EEVblog
HTML-код
- Опубликовано: 7 май 2017
- Unboxing an ebay Olympus BHM wafer inspection metallurgical microscope. Did it survive the atrocious ebay packaging?
Will a custom Cree LED light replacement work?
PART 2: • EEVblog #992 (Part 2) ...
Forum: www.eevblog.com/forum/blog/eev...
EEVblog Main Web Site: www.eevblog.com
The 2nd EEVblog Channel: / eevblog2
Support the EEVblog through Patreon!
/ eevblog
EEVblog Amazon Store (Dave gets a cut):
astore.amazon.com/eevblogstore-20
T-Shirts: teespring.com/stores/eevblog
💗 Likecoin - Coins for Likes: likecoin.pro/@eevblog/dil9/hcq3 
Наука
Maybe that Mitotoyo Z scale is for thickness measurement. Focus on one layer, set it to 0, then focus on the next layer and read the display.
Yep, could be, if it has the resolution.
Chip Guy Vids , long time ago I used to work @ Stmicroelectronics and the focus is used to measure the Z axis or the height and also the actual focus knob it should have a metric scale for measurements, the table y and x axis they all have little play but you can get models that have a better setup for the x/y axis measurement, @st they us them for QC, wire bonding heights,bonding on the pad of the die etc, unfortunately these delicate equipment is better to use them in a clean room because dust is their worse enemy. but definitely it is a beautiful toy :).
Now that you mention it. There is a scale on the fine focus knob. However that Z axis meter would be much more convenient, just hit the 0-set button :)
Chip Guy Vids Would be interesting to see if the fine Z adjust scale corresponds to microns...
all the good comments are coming from my subscribers :-)
You can see the indents of the optics in the box lid :(
I missed that!
Michiel van der Meulen Yeah, I noticed that too. That scope was lucky to make it in the state it was, any lesser scope would have been in many pieces.
Agreed. It's built like a tank.
noticed that. there must have been a lot of g-force on the instrument. that packing was awful. You can ship balloons like that, but not a scientific instrument. Dave, how much did you pay for shipping, and the ebay item itself?
Repacked to GET THE WEIGHT DOWN!
Dave... Please show us more wafers up close! I could watch this for hours.
JacobCZLP (Sorry Dave if you're against advertising other channels)
Check out a channel called Electronupdate!
I can do that.
I'm never against advertising other channels
Wafer Wednesday?
I *REALLY* want to hear the response from the "eBay Global Shipping Program." A follow up video would be helpful for others that are considering their "service". 1) extremely poor packaging 2) punctured and torn box during shipping 3) no "glass" or "this side up" stickers 4) it should have been at least double boxed with better / more use of foam sheets or bubble wrap. I mean throwing in a few foam chips doesn't do *squat* to protect the extents of the scope. I would want some major compensation even if the light source wasn't broken. If it is packaged by "eBay" they should know all this and how poorly their shipping would treat it.
I agree. the packing was improper. I don't believe PitneyBowes did the packing themselves. They probably "repaired" the box or re-filled the cushing material. The seller is responsible for proper packing. I am an ebay seller myself...
+ElmerFuddGun I doubt we'd be seeing a response video by Dave, since that would imply a response from Pitney Bowes or eBay. And since Dave's channel is mostly popular in the electronics niche, both companies will probably promptly ignore him. Yeah, I might sound cynical, but unfortunately that's the case 99% of the time. And actually the light source WAS broken (it wasn't shattered, but the rough treatment that has caused the lamp body to crack was more than enough to break the fragile filament inside). Also, since packaging is usually done by the sellers themselves, it's probably the seller's fault anyway. Punishing him/her in any was just wouldn't make the whole ordeal a more pleasantful experience either.....
From here AT&T isn't upside down
HAHAHAHA
Is here. Our ISP uses smart Ethernet ports for antipodean connections. Y'know, the kind with automatic X-over...
ǝǝs I ʎɔuɐʇʃnsuoↃᴚMI@
I don't know where you bought it, but I think it would be completely fair to simply state "Seller allows other people to pack items for him/her."
