Help needed: OSC camera images totally green (but it's not what you think)

[Lee_P]

I know I know, you hear about an OSC camera producing green images and you figure the user doesn't understand that the Bayer Matrix is weighted to the green. Pretty sure that's not the issue here though, but I'd he happy to be proven wrong. Last night I logged into my ASIAIR to check on the progress of my latest epic multi-week project, and noticed that the image previews looked very green. Never had that before... I rebooted everything and reseated all the cables, but no joy. I fear the camera may need repairs, but maybe someone here has a useful suggestion? Thanks in advance!

Note I'm using an ASI 2600MC Pro camera, ASIAIR Plus, regulated power supply, Lindy USB cables, and Lynx Astro power cables (all from FLO).

Here's an example source file: https://drive.google.com/drive/folders/1g-qh43kK8STia740nMx6pex8Mo2A-1zH?usp=sharing
They were green like this all night.

Screengrabs for quick reference...

Debayered with a linked STF:

 

Debayered with an unlinked STF:

 

After running SCNR to remove green, there's no colour:

Edited August 17 by Lee_P

[Vroobel]

There is nothing about the cables if you can see images. I also use the 2600MC and ASIair Plus (v.1) and never saw green image as preview, they are more or less affected by LP. 

Try to connect the camera to another platform and program like  SharpCap, don't forget about powering the camera. 

Also, as you have the images debayered, try to separate RGB channels, it may give you some clue. 

[vlaiv]

I just looked at the posted file and here are findings.

Red and blue pixels are all "dead".

Only green pixels captured signal, other two are simply all the same value of 501 (which might be offset value). Not a single pixel is different - look at the stats:

first two are red and blue while last one is green (or rather two green sub images - I debayer by splitting image up).

Either something is horribly wrong with the camera - or there might be explanation, but I'm not overly confident that it is the cause (because I might expect some read noise at least, but it depends on how it's implemented).

I've seen ZWO color cameras to have two parameters to help with color balance during the capture - Red weight and blue weight parameter. These should be ideally set to 50 (as they are percentage) - which should mean leave as is and don't either increase or decrease value with respect to green channel.

I'm suspecting that if both of these weights are set to 0 - similar result to above might occur - but only if this per channel gain is applied at the end before adding offset (so that read noise is reduced as well).

Here is screen shot from SharpCap showing image controls for color camera and there are two sliders White Bal (R) and White Bal(B) that go from 0 to 100% (or 99 in this case). Those are the values you should be looking at - but not sure where they might be in your capture software.

HTH

 

 

[tomato]

Not seen this before, hopefully some parameters have got changed in the software but I can't see how. I presume you are using the ASIair software so there shouldn't be any glitches in the camera communication.

I have found this thread on CN, looks like there was a product recall on some ASI2600MC pro cameras which exhibited the same problem, they had to go back to China for repair...😟

Green images from ZWO ASI2600MC Pro - Page 3 - Experienced Deep Sky Imaging - Cloudy Nights

[Lee_P]

Thanks for your help, everyone. I'm pretty sure the camera will need to be shipped back to China for repairs, but I know from experience that I'll have to jump through some troubleshooting hoops before that gets authorised. So, I thought I'd get a pre-emptive start on that tonight. The plot thickens though... the green issue has gone, replaced with pink lines appearing on some (but certainly not all) images.

 

I plugged the camera into my laptop running ASIStudio, but the issue remained (ruling out the ASIAIR Pro as the problem). Next I replaced my FLO-supplied Lindy USB cable for the original ZWO ribbon-style one. So far it's actually all been fine, but I'll leave it running all night and check in the morning.

 

 

 

 

[gorann]

I think it is a good chance the problem was with the USB cable, at least lets hope so!

[Lee_P]

It worked perfectly all night (using the original ZWO USB cable instead of the Lindy one). I really hope it is just the cable, but I'll need more tests before I can be confident in that diagnosis... 

[Vroobel]

I don't understand it. As a person with a decent IT and tech in general background I know that any transmission, including the serial transmission like the USB, cannot loose 'something' by the way and mark the package as properly transmitted. The single file is much bigger than than anything 'allowing' for accidental problem, please take into account that whole images were green. So, who doesn't agree with me? 

 

[vlaiv]

25 minutes ago, Vroobel said:

I don't understand it. As a person with a decent IT and tech in general background I know that any transmission, including the serial transmission like the USB, cannot loose 'something' by the way and mark the package as properly transmitted. The single file is much bigger than than anything 'allowing' for accidental problem, please take into account that whole images were green. So, who doesn't agree with me? 

 

Actually, I'm also perplexed by this.

 

[TiffsAndAstro]

40 minutes ago, Vroobel said:

I don't understand it. As a person with a decent IT and tech in general background I know that any transmission, including the serial transmission like the USB, cannot loose 'something' by the way and mark the package as properly transmitted. The single file is much bigger than than anything 'allowing' for accidental problem, please take into account that whole images were green. So, who doesn't agree with me? 

 

usb doesn't use a packet type system like ethernet as far as i  knoq. also error correction could result in a 'full' image reconstructed from limited info from a broken or partially broken cable?

i could be extremely wrong however.

[vlaiv]

1 minute ago, TiffsAndAstro said:

usb doesn't use a packet type system like ethernet as far as i  knoq. also error correction could result in a 'full' image reconstructed from limited info from a broken or partially broken cable?

i could be extremely wrong however.

