The X-Ray image thread by Aurora - various electronics via X-Ray imaging

[Fraser]

More pics of DUO R......

[Fraser]

Visible light pictures of the DUO R camera as a reference for the above X-Ray images......

[KE5FX]

Out of curiosity, Fraser, what do you do for tube warmup on your MX-20?  The manual that came with mine advises an initial 300-second exposure at 20 kVp before running at higher voltages.  But it doesn't say how often this should be done, such as after every X hours of idle time. 

If the idea is to reduce thermal shock to the anode, then I would think it would need to be repeated frequently.  But if the goal is to maintain vacuum integrity then I'd expect it would only be needed at much longer intervals (days or weeks).  Do you know anything about that, beyond what little the manual says?

[BD1QMP]

Hi KE5FX
   
   I am a designer of Medical X-ray system, from my experience, some tube manufactures suggest the user to conduct seasoning (warm up)process when not used for 3 months.  The goal is to maintain vacuum integrity. The warmup shot will be from lowest kV to higher kV step by step.( we use 5kV step).

[KE5FX]

Thanks, that does agree with what I've read elsewhere.  It's strange that the documentation says to do it every time the machine is powered up... or at least, that what it seems to imply.

[mikeselectricstuff]

Hi KE5FX
   
   I am a designer of Medical X-ray system, from my experience, some tube manufactures suggest the user to conduct seasoning (warm up)process when not used for 3 months.  The goal is to maintain vacuum integrity. The warmup shot will be from lowest kV to higher kV step by step.( we use 5kV step).

Presumably this is to avoid flashover, but is this likely to be an issue on a tube like the one in the  MX20 that only runs at 35kv?

[Psi]

I can confirm that Rapiscan x-rays also have this incremental warm-up procedure if the generator has not been run for some time. However it's a manual process you choose to do by running through service manual steps and tweaking the pots on the x-ray driver PCB. The warm-up is not enforced by the system at all.

L3 x-ray machines also have this incremental warm-up, but the control PC does it automatically at startup if it detects the machine has been off for too long. You just have to wait for it to finish before it will let you do anything.

[Fraser]

Just in case anyone is interested here, I am selling a digital X-Ray imaging plate here:

https://www.eevblog.com/forum/buysellwanted/fs-digital-x-ray-imaging-plate-plus-information-on-how-to-connect-to-pc-and-use/

Fraser

[Treehouseman]

I guess I'll try this again, I think I tried putting too many pictures in one post and it just failed out... But I recently picked up a Siemens Heliodent MD (60-70kV 7mA) and some 400 speed cassettes and I've been messing around trying to learn how to get the most out of it until I can afford to get a digital sensor. I'll start with the first pictues and go from there. My initial setup was shooting off an an angle with the sheet facing the object at a distance so I could get a view of it, but now I'm using a mirror shooting from behind with some macro filters. The camera I'm using is a Nikon D7000. I'm going to see about getting the raw files hosted somewhere.

Starting off with the angled shots:

Fluke 179
Hitachi 6GB Microdrive

[Treehouseman]

That filesize limit is killer, I'll just have to get these elsewhere in full resolution.

2.5" HDD 640GB
55-200mm Nikon DX kit lens
Fluke 179
Sony Watchman FD-10A

[Treehouseman]

Continued

Zelda Oracle of Ages for GameBoy Color
EL34 Tube
Sony PS One display (The one at attached to the back of the later model slim ps one)
Benchmarq 64k NVRAM bq4010MA-85

[Treehouseman]

Continued

Philips XP2072/B Photomultiplier tube
GL-2C44 Lighthouse Triode
Cosel ZUS60515 5v in 15v out isolated DC-DC converter
6GB Hitachi Microdrive

I suppose I'll make this the last photos uploaded for now, that's probably enough bumping of an old thread, but I'll probably make one more post with a link to the full size/raw images later.

[Ultrapurple]

At the risk of resurrecting a zombie thread, I thought it may be of interest to see inside a couple of high-capacity microSD cards.

On the left is a 512GB card of indeterminate brand - it may look like a dodgy fake but it genuinely is 512GB. On the right is a SanDisk 1TB microSD - again, genuine.

I took the 'easy' X-Rays first, with the cards flat on the imaging plane, then the much harder transverse view with the cards stood on end or on edge. Due to the way the bond wires are placed, the 512GB card X-Rayed better when stood on its long edge, while the 1TB card was better viewed when standing on its short edge. If you look really carefully at the 1TB transverse view you can just see the gold contacts (at the bottom) as an undulation of about 1 pixel thickness.

Those who know more than me are welcome to correct me, but looking at the side-on view of the 512GB card, it appears to have four layers (dies) stacked on top of each other and wire-bonded at the 'pins' end of the card. The 1TB card, on the other hand, appears to have three separate stacks of dies - two of five and the bottom one of six, I think. It's hard to see clearly on the X-Ray, but do bear in mind that the cards are only about 900µm thick except at the 'bottom', so being able to resolve any detail is good going!

Both of the cards have a much smaller driver die in the bottom right. In the 1TB card it's actually tucked under the overhang of the bottom stack of memory dies!

I think the reason we see the individual dies in the stacks as thin lines separated by voids is that the lines may be the metallisation layer(s) on the dies, which are more opaque to the 35kV X-Rays than bulk silicon. But on the 1TB card the dies have been thinned to such an extent that there's precious little silicon left under the metallisation. Assuming the stack of 16 dies is 800µm high then each die can only be 50µm thick. I think it's amazing that I can (just) resolve details that fine - but not as amazing as the engineering that has gone into making it possible to store a terabyte of data in an overall package of 15 x 11 x 1mm (165 cubic mm, according to Wikipedia). Put is another way, the overall storage density is something like 6GB per cubic millimetre.