That's absurd to put something like that in the hands of the guy at the shipper's store. Totally, completely, utterly absurd. And repack? My ass.
Geologist here with plenty of petrographic microscope experience. The optics on that Olympus look fine to me, no obviously glaring defects. The colour with your bodged LED look pretty darn good. A trick I’ve used from the camera side to get optimal white balance is to stick something that’s a neutral grey under the microscope, white balance to that using photo/video editing software or the custom white balance feature in the camera. Now you’ve got a white balance setting for that camera on that microscope that you can reuse whenever you want. But honestly the colour looks more than useable as it’s setup now.
Great video Dave!
Bloody cool that is. Cheers for sharing, Dave.
Impressed you got it working.
Very very cool, I've always wanted a scope but couldn't justify the expense on something even half decent that would get limited use. Cheers for the heads-up on the AmScope, they look alright for the price!
+Simowill75 i got lucky when they closed down a school. i payed 80 € for the microscope(the one for the teacher i guess because it had a camera mount and all kinds of filters and stuff) and 2X25€ for 2 different sets of lenses(realy realy nice carl zeiss ones).
schools often buy quality, dont use them so extensiv and dont know or care for the value of the stuff they sale when they close down.
Looks amazing dave :-D, worth every penny.
Hi Dave, I have a trinocular Olympus made in Japan scope still in original wood box with spare lenses. Diffuser intact also 11watt bulb haha. Don't know much about these scopes so glad you made this video will have a go at servicing it but it looks in excellent condition all solid metal precision engineering. Thanks for the video.
The stage can be retightened a little. Both a linear bearings and a gearset. Just undo the bolts, apply a light hand pressure to bar and tight again under pressure, the slack will gone.
Dave, now all you need is a Soxhlet extractor and condenser to decap your own chips in the lab.
Nitric acid
I kind of threw that out there, but thinking about it now... Maybe Sulphuric acid could work in a distillation apparatus. Nitric acid is supposedly the most "effective" at decapping though I'm not sure it would really work in this configuration?
The difficulty of nitric acid is why i suggested sulphuric acid. But regarding the problems with doing it the soak method is demonstrated in Applied Science's video on decapping. I figured a soxhlet extractor would help solve the problems he ran into.
Googling "Soxhlet"...
I got the idea from the NerdRage video "Make a Tritium Nuclear Battery or Radioisotope Photovoltaic Generator" where he dissolves the plastic off of some tritium keychain things.
I recently purchased a $160 amscope brightfield microscope from ebay. I was blown away by the quality.
"I had a box of wafers"
That's funny. I had just put a bag of chips back in the cupboard when I heard you say that.
From some of the markings it looks like it was stepped on a 4x XLS platform. I think the combs were verniers for physical inspection. The probing in the scribe lines looks like you happened upon some slm or etest structures that were probed with a wafer probe card.
When I was a child I opened the can of a µPD416 NEC RAM to look inside with the microscope at my dad's laboratory, there's so many "things"in such a tiny chip, almost unbelivable. But you're missing a very cool sight there: the bonding of the wires to the chip! Go find a ceramic "can" chip and open it asap to see them!
It amazes me that we can fit over 100 billion transistors into something like a 32GB MicroSD card. It's damn near magic.
HalfSpeedMastering - Sure. Can't you?
Actually, you're wrong. 32 GB is 32 Giga BYTE... A byte is 8 bits. 32x8=256. Giga is a billion. That means you have 256 billion bits. If it's SLC NAND flash, it's one transistor per bit, per memory cell. If you have a MLC NAND flash, then it's 2 bits per transistor. TLC is 3 bits per transistor. This doesn't count control logic.
I filed the top off a plastic 4164 to try and make a cmos video sensor, and it nearly worked except I took a corner off the die right at the end. I'm going to try that again with a milling machine soon.
MattOGormanSmith did you try that with a genuine ic or a cheepy from ebay ?
