Something shooting the sun? NASA removes images

Rob48

You are reading and listening to people who were not actually there back in the day...they were somewhere else..."on the outside looking in"; and you think they are reporting everything correctly...you are truly funny!

The SOHO team were certainly there back in the day, when they wrote that handbook. And James Janesick seems to have been there too, based on that Sky & Telescope piece from 1987, written by one James Janesick of Caltech and Morley Blouke of, erm, Tektronix.

edit on 17-4-2014 by Rob48 because: (no reason given)

Amazing how he absolutely refuses to address this and admit he was wrong. It really exposes the lie that he would be "the first" to admit it if he was wrong about anything.

tanka418

Rob48

In basic terms, energetic particles striking the silicon atoms in the CCD and "knocking the electrons out of orbit", liberating them into the pixel sites along the track of the particle. Electrons, of course, being what a CCD is recording. (Yes I know electrons aren't really "orbiting", but if an analogy is good enough for my chemistry professors it is good enough for me.)

(By the way, the CCD pictured is not the one used in SOHO. The ones used in the LASCO imagers are less than half that size, being 1024 x 1024 pixels with 21μm pixels.)

Thank you...21um technology...

Now, can you find the penetration depth of your typical atomic nuclei?

Then explain just how there are "streaks" longer than 2 pixels.

Homework time, is it?

OK, well, the penetration depth will depend on the energy of the particle, of course.

I've found a cite giving the mean rate of energy loss for cosmic ray particles in silicon as 387 eV per μm. [1]

So the energy loss across one pixel should be of the order of 387 eV x 21 = 8.1 keV.

Theoretically, assuming the particle came in completely parallel with the plane of the CCD, it seems to me that a particle with an energy above about 1024 x 8.1 keV = 8.3 MeV could make it across the entire CCD. Given that SOHO regularly encounters particles with energies well above 10 MeV, I'd say there should be plenty of streaks longer than two pixels.


[1] Radiation events in astronomical CCD images
Alan R. Smith ; Richard J. McDonald ; Donna C. Hurley ; Steven E. Holland ; Donald E. Groom ; William E. Brown ; David K. Gilmore ; Richard J. Stover ; Mingzhi Wei
Proc. SPIE 4669, Sensors and Camera Systems for Scientific, Industrial, and Digital Photography Applications III, 172 (April 26, 2002); doi:10.1117/12.463423

tanka418
Here is where your misunderstanding is coming from: Tektronix has never been a "chip maker", they are however, one of the best / most famous instrument makers n the industry. They have traditionally designed and manufactured instruments across a wide variety of industry and discipline. Thus they would be quite likely, and in fact seem to have designed and produced imaging systems using CCD devices. However, like most companies like that; all they really do is integrate existing technology into a useful system. That's what I do, along with virtually every other technology company (I own my own software house).

So...while SOHO and probably many other imaging systems may be based on an imaging system produced by Tek. Tek still did not product the actual CCD device, even IF it was originally a custom build for Tek.

The largest charged coupled-device currently available is made by Tektronix, Inc.

www.phy.duke.edu...

Tek have had problems with the flatness of their and devices. An approximately bow has been observed in their devices, this bow originating in the wafer itself during processing. The wafers take on the shape of part of a spherical surface and this results in a centre-to-edge change in flatness of, at worse, . This curvature should be easily seen in all our instruments, especially in optical systems with fast beams.

www.ing.iac.es...

How much clearer does it need to get for you? You really don't get it. You were wrong. You can't even understand something so simple as the fact that Tek really did make CCDs, so you're nowhere near the level of being ready to understand cosmic ray backscatter radiation, which itself makes up essentially the entirety of cosmic ray radiation received by CCDs on the ground on earth. And yes, the result tracks can be many dozens of pixels long, not just 2 or 3. And yes, by going underground and shielding the CCD from cosmic radiation (including secondary backscatter) you can effectively eliminate these events.
www.astronomy.ohio-state.edu...
QED

Rob48

tanka418

Rob48

In basic terms, energetic particles striking the silicon atoms in the CCD and "knocking the electrons out of orbit", liberating them into the pixel sites along the track of the particle. Electrons, of course, being what a CCD is recording. (Yes I know electrons aren't really "orbiting", but if an analogy is good enough for my chemistry professors it is good enough for me.)

(By the way, the CCD pictured is not the one used in SOHO. The ones used in the LASCO imagers are less than half that size, being 1024 x 1024 pixels with 21μm pixels.)

Thank you...21um technology...

Now, can you find the penetration depth of your typical atomic nuclei?

Then explain just how there are "streaks" longer than 2 pixels.

