Viewing an active electronic circuit with a scanning electron microscope

[@scottboddicker6549]

As an IC failure analysis engineer for many decades at TI, nothing was more impressive than watching a circuit function inside of an SEM. Of course, we never built our own SEM, our old Cambridge S150 came equipped with a electrical feedthrough connector to power the IC. My first experiment was with an old military grade 5400 quad nand gate that definitely had an SiO2 protective overcoat and HF would move the SiO2. A more modern IC would have an silicon nitride overcoat and would resist the HF etch. It would undercut into the SiO2 under the SiN and would eventually lift off the SiN but would destroy the circuit. With your hardware genius, you should build a fluorine based plasma etcher to remove the SiN overcoat.

[@abidhussain7552]

Scott Boddicker thank you for the important tip. I hope to see the results soon.

[@drakethedragon457]

Did anyone understand this? I didn't

[@scottfranco1962]

"build a fluorine based plasma etcher" should go nicely with the DIY particle accelerator...

[@pahvalrehljkov]

i dont know what you said, but i like it... +1 for a fluorine plasma etcher build idea...

[@lambmaster27]

What language am i supposed to read this comment in? Because it isn't english.

[@Od3ri]

"I used my DIY scanning electron microscope to--" What? >DIY scanning electron microscope >DIY Subscribed.

[@ramade9040]

@oderi me too

[@mrtracyut]

Ben you are a seriously amazing guy. Not only is you endless equipment list impressive but your simple presentation style is brilliant. You are worthy of a Professorship!

[@lachnloaded2]

He works for Google X FYI

[@francoisdastardly4405]

it's been 6 years and this still the most awesome KZbin homemade science proyect !!

[@EddyFong]

Amazing! As another poster commented, I don't always fully understand everything you do. But that is unimportant because your experiments are so fascinating, well presented and informative. Respect from Australia, and thank you for taking the time to create and upload this most stimulating and educational series.

[@sciencoking]

Apart from your actual discovery, which is like a dream come true for anyone who ever debugged an electronic circuit, what amazes me is that you can just have nitric acid eat away the cover and leave the die perfectly intact, and also that the manufacturers actually print their logo on the silicon.

[@robertbilling6266]

When I was an undergrad in Cambridge about 1975 I went to some lectures by Prof Bill Beck, who had not only made this work, but had a trick of chopping the electron beam at a frequency slightly below that of the clock to the chip, giving a strobe effect.

[@atheos2050]

Hi Ben :) Love the videos! I think it's more likely the chip killed itself than the e beam - as the vacuum reduced thermal conduction, the chips wires overheated? source: have SEM'd 50 nm silver nanowires without destroying them + killed a few devices before realising thermal conduction in vacuum was the problem.

[@gary.richardson]

Thanks for the insight!

[@davet11]

I remember having to drag an old electron microscope out of mothballs for e-beam probing about 20 years ago. I was ready to throw in the towel when I finally realized the image is the inverse of what I was naively expecting. Instead of higher voltages being associated with brighter metalization, the 5V/3V metal lines were actually darker. Once you realize that and adjust the contrast accordingly, you should even be able to see voltage differences of one or two volts pretty easily....ie. reasonable analog voltage capability. btw this was all without removing passivation from the die. Fascinating video.... I'm sure e-beam probers have much better resolution nowadays but pretty quick way to find out where a chip is broken. Thanks and love all your videos.

[@solarwarriorkek]

You are the coolest person ever, bro!

[@MrGodofcar]

Why?

[@acaaew]

Who the hell else do you know that built their own scanning electron microscope? Ben has absolute mad knowledge and skills and even though I am highly skilled in my field as a Data Center engineer he makes my knowledge of skills feel rather insignificant. What amazes me the most is his diverse skill set including electronic engineering, chemistry and his mad CNC skills. He is the nerd of all nerds and I would love to hang out with him in person screwing around in his workshop which is also the ultimate nerd cave.

[@Nick_Tag]

Andrew you know those “Masterclass” courses with DeadMau5 and Aaron Sorkin... I’m sure people would love to have Ben start from the beginning and bring us mortals up to speed! I finished a Masters Degree in Physics and his knowledge in *just* that specialism easily eclipses mine..how are you doing this Ben!? I first found this channel when I had an unrealistic and wacky idea to try and make Aerogel: only NASA spinouts, and this guy...

[@Yannique1]

I really just understand like half the things you are telling in these video's but i like to watch them anyway, just because of the fact that you are so passionate about the things you do and the reason and intelligence you look at them, keep the video's coming! Greets from Belgium!

[@ScoopDogg]

Aaah the good old 555 timer chip : ) my first electronic project

[@rickpercy87]

Top notch human being...unbelievable work. All of you're videos are indispensable to mankind.