What a nice instrument! Are you going to make a video about dumping ROMs of decapped chips?
You need a can of "spray air" (which is usually some other clear gas) to clean off the wafer after you dip it in a cleaning fluid. That should eliminate the little gremlins that you see. Don't use ordinary compressed air as it probably had gunk in it. Good old iso-propal alcohol is probably a good solvent.
I'm amazed that made it. I would never ship something globally unless it was worth a bunch of money and had a very small customer base. But then it would be shipped using a courier or something like that. As an eBay seller I have to put up with plenty of shipping bs just here in the States.
My experience is in a medical laboratory, but the microscopes we use are pretty similar except for the light source being beneath the stage. Your lighting looks pretty good to me. The problem with the filament lamps is that they make everything very yellow so you often have to put in a certain blue filter. Depending on the microscopes people trained on, some people had to use the filter and some people had to remove it since accurate colour is essential for diagnoses when everything is a scale in shades of purple. If we switched to LED lights, we'd have to retrain our eyes.
One interesting point is that most of the important adjustments (at least on the type of microscope I use) are made to the light *before* it hits the specimen. Intuitively you'd think light is light and if it's illuminated, what's the problem, but it makes a big difference. e.g. The extra little numbers on the objective lenses tell you what to set the aperture iris to for each lens; too closed and it's dark or you only get partial illumination, too open and you lose sharpness of focus and everything seems washed out, it's only a subtle effect though.
In terms of brands, I have found Olympus to be the best. I like Nikon cameras but their microscopes aren't nearly as good. Leica microscopes are just shit.
Also, that depth of field seems appropriate for the intended purpose. Wafer processes mandates small differences in the topology, and the depth of field you remarked on is actually needed for wafer inspection.
Most certainly.
Time to break out the fuming nitric acid
Inside his lab which probably has little ventilation? Riiight.....
The AMscope's are just rebranded china microscopes which you can get direct from aliexpress yourself.
I bought one a while ago for work which cost ~US$400 inc DHL shipping. 7-45x with trinocular and 720p HDMI camera.
I was so impressed i got my own one at home and so did another employee. Work bought a 2nd one too.
Really amazing for the price. The camera isn't the best but not too bad. You want the $20 0.5x barlow lens for electronics work (much more working distance) Search aliexpress for LuckyZoom thats the seller we all used and didn't have problems.
wow that's beautiful!
Hey Dave, Pls make the video of the Flir Thermal Imaging Sensor from your recent tear-down.
Get a marble block with rubber feet to stand it on, that will significantly reduce the wobble when you move your mouse, better mounting of your camera would improve it further.
You Can still buy the Globes i put one into a microscope not to long ago part no LS-15 or LS-30 number is Watts.
Economy shipping. DOH!
eBay global shipping is NOT economy. It's bloody expensive!
For GLOBAL (i.e. international) shipping, it is.
Hi Fran. Could have been cheaper I guess. But I've never had a single bad experience with ebay's global shipping program. And it's not too pricey. I guess PitneyBowes had to repack, because of improper packing of the seller.
Ing. Max Koschuh But why repack to something (at least) equally improper...?
Seller probably told 'em to stamp it 'Fragile'
In the lighting-pointing-at-a-stage world, this hot spot is referred to as the "peak". It is related to the physical position of the virtual point of light, within the reflector/lens system. - Does your illuminator have a reflector on it?
Wafers are very reflective, you will need much more light when inspecting dark and rough materials.
Looking forward to any comparisons you make with an Amscope
For E-mount adapters and the like, check out a telescope shop. Someone like Bintel in Sydney.
Just to echo, you really want a polarizer after your light source, to work
in conjuction with the analyzer. Use glass polarizers, not plastic.
Also, you will want calibration slide (AMscope will do) and eyepiece
reticle (Klarmann Rulings are the best). Would love to see you motorize
the stage ala microgigapan.
Dude, soo cool! I've been looking for the last two weeks for a metallurgical microscope for home, haven't seen these yet on ebay! I actually plan to use it for micrographs so a lot of other stuff on ebay just won't do.
