ElphelWiki - User contributions [en]

Talk:Diagnostic & repair

2006-01-01T09:09:07Z

Bernardc: HACKING File

Ilya, it is not exactly "User information" - users will can get cameras with capacitors that need to be re-soldered :-)

Diagnosis of the bad boards is helpful only for the production - to find out what was wrong in the process.

So it is good to have a list of encountered failures (we do not have to keep it secret)- some can show up later and that needs improving production testing. But it should clearly be separate from the user docs.
--[[User:Andrey.filippov|Andrey.filippov]] 01:50, 30 October 2005 (CST)

----
OK, I understand it now. I will delete that chapter from Cameras Manual.

--[[User:Ilya|Ilya]] 08:45, 30 October 2005 (CST)

== HACKING File ==

Is there a file that gives a reasonably thorough introduction to the software? I understand that for GPL systems this is normally in a file called HACKING. It should include information such as a list of the major source files, functions, block diagram indicating the basic program flow, how data is passed from web client to Axis to FPGA, and so on. Since it is open hardware, a tutorial on the architecture and key circuit design features would be useful too. It would be very helpful in reducing the potential barrier (i.e. the cost of entry) to making improvements or customization of the software or hardware.

Establishing Basic Exposure with 323

2005-12-19T06:22:55Z

Bernardc:

It is useful to establish a correct exposure when unlimited light is available. In digital cameras with viewfinders (i.e. those that replaced film cameras), this exposure is sometimes referred to as the base ISO setting. Since Kodak only publishes the CCD's sensitivity in terms of QE, the following is an empirical method of arriving at the base ISO.

According to Kodak's specification for the KAI-11000, the vertical transfer collumn, or VCCD, has a charge capacity of 90 ke. When Vsub is set to its minimum value of +8 volts, Kodak states that there is no blooming protection because the photodiode capacity is higher than this. Thus, the threshold of blooming indicates an exposure of 90 ke. Blooming is caused by charge spilling into adjoining pixels, and appears as elongated vertical smears starting in bright areas and spilling into dark areas. Here is an example of a GreytagMacbeth chart with bloom:
[[Image:121805-03 Bloom.jpg]]
Notice the brighter squares have spurious extensions out of their squares, down into the black divider.

The presence of blooming is independent of the parameters on the main CCD acquisition parameters page because it occurs before charge-to-voltage conversion, and before any amplification. (Plus, the amplifier gains in the system are constant. The gain settings are only multiplicands.) Thus, it cannot be confused with other types of saturation.

So how is Vsub set? The number in the Vsub box has a relationship with Vsub as shown:
[[Image:Vsub graph.jpg]]

The equation (at least on the camera that I measured) is Vsub*200 + 85 = Parameter value. So +8v is represented as 1672.

Kodak also states that they label their CCD containers with the Vsub that gives a photodiode capacity of 60 ke, with at least 100x antiblooming protection. This is only about half a stop down from the full well capacity, and 100x is about seven stops brighter than the scene highlights. Thus, as its special measurement and designation suggests, this is a good compromise between well capacity and blooming protection, and thus this is the exposure value to aim for, at least as a good starting point.

On my cameras, the label was apparently transfered to the inside of the camera housing. It is marked "VAB = 11.2 V". It seems likely that this is the Vsub value that they are talking about. 11.2 volts corresponds to 2307.

So here is the method: Set Vsub to 8 volts, adjust exposure in any of the ordinary ways, and find the threshold of blooming. Record that light reading with a separate meter. Then set Vsub to the suggested value, 11.2 in my case, and reduce exposure by about a half to two-thirds stops. Then set the gain and zero values so that the bright-limit highlights saturate at 255,255,255 or close to it, and the dark-limit shadows read 0,0,0. The gamma value also needs to be set so that intermediate values are rendered correctly; so far I've found 0.45 is optimum.

Finer color adjustments can be made with a custom color profile.

Using this method, I've arrived at ISO 40 as the base value. This seems reasonable compared to other large CCD's which are typically ISO 50 to 100. Being an interline transfer CCD, the KAI-11000 fill factor is significantly lower than full frame chips because half the chip is covered with the light-shielded VCCD.

File:Vsub graph.jpg

2005-12-19T05:41:47Z

Bernardc:

File:121805-03 Bloom.jpg

2005-12-19T05:34:22Z

Bernardc: Bloom example.

Bloom example.

Flash Illumination with the 323

2005-12-18T16:28:04Z

Bernardc: /* Using the 323 with Flash Illumination -- Suggested settings */

== Using the 323 with Flash Illumination -- Suggested settings ==

I photographed a GretagMacbeth Color Rendition Chart with a 323 camera equipped with a Kodak KAI-11000CM Grade 2 color microlens CCD. The lens was a 60mm MicroNikkor, set at its wide open aperture of f/2.8. Illumination was from a single Elinchrom 600S set to about 25 watt-seconds, equipped with a 14x20" Chimira softbox with both diffusing panels in place. With the target in a vertcal plane and the camera aligned on the center normal, the bottom of the softbox was positioned about 6" above the back of the 323, angled down at the target. The CCD to target distance was about 0.8m. The soft box to target was about 0.95m along the center to center beeline.