I'm not sure what type of error correction is included in USB protocol at physical level, I'm sure there is at least some as speed can decrease from theoretical max under unfavorable conditions, but what I'm perplexed about the most is the fact that exactly every other pixel was transferred as 0 with even parity switching on consecutive rows. That simply can't be due to usb cable in traditional sense.

What can happen - but I'm far from properly explaining how, would be that usb cable and electrical wiring have some sort of effect on processor on camera - making it malfunction in some weird way (say some registers used to store and forward red/blue values could not hold logic level voltage to indicate difference between 1 and 0 due to something related to usb cable and voltage on it or similar).

[TiffsAndAstro]

16 minutes ago, vlaiv said:

I'm not sure what type of error correction is included in USB protocol at physical level, I'm sure there is at least some as speed can decrease from theoretical max under unfavorable conditions, but what I'm perplexed about the most is the fact that exactly every other pixel was transferred as 0 with even parity switching on consecutive rows. That simply can't be due to usb cable in traditional sense.

What can happen - but I'm far from properly explaining how, would be that usb cable and electrical wiring have some sort of effect on processor on camera - making it malfunction in some weird way (say some registers used to store and forward red/blue values could not hold logic level voltage to indicate difference between 1 and 0 due to something related to usb cable and voltage on it or similar).

I wouldn't be shocked if the camera sensor was powered by usb, the tec cooler and fan by the PSU.

Some freyed strands in the usb might result in an intermittent not enough power to the sensor and possibly this end result.

I admit I'm grasping at straws though.

[tomato]

The ASI 2600 camera needs it’s own 12v supply to operate I believe, it’s not just to power the cooler.

[TiffsAndAstro]

13 minutes ago, tomato said:

The ASI 2600 camera needs it’s own 12v supply to operate I believe, it’s not just to power the cooler.

im not 100% certain but this might not be true in fact, just true how it operates? eg my non zwo camera works in sharpcap with no power supply. the zwo version only works in sharpcap when the psu is connected. i think

obviously different camera to the 2600 but i think there's a possibilty its the same. 

[tomato]

I must admit I have never tried to connect my cameras without first providing the 12V supply, but having blown one up when trying to plug in a live lead, that has made me cautious. Most (if not all?) uncooled cameras only have the USB connection, so that must be where they get their power from.

[Elp]

Some cooled cameras can operate without the 12v, some cannot. My 183 does, but the 533 I had didn't.

[Lee_P]

It's a puzzle isn't it. I wonder if there's a loose connection or wobbly circuit board in the camera, and the physical act of swapping the USB cable somehow pushed it back in place. In which case I'd expect it to wobble out again, and the issue to return. Clear skies tomorrow, let's see what results I get...

[TiffsAndAstro]

2 minutes ago, Lee_P said:

It's a puzzle isn't it. I wonder if there's a loose connection or wobbly circuit board in the camera, and the physical act of swapping the USB cable somehow pushed it back in place. In which case I'd expect it to wobble out again, and the issue to return. Clear skies tomorrow, let's see what results I get...

More likely usb cable

[Lee_P]

Just now, TiffsAndAstro said:

More likely usb cable

That's the cheaper fix, so I hope you're right 😂 It's one of these Lindy CROMO cables though, which I'd expect to be robust and reliable. I was also suspicious of the USB ports on my ASIAIR, but the fact that the issue remained when I completely bypassed those (by using ASIStudio on a laptop) knocked that theory on the head. It's a mystery worthy of Sherlock Holmes.

[Elp]

I've got a specific cable I use with my HS setup, for some unknown reason if I plug it into an air with a different cable above (or below it) in the vertical port arrangement, the camera doesn't get detected. Why a cable would do this is beyond me.

[TiffsAndAstro]

16 minutes ago, Lee_P said:

That's the cheaper fix, so I hope you're right 😂 It's one of these Lindy CROMO cables though, which I'd expect to be robust and reliable. I was also suspicious of the USB ports on my ASIAIR, but the fact that the issue remained when I completely bypassed those (by using ASIStudio on a laptop) knocked that theory on the head. It's a mystery worthy of Sherlock Holmes.

It looks a nice cable, but who knows.

If it's any consolation the usb3 and power socket on my sv605cc don't inspire confidence. So much so I now unscrew my lens hood rather than the camera/cables to store it between sessions

[CraigT82]

22 minutes ago, Lee_P said:

That's the cheaper fix, so I hope you're right 😂 It's one of these Lindy CROMO cables though, which I'd expect to be robust and reliable. I was also suspicious of the USB ports on my ASIAIR, but the fact that the issue remained when I completely bypassed those (by using ASIStudio on a laptop) knocked that theory on the head. It's a mystery worthy of Sherlock Holmes.

Have you inspected the Lindy cable? Any damage at all? May have been caught or pinched by something

Edited August 19 by CraigT82

[Lee_P]

5 minutes ago, CraigT82 said:

Have you inspected the Lindy cable? Any damage at all? May have. Been caught or pinched by something

I have, and it looks fine. I'm really hoping it is the cable, but I'm not convinced...

[chrisecurtis]

There are many and strange issues that can be caused by cables.  If putting the "bad" cable back causes the issue to recur, and swapping it out fixes it again, I'd rely on the diagnosis. If it still works with the "bad" cable it's more likely to be a fault that inserting or removing cables might affect (e.g. moving something on the main board)

[gorann]

Yes, do you feel the pressure from us curious SGLers for putting in the old cable so we can finally conclude?😅