[00:00.620 --> 00:03.860]  Thanks for joining us today at the Tool Lockpicking Village.
[00:03.860 --> 00:07.480]  The talk I'm giving today is key duplication. It's not just for the movies.
[00:08.480 --> 00:13.120]  My name is Tony Varelli, aka Byte Stealer, or some people just call me Byte. You can
[00:13.120 --> 00:18.380]  find me mostly on Twitter if you want to talk. And that's twitter.com at Varelli.
[00:20.700 --> 00:24.540]  I am a member of the Open Organization of Lockpickers, or Tool.
[00:25.000 --> 00:28.460]  And I'm based out of Iowa, and I'm a member of the Des Moines chapter there.
[00:29.280 --> 00:35.000]  So, I highly encourage you that after this talk today, stick around, keep checking out
[00:35.000 --> 00:40.160]  the Lockpicking Village in DEF CON safe mode. We're going to have a lot of great talks going on.
[00:40.620 --> 00:45.460]  A lot of really smart people when it comes to locks, bypassing them,
[00:45.460 --> 00:49.800]  and physical security. So, I highly suggest you stick around for those talks as well.
[00:50.600 --> 00:55.160]  Today, I'm going to cover two types of key duplication. The first one I'm going to cover
[00:55.160 --> 01:00.340]  is how you can 3D print a key. And the second one will be molding and casting a key. This gives you
[01:00.340 --> 01:05.080]  two different attack vectors, specifically because in option one of 3D printing a key,
[01:05.080 --> 01:10.500]  you never need access to that key. And with option two, we can go a little bit further
[01:10.500 --> 01:15.940]  with what we can do on complexity of keys. You're not limited to the key type, we'll say.
[01:17.520 --> 01:21.160]  So, first we're going to cover 3D printing, and it's way easier than you think.
[01:21.580 --> 01:24.540]  There's a few things you're going to need, obviously, which is a 3D printer,
[01:25.180 --> 01:28.720]  the material you need, and I'm going to go through that entire list now of what you need.
[01:29.480 --> 01:33.480]  So, first of all, you're going to need a 3D printer. The 3D printers I work with are an
[01:33.480 --> 01:39.900]  Ultimaker 2 Plus, and I also have a Prusa i3 MK3S. These are just the printers I own,
[01:39.900 --> 01:45.120]  that's why I use them. If you have your own printer, it should work well. Some lower end
[01:45.120 --> 01:51.140]  printers, sub 200, sub 100 printers might have problems with this. But if you're experienced
[01:51.140 --> 01:56.160]  with 3D printing, and you know how to really dial your settings in and get it configured
[01:56.160 --> 02:01.000]  just properly, you might have success with even some lower end, cheaper printers.
[02:02.480 --> 02:07.340]  A lot of questions I get about the 3D printers are, what size nozzle diameter do I use? It's a
[02:07.340 --> 02:13.800]  0.4 millimeter nozzle, and I normally print at 0.1 millimeter layer heights. You could get away
[02:13.800 --> 02:18.000]  with 0.2 millimeter layer heights, I just go for the 0.1, especially if you're doing higher security
[02:18.000 --> 02:26.180]  keys. And I use PLA plastic all the time. You could use something stronger if you wanted to,
[02:26.180 --> 02:31.220]  but PLA works perfectly for what we're doing here. The next thing you're going to need is
[02:31.220 --> 02:37.740]  the key you want to duplicate. Now, you absolutely do not need to have access to this key. I can't
[02:37.740 --> 02:42.180]  be the only person who has a friend on Facebook or other social media who is not security conscious,
[02:42.800 --> 02:46.880]  and they're new homeowners or something along those lines. And you see this picture where
[02:46.880 --> 02:52.720]  they're taking a big smiley picture and holding their keys up. If you get the right angle on that
[02:52.720 --> 03:00.120]  key, you can take that photograph and use it to duplicate a key. If you are a real world pen
[03:00.120 --> 03:06.420]  tester, and you can get a photograph of a key, for example, if you have the ability to walk up to a
[03:06.420 --> 03:11.480]  security desk, and on the background of the wall, there's keys hanging there, and you can get a
[03:11.480 --> 03:16.480]  photo of those. That's all you need. If you can get a photo and you can identify the type of key,
[03:16.880 --> 03:22.420]  you can use those to duplicate the key from a photo, which is a really cool attack vector
[03:22.960 --> 03:30.180]  compared to casting and molding a key, because that you do require access to the key for at
[03:30.180 --> 03:35.020]  least a short period of time. For the examples I'm going to show today, I just did a basic
[03:35.020 --> 03:44.860]  quick set KW1 key way. And then you're going to need a measuring tool. This isn't 100% necessary,
[03:45.660 --> 03:51.520]  but it makes life a lot easier, which is a key gauge. If you have physical access to that key,
[03:51.640 --> 03:55.720]  a key gauge lets you very quickly just move the key through the key gauge to get the
[03:56.400 --> 04:00.680]  bidding depths, the cut depths of that key. So you can quickly just know, okay, that's a one,
[04:00.680 --> 04:03.640]  that's a two, that's three, so on and so forth, depending on your key type.