Homework time, is it?

OK, well, the penetration depth will depend on the energy of the particle, of course.

I've found a cite giving the mean rate of energy loss for cosmic ray particles in silicon as 387 eV per μm. [1]

So the energy loss across one pixel should be of the order of 387 eV x 21 = 8.1 keV.

Theoretically, assuming the particle came in completely parallel with the plane of the CCD, it seems to me that a particle with an energy above about 1024 x 8.1 keV = 8.3 MeV could make it across the entire CCD. Given that SOHO regularly encounters particles with energies well above 10 MeV, I'd say there should be plenty of streaks longer than two pixels.


[1] Radiation events in astronomical CCD images
Alan R. Smith ; Richard J. McDonald ; Donna C. Hurley ; Steven E. Holland ; Donald E. Groom ; William E. Brown ; David K. Gilmore ; Richard J. Stover ; Mingzhi Wei
Proc. SPIE 4669, Sensors and Camera Systems for Scientific, Industrial, and Digital Photography Applications III, 172 (April 26, 2002); doi:10.1117/12.463423

Very nice citation. I think that settles it, both from the theoretical side, and from the observational side. I seriously doubt he'll ever admit he was wrong about it though. It's amusing that he thinks he'd "be the first" to admit he was wrong.

originally posted by: ngchunter

Rob48
Very nice citation. I think that settles it, both from the theoretical side, and from the observational side. I seriously doubt he'll ever admit he was wrong about it though. It's amusing that he thinks he'd "be the first" to admit he was wrong.

Yes, indeed it is a very nice citation.

Too bad I had to lead y'all by the hand so that you could find the proof that those streaks could be cosmic rays.

Y'all went round and round through so many misconceptions, misunderstandings...hell I bet y'all still don't quite understand what happened...that's okay...I do.

So, now that those streaks "might possibly" be cosmic rays, all you need to do is get past overall systems reliability, and of course probability.

originally posted by: tanka418
Yes, indeed it is a very nice citation.

Too bad I had to lead y'all by the hand so that you could find the proof that those streaks could be cosmic rays.

Holy crap, all that and this is your response? So now you're going to pretend you knew all this already and you just spewed a bunch of false claims to "lead us around." All so that what, we'd come back to the same point we were at pages ago? Yeah, you must think we're idiots if you expect us to believe that. I want you to say the following words, "I was wrong, I'm sorry." Is that really so impossible for you? I guess so. Well gentlemen, I think this is as close as we're ever going to get to an admission from tanka that he was wrong and we were right. Adios, have yourselves a good easter.

originally posted by: Rob48

Homework time, is it?

OK, well, the penetration depth will depend on the energy of the particle, of course.

I've found a cite giving the mean rate of energy loss for cosmic ray particles in silicon as 387 eV per μm. [1]

So the energy loss across one pixel should be of the order of 387 eV x 21 = 8.1 keV.

Theoretically, assuming the particle came in completely parallel with the plane of the CCD, it seems to me that a particle with an energy above about 1024 x 8.1 keV = 8.3 MeV could make it across the entire CCD. Given that SOHO regularly encounters particles with energies well above 10 MeV, I'd say there should be plenty of streaks longer than two pixels.

Yes well somebody needed to help yall out with YOUR homework. I already know this...

Anyway;

That wasn't so hard now was it? Problem is that this old engineer, who you doubt so much, had to show you how to demonstrate that the streaks could actually be cased by cosmic rays. Of course, that does not mean that they are caused by cosmic rays, only that it is in the realm of probability. But, don't worry...probability does suggest that they indeed are; cosmic rays.

And as for the rest of it; gentlemen, I have been in the electronics industry for something around 40 years, you have not. So, if you please, don't try to tell me how things are / were done.

originally posted by: ngchunter

Holy crap, all that and this is your response? So now you're going to pretend you knew all this already and you just spewed a bunch of false claims to "lead us around." All so that what, we'd come back to the same point we were at pages ago? Yeah, you must think we're idiots if you expect us to believe that. I want you to say the following words, "I was wrong, I'm sorry." Is that really so impossible for you? I guess so. Well gentlemen, I think this is as close as we're ever going to get to an admission from tanka that he was wrong and we were right. Adios, have yourselves a good easter.

No pretense, no false claims. All that run around is a direct result of YOUR misunderstanding of the nature of the electronics world, and apparently everything I was trying to say. Sorry man...but that is your bad.

a reply to: tanka418

Well, I wasn't going to bother to dignify this ridiculousness with a reply, but just on the off chance anyone doesn't recognise that this "old engineer" (who I suspect has never been within a hundred yards of a naked silicon chip) is an old troll, I'll leave this here.

