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

[Fraser]

For those interested, a picture of the Faxitron software front end.

The Magnify function is activated in second picture. You can move the magnified square around the image to quickly see more detail or determine if the exposure was OK in intricate areas.

There are menu options down the left hand side. They include tools to adjust the brightness and contrast, sharpness, measurement rulers, Regions of interest, Histograms and all manner of analytical stuff. You can even adjust the contrast in one specific area without changing the rest of the image 

I have barely scratched the surface of the menus as I normally just need a quick image in order to make sense of a situation. I shall have to play with the options sometime.

I have several high resolution Mitsubishi and Sony thermal image printers that I bought for use with this machine. I have yet to hook one up but I have a choice of either 4" or 12" paper width models. The 4" is brand new in the box and just awaits me finding time to plumb it into the rack 

Thats all for tonight. Bed beckons 

Aurora

[tautech]

For those interested........

Who wouldn't be? 

[ez24]

Do these x-rays destroy the DUT ?  Can you take an x-ray of your hand?
Nice toy.

[Ysjoelfir]

A few more images to end the day.

First of all Thermionic Valves (or Tubes for US readers)

The MX-20 cannot penetrate the metal parts but then I did not expect it to.

If you look closely it actualy did penetrate the metal parts quite good! You can clearly see the heater inside the cathode tube, beautiful! thanks!

[sleemanj]

If you look closely it actualy did penetrate the metal parts quite good! You can clearly see the heater inside the cathode tube, beautiful! thanks!

I often find it interesting how much more is actually recorded in an image than we can readily see.  Just some quick fiddling with hue, contrast, brightness and inversion brings out quite a bit more.

[HighVoltage]

This is one of the best "toys" I have ever seen on eevblog.
Pretty cool.

Can you show a picture of where you place the DUT to take the x-Ray?
Is it enclosed or wide open?

[Molenaar]

Very interesting and beautiful as art!

Do these x-rays damage the finer semiconductors, such as the cpu you imaged?

[Fraser]

To answer some of the questions...

The DUT is placed in a large chamber that has the X-Ray tube at the top, firing downwards (the safest direction usually). At the bottom of the chamber there is a 100mm x 100mm carbon fibre window. Below the window there is the imaging array. A photo diode array made up of 8  rectangular panels. You can often see the panels in the pictures as I have yet to optimise the imaging and some fresh camera calibration may be needed. Fortunately I have the service manual.

The chamber is fitted with a door that incorporates safety switch to prevent X-Ray emission with the door open.the door seals the chamber against X-Ray egress.

Whilst it is possible to X-Ray live animals in the MX-20, I am not intending to do so. You will note on the images that It produces the X-Ray radiation dose is details in mAs. This is for keeping a record of a live DUT's exposure to non ionising radiation. In some experiments, live DUTs are deliberately exposed to certain levels of X-Ray to irradiate them. I am not into vivisection though 

I would not x-Ray any part of my body in my unit as it has not been recently calibrated and anyway, the door would prevent such experiments. I advise against experimenting with X-Ray on yourself or anyone else.

I shall post a picture of the X-ray chamber later.

Aurora

[dom0]

No, x-rays, even with much higher energy (~acceleration voltage) don't damage semiconductors.

[Fraser]

With regard to damage to the DUT

There are soft X-Rays and Hard X-Rays. Soft X-Rays are known to potentially corrupt data cells in memory devices if the exposure is long enough. The memory is not physically damaged though and may be re programmed. Hard X-Ray is far less harmful to memory cells, hence why airport X-Ray machines are relatively harmless to laptops and memory sticks etc.

In many X-Ray machines an aluminium filter is fitter to the X-Ray generator to stop emission of the soft X-Ray energy. In medical equipment this is done to reduce the non-ionising radiation exposure of the patient as soft X-Rays add nothing to such imaging yet add to the exposure dose.

My MX-20 can drive the tube with a voltage between 15KVp and 35KVp. This covers the range of soft X-Ray production and enters Hard X-Ray production at its higher voltages. I tend to work at around 30KVp for penetration and harder X-Ray emission. The MX-20 does not contain an aluminium Soft X-Ray filter as soft X-Rays are used for some tissue sample images.

Non memory cell related electronics are not harmed by the levels of X-Ray that most medical and industry QC machines produce. 160kVp produces much hard X-Ray energy yet is harmless to a CPU. You can ruin electret microphones with long durations of X-Ray exposure though. It discharges the charge held within the microphones components rendering it useless.


Aurora

[BravoV]

No, x-rays, even with much higher energy (~acceleration voltage) don't damage semiconductors.

Although not all, but certainly yes for certain unique semiconductors.

 -> X-Ray Effects on Intersil FGA References

PS : Also read the references regarding x-ray vs semiconductor, referred at the last page of above Intersil document.

[Fraser]

@sleemanj

Nice work.

Yes image enhancement certainly brings out the detail. The MX-20 software does have lots of image adjustment capabilities but not false colour. If there is a challenging DUT, I believe that images can be tweaked to reveal otherwise hidden details. This is a laboratory equipment after all. Lots of adjustments, not just a full auto happy snappy machine 

Aurora

[Fraser]

I needed to X-Ray my new FLIR ONE G2 camera to see if the power was connected to the PCB connector (my unit is totally dead) It is going back to FLIR tomorrow so I do not want to open it.

