Incredible how tiny all those L2 cache, 8 core CPU is and crammed right next to it is the 14 CUs Radeon GPU and eSRAM in such a tight space in that die. This is what powers all our gfx and physics intensive, awesome looking games.
I have a standalone dead graphics card (MSI R7 360 OC )whose GPU(AMD Tobago) has the same specs(768 stream proccesors:48 TMUs:16 ROPs (although it had a higher clock ) that the XBone GPU (835MHz vs 1100) I'd like to compare this dGPU to the XBOnes' iGPU=
Hiho, danke dir! Das Video ist mittlerweile schon wieder überholt. Dieses Jahr wurde kein einziger Chip mehr geläppt. Ein Ultraschallbad mit nicht ganz ungefährlichen Zutaten bewirkt wahre Wunder. Gibt da demnächst an anderer Stelle ein etwas ausführlicheres How-To.
Oh ha. Du geizt da ja nicht. Womit kann man das herauslösen? Schwefelsäure? Bin jedenfalls von deiner Gallerie sehr beeindruckt. Darf ich ein paar Bilder von dir mal im Original haben? Mich interessieren ja nur Grafikchips und CPUs. :) Kann dir meine mail im 3dc nennen. LG huebietheguru
Glasätzpaste basierend auf Ammonium Bifluoride und Sodium Fluoride plus ein paar Tropfen Wasser ergeben eine ziemlich gefährliche und giftige HF-Lösung. Gibt es praktisch fertig für jedermann in den USA zu kaufen. Damit ätzt es sich hervorragend bis zum letzten gewünschten und sogenannten polysilicon-layer durch. Du kannst auch Bilder von mir im Original haben wenn du zeitig fragst. Die RAW Bilderserien werden nach einiger Zeit gelöscht. Archiviert werden nur JPGs in 100% die du so auch bei Flickr unter Original herunterladen kannst. Aber ich kann dir auch gerne eine dicke Zip Datei schicken wenn du magst und nicht die Muße hast dich durch die Gallerie zu klicken.
Phenomenal results. I'm staggered at how well you exposed the tiny structures! Thanks for watching? Thanks for posting! Have you ever tried using a Dremel with a buffer pad and some polishing compound?
Thank you, yes I have tried. Look at some of my very first attempts, there you can still see sanding marks from the Dremel. www.flickr.com/photos/130561288@N04/albums/72157650403404920/page18
Too bad I cant find my old AMD K5 ... this gem processed tons of CDs of porn at 240p quality back when i was younger. Would be such an honor having this little gem on my keyring, reminding me the old, gold, simple days !!
This looks labor intensive! If you don't have the time to wet-sand, you could just buy an entire reject wafer off ebay (yes an entire wafer) for pretty cheap!
Hello! Thank you for the video, the instructions are very valuable! Have you tried to check the structures on a microscope? I wonder if the picture will be clean or full with scratches from the sanding... Could you also please put the description of the sanding/lapping papers on the description? Every time I have to look for it I run through all video to get them, xD
In this video just two or three hours lapping time. Some chips needs longer and some are much faster. Btw. Glass etching cream is much faster and need less than a hour. This is the only stuff you can use at home. Nevertheless, glass etching cream is very dangerous.
yep, no epoxy, just sanding to the first copper layer, put the chip in a plastic cup or borosilikat glas cup, some drops etching cream, some drops water, mixing it and all together in a ultra sonic cleaner for 30-120 minutes (works for the mosts chip technologies, don't work for the SOI process). It will etching down to the last polysilicon layer.
Sanding is necessary for all kind of chips to remove solder balls and the first connection layers (you can't start etching before it). Only SOI chips like Athlon 64 CPUs or some IBM chips need the full program of sanding and lapping.
