ICX204 v ICX445 CCD questions

[Leedsgreen]

Dear All,

A few months ago I wrote with a question re: good lunar CCD camera and received some great advice. If OK, I would like to ask some further questions, now that I’ve done some more research?

At the moment I use a SPC900NC modified cam with the ICX098BL chip. My reason for upgrade is to get more real-estate and sensitivity. The CCDs I have been considering are:

ICX204AL (DMK31 and new Opticstar camera PX-75M)

ICX445ALA (Point Grey Chameleon)

With regard to the ICX204AL, I recently found the new Opticstar camera PX-75M, which offers 20fps at full resolution but also 40fps and 80fps using ROI. The camera is fan cooled at £80 cheaper than the DMK31 at £399:

http://www.opticstar.com/Run/Astronomy/Astro-Imagers-Opticstar.asp?p=0_10_0_50_110

Point Grey have announced the Chameleon (ICX445ALA) at ~£425 (with USB lead) max rate 18fps. If my research is right, this can also do ROI and higher frame rates?

http://www.edmundoptics.com/onlinecatalog/displayproduct.cfm?productID=3356

Now for my questions:

1. ICX204AL has pixels of 4.65 microns compared to 3.75 microns in ICX445, am I correct in assuming that for the same focal length, the ICX445 will resolve more detail on lunar surface?

2. Is there a relationship between focal length and pixel size (i.e. will I still be able to work at 3000mm using ICX445ALA as I currently do with my ICX098BL chip)?

3. Whilst very tempting, does anyone have experience of the software bundled with the Point Grey Chameleon (i.e. my sense is that DMK software is well tested and Opticstar appear confident with their Opticstar View software)?

4. It seems the ICX445ALA might offer the same sensitivity as the ICX618ALA and with smaller pixels (and ROI) could even be almost as good as the latter for planetary work:

http://www.ptgrey.com/support/downloads/documents/TAN2008006_Sensor_Response_Curve_Comparison_for_ICX445.pdf

Would this be a fair assumption?

Thanks very much for any thoughts/advice,

Nick

[riklaunim]

ICX204/205 is old HAD CCD. ICX445 is the successor - ExView HAD which gives much higher sensitivity in general plus the "ExView" keeps the high sensitivity in red and infrared part of the spectrum (important for H-a solar imaging, lunary/planetary imaging in Infrared bands).

Caméras industrielles Basler ACE pour les applications scientifiques, expérimentales et astronomiques

There are few people with ICX445 and ICX618 Baslers and for planetary they preffer the 618 sensor as it gives better sensitivity for ~the same scale. 445 rocks at lunar mosaics.

And for such cameras more aperture is welcomed

[Pablito]

I don't know if it helps but you can find some of my solar shots using PTgrey Chameleon HERE

Resolution is beautiful but provided software is awful

I have problems in high speed and have to lower it to get stable frames in avi. So far had no time to check what's wrong but the images are devided in 4 parts in most of frames (parts change place...) and are completely unusable in stacking.

Will play with FireCapture soon

[Leedsgreen]

Pablito and Riklaunim,

Thanks for your useful replies. I have to say, I'm very tempted by the ICX445. Pablito, will be interesting to hear how you find Firecapture software for the Chameleon. Please will you post results if you get the chance! What fps do you manage at the moment? Also, does the Chameleon offer ROI? What size options are possible and how does this effect fps?

Rikluanim, can you give any indication of acceptable aperture? At the moment I only have 127mm (5") Mak. Is there a relationship between f/number and pixel size (i.e. is there a benefit of going above f/30 with 3.75 microns)? Would appreciate any guidance you can give.

Thanks again,

Nick

[riklaunim]

Each aperture has max resolving power. With smaller pixels you get it at smaller f-ratios. Theoretical for 5,6 micron pixels is f/20, while for 3,75 is f/14. If you hit the max resolving powers of a scope - then the only way to get more resolving power is to get a bigger scope :] "Cheap" way are Newtonians, and for example second hand SCTs from the US market (in good shape).

Basler and Point Grey cameras support ROI, binning, but usually don't have very long exposures... as they use GigE and Firewire 800 for very fast framerates. Point Grey is popular in USA (as it's made there), Basler in Europe. Those companies produce competiting models (Ace vs Flea 3, Chameleon vs Pilot/Scout). PGR is supported by FireCapture and a Linux application (I think Anthony Wesley knows more about that). Basler also runs via FireCapture, but there is Genica application comming (may not be free), and SharpCap is getting some Ace support too.

Basler uses ethernet, PGR uses Firewire 800 (9-pin) which offer fast framerates (100-120 FPS of full frame ICX618). This however is more than some laptops can handle. In my case I have to use better ethernet card (on smartcard) that is capable of doing gigabit ethernet without loosing packages And the HDD can do full frame at around 90 FPS. Can't save faster. However 400x400 ROI on Jupiter and on green channel I could go over 100 FPS at f/20 This is for 8-bit format. The 12-bit is more data -> slower -> more demanding on the computer. FireCapture can lag the laptop a bit. Genica works much better (SER format). For planetary imaging the 12-bit isn't usually needed to anything. In solar H-alpha imaging there are bright parts (the sun), and darker (things around the sun) and that can benefit from 12/14 bit imaging.