[@Mezgrman]

4:15 I didn't know that they put their logo even on the die itself :D (you can see the ST Microelectronics logo on the right side)

[@MJ-em2ix]

Mezgrman It's an Easter egg that few will ever find.

[@gentlemanfarmerPNW]

Mezgrman I used to work at Maxim integrated circuits and when I would do Post lithography visual inspection on almost every layer in the Alleyways in between die there would be Dilbert comic strips. Turns out someone on the radical design team was a big Dilbert fan.

[@ThriftyToolShed]

Mezgrman, I thought that was really neat also. I was thinking the logo? and is that the trademark and copyright, OMG!

[@nigeljames6017]

gentlemanfarmerPNW Did you ever get to test the Maxim product ? Did you use a Credence or a T.M.T. tester ? I only ask because I was a founder of T.M.T. and we had testers running in all of your production sites. Our office was just around the corner from you in Sunnyvale CA.

[@hardscorerockkssss]

can i ask question..what point doing die large if die is only few mm large..what is that black stuff near die wires

[@MrJrbryan]

many of us appreciate these type of experiments..you are giftedthx

[@evenealy5540]

Wow. More please. This is extremely impressive content. Thank you for sharing.

[@pcrengnr1]

Thx for the video. Sorry you lost a 555 timer IC. It gave it's life in the name of science. Since the rates are so low a low bandwidth/low sample rate data acquisition system plus PC would be a great display.

[@gustavgnoettgen]

Science' long arm reaches out again to see inside something. This view deserves to be believed and remembered.

[@falldeaf]

I was pretty pumped that I got a change to meet you at this years SF maker faire, your presentation was really mind blowing and this video was fun! I can't wait to see if removing the layer on the IC works! Nice job and thanks for making the video :)

[@smellycat249]

Where is this magical work shop with every gadget I couldn't dream of?

[@SECONDQUEST]

Wow, this is my first video from this channel and I'm very impressed. Very interesting!

[@thembaguy]

I am so glad I found your channel. KZbin has never been so interesting before.

[@en2oh]

Hey Ben, not to guild the lily here, but you are an incredible teacher. I hope that your "real job" at Google brings you as much joy as I think this outlet does..... keep at it.... keep loving it! Doug

[@astrialkil]

if you change the oscilloscope's crt refresh rate you can get rid of the lines in the video so we could see it better. this was very cool ! Thank You for sharing this ! 8)

[@anullhandle]

astrialkil The scope just follows the SEM scanning. you'd have to monkey with the SEM

[@jamesharrell4360]

Jesus Christ. Is there anything more awesome than what you are posting? Anything I dream up has a precursor video on KZbin full of knowledge. You, Cody, AvE. ToT, plasmaCH, etc. Would love to brainstorm with ya and put in some hard hours. Thank you for everything. Hope you make it even bigger.

[@edmarkham2632]

Great idea. Keep it up. As you learn what works and what doesn't work, so are we. Thank you much.

[@doceigen]

In my very early engineering experience as a DPA (destructive parts analysis) engineer at a major aerospace, we did this regularly to add data to the quality control of vendor supplied chips. We would get 32 chips out of a lot of perhaps 250 pieces, and bang them with piezo shock to make the dust dance, decap just like this guy,until removing the oxide he is about to do, and then viewing the chips in an ISI-60 SEM, then X-ray, then rip the silicon and pull the bonds to test strengths, etc

[@qixxxz]

I don't care WHAT that beer commercial says. YOU sir, are the most interesting man in the world.

[@RobertBaindourov]

Nitric Acid and Decapping has returned into my life again.

[@GoldSrc_]

that was one of the best things i've seen.

[@punishedexistence]

You're like the male version of the character Gadget, from a show I remember long ago that could build just about anything. It's like, Hmm, I'll just reach in the closet n pull out the ol' electron microscope. That's totally awesome and now I have to watch the video on how you made it. You rock!

[@PixelCortex]

This has to be the best IC diagnostic tool ever. Just needs a little work and it'll be able to fit into my pocket.

[@levogiro]

Spark plugs has 'pole' HV insulators... *clap clap clap*... I just love the straightforwardness of your solutions to problems

[@davidmintun]

First video of yours i've seen..., you've got skills brother.

[@SkansgardCNC]

You built your own scanning electron microscope?! wow... just... wow.. That is amazing! O.O

[@hab2856]

Back in the 1980s I levered the top of some chips (74 series logic and some memory) looked with a powerful optic microscope. As well as the circuit I remember seeing some very small writing on the edge of the chip. These were the ceramic type made of a ceramic and epoxy sandwich. They came apart without too much trouble and reasonably in tact.

[@oliveroberdingkokron1117]

Did you do the video without the oxide layer? (Can't find it)

[@NickMoore]

I always thought the business with static electricity and ICs was bunk but bombarding the die directly with a beam is a bit of an extreme case. Can't wait to see the next step.