Hey, wollte mal fragen, ob du immer noch weitere die shots machst oder damit aufgehört hast? Habe gestern selbst mal einen Versuch gemacht einen die zu schleifen. Erster Versuch kam nur so semi-gut raus. Von wo kommst du an die Prozessoren und Grafikkarten ran? Ich selbst weiss leider nicht genau von wo ich die Prozessoren bekommen soll um mehr versuchen/testen zu können. Ausserdem würde mich mal interessieren, wie du darauf gekommen bist Glasätzcreme zu verwenden um die Chips wegzuätzen? Hast du geplant evtl. mal ein neues Video zu machen wie du die Glasätzcreme verwendest? Habe auch ein Bild auf flickr von dir gesehen, wo Salzsäure zu sehen ist. Für was genau hast du die Salzsäure verwendet? Nur zum reinigen oder auch wie die Creme zum wegätzen? Und zum Schluss würde mich auch noch interessieren, wie du darauf gekommen bist Münzkapseln vom Münzsammlen zu verwenden um die Chips zu lagern? Sieht einfach richtig genial und gut aus :)
Hiho, danke der Nachfrage, habe nicht damit aufgehört weitere Shots zu machen, manchmal braucht es auch Zeit für andere Dinge. An alte Prozessoren und Grafikkarten komme ich primär mittels Ebay. Alt, defekt, Lot-Sammlungen oder Goldcap-Recovery sind da die Anlaufstellen. Die ungefähre Richtung mit dem Ätzen kam aus dem CPU-World Forum, das verwenden von Glasätzcreme war dann eine spontane Eingebung. Ob wirklich mal ein Video zum Ätzen kommt, weiß ich noch nicht genau aber der8bauer hatte die Methode mal in einem seiner Video vorgeführt (ruclips.net/video/xFsi2HSYwk4/видео.html). Das nächste Video erklärt sehr eindrucksvoll, welche Fehler dazu führen, wie und warum das mit dem Schleifen und Polieren nicht so funktioniert wie man es gerne hätte (absolut empfehlenswert! = ruclips.net/video/i7Pf6Y02Q4I/видео.html ). Die Salzsäure (örtlicher Baumarkt), die da manchmal zu sehen ist, wird hauptsächlich zum Ätzen der Lotverbindungen und Aluminium-Layer älterer Chips verwendet. Mit einem Schuss Bleiche aus der Waschmittelabteilung lassen sich wunderbar auch andere Metalle prima ätzen. Trotzdem nicht nicht nachmachen wenn man nicht weiß was man tut. Der Umgang mit solchen Materialen darf nicht auf die Leichte Schulter genommen werden, denn diese sind nunmal sehr giftig und Lebensgefährlich. Die korrekte Entsorgung ist ebenfalls eine heikle Angelegenheit. Münzkapseln waren ebenfalls eine spontane Eingebung da mir Schraubenkästen irgendwie zu blöd waren ;)
@@FritzchensFritz Vielen dank für die schnelle Antwort. Hoffe ich bekomme irgendwann einen schön anzusehenden die-shot hin. Wird aber evtl. noch ein bisschen dauern, da dieses doch etwa speziellere Hobby für mich erst gerade anfängt :) Aber deine Fotos auf flickr haben mich richtig gepackt es selbst zu versuchen. Für die Verwendung von der Glasätzcreme werde ich definitiv noch einen Chemiker nach Rat fragen.
I just heat it up enough so the substrate "burns" off and then I glue the opposite side of the chip to a clean, flat, piece of aluminum or something as to not crack it and also for a msntle piece. Then I scrape evenly with an exacto knife.
Yes indeed, you get a lot of tiny scratches when sanding by hand. This is nothing for a microscope, only for cameras with a normal macro lens. If you want, you can take a look at the microscopic image of exactly this chip here.: flic.kr/p/RbraLK
Great video!! Thank you :) What is the exact reason for the epoxy? Is it to protect the edges? I tried to do this myself a few years back with just some very fine sandpaper and I had problems with the edges wearing away faster than the middle. And tips for that? About the glass etching cream, will it not attack the silicon itself, but just the metal layers?