Too bad I had to lead y'all by the hand so that you could find the proof that those streaks could be cosmic rays.

Sometimes, there just aren't enough faces and enough palms.

I have a modern day example of your "Tektronix" CCD...

If you open your PC and look at the video card...especially if it is an older Nvidia. On the GPU chips it will clearly say "Nvidia". However, Nvidia did not manufacture those parts. They are a custom job by either a well known "chip maker", or one of several custom chip makers.

Now you can argue all you like, but especially "back in the day" setting up a chip manufacturing facility was quite expensive, so...many companies, even large ones, contracted someone else to do the actual manufacture... even the design aspect. And then of course the part would have their mane on it.

Another example would be some early microcomputers; Many companies had their micro and mini computer systems...their name was right on the front...yet inside was someone else's hardware. Digital Equipment Corp. for a while produced their LSI-11 chip set...at first not manufactured by DEC.

I hope you are beginning to see where your confusion stems from on these issues. In addition to the fundamental misconception that a system containing a Charge Coupled Device is a "CCD".

Y'all have taken us around the mulberry bush a number of times simply because you don't grasp some rather fundamental concepts that I always thought of as common.

a reply to: Rob48

Yes, and after all that...you are like the math student who always has the correct answer; and fails the course.

Have you figured out "why" not all that are cosmic rays? Do you know why many streaks, especially long ones probably are not cosmic rays?

By the way; I never said that the streak in the present image was not a cosmic ray; I only remarked about the report and the synchronicity of the events.

You proceeded to misconstrue what I said, and did a very good job too!

originally posted by: Rob48
a reply to: tanka418

Well, I wasn't going to bother to dignify this ridiculousness with a reply, but just on the off chance anyone doesn't recognise that this "old engineer" (who I suspect has never been within a hundred yards of a naked silicon chip) is an old troll, I'll leave this here.

Too bad I had to lead y'all by the hand so that you could find the proof that those streaks could be cosmic rays.

Sometimes, there just aren't enough faces and enough palms.

I can't take it any more, I seriously can't. This [snipped] needs to clam it or I'm going to say something that will get me banned, which is probably precisely what he wants.

a reply to: tanka418

Now you can argue all you like, but especially "back in the day" setting up a chip manufacturing facility was quite expensive, so...many companies, even large ones, contracted someone else to do the actual manufacture... even the design aspect. And then of course the part would have their mane on it.

Yes, many did. I am not disputing that. We all know what OEM manufacturing is, thank you, so there's no need for the OEM for Dummies lesson.

However, Tektronix didn't. It built its own CCDs, from scratch, as has been conclusively demonstrated in this thread. It was, in fact, the first company to do so at full-wafer scale, anywhere.

BTW I tracked down what happened to the Tektronix CCD spin-off. The source I quoted at first had the name wrong, which led me down a blind alley. It didn't become Silicon Imaging Technologies, it became Scientific Imaging Technologies (SITe for short) and no longer appears to be in business under that name, but it was active well into the 2000s.

Here's a sample spec sheet, because I know you love those. (Edit: and here's another, for the CCD chip itself, in case you get all excitable because the first one is for an assembled CCD imager.) Notice in particular the small print near the end:

With its focus on scientific-grade CCD imaging components and modules, SITe provides standard designs, user defined custom CCDs, and foundry services. SITe’'s engineering and manufacturing team builds custom CCD
imagers for use in the most demanding applications including NASA programs, satellite platforms, and other
research projects. Device formats are available as front illuminated or thinned, back illuminated CCDs.
Innovation, process development, and design experience date back to the founding of the group in 1974.

Foundry services, user-defined CCD manufacture? Sounds awfully like a bona fide manufacturer, doesn't it, and that's because that is precisely what it was.

Here's more evidence for you: www.photonicsonline.com...

In September 1999, the company completed construction of a new 2,600 square foot Class 100 clean room for packaging and testing of imaging sensors. Currently, the 36,000 square foot Tigard manufacturing facility is being expanded to include a Class 10 clean room environment capable of thinning four- and six-inch and larger wafers.

What would a company that outsources all its manufacturing be wanting with all those clean room manufacturing spaces?

Actually, it's more likely that other CCD manufacturers were outsourcing their manufacturing to Tektronix/SITe!

SITe focuses on the design and manufacture of scientific grade CCD imaging components and modules, providing both standard designs and custom CCDs for user-defined performance specifications. SITe also offers CCD fabrication services for customers with specialized imager designs.

