaps of IR.

The IR is of course present, but its scattering is far lower than that of the visible light. Thus it shouldn't cause problems. What little of the scattered light will remain can probably be later filtered out by the software.

Originally Posted by sil

100mm diameter lens? Are you sure you don't mean focal length? The Nikon 105mm f2.8 Macro is the only choice. Fantastic lens for astro, been using it for years, Canon do make an equivalent one. BTW the f2.8 Nikon lenses are great, pin sharp, i use 14-24 and 70-200 as well.

Yes, I meant lens diameter. I’m going for a big one as I want as low exposure times as I can get.
What is the min/max viewing angle of these lenses?



@glend, I will check it out, thanks.



As I have done some additional research and calculations, it seems that 16 MPix chips might be accurate enough. This decreases the price of the cameras, but I will need to upgrade my algorithms.
Could someone send me a picture of starry sky, together with the values of exposure times and an approximate viewing angle (the angle the camera and lens registered on the picture – I can calculate this angle accurately, but I need an estimation first to base the calculation on)?
This would really help me with the decision on how much to invest in the camera for the first tests.

Thank you,

T