The FLIR ONE G2 provides a good example of the limitation of low energy X-Ray as produced by the MX-20. I can penetrate the case, but not the battery or denser parts of the camera. Whilst I could see the cables from the battery, the connector is hidden under it and cannot be imaged. Such is life. X-Ray is not the panacea to all Reverse Engineering and inspection Challenges.

I include the images here for interest only as they will not win any prizes 

[Fraser]

More

[Fraser]

More

[wkb]

Professional GPS Antenna

Looks like a mini QFH 

[Fraser]

WKB,

That is exactly what it is. A very common antenna type for such systems. I used to use one on 137MHz for the LEO weather satellites such as NOAA etc.

In this case the antenna side elements are likely copper foil wrapped around a foam or polystyrene former. Again common practice for such antennas used at higher frequencies such as GPS and satellite phones.

Aurora

[Fraser]

By readership request, here are some pictures showing the business end of the Faxitron MX-20 

By way of commentary for the pictures....

The top section of the machine (its head) contains the X-Ray generator and the control panel. The MX-20 can produce images using its built in digital camera or using conventional X-Ray photographic plates (film). The latter can be done via the front panel with no need of a PC. Indeed, early MX-20's were film only and had no digital camera fitted.

The X-Ray tube is a high performance 12W Microfocus type with Beryllium window. Its beam is shaped and is conical spreading out from the target spot.

At the bottom of the cabinet there is a BIOPTICS 2048 x 2048 X-Ray imaging array that is connected to the PC as a Camera via a custom PCI interface card.

In order to assist in positioning of the DUT in the field of view, there are two lasers producing a cross target on the cameras face and on any DUT placed in the cabinet.

The shelf on which the DUT is placed is made from perspex and has a >100mm x 100mm hole in it. This is to reduce attenuation of the X-Ray beam after it has passed through the DUT On top of the shelf there is a thin low X-Ray absorption sheet on which the DUT actually rests.

The height at which the shelf is fitted in the cabinet dictates the physical magnification that occurs. With the shelf lying on the bottom of the cabinet, the magnification is X1. Magnification via movement of the DUT towards the X-Ray generator provides a larger DUT image whist maintaining resolution at the camera, unlike when using digital zoom techniques. There is a trade off between the magnification of a DUT and the sharpness of the image produced however. This is just basic physics and not a sign of poor performance on the part of the MX-20.

Hope this is of interest.

Aurora

[free_electron]

As an example of just how useful X-Ray can be when reverse engineering a sealed product......

The Hewlett Packard Logic Dart was a very clever little Logic 'analyzer' with just 3 channels.

I got one of those. use it quite often. They had an ad campaign saying "This is a close as a tricorder as you will ever come"

http://www.pd.infn.it/elettronica/Strumenti/HP%20E2310A.pdf
http://www.keysight.com/en/pd-1000002632%3Aepsg%3Apro-pn-E2310A/advanced-logic-probe?cc=US&lc=eng

[Fraser]

@free Electron,

I am greedy.... I own four of them complete with leads and the IR linked printers 

They were being sold so cheaply without their leads that I could not resist them. A huge haul of leads were sold on eBay and I did a deal with the buyer to buy some lead sets at a very reasonable cost of IIRC $36 per set of 3 leads. I later found a complete Logic Dart lead set with the explorer probe etc. I bought the forth Logic Dart quite recently and at a great price. As new with all leads still in their sealed bags 

I actually like the design and it sits with my multimeter very well. Convenient for quick checks on data lines.

Aurora

[Fraser]

The X-Ray party will have to take a short break now as I am moving kit around in the garage as part of a major sort out. I will have neither the room nor time to play for a while. I also need to do some maintenance on the MX-20 that has been due since I bought her. The camera communications cable has a connector that needs some attention, so that will get done while the machine of down.

Interestingly the cable is marked up as for Universal SCSI. I will check how it is wired as it would be great if its a COTS cable.

A quick check shows that Madison Extreme make standard SCSI cables and my cable is from Madison Extreme.... looking promising.

Back here soon with (hopefully) more interesting images

Aurora

[dom0]

Looks like an IBM-style SCSI-2 cable to me. Should have 60 pins or so.

[Fraser]

Close but not quite close enough.

It is in fact a SCSI 4 VHDCI male to male cable. Its hard to tell without having a scale next to the plugs in my pictures. The D connector part is 32mm wide and contains 68 contacts. SCSI VHD for sure.

I have just ordered a couple of high quality replacements, so I will not need to mess around with the damaged one.

Thanks for your comment though 

I do like it when an OEM uses COTS cables. So much easier when the cable gets damaged.

Aurora

[SeanB]

Mealworms might not have been harmed in the Xray, but I doubt that all are enjoying the visible meal of corn starch they are easting, seeing as some surely went down the beak of a very happy dinosaur relation.

[mikeselectricstuff]

For more info on this machine (mine has the smaller 2" sensor )


The biggest limitation is the low kvp, which has minimal penetration through any significant thickness of metal or ceramic.
Another issue on mine is the small sensor - at some point I plan to make an X/Y stage so it can automatically do multiple tiled shots.

Here are a few random pics