This image is reasonably close numerically (+- 5% typical) to the RGB specifications for the target, and subjectively looks very good.
[[Image:121705-13.jpg]]

The chart is rotated to avoid the huge horizontal QE falloff of the CCD.

With a Minolta IVF light meter in non-corded mode, the (incident) illumination reads f/2.8 at ISO 32.

Screenshot of the main parameters page:
[[Image:121705-13 Screenshot.jpg]]

This camera uses a test version of software that allows capture to be triggered by pulling CTS1 low on the master compressor. This pin is available on the power supply header connector. The flash is triggered via Pocket Wizard, whose Port 1 input is paralleled to CTS1. Both inputs float at 3.3v. The exposure period must be set long at 2 mS because the Pocket Wizard radio link takes about 1 mS. (When both camera and flash are radio triggered, a 1 mS exposure syncs reliably as long as the CCD analog supplies are already active, and no sync delay is needed--astounding.)

From this example, it appears that if you can nail the exposure, you'll get good photos. Until higher resolution raw data becomes available (preferably in [http://www.adobe.com/products/dng/main.html Adobe DNG format]), this is a good workaround.

...by BernardC

Flash Illumination with the 323

2005-12-18T02:53:42Z

Bernardc:

== Using the 323 with Flash Illumination -- Suggested settings ==

I photographed a GretagMacbeth Color Rendition Chart with a 323 camera equipped with a Kodak KAI-11000CM Grade 2 color microlens CCD. The lens was a 60mm MicroNikkor, set at its wide open aperture of f/2.8. Illumination was from a single Elinchrom 600S set to about 25 watt-seconds, equipped with a 14x20" Chimira softbox with both diffusing panels in place. With the target in a vertcal plane and the camera aligned on the center normal, the bottom of the softbox was positioned about 6" above the back of the 323, angled down at the target. The CCD to target distance was about 0.8m. The soft box to target was about 0.95m along the center to center beeline.

This image is reasonably close numerically (+- 5% typical) to the RGB specifications for the target, and subjectively looks very good.
[[Image:121705-13.jpg]]

The chart is rotated to avoid the huge horizontal QE falloff of the CCD.

With a Minolta IVF light meter in non-corded mode, the (incident) illumination reads f/2.8 at ISO 32.

Screenshot of the main parameters page:
[[Image:121705-13 Screenshot.jpg]]

This camera uses a test version of software that allows capture to be triggered by pulling CTS1 low on the master compressor. This pin is available on the power supply header connector. The flash is triggered via Pocket Wizard, whose Port 1 input is paralleled to CTS1. Both inputs float at 3.3v. The exposure period must be set long at 2 mS because the Pocket Wizard radio link takes about 1 mS. (When both camera and flash are radio triggered, a 1 mS exposure sychs reliably as long as the analog supplies are active, and no sych delay is needed--astounding.)

From this example, it appears that if you can nail the exposure, you'll get good photos. Until higher resolution raw data becomes available (preferably in [http://www.adobe.com/products/dng/main.html Adobe DNG format]), this is a good workaround.

...by BernardC

Talk:Using the cameras

2005-12-17T22:59:13Z

Bernardc:

== Using the 323 with Flash Illumination -- Suggested settings ==

I photographed a GretagMacbeth Color Rendition Chart with a 323 camera equipped with a Grade 2 color microlens CCD. The lens was a 60mm MicroNikkor, set at wide open aperture. Illumination was from a single Elinchrom 600S set to about 25 watt seconds, equipped with a 14x20" Chimira softbox with both diffusing panels in place. With the target in a vertcal plane and the camera aligned on the center normal, the bottom of the softbox was positioned about 6" above the back of the 323, angled down at the target. The CCD to target distance was about 0.8m. The soft box to target was about 0.95m along the center to center beeline.

The following image is reasonably numerically (+- 5%) close to the RGB specifications for the target, and subjectively looks very good.
[[Image:121705-13.jpg]]

The chart is rotated to avoid the huge horizontal QE falloff of the CCD.

With a Minolta IVF light meter in non-corded mode, the (incident) illumination reads f/2.8 at ISO 32.

Screenshot of the main parameters page:
[[Image:121705-13 Screenshot.jpg]]

This camera uses a test version of software that allows capture to be triggered by pulling CTS1 low on the master compressor. This pin is available on the power supply header connector. The flash is triggered via Pocket Wizard, whose Port 1 input is paralleled to CTS1. Both inputs float at 3.3v. The exposure period must be set long at 2 ms because the Pocket Wizard radio link takes about 1 mS.

From this example, it appears that if you can nail the exposure, you'll get good photos. Until higher resolution raw data becomes available (preferably in [http://www.adobe.com/products/dng/main.html Adobe DNG format], this is a good workaround.

File:121705-13 Screenshot.jpg

2005-12-17T22:46:34Z

Bernardc:

File:121705-13.jpg

2005-12-17T22:33:47Z

Bernardc: GretagMacbeth color checker with 323

GretagMacbeth color checker with 323