Wow.

Please note: I wish to retain the copyright of this image but you can use it under the terms of CC BY-SA, credited to Ultrapurple.

[Gyro]

It's a zombie thread that needs resurrecting. 

[Ultrapurple]

It's a zombie thread that needs resurrecting. 

In that case...

SIM card

Raspberry Pi Pico

Leadframe, bond wires & chip

Bluetooth earbuds

[Ultrapurple]

18-pin DIP (probably a plastic packaged PIC)

128GB microSD

Automatic lancet (finger-pricker)

Main board, battery protection board & sundries in a cheap tablet

[Ultrapurple]

Not exactly 'engineering' as we know it, but a proof-of-concept fusion of an X-Ray and colour photo of the same item.

[Ultrapurple]

Finally, for now, a fusion of LWIR, visible and X-Ray.

[_Wim_]

Finally, for now, a fusion of LWIR, visible and X-Ray.

That's my favorite combination! Should invest in an X-Ray machine also, but to damn expensive just to play around with the technology...

[CatalinaWOW]

Interesting photos.  I have long been fascinated by the possibilities of using false color sub-bands of various spectral regions.  Regions which behave more or less uniformly over the total band covered so that the process is analogous to our color vision in the visible.  I haven't done much with it, and to be honest the few attempts I have made have not generally proved very exciting. 

The three spectral bands you are using have very different qualities (one dominated by self emission, one dominated by reflected light and one dominated by optical density).  Have you found these composites of spectral regions with widely different properties useful, or is it purely an artistic endeavor?

[KE5FX]

At the risk of resurrecting a zombie thread, I thought it may be of interest to see inside a couple of high-capacity microSD cards.

Great shots.  What machine are you running? 

[Ultrapurple]

@CatalinaWOW - Thank you for your kind words.

I presently own equipment to image in LWIR, MWIR, SWIR, NIR, visible (duh!), near-UV and X-Ray and have done many experiments over the years. Most has been LWIR, NIR, visible and near-UV, simply because my capabilities have only extended to X-Ray, MWIR & SWIR relatively recently.

It can be a challenge to get (subjectively) 'good' results, especially when dealing with imaging systems of vastly different types. Mostly I use the decades-old stalwart Paint Shop Pro X for image combination and editing, though other software comes in handy too (eg Microsoft Image Composite Editor to make hi-res thermal images). It really is a matter of 'suck it and see', though with experience comes a better feel for what may, or may not work. Thermal, visible and X-Ray do work reasonably well together, for example




but of course there is a physical limitation on the size of things that can be X-Rayed. It would be difficult for an amateur experimenter to X-Ray the Pyramids and Sphinx of Giza, though I have made visible and high resolution thermal images of them.



I can't help thinking that X-Raying (or, more likely, gamma-raying) the Pyramids might reveal some interesting information.

[Ultrapurple]

@KE5FX - I am fortunate to have two cabinet-type X-Ray systems. The first was a ScanMax 20 mail scanner, which was very low resolution when it came to me but I modified it with a higher resolution imaging system (without, I hasten to add, affecting the cabinet's X-Ray containment). This is the sort of resolution I got from the original setup (the subject is a Nexus 10, 10" Android tablet, and I am aware there's an aspect ratio error):



and once I'd 'breathed' on it, I was able to get images like this (a Nexus 5 phone, very roughly the size of one of the batteries in the tablet):



More recently I have acquired a Faxitron MX-20 (rather like the one Fraser / Aurora describes earlier in this thread) with a 100mm square 4Mpix imaging sensor, which lets me produce much higher resolution images - not least because the MX-20 has a microfocus tube with a very small spot size, far smaller than that of the Scanmax. Spot size is a limiting factor on image sharpness and resolution - if you look closely at a shadow in sunlight you'll notice that the edges are never actually sharp because the sun appears about half a degree across. If we were on (say) Pluto, much further away, shadows would be very much sharper because the sun would approximate a point source. 

Why keep two X-Ray machines? Well, the Faxitron MX-20 is incredibly convenient to use and has exquisite resolution, but tops out at 35kV and can't penetrate metals very well. The older ScanMax 20 produces higher energy X-Rays (55kV I think, though possibly higher) and thus offers better penetration.

[Ultrapurple]

One last point whilst I'm feeling garrulous. I don't claim to be anything like an expert on X-Ray imaging, technology, or - most importantly - safety. The machines I have were designed for safe use by relatively untrained personnel in what amounted to an open office environment. There are many new and used X-Ray machines on sale here and there that don't have shielding (sometimes by design, like hospital X-Ray machines) and/or require significant training in order to use them safely. Don't forget, too, that X-Ray machines necessarily involve high voltages - sometimes above 100kV and rarely below 10kV - from power supplies designed to produce current.

If you don't know what you're doing, stay away from X-Rays. Like other forms of ionising radiation you can't hear, see, smell or taste X-Rays but they are harmful and the damage they do to you is cumulative. By all means get yourself some training (or at least read around the subject, from authoritative sources), but don't just rely on an internet forum for safety advice!

[Mecanix]

The Unofficial Where-to-Conceal Enthusiast Terrorist thread!!!

In all seriousness, great and fascinating pics