[@MrControColpo]

Can't describe how amazing i find this. Genius

[@scottfranco1962]

They actually make machines that can use this to probe ICs to debug them. They use a map of the IC and can probe multiple signals at once and present the signals in real time on a "scope" display. Needless to say the price of this is truly breathtaking.

[@KowboyUSA]

Man, you have the most awesome workbench.

[@BrekMartin]

Nice :) A shift register would be cool.

[@spidermcgavenport8767]

thank you for this it's something I've been curious about since 386sx processors. Just never had the resources in college for much of anything more than pro-e and rapid prototyping.

[@spidermcgavenport8767]

you most definitely will find in vintage hardware Easter eggs and graphics by the actual engineer.

[@TheSenorTuco]

"I can see electricity" If this would come from anybody else then you, Cody and other badasses then it would be automatic dislike. And I like the irony that with this i don't care about resolution and it's perfect resolution, I can see everything and the information content is great. While some other KZbinrs have 4K videos and 1minute intros and outros and the content is false and misleading information. Just waste of traffic compared to channels like this.

[@datsquazz]

I think your next video is now the most anticipated video for your subscribers yet

[@Blackmage50]

So.... mind blown at this point but having a blast. Keep it up! 👍

[@120ohm]

Have you ever considered writing a book? Your projects are so unique and you explain them with such clarity! If you ever do I'd love to get my hands on a copy!!!

[@AppliedScience]

Maybe. The electron beam may alter the state of the bits in the memory before its possible to read them. It's definitely something that I would like to try soon.

[@MuradBeybalaev]

This is kinda like appreciating an ice sculpture garden by way of exposing it to a hailstorm.

[@SECONDQUEST]

Murad Beybalaev yes

[@gary.richardson]

Yes

[@TGiFoosday]

Yep, agreed on that conclusion and a great analogy. But just to see that Oscope display cancelled all sympathy I had for the 555 chips sacrifice. Upon thinking about it, I would also like to see the ice sculpture garden go out that way too! lol ¶8^)

[@peterjameson321]

Wow! What a brilliant experiment. Thank you for posting. I'm rather surprised that the double sided tape didn't outgas and spoil your vacuum though. I seem to remember that someone did an electron micrograph of a pieces of magnetic audio tape once and you could see the magnetic pattern on it. Perhaps that's a simple experiment you could try sometime.

[@SafeAndEffectiveTheySaid]

Pretty cool, I can't wait to see the end of this

[@occamsrazor1285]

Did...did this guy build a scanning electron microscope in his GARAGE?! Fucking awesome

[@NicholasAarons]

That Was Absolutely Amazing Dude Both Video 1 & 2. Keep up the great work. Nick.

[@TRX450RVlogger]

It amazes me how electronic work!!!

[@felixar90]

You have the coolest man-cave in the world. You have a friggin electron microscope in your friggin cave!

[@Fifury161]

So this popped up in my recommended list - I gotta know - did you make follow up videos?

[@m3sca1]

has anyone else done this?could be a ground breaking new way to make new designs by seeing the voltage field,so as to manipulate it. i bet new transistors could be conjured up from being able to see the voltage field,just move some electrodes around see how they effect each other,imagine graphene lines deposited a few carbon atoms at a time,on a surface that becomes the whole circuit board,printed from touch screen devices loaded with schematics,thats where your idea has taken me,awesome video

[@LFTRnow]

I believe what you are getting here is a technique called "voltage contrast". The SEM detects charge differential and displays it as light or dark areas. The link contains some info on it and very cool pictures. www.highfrequencyelectronics.com/images/whitepapers/downloadables/1404_Voltage%20ContrastIn%20Microelectronic%20Engineering.pdf

[@CaioPeixoto1]

awesome!

[@7ngaf]

i never knew Greek was this easy to understand

[@deefdeefdeef]

This is, as usual, very interesting. Great work, Ben!

[@Kalkaekie]

Enjoyed that ! Thanks for sharing !

[@riteshpuley2891]

I was very keen on getting info. U killed it awesome

[@etmax1]

This principal is used in e-beam probers, a company I worked for had one and you could place a virtual probe on a schematic node and see the waveform on a virtual scope on screen

[@markussteinbacher8807]

the good ol´ 555 again. When he will replaced by a TI DSP ;-) BTW - could watch your stuff all day ! Thanks a lot 4 that

[@robprebil]

That is one of the most interesting things I have seen. Keep going please!

[@VolksTrieb]

Sir I have to say.. you do a lot of awesome awesome AWESOME STUFF!

[@CNLohr]

Doesn't the SEM work by receiving the reflected electronics from the beam? If so, wouldn't active areas of the chip just radiate electrons? How can you get the directionality/locality?

[@AsymptoteInverse]

You lead a very interesting life, sir.