By the way, at higher magnifications Helicon Focus works fantastically to help get the depth of field you would get with computer controlled motorized scopes, just take a few images at a couple different focus points, drop them into Helicon Focus and as you would say "bob's your uncle".
I'll have to try that out.
Just a suggestion but a 10W COB might be a more suitable light source if it will fit the holder. Cree LEDS are good but do suffer from colour differences over the field of illumination. The COB gives a more uniform light over a very wide angle.
In today's episode, Dave miraculously dodges a shipping bullet.
Buy a sack of tennis balls; spread them out in a frame under a slab of wood (or granite); put the microscope on it. That should get rid of the desk coupled vibrations. You still need a light touch for the stage adjustments, or you can fit some steppers and automate the stage movements.
I think you can remove the camera jitter buy taking it to a machinist and having them bolt on a sturdy gusset that connects to your camera's tripod mount. They should also be able to tighten up ways on that stage to make it move properly.
I assume packing a delicate instrument would require some higher density foam, custom cut to shape. and then for extra protection, those packing nuts.
THose wafer shots, nice!
I had exactly the same problem a few months ago when I purchased a trinocular microscope with a boom and HDMI camera from china. They just chucked the parts into a box with a bit of bubble wrap and the heavy boom smashed the eyepiece mounts off the head. No one seems to know how to ship things these days.
Would be fun to measure the transistor size on some various old chips.
thank you for the insight, btw, your vids on opanps helped me ace my circuits 1 class. you are a real treasure for students like me. please keep it up! you are helping future engineers more than you might realize.
Glad to hear, thanks.
GSP indeed repack stuff, I recently get quite a lot of stuff by GSP. It may be bad, but sometimes it's the only way to get stuff, as in pre-GSP era many US-based seller would not even consider sending stuff internationally.
may have been mentioned but the Mitutoyo is an indicator, not a micrometer
Looks pretty nice... great score, curious as to what you ended up paying. I ran a Leitz wafer inspection 'scope for a year at Motorola MOS4 in the early '80s, in the 02 and 03 (early) layers... building 6809's, 6800s, and 64Kb EPROMs. A good scope holds focus all the way across the stage... if it goes out, then the stage is drooping... not that you're trying to do 1000wafers a shift! Still in the industry, now we maintain/refurbish/augment AMAT equipment.
in this episode Dave gets an awesome microscope and gets us to guess the guest!
It would be interesting to see if an ultra-violent light source could be used on this.
Dave that reminds me of a lecture I attended many years ago by a PYE telecoms
They showed some chip scans and alongside of a delay line was a steam engine and carriages complete with steam from the funnel and tracks it would have been about 35 years ago
Nice score shame about the packing
It's a mixed bag with Ebay Global Shipping: I have had both abysmally bad packing and excellent, double carton, zero damage packing through them. At least the microscope was not badly damaged.
People seem to have developed amazing faith in the miraculous protective power of bubblewrap... yet everyone enjoys popping the bubbles between their fingers.
Thank
Dave, I might be wrong but to me the features at 15:30 look like alignment marks for the stepper/mask aligner.
Great optical setup, BTW!
Ah, interesting, could be.
I can corroborate. I am a VLSI/Wafer design grad student and we use similar structures for mask alignment all the time. But that's not all - sometimes, we use similar fingers-like structures with varying widths to judge etching quality as well. Hope this helps :)
Expanding foam , double boxed how hard is that. Is it really that much more expensive? And take stuff apart as much as possible, and wrap all pieces separately.
Dave's Apple Laptop Repair Videos coming soon.
Dave I think the description is missing the forum link.
Hi Dave, I have a times 100 objective for this microscope. have to look for it at the attic but I will send it you! only down side to the 100x objectives ar that you need immersion oil with it... still usefull I think
That would be cool, thanks. I'm looking for a x60 and x80 to complete the set
You're welcome, it could be that I have also x60 or x80... not sure. I must have a set of 3 but can't remember the other values. will send the whole set
Dave, you should try one day some funny things like CCD sensors, printer heads...
you need a engineers slab maybe half ton to cut down on vibrstion
Can't you adjust the apature of the microscope an put some more light in? It seems t have a lot purple fringing (you cam see it clearly in the last shots at the edges of the gold pads).
