[dmach47]

I work with three currently on-orbit CCD sensors looking at lightning (GLM-16, GLM-17, and ISS-LIS). Now I own my very own CCD sensor system, the K-1ii. I'm very curious about how the CCD sensor on the K-1ii works. Are there any articles on the trade-offs and such made to get the sensor in the camera and into production? I'm most curious about how they change the sensitivity of the CCD. On our instruments, we more or less made them as sensitive as possible with enough dynamic range to detect lightning from geostationary orbit in broad daylight. So obviously, we made a very different set of compromises than Pentax/Ricoh did. And yes, I'm looking for the gory technical details. I also understand the concept of "proprietary" as we deal with that all the time with our sensors (and something even more important, ITAR).
So does anyone have links to such articles about the K-1ii sensor and system?
Thanks!

[Bertrand3000]

I wish I could give you technical details, but the K1 is supposed to get a CMOS sensor, not a CCD sensor.

[jatrax]

does anyone have links to such articles about the K-1ii sensor and system

Sensor in K-1 (and indeed most modern DSLRs) is CMOS, not CCD. Also it is a Sony sensor so perhaps a search for Sony technical documentation would be helpful.

[SelrahCharleS]

I believe it is the same Sony made sensor from the Nikon D800, so you might be able to use that information to help find some additional details. As others have mentioned it is CMOS and not CCD.

[UncleVanya]

As others have said the sensor in the k1 and k1 mk II is CMOS not CCD. The 645D is the most modern Pentax I can think of that has a CCD sensor.

As for sensitivity, if you ignore the use of the accelerator the sensor is isoinvariant which means that undercoating and lifting the levels later gives roughly the same result as raising the iso.

The accelerator changes that - but that’s not terribly important for this discussion.

[c.a.m]

And yes, I'm looking for the gory technical details.

So does anyone have links to such articles about the K-1ii sensor and system?

Interesting question!

I also have a technical/scientific background (including 16 years in space/satellite R&D), and understand what you're looking for. I don't recall having seen any detailed technical descriptions, documents, or white papers, from either the OEM sensor mfr (Sony) or Ricoh Imaging. As you probably know, detailed technical designs or implementations are usually guarded. I haven't seen even a meaningful simplified block diagram.

You might have already seen the sparse info available at Sony Semiconductor's site: Products?Image Sensor?Consumer Camera?products?Sony Semiconductor Solutions Group

Prior to the launch (pun intended) of the Pentax K-3 Mark III, Ricoh Imaging presented several 'product stories' that explored some of the salient design aspects of the camera. Not much on the sensor, though.
Product stories of new PENTAX K-3 Mark III, APS-C flagship model | RICOH IMAGING

K-3 Mark III Challengers looks at several key design and integration issues: Challengers | PENTAX K-3 Mark III | RICOH IMAGING

And there is the K-1 'Challengers' series: Challengers | PENTAX K-1 Special site | RICOH IMAGING

- Craig

PS. I looked at the GLM website. Great stuff! The GOES series has been very successful over many years, carrying important primary and secondary payloads.

[dmach47]

I wish I could give you technical details, but the K1 is supposed to get a CMOS sensor, not a CCD sensor.

Well, poop . So it's like the OTHER sensor I am working with, the EUMETSAT Lightning Imager...

---------- Post added 06-15-21 at 06:28 PM ----------

Thank you everyone for the links and the information. I'll be checking them out.

[pentaxus]

I like CMOS sensors better anyway.

[ProfessorBuzz]

Dmach47, sorry to say you're mistaken on both counts.
You might enjoy reading this article on the CMOS sensors supplied by Teledyne e2v for the Eurmetsat::
Teledyne e2v delivers first flight module for EUMETSAT?s Meteosat Third Generation (MTG) | Teledyne Imaging

and here's an interesting article touching on some of the differences between CMOS Active Pixel Sensors and Charge Coupled Devices, in the context of scientific imaging (astronomy):
CCD versus CMOS: Which is Better? - Astronomy & Scientific Imaging Solutions

Both CCD and CMOS APS have photo-sensitive pixels; the main difference is how to convert the charge or voltage to a digital number, and how fast you can do it.
CCD's usually have the charge clocked across the row and out a corner of the chip through an amplifier, and then an analog to digital converter (ADC) chip on the circuit board converts it to a number. There are usually 1-4 readout channels and ADCs, allowing up to 4 quadrants of the chip to be read out at a time.