Indeed, epoxy is for protecting the edges. Without epoxy, the edges are sanded much faster, just as you said. But with epoxy you get a new effect. The Chip is going more and more concav. After a while the chip sucks the lapping film and has a vacuum effect, the chip are at this point to much concav and edges lapped much faster again. A heatgun and a cold metal plate can fix that problem and make the lapped side plan or convex again. About glass edging cream: It works good on modern chips 28nm, 20nm (lower and higher not tested yet) and very old chips (Athlon K7, Pentium 4, Ultra Sparcs, graphic cards etc). The last layer is attacked by the glass edging cream very slowly, it's almost impossible to destroy these chips. This weekend i tested some old Athlon 64 and it didn't work for these chips. The glass edging cream attacked all layers (even the last). I think the old SOI process is the problem.
Where did you buy that diamond lapping films? I can only find aluminium oxide films on ebay (or is it also possible with them?) and diamond film in packages of 25 or more for lots of money.
I bought the diamond films in eBay. Indeed diamond films are expensive and less than 25 you get barely on ebay. Lapping with aluminium oxide films don't give good results and the wear is very high.
Hello! First of all I´ve loved the video and your work is amazing. I was wondering if you could recommend me some alternatives for these microfinishing films: presanding with: 3M 268L Microfinishing Film (30µm) prelapping with: 3M 662XW Diamond Lapping Film (9µm) endlapping with: 3M 661X Diamond Lapping Film (3µm) I live in Argentina and it would be imposible to get those. I know it may not be as good, but I'm willing to try some alternatives. Thank you for your time and this amazing video!
There are other lapping films from other manufacturers. Most of them are very cheap and ok for first attempts, but these films have less durability. You can also pour the chip into larger epoxy molds and then try it with regular sandpaper.
@@FritzchensFritz oh, thank you! I was thinking of using sand paper, which grit do you recommend for each step? I've been looking and I didn't find much information. Thank you for your reply and your time!
thanks also heard de8auer talk about the other method with glass and copper etchant (can't find the Martha Stewart brand one, id assume there's not much difference between brands?). The Etchant method looks to be more cost-effective imo(didn't look like de8auer used much of either etchant but do correct me if that's a bad assumption) and also offered better results, you planning to make any tutorials for that method?
i use epoxy like you, but my corners are still getting sanded off first.... what am i doing wrong? im also using 9micro, 3micron and 1micron diamond films... please help
Check it with a blade and the light-gap method. Is it concav you must heaten up the chip from both sides with a heatgun (100°C - 120°C) and putting the chip with the top and non lapped side on a cold metal plate. Check again the chip with the light-gap method. The chip must be on the lapped side convex or plan.
thanks, i dont know what light gap is... but i have a heat gun, so i will try to heat it and then put it on a cold metal plate with the non-lapped side facing on the metal plate.
I don't know whats the correct translation for the light-gap method with a straight edge and light source behind the edge/blade. If you look this picture you understand.: de.wikipedia.org/wiki/Lichtspaltverfahren
hi, it still doesnt work... i heated up to 120c and put it on the cold plate like you said, but the corner are still convex and they get sanded first. what else can i do?
Is your epoxy and silicon chip like same the convex picture? :www.grafik-feti.de/ftp/Die-Shots/how_to_make_dieshots_at_home/sanding_and_lapping_flipchip_cpus/convex_concav_or_direct_epoxy.png While the lapping do you have a vacuum effect? Is is hard to lift the chip up or very easily? When it's hard, you have a direct or a concav silicon chip.
Wouldn't this just make the heatsink not fit properly anymore? If so what heatsink do you recommend if I did this with my i9? What's the thermals look like on your xbox one?
Have you tried open old type chips seen on 80's and early 90's cards? I mean these with contacts on sides..... Have 2 full boxes of these with no use for them.... Something like this - vgamuseum.info/images/vlask/cirrus/cl-gd5401-avga1logochips.jpg
yep, i open some old ati and s3 cards with these ceramic packages. I cracking it with heat (400°C +) and cold water. The ceramic packages can then be easily bent at these temperatures (bonding wires will be destroyed). Works in under 5 minutes, much faster than acid, but the smoke is dangerous.