Anyway, I like these kinds of things. Would be interesting to see som ESD damaged chips or test video on how much it will take to destroy a chip...
By the way it is not only important that the optics be Plan, but also need to be of M type (metalurgical), which is to say that the glass has the approriate coating to allow the light from the vertical illuminator to go through it and not reflect before hitting the object. Standard biological objectives are not suitable.
Can't find second video about 3d-printing.
Aww Daveee i was waiting to see a PCB under that :(
How is there color on the wafer? Aren't those microscopic structures you're looking at how would they be colored?
huh, interesting.
And of course the longer working distance these must be designed for (don't expect oil immersion objectives here :) )
Ohhh that package... I had the same horrible experience (many times), don't know why but always happens with expensive and heavy equipment... clearly awfully packed exactly like your box, bubble wrap protecting nothing, equipment completely lose inside the box, holes everywhere and pieces of equipment sticking out.
The box is more like a soft paper ball rather than a box... ohh.. it's heart breaking.. what horrible sight.... =/
bubble wrap is fine for smaller items up to 0.5kg, would have used styrofoam, double box, multilayer cardboard....
Holy shit, I'd be sooo pissed at that packaging. File a complaint for sure. Hell, Olympus might even take the opportunity to throw you a new unit just for the hell of it since it'd be a great promo.
General tips if you're going to be doing decapping IC's :: You generally want to have at least a 50% overlap for stitching your images. Chris Tarnovsky (arguably the 'founder'), Joe Grand and Karsten Nohl all have talks from CCC's in the past which list their processes. Ken Schriff (the righto.com guy) also had a nice talk from the HACKADAY on just general MOS and BJT analysis (which was very interesting to hear --- you draw out a BJT and think -- base,collector, emitter -- but some gates are base/emitter/6 collectors so you can source 6 different blocks with their own constant current... stuff like that!)
Anyways, good deal on that. Not sure how much you paid for it, but the 80s-90s were the hay day for all Japanese and German optics. Japanese Nikon (Nikkor) line, the Olympus glass, Carl Zeiss, Leica, etc. All absolutely amazing. If you have glass that's scratched up on accessible regions (exterior surfaces), aluminum oxide slurry at like 3 micron is the abrasive of choice in the DIY telescopy community. Obviously don't do this on any contoured surfaces, but most objectives that suffer scratches are almost always on the 2 faces that are exterior-side, and you can essentially 'buff' the imperfection out as long as you have a reference planar surface that's true. Unless the enclosure itself had foreign particulate the assembly and then abrade internally (which does happen but not that often), you have a very good chance of restoring the glass to very good condition by surface lapping. (See Tom Lipton from oxtools/UC Berkeley for a fairly good introduction).
Sadly, Modern chips are all at like 22 nm (diffraction limit ~300 nm...) so you have to go to AFM/SEM/TEM sadly for more modern processes. Active rev engers I talk with all have to rent out time on SEMs. The downside is the cost, the upside is most of the SEMs have FIB's on it, so you can do things like mill through layers of poly with amazing precision, or do additive manipulation (chemical vapor deposition of platinum usually) to literally alter the chip at the transistor level.
I'd even be curious to know how fab houses do things like mask alignment these days, since equipping a 100k SEM to each calibration/alignment/wafer inspection station sounds economically challenging...
Thanks Dave. I didn't need one of these, but now I've got one in the mail. Probably gonna get thrown out of the house.
And just like that the seller canceled on me because the scope was "damaged or lost." Easy come easy go I suppose.
Hi Dave how do you increase working distance???
I'm surprised nobody uses foam packing that forms around the item. It's two chemicals in a bag, you break the barrier between them and you have foam.
instapak is ridiculus expensive. but the first choice for items like this. even though, if you are well experienced with delicate items, you can do even better than instapak
You can try to inspect some old EPROMs with a glass window to see the die.