[Leedsgreen]

Riklaunim,

Thanks very much for the detailed reply. That's just the sort of info I am after (but which seems hard to find)! Mmmm, with relation to resolving power I have noticed that my ICX098BL (5.6 microns) produces great images at f/23.6 (i.e. my f/11.8 MAK with 2x Barlow) but when I push it to f/29.5 (with 2.5x Powermate) there appears little (if no) improvement on lunar resolution.

In your experience how rigid is the theoretical resolving limit? For example, could you ever imagine good results with 3.75 microns at f/23.6 or would that just not work? If I stick with my MAK (and for the sake of portability it's currently a preferred option) then it seems that it terms of resolution there would be little to choose between ICX098 at f/20, ICX204 at f/16 and ICX445 at f/14 or am I missing something? (I realise in terms of lunar real estate there will be a noticeable increase in coverage).

Thanks again,

Nick

[riklaunim]

Things like Mars and Moon like to overscale. Jupiter less. Nevertheless no matter what sensor you have - you won't get more max. resolution 8-10" gives a lot of resolving power and still is easy to use. 11" or more is... demanding

I did a Jupiter image at around f/44 but that was only one at once-a-year epic seeing. It was big, sharp, but did not contain as much details as images of similar size from bigger scopes.

f/34 C8 (estimated by FireCapture)

f/44 C8 (estimated by FireCapture)

[Cornelius Varley]

Riklaunim,

Thanks very much for the detailed reply. That's just the sort of info I am after (but which seems hard to find)! Mmmm, with relation to resolving power I have noticed that my ICX098BL (5.6 microns) produces great images at f/23.6 (i.e. my f/11.8 MAK with 2x Barlow) but when I push it to f/29.5 (with 2.5x Powermate) there appears little (if no) improvement on lunar resolution.

In your experience how rigid is the theoretical resolving limit? For example, could you ever imagine good results with 3.75 microns at f/23.6 or would that just not work? If I stick with my MAK (and for the sake of portability it's currently a preferred option) then it seems that it terms of resolution there would be little to choose between ICX098 at f/20, ICX204 at f/16 and ICX445 at f/14 or am I missing something? (I realise in terms of lunar real estate there will be a noticeable increase in coverage).

Thanks again,

Nick

I found this website a little while back whilst trying to find the best f number to shoot at Astrophoto.tips. About halfway down the page is a section entitled "What digital sensor for your telescope". There is a formula in that section which allows you to calculate the arc-seconds/pixel ratio for astrophotography.

Peter

[Leedsgreen]

Riklaunim and Peter,

Thanks for further info. The Astrophoto formula is really useful (and will help in my decision)!

Nick

PS Lovely Jupiter images BTW! I love the smoothness of the f/44 version.

[Leedsgreen]

Pablito,

I notice the Chameleon (mono) offers Y8 and Y16 formats, which seems to effect fps. Which format do you use and is there much visible difference?

Thanks,

Nick

[Pablito]

In Y16 speed is getting down by half. So I use Y8 mode. Due to my problems with torn images I have to use slower mode.

I found today that my problem with torn frames (verticaly and horizontaly) comes from my laptop. Hardware or settings. Still don't know what...

[Leedsgreen]

Pablito,

Please could you tell me what laptop you have? I've read the Dell Latitudes can cause this problem but things can be tweaked (as in the PGR help sheets).

Thanks,

Nick

[Pablito]

I have HP dv5 series (2.0GHz 3GB RAM).

But I think I solved the problem. How? I don;t really know I downloaded newest software and firmware from PGR site and some other soft reccomended if the problems occure.

I took two sessions, yesterday and today. Yesterday I used FlyCap (provided by PGR) and in the fastest speed still had torn images. I was able to take 14 fps without any problems.

Today I used FireCapture and have no problems AT ALL. I managed to take even18fps.

FireCapture enables to give fps rate but forgot to test it... In FlyCapture you can also give fps but the camera does what it want without control, looks like it is max. fps rate and depends on the hardware ability at the moment.

[Leedsgreen]

Thanks for the further information Pablito, much appreciated. Looks like the Chameleon is behaving itself now!

Nick

[cameraman4ever]

Hi Everyone.

For sure the 445 is better than the 204 if you want NIR response. In fact, there is a rumor that the 204 will be discontinued in the future.

As for the problem with the ICX618 you sometimes see (noise in the image) it is because the cameras overclock the sensor to 120fps, bringing the dark noise much higher. The cameras were designed for industrial use.

The ICX445 binned 2x2 with a long exposure mode can do wonders for low cost sensitivity. I would suggest looking at the AVT Manta G-125, you will get excellent results and it will run much cooler than the POE camera you show below.

It also allows for longer exposure times, I really enjoy this camera.

I will soon be testing the AVT GX1920 in my set up as the ICX674 will allow for HD photos, NIR response and much more sensitivity.