Thanks for advice, used heat gun + water and it worked. Sadly i destroyed one corner, but luckily core was exposed. So this is mine 1st attempt of dieshot - its Cirrus Logic CL-GD5434 - need work more on sharpness.. vgamuseum.info/images/vlask/cirrus/cl-gd5434die.jpg
Thank you for your video. I tried heating ceramic chips with a heat gun (1000F) and dunked it in ice water, but was unable to separate the layers. The chips I refer to are DIP EPROM chips with the round quartz window. Can you recommend the steps needed to separate the layers cleanly? I can knock them apart with a small chisel, but would like to do a cleaner separation.
@@FritzchensFritz i do it like this: 1st heat up the cpu to release the die 2: little bit sanding to get rid of solder 3: put die in iron-3-chloride 4: check the die and maybe repeat step 3 5: put die in glass atching paste 6: check die and maybe repeat step 5 7: clean the die and take the shots🙂 The colors of the die's are just to beautiful😅 I wish there were a way to print something holograpic like this and i would asap put it one wraping film and put it inside of my ps case😀😅
@@swealer yep, thats the way for 12nm and older process nodes. For 10nm and smaller nodes, you can't put the chip in glass etching paste anymore, it will result in a colorless mess.😔 Indeed colors of the dies are just beautiful. Have you tried printing on Alu-Dibond, this looks also very good. Maybe lenticular posters are a good option to get a nice 3D effect, unfortunately, I haven't tried this yet.
@@FritzchensFritz the lenticular print is an absolut genius idea! Never tought about that😅 i am just now looking for a print service and will give it a try😉
Will this void my warranty?
😂😂😂😂
na
Not in the slightest.
I lolled
lol
the work of an artist!! full respect to your precision and patience
Q
Incredible how tiny all those L2 cache, 8 core CPU is and crammed right next to it is the 14 CUs Radeon GPU and eSRAM in such a tight space in that die. This is what powers all our gfx and physics intensive, awesome looking games.
I have a standalone dead graphics card (MSI R7 360 OC )whose GPU(AMD Tobago) has the same specs(768 stream proccesors:48 TMUs:16 ROPs (although it had a higher clock ) that the XBone GPU (835MHz vs 1100) I'd like to compare this dGPU to the XBOnes' iGPU=
I love silicon shots so beautifully crafted engineering
Echter Nerdporno :D Interessant wie du dein Verfahren mittlerweile raffiniert und perfektioniert hast. Wirklich erstklassige Arbeit. :)
Hiho, danke dir! Das Video ist mittlerweile schon wieder überholt. Dieses Jahr wurde kein einziger Chip mehr geläppt. Ein Ultraschallbad mit nicht ganz ungefährlichen Zutaten bewirkt wahre Wunder. Gibt da demnächst an anderer Stelle ein etwas ausführlicheres How-To.
Oh ha. Du geizt da ja nicht. Womit kann man das herauslösen? Schwefelsäure? Bin jedenfalls von deiner Gallerie sehr beeindruckt. Darf ich ein paar Bilder von dir mal im Original haben? Mich interessieren ja nur Grafikchips und CPUs. :) Kann dir meine mail im 3dc nennen. LG huebietheguru
Glasätzpaste basierend auf Ammonium Bifluoride und Sodium Fluoride plus ein paar Tropfen Wasser ergeben eine ziemlich gefährliche und giftige HF-Lösung. Gibt es praktisch fertig für jedermann in den USA zu kaufen. Damit ätzt es sich hervorragend bis zum letzten gewünschten und sogenannten polysilicon-layer durch. Du kannst auch Bilder von mir im Original haben wenn du zeitig fragst. Die RAW Bilderserien werden nach einiger Zeit gelöscht. Archiviert werden nur JPGs in 100% die du so auch bei Flickr unter Original herunterladen kannst. Aber ich kann dir auch gerne eine dicke Zip Datei schicken wenn du magst und nicht die Muße hast dich durch die Gallerie zu klicken.