Hey Dave, how about checking out some Wifi/Bluetooth chips on the microscope? so we can get some on chip RF goodness.
If I had the raw dies, sure.
what's up with the one you were discussing with David2 on a previous video
How full is the glass Dave?
Why are the instuments in the background upside down? 0:36
do a video where you put plant leafes and such under it :D
Did you find that one in the dumpster too?
IMO, it would have been in a better condition if it was found in the dumpster :)
Hi Dave, long time viewer, first time commenter. I am a student at R.I.T, U.S.A campuse in the Microelectronics engineering program. I spend a lot of time in the clean room, and use inspection microscopes all the time. It broke my heart to see what the shippers di to your microscope, but I just wanted to point out that typically, wafer test probes (at least using the parameter anylizer that I use) won't leave thos ugly marks on the wafer test pads. There is a precharge function on some parameter analyzers, that allows for precharging before the probes hit the test pads. This can cause a small amount of arcing and burning. I was wondering if this might be the case with your wafer? Again, I am a student, and by no means an expert.
I assume it was simply scratch marks from the probe top, not burn marks as this is not a power device and the wouldn't probe powered up anyway.
SMT resistor missing at 15:00 xD
I wonder what images would look like with a dynamically adjustable polarized filter over the light source?
You can get polarising and colour filters for this, it would be interesting to see.
Kevy ... I cut out a small circular piece from the lens from plastic Polaroid sunglasses. It worked quite well in the light path of my microscope.
Dexxter interesting.
As soon as I seen the box, I thought "Oohh Man". Those packing peanuts are not acceptable in my eyes. I absolutely hate sellers who use that global shipping program, its insanely expensive, way more than regular shipping. Ive had so many good deals turn into ok deals because of those import fees. The whole idea of global shipping is supposed to be to help prevent these things from happening. Beautiful Scope and great video as always. Happy to see you came up with a quick fix, but Id be filing a claim for sure.
They must have a quota for shipping material!
As a grad student doing analog IC design, I am very happy to see some IC imaging on here!
How close are you getting the 40x objective to the wafer? On the scope we have at uni, it's always so close that I'm afraid I'm going to bang the objective into my die while changing objective or refocusing it (like you, I've found changing objectives doesn't quite maintain focus)...
(Also, I'd like a box of wafers. =< Need to figure out how I can find some throwaway wafers!)
You can buy some on ebay, it is not expensive
...good point. I consistently forget about ebay when it's most useful. Thanks.
I was just watching another unboxing video when this one came up (-:
Where's part 2?
we have olympus scopes where i work. they're good, but they're not nikon quality!
Dave, I have been watching your videos for a while... I have received 2 out of 3 items damaged because of GSP / Pitney Bowes... Ebay has to take the cost and do something about this. So we have to protest. Give a call to Ebay, you are insured by the GSP, get your money back 100%... that is what I did.
I keep wanting to drag the vid around :D
Now Shahriar needs to find something even bigger and better :)
Awesome!
Se fosse aqui no Brasil iria chegar todo quebrado se não fosse roubado ou perdido.
Ah! And how does the microscope image jump onto the computermonitor?
What camera? I see none in the video.
Pro tip: Jump to 4:10 to avoid several minutes of Dave complaining.
The fact that they pack it that crappy is one thing. The fact that you can't blame (sue) to be responsible for it completely blows my mind! You trust someone to ship your gear properly and on a responsible way. If they fail to do so (don't care about the reason), THEY failed their responsibilities, it's as simple as that. Lots of other fields that's normal procedure. Still don't understand that every post/mail company gets away with it. To bizarre for words.
My biggest request for improvement would be for stabilization of the tray.
A much finer stage adjust would be my top pick too
the sliding rails are probably worn out, or damaged during shipping
get a granite surface plate as a proper stand for your new microscope
Now get some really nasty chemicals and decap some chips!
Suppperrrr veriiuguut