So, if the lesson was meant to shift from cosmic rays to CCD manufacture, it has utterly failed on that level, too.

I think that warrants a final:

originally posted by: Rob48

tanka418


Thank you...21um technology...

Now, can you find the penetration depth of your typical atomic nuclei?

Then explain just how there are "streaks" longer than 2 pixels.

Homework time, is it?

OK, well, the penetration depth will depend on the energy of the particle, of course.

I've found a cite giving the mean rate of energy loss for cosmic ray particles in silicon as 387 eV per μm. [1]

So the energy loss across one pixel should be of the order of 387 eV x 21 = 8.1 keV.

Theoretically, assuming the particle came in completely parallel with the plane of the CCD, it seems to me that a particle with an energy above about 1024 x 8.1 keV = 8.3 MeV could make it across the entire CCD. Given that SOHO regularly encounters particles with energies well above 10 MeV, I'd say there should be plenty of streaks longer than two pixels.


[1] Radiation events in astronomical CCD images
Alan R. Smith ; Richard J. McDonald ; Donna C. Hurley ; Steven E. Holland ; Donald E. Groom ; William E. Brown ; David K. Gilmore ; Richard J. Stover ; Mingzhi Wei
Proc. SPIE 4669, Sensors and Camera Systems for Scientific, Industrial, and Digital Photography Applications III, 172 (April 26, 2002); doi:10.1117/12.463423

And the [snipped] goes right back to claiming the "long ones aren't cosmic rays" so I will again quote this awesome post showing that indeed it would not be abnormal for cosmic rays to even go through the ENTIRE image. I guess he realized he wasn't happy with coming as close as he did to admitting defeat and admitting he was wrong. At first he thought he would end by pretending that he never actually believed his original claim and it was all a big "lesson" to "teach us" so that he could save face and retain his superior attitude despite everything he said being shown to be wrong. Now he realizes how much he exposed himself, so he goes right back to his old claims and absolutely nothing has been accomplished. Unbelievable. This thread just gets worse and worse. Oh well, it's worth quoting this awesome post one more time.

originally posted by: ngchunter
And the troll goes right back to claiming the "long ones aren't cosmic rays" so I will again quote this awesome post showing that indeed it would not be abnormal for cosmic rays to even go through the ENTIRE image.

Your reading and comprehension skills really need developed.

I did not say that! Yet for some reason you think I did...

Again...not all the streaks are cosmic rays, especially the long ones. Yet you have no idea what I'm talking about..to you, it doesn't matter "what" they are; they're "Cosmic Rays"...

Some of what you call cosmic rays are in fact system noise. Some of the Cosmic rays seen in NASA images are the result of image processing...that was why I asked if you had viewed raw data...apparently not. If you had you should know about the amplification process NASA uses and the fact that it can turn a black cell grey (change a pixel value from 0 (zero) to some value greater than 0. Then of course once you know that they do these things, it is important to understand "why".

And then there is still the continued operation of the overall system...but there again you will fail utterly to explain...and again from lack of understanding...

a reply to: Rob48

Okay...I see that...built all that in 1999...some time after SOHO was launched. And, a bit after the time frame we were discussing as well.

originally posted by: tanka418

Some of what you call cosmic rays are in fact system noise. Some of the Cosmic rays seen in NASA images are the result of image processing...that was why I asked if you had viewed raw data...apparently not. If you had you should know about the amplification process NASA uses and the fact that it can turn a black cell grey (change a pixel value from 0 (zero) to some value greater than 0. Then of course once you know that they do these things, it is important to understand "why".

Could you explain how random system noise would produce sharply defined, bright white streaks on the image? Way back, I quoted the noise levels of the SOHO CCD. About 25 electrons per pixel in total, compared to a full well capacity of 150,000 - 250,000 electrons.

originally posted by: tanka418
a reply to: Rob48

Okay...I see that...built all that in 1999...some time after SOHO was launched. And, a bit after the time frame we were discussing as well.

Your reading comprehension really is very poor indeed, isn't it?

In September 1999, the company completed construction of a new 2,600 square foot Class 100 clean room for packaging and testing of imaging sensors.

That's packaging and testing, not manufacturing. The manufacturing facility already existed. The new facility was 2,600 sq ft; the existing one was 36,000 sq ft. One of these things is not like the other.

Currently, [i.e. in 2000] the 36,000 square foot Tigard manufacturing facility is being expanded to include a Class 10 clean room environment capable of thinning four- and six-inch and larger wafers. The project is due for completion by August 2000.

The expansion was due for completion in 2000. The facility already existed. You cannot expand something which does not already exist.

Closed for review.