[@zyzzyva303]

Interesting bit about your vacuum gauge. It makes me wonder if the electrolytic cap was outgassing? Cool stuff.

[@uriituw]

Now, I just need my own scanning electron microscope.

[@Qui-9]

Side question... Why make the 555 chip so small, if it's meant to fit in an 8-pin DIP? If they made the die larger, they could increase it's discharge and output current capacity.

[@Antagon666]

Cost and power requirements.

[@AureliusR]

It's all about how many die they can fit on a single wafer. The smaller the die, the more per wafer, and the less each chip costs to manufacture.

[@jeroenvisser4616]

Wow!! Thanx for that, always wondered what magic happens in there. Sure gonna follow the next try 👍

[@WoodysAR]

Bacteria on the chip: _"High solar activity today... I think the Sun did a CME or something, What the... Holy crap ITS GETTING CLOSER!!"_

[@spreepin]

You have some impressive equipment right there!

[@SuperHddf]

If I won't be a poor student, I would pay you money to make such awesome videos ♥

[@johnsweeney6072]

I understood every word you never said 😂 Nicola Tesla but I found it entertaining 👍🇦🇺

[@dr.ofdubiouswisdom4189]

That's badass applied science right there. Thanks!

[@Incrue]

Found your channel today. You rock!

[@TheRobotMaster1]

For your next attempt, would it be possible to film at a different frame rate to reduce the scrolling of the oscilloscope screen? It made it difficult to see what you were referencing.

[@frabert]

Would using a bipolar (non CMOS) version be more resistant to the high voltage electron beam? Much like MOSFETS being more prone to be destroyed by ESD than BJTs

[@robertgift]

Fascinating! Well done! Thank you.

[@briitch3131]

what about, rather than plugging into an oscilloscope, hooking it up to a CRT display? 3:08

[@andymouse]

all my dreams come true a 555 with nothing on !

[@PhilJonesIII]

That chip has been around long enough to have its own church.

[@godfreypoon5148]

all my dreams come true, a 555 snuff film

[@wa4aos]

You and your experiments are the envy of us mere mortal engineers and scientist. What Star Cluster did you travel from to reach Earth???

[@gandsnut]

Was the 555 destroyed by the electron beam affecting the semiconductor material? Would this experiment bear out those who claim a malicious EMP could wipe out electronic devices? What was being seen during circuit operation - could that have been temperature variations rather than 'charge'? Just curious... Thx.

[@fabts4]

Amazing, that was video number 555 in my watchlist. Not kidding! And well, nice video too. But it was #555 in my list, come on!

[@MikeGervasi]

I am totally geeking out right now

[@armenvegas]

Wow. Wow and wow. Best I could do was use a seek thermal cam with a zenon lens to magnify but it doesn't get anywhere as cool as this

[@GamingFromFinland2]

Oh man. I would just looove to get hold of your oscilloscope video. I think it is possible to clean the flicker, do some filtering and perhaps observing the logartimic scale to make the oscilloscope observations more visible. It is not so clear on the video, but I understand that in real life you can see it, so it would be possible to make more visible. Anyways, nice video. Thank you for sharing.

[@theteslasway4297]

DIY scanning electron microscope COOL! :D Now I'm going to have to build me one! :D

[@michalchik]

interesting idea but I an amazed that you didn't instantly fry the chip. I bet you got lucky. I would just build a simple oscillator circuit out of high voltage parts and not even try an Ic.

[@MartyOtzenberger]

HF! Your a braver man than I.

[@joeestes8114]

Wow!, who just happened to have a electron microscope !

[@RimstarOrg]

Scanning damages the chip so you can't observe what's going on? Ben's Uncertainty Principle? :) (I know... you think it's due to the charge buildup on the oxide layer but it sounds like a good principle for now.)

[@hugobalder1866]

im so happy i was able to see this

[@TheHolyReality]

Omg how did u get that epic laboratory? I want to work there forever

[@MuradBeybalaev]

That's what San Francisco money can get you.

[@gcbzzzz]

Murad Beybalaev if you live somewhere with a huge commute from San Francisco.

[@PurpleLion35]

Truth, Its expensive as shit here lol

[@occamsrazor1285]

Yeah, just try to find housing. You have to go out to Brentwood or Modesto.

[@DoubleM55]

CMOS technology is very sensitive to high voltage / static charge, why not try some chip from TTL series. I think they are much more robust and can handle more. Anyway, that was a brilliant idea, being able to actually look at circuit working, transistors switching on and off, awesome :)

[@TheIAMINU]

Tunneling electron microscope , works in a voltage range the chip could tolerate..

[@PabloPazosGutierrez]

I love these videos, thanks for sharing!

[@oturgator]

Why don’t you use a ceramic chip, you can easily either pop the metal cover off, or split it into half from the horizontal plane, without damaging the IC?