Ja flickr ist mir zu fummelig :D Ich hab dir ne PN im 3DC geschrieben.
Phenomenal results. I'm staggered at how well you exposed the tiny structures! Thanks for watching? Thanks for posting! Have you ever tried using a Dremel with a buffer pad and some polishing compound?
Thank you, yes I have tried. Look at some of my very first attempts, there you can still see sanding marks from the Dremel. www.flickr.com/photos/130561288@N04/albums/72157650403404920/page18
GREAT VID!!
Oh by the way, der8auer sent me ;)
Too bad I cant find my old AMD K5 ... this gem processed tons of CDs of porn at 240p quality back when i was younger.
Would be such an honor having this little gem on my keyring, reminding me the old, gold, simple days !!
This looks labor intensive! If you don't have the time to wet-sand, you could just buy an entire reject wafer off ebay (yes an entire wafer) for pretty cheap!
You can't buy scrap wafers of modern CPUs, they're all pattern test wafers. I've been looking for years.
@@mariusfrost640 What's a pattern test wafer?
I’m a working On Extracting Itanium Dies.
Using and Old VCR Motor Head Makes A Nice Precision Sanding/Buffing Wheel.
For the inpatient people here - You can always just throw it into a campfire, wrapped in a steel mesh and collect the silicon after 5 minutes...
Great Video, as a surface to hold the lapping film on, does any regular glass pane suffice?
Yes, regular glass at the present time is almost always very smooth "float glass".
Hello! Thank you for the video, the instructions are very valuable!
Have you tried to check the structures on a microscope? I wonder if the picture will be clean or full with scratches from the sanding...
Could you also please put the description of the sanding/lapping papers on the description? Every time I have to look for it I run through all video to get them, xD
Hi, it's now added to the description. A microscoped image of the chip can you view on flickr (flic.kr/p/RbraLK).
Hope you'll do this with PS4 or PS4 Pro's APU someday. Subbed. :)
PS4 and Pro are still on my list
They don't use APUs
Wow. You've just showed us the complete GDS2
Trying the same thing and wondering why you sand past the first layer and go to the second. Can you clarify?
This is gorgeous, trying to do same thing :D Please tell me, how long are you lapping ? :)
In this video just two or three hours lapping time. Some chips needs longer and some are much faster. Btw. Glass etching cream is much faster and need less than a hour. This is the only stuff you can use at home. Nevertheless, glass etching cream is very dangerous.
Is the process for using the glass etching different?
yep, no epoxy, just sanding to the first copper layer, put the chip in a plastic cup or borosilikat glas cup, some drops etching cream, some drops water, mixing it and all together in a ultra sonic cleaner for 30-120 minutes (works for the mosts chip technologies, don't work for the SOI process). It will etching down to the last polysilicon layer.
So sanding would still be needed on anything that uses SOI?
Sanding is necessary for all kind of chips to remove solder balls and the first connection layers (you can't start etching before it). Only SOI chips like Athlon 64 CPUs or some IBM chips need the full program of sanding and lapping.
I did this on an OLD intel flip chip, looks HELLA COOL!
I would really like to do this, how expensive is the special sandpaper you use?
usually between 10$ and 100$ (ebay or some online-shops), depending on size and quantity
Thank you for a hard work! That is awesome!
Hey, wollte mal fragen, ob du immer noch weitere die shots machst oder damit aufgehört hast? Habe gestern selbst mal einen Versuch gemacht einen die zu schleifen. Erster Versuch kam nur so semi-gut raus. Von wo kommst du an die Prozessoren und Grafikkarten ran? Ich selbst weiss leider nicht genau von wo ich die Prozessoren bekommen soll um mehr versuchen/testen zu können.