CMOS APS devices have the voltage passed to a column ampliifier to an on-chip ADC; there are many many parallel ADCs allowing simultaneous readouts; this makes the APS device much faster to read out the entire array, as compared to a CCD.
Both devices may have a micro-lens array which concentrates light in about a 30 degree cone onto the pixel. On APS devices, this can be really critical, as the pixel is small and surrounded by 4-5 transistors, and the pixel doesn't completely fill the square with it's neighbour. (Fill Factor).
If it's a Bayer colour sensor or a Sparse Colour, there is a micro-filter array of RGGB or LRGB fitlers above the pixels, sometimes incorporated with the microlenses.
Both devices can be front-illuminated (FSI), meaning that the light may have to pass through some semi-transparent electrodes that make up the circuitry, or back-side iluminated (BSI), where the rear of the chip gets polished down thin enough to allow light to hit the pixels without going through the logic. BSI costs more to make, as it's harder to get good yields.
Sensitivity is a function of the wavelengths of light that the semiconductor responds to; there is an Absolute Quantum Efficiency curve used to figure out how sensitive the chip is to light of different colour, eg 400nm to just under 1000nm for Sony chips.
The Teledyne e2v parts are even more sensitive (and very expensive). If there are colour filters and microlenses, that impacts it. Then the chip may have an amplifier gain control - the readout amplifier (CCD) or AP transistors and column amps (CMOS APS) may boost the signal level a bit. The Sony and Gpixel CMOS APS devices have a dual-gain mode for High-Dynamic Range (HDR) operation.
So there's a lot that goes into it. (I haven't mentioned calibration!)

[UncleVanya]

and here's an interesting article touching on some of the differences between CMOS Active Pixel Sensors and Charge Coupled Devices, in the context of scientific imaging (astronomy):
CCD versus CMOS: Which is Better? - Astronomy & Scientific Imaging Solutions

Thank you! I was not the originator of this thread but your post really tightened up my understanding of the differences.

[DWS1]

This discussion has gone way over my head.

[dmach47]

I like CMOS sensors better anyway.

So do the Europeans.... And I've been told our next generation GLM will be CMOS.

---------- Post added 06-16-21 at 02:58 PM ----------

This discussion has gone way over my head.

Don't feel bad. I'm on more of the "results" side, so I don't build sensors. I just look at their output and write papers on what I see. In getting our GLM built and launched, however, I did show up to a lot of VERY technical meetings where i actually learned stuff (much of which I didn't understand and the bits I did understand I can't write about due to "proprietary" and "ITAR" stuff). So I was interested in how my new K-1ii sensor differs from the one I've been working with for more years than I'd care to think about. Obviously, at the most basic level, both sensors are doing the same thing: measuring light. The way they go about it, and the specific goals are very different. So I was looking for some mid-level information about the K-1ii sensor. There are enough links and such in this thread to at least for now, satisfy my curiosity.

[ProfessorBuzz]

This discussion has gone way over my head.

Sorry about that... I've been dealing with detectors since the mid-90's off and on.
The important thing is that both types of chips are a lot alike in how they collect light, but they differ in how (and how fast) it gets converted to a digital number, and how noisy the chips are (which looks a bit like film grain).

[dmach47]

You might enjoy reading this article on the CMOS sensors supplied by Teledyne e2v for the Eurmetsat...

"My" sensor on the MTG is the other one (LI, or Lightning Imager).

[ProfessorBuzz]

"My" sensor on the MTG is the other one (LI, or Lightning Imager).

Cool... that's really awesome.