Ausserdem würde mich mal interessieren, wie du darauf gekommen bist Glasätzcreme zu verwenden um die Chips wegzuätzen? Hast du geplant evtl. mal ein neues Video zu machen wie du die Glasätzcreme verwendest? Habe auch ein Bild auf flickr von dir gesehen, wo Salzsäure zu sehen ist. Für was genau hast du die Salzsäure verwendet? Nur zum reinigen oder auch wie die Creme zum wegätzen? Und zum Schluss würde mich auch noch interessieren, wie du darauf gekommen bist Münzkapseln vom Münzsammlen zu verwenden um die Chips zu lagern? Sieht einfach richtig genial und gut aus :)
Hiho,
danke der Nachfrage, habe nicht damit aufgehört weitere Shots zu machen, manchmal braucht es auch Zeit für andere Dinge. An alte Prozessoren und Grafikkarten komme ich primär mittels Ebay. Alt, defekt, Lot-Sammlungen oder Goldcap-Recovery sind da die Anlaufstellen.
Die ungefähre Richtung mit dem Ätzen kam aus dem CPU-World Forum, das verwenden von Glasätzcreme war dann eine spontane Eingebung. Ob wirklich mal ein Video zum Ätzen kommt, weiß ich noch nicht genau aber der8bauer hatte die Methode mal in einem seiner Video vorgeführt (ruclips.net/video/xFsi2HSYwk4/видео.html). Das nächste Video erklärt sehr eindrucksvoll, welche Fehler dazu führen, wie und warum das mit dem Schleifen und Polieren nicht so funktioniert wie man es gerne hätte (absolut empfehlenswert! = ruclips.net/video/i7Pf6Y02Q4I/видео.html ).
Die Salzsäure (örtlicher Baumarkt), die da manchmal zu sehen ist, wird hauptsächlich zum Ätzen der Lotverbindungen und Aluminium-Layer älterer Chips verwendet. Mit einem Schuss Bleiche aus der Waschmittelabteilung lassen sich wunderbar auch andere Metalle prima ätzen. Trotzdem nicht nicht nachmachen wenn man nicht weiß was man tut. Der Umgang mit solchen Materialen darf nicht auf die Leichte Schulter genommen werden, denn diese sind nunmal sehr giftig und Lebensgefährlich. Die korrekte Entsorgung ist ebenfalls eine heikle Angelegenheit.
Münzkapseln waren ebenfalls eine spontane Eingebung da mir Schraubenkästen irgendwie zu blöd waren ;)
@@FritzchensFritz Vielen dank für die schnelle Antwort. Hoffe ich bekomme irgendwann einen schön anzusehenden die-shot hin. Wird aber evtl. noch ein bisschen dauern, da dieses doch etwa speziellere Hobby für mich erst gerade anfängt :) Aber deine Fotos auf flickr haben mich richtig gepackt es selbst zu versuchen. Für die Verwendung von der Glasätzcreme werde ich definitiv noch einen Chemiker nach Rat fragen.
I just heat it up enough so the substrate "burns" off and then I glue the opposite side of the chip to a clean, flat, piece of aluminum or something as to not crack it and also for a msntle piece. Then I scrape evenly with an exacto knife.
Does this method damage the IC if you were to photograph it under a microscope would it be badly damaged?
Yes indeed, you get a lot of tiny scratches when sanding by hand. This is nothing for a microscope, only for cameras with a normal macro lens. If you want, you can take a look at the microscopic image of exactly this chip here.: flic.kr/p/RbraLK
Great video!! Thank you :)
What is the exact reason for the epoxy? Is it to protect the edges?
I tried to do this myself a few years back with just some very fine sandpaper and I had problems with the edges wearing away faster than the middle. And tips for that?
About the glass etching cream, will it not attack the silicon itself, but just the metal layers?
Indeed, epoxy is for protecting the edges. Without epoxy, the edges are sanded much faster, just as you said. But with epoxy you get a new effect. The Chip is going more and more concav. After a while the chip sucks the lapping film and has a vacuum effect, the chip are at this point to much concav and edges lapped much faster again. A heatgun and a cold metal plate can fix that problem and make the lapped side plan or convex again.
About glass edging cream: It works good on modern chips 28nm, 20nm (lower and higher not tested yet) and very old chips (Athlon K7, Pentium 4, Ultra Sparcs, graphic cards etc). The last layer is attacked by the glass edging cream very slowly, it's almost impossible to destroy these chips. This weekend i tested some old Athlon 64 and it didn't work for these chips. The glass edging cream attacked all layers (even the last). I think the old SOI process is the problem.
FritzchensFritz Thank you! :)
What kind of epoxy did you use?
This made me want to try this again, it looks awesome! :D
Try google:
EAN 4016138629644
or
EAN 4048500600579
Where can i find your results?
FritzchensFritz Thank you. I will upload some pictures soon.
I am Dutch and live 10km from the German border btw ;) But my German isnt that good.
Once I get the hang of it I will try the 2 G80 GPU's that I have, of almost 500mm²
have you tried sanding top to bottom?
Super Arbeit! HAst du echt toll hinbekommen!
does it have any art on the die under a microscope?
How feasible is it to redo a underfill material?
Is the Balgen you used really necessary for the pictures? I guess it depends on the Macro objective you use?
Sometimes the chips are much smaller than the sensor. That's why I need a balgen
Where did you buy that diamond lapping films? I can only find aluminium oxide films on ebay (or is it also possible with them?) and diamond film in packages of 25 or more for lots of money.
I bought the diamond films in eBay. Indeed diamond films are expensive and less than 25 you get barely on ebay. Lapping with aluminium oxide films don't give good results and the wear is very high.
Hello!
First of all I´ve loved the video and your work is amazing.
I was wondering if you could recommend me some alternatives for these microfinishing films:
presanding with: 3M 268L Microfinishing Film (30µm)
prelapping with: 3M 662XW Diamond Lapping Film (9µm)
endlapping with: 3M 661X Diamond Lapping Film (3µm)
I live in Argentina and it would be imposible to get those. I know it may not be as good, but I'm willing to try some alternatives.
Thank you for your time and this amazing video!
There are other lapping films from other manufacturers. Most of them are very cheap and ok for first attempts, but these films have less durability. You can also pour the chip into larger epoxy molds and then try it with regular sandpaper.
@@FritzchensFritz oh, thank you! I was thinking of using sand paper, which grit do you recommend for each step? I've been looking and I didn't find much information.
Thank you for your reply and your time!
@@facundomusso Try 800 grid for the first layers and then go up to 2000
jw would it need to be diamond lapping film or would Aluminium Oxide work or any other alternatives, seems near impossible to find any
aluminium oxide lapping film would work, but has a low duration. You need many of them.
thanks also heard de8auer talk about the other method with glass and copper etchant (can't find the Martha Stewart brand one, id assume there's not much difference between brands?).
The Etchant method looks to be more cost-effective imo(didn't look like de8auer used much of either etchant but do correct me if that's a bad assumption) and also offered better results, you planning to make any tutorials for that method?
Gotta love the contrast between the generic how-to video BGM and the guy showing you _how to expose the structures on a freaking CPU die_ lmfao
hi m8, would be fantastic also can see the APU of Xbox One Slim made in 16nm =D
What are you doing with the erasers?
use it to apply pressure to a small part of the chip
Beautiful 😙 ! Great technique.
The real challenge is to solder it back to pcb and make it work again
haha, let's start the impossible challenge. You begin
Das muss eine Arbeit sein!
ElooKoN längst hinfällig das Video, gibt da mittlerweile einen Weg der da weniger Arbeit macht.
Kommt da auch ein Video von? : )
Using the pliers to set the chip on burner was super not necessary
i use epoxy like you, but my corners are still getting sanded off first.... what am i doing wrong? im also using 9micro, 3micron and 1micron diamond films... please help
Check it with a blade and the light-gap method. Is it concav you must heaten up the chip from both sides with a heatgun (100°C - 120°C) and putting the chip with the top and non lapped side on a cold metal plate. Check again the chip with the light-gap method. The chip must be on the lapped side convex or plan.
thanks, i dont know what light gap is... but i have a heat gun, so i will try to heat it and then put it on a cold metal plate with the non-lapped side facing on the metal plate.
I don't know whats the correct translation for the light-gap method with a straight edge and light source behind the edge/blade. If you look this picture you understand.: de.wikipedia.org/wiki/Lichtspaltverfahren
hi, it still doesnt work... i heated up to 120c and put it on the cold plate like you said, but the corner are still convex and they get sanded first. what else can i do?
Is your epoxy and silicon chip like same the convex picture? :www.grafik-feti.de/ftp/Die-Shots/how_to_make_dieshots_at_home/sanding_and_lapping_flipchip_cpus/convex_concav_or_direct_epoxy.png
While the lapping do you have a vacuum effect? Is is hard to lift the chip up or very easily? When it's hard, you have a direct or a concav silicon chip.
Can you try this with an Intel Xeon E7-8890 v4 ??
do you have an Xeon E7-8890v4 (24 core Broadwell) for me?
asking intel for a social media experiment :D
einfach geil!
awesome work!
Would you sell one?
very beautiful!
Oh May god, thats so awesome. like
Wouldn't this just make the heatsink not fit properly anymore? If so what heatsink do you recommend if I did this with my i9? What's the thermals look like on your xbox one?
Um this basically kills the cpu
tht looks like a core i7. is it?
this is amd apu Jaguar for Xbox One
So much work but I always wanted to do my Cell Processor
Have you tried open old type chips seen on 80's and early 90's cards? I mean these with contacts on sides..... Have 2 full boxes of these with no use for them.... Something like this - vgamuseum.info/images/vlask/cirrus/cl-gd5401-avga1logochips.jpg
yep, i open some old ati and s3 cards with these ceramic packages. I cracking it with heat (400°C +) and cold water. The ceramic packages can then be easily bent at these temperatures (bonding wires will be destroyed). Works in under 5 minutes, much faster than acid, but the smoke is dangerous.
Thanks for advice, used heat gun + water and it worked. Sadly i destroyed one corner, but luckily core was exposed. So this is mine 1st attempt of dieshot - its Cirrus Logic CL-GD5434 - need work more on sharpness.. vgamuseum.info/images/vlask/cirrus/cl-gd5434die.jpg
Thank you for your video. I tried heating ceramic chips with a heat gun (1000F) and dunked it in ice water, but was unable to separate the layers. The chips I refer to are DIP EPROM chips with the round quartz window. Can you recommend the steps needed to separate the layers cleanly? I can knock them apart with a small chisel, but would like to do a cleaner separation.
Passion!
wow!
Very nice :)
Or you could just use iron-3-chloride instead of sanding 4 days straight😅
I haven't sanding for a long time, that's how it's done now:: flic.kr/p/2hPpMZc
@@FritzchensFritz i do it like this:
1st heat up the cpu to release the die
2: little bit sanding to get rid of solder
3: put die in iron-3-chloride
4: check the die and maybe repeat step 3
5: put die in glass atching paste
6: check die and maybe repeat step 5
7: clean the die and take the shots🙂
The colors of the die's are just to beautiful😅
I wish there were a way to print something holograpic like this and i would asap put it one wraping film and put it inside of my ps case😀😅
@@swealer yep, thats the way for 12nm and older process nodes. For 10nm and smaller nodes, you can't put the chip in glass etching paste anymore, it will result in a colorless mess.😔
Indeed colors of the dies are just beautiful. Have you tried printing on Alu-Dibond, this looks also very good.
Maybe lenticular posters are a good option to get a nice 3D effect, unfortunately, I haven't tried this yet.
@@FritzchensFritz the lenticular print is an absolut genius idea! Never tought about that😅 i am just now looking for a print service and will give it a try😉
تو ایران میکرو فینیشینگ پیدا میشه؟
I'm assuming the Xbox went RROD or the chip is no good... That chip is most likely made by TSMC... Taiwan Semiconductor Manufacturing Company.
awesome!! awesome!!
Думал в конце припаяет обратно и всё заработает
WOW! AWESOME!!!!
wow
(Y)
Why you waste your Time? Google Picture is your friend!