[04:04.440 --> 04:08.440]  When you're working with photos of a key, though, you're going to need a pair of calipers.
[04:08.580 --> 04:13.120]  Calipers make your life super simple. You can get a cheap pair for like $20 or less on amazon.com.
[04:13.120 --> 04:18.060]  And digital calipers are... I prefer digital calipers because they just read off very quickly.
[04:20.120 --> 04:24.940]  And then if you're going to do photos, I highly suggest taking a look at Deviant Olaf's
[04:24.940 --> 04:30.740]  GitHub repository. He has a repository called Decoding. And on here, he has photographs for
[04:30.740 --> 04:34.660]  all these different key measurements. If you look on the screen, you can see just a couple like
[04:34.660 --> 04:41.520]  American, MasterQuickSet, Schlage, Sargent, EL, a ton of them are up there. It's a very awesome
[04:41.520 --> 04:48.740]  resource for decoding keys. Here's an example of a QuickSet one he has. It shows the entire key
[04:49.420 --> 04:56.860]  from tip to bow. And on that, it shows the spacing you need for the cuts. And then on the bottom
[04:56.860 --> 05:03.680]  right hand side, it shows the depths you need to make for each of the depths. So one depth
[05:03.680 --> 05:09.440]  is 8.36 millimeters, so on and so forth, all the way across. So it's very cool that you can
[05:09.440 --> 05:15.660]  use this for a reference. I'm not going to cover filing keys down, but this would be very good for
[05:15.660 --> 05:21.780]  that kind of reference if you're measuring to have that information. He also has photos on his
[05:21.780 --> 05:27.280]  repository that look like this. And where these come in use is when you have a photo of a key.
[05:27.540 --> 05:33.020]  So for this example, for this presentation, I just went out and searched QuickSet key on Google
[05:33.020 --> 05:37.100]  Images. And I got this key, I brought it into Photoshop. Now you can use any photo editing
[05:37.100 --> 05:41.780]  software you want. I happen to have Photoshop, but you can use GIMP or anything else. And what
[05:41.780 --> 05:47.280]  you do is you bring the picture in of the key and then you cut out the key profile for the most part
[05:47.940 --> 05:54.640]  and you scale it down to fit on the photo here that you get from his repository.
[05:54.880 --> 06:01.740]  So here's a picture that I brought in and scaled down off Google Images. Now it may not 100% line
[06:01.740 --> 06:05.800]  up all the time, depending on how much you fiddle with it. But the good news is that it doesn't have
[06:05.800 --> 06:10.700]  to be that precise to get a basic idea of your depths. So if you start at the bow of the key,
[06:10.700 --> 06:14.690]  what someone might call the back or the handle, and we work our way to the tip of the key,
[06:15.960 --> 06:21.340]  we can see here that the first depth cut is about a seven. Now I don't see the line for the seven,
[06:21.340 --> 06:25.500]  but I can see it's shorter than the six. So it's easy for me to assume that's a seven cut.
[06:25.520 --> 06:30.200]  And if I move to the next cut there, we can see that that's a five cut. The one next to it's a
[06:30.200 --> 06:35.500]  five cut. We jump up to a four cut, and then we go back down to a five cut. So we have the code
[06:35.500 --> 06:44.120]  for this, which is 75545. This is a horribly bitted key, except for that far back deep cut
[06:44.120 --> 06:51.360]  on the seven. But now that I have this information, what can I do with this? This is a great way to
[06:51.360 --> 06:58.000]  get a photo and get the depth cuts, but how do I take this and model a key? So a lot of people
[06:58.000 --> 07:04.900]  don't have the skill or ability to sit there and get the warding cut into a key and the proper size,
[07:04.900 --> 07:09.300]  height, and all this, the right angles. That would take a lot of work and using something like
[07:09.300 --> 07:16.360]  Fusion 360, AutoCAD, SolidWorks, whatever you want to use. Well, that's where a great talk that I
[07:16.360 --> 07:21.620]  believe happened at DEF CON 21 occurred. In the credits, I'll have a link to that talk. That's
[07:21.620 --> 07:25.840]  on YouTube, I believe, and all the people who are part of that talk. But it was something along
[07:25.840 --> 07:32.440]  the lines of key duplication and defeating high security Primus locks.
[07:33.960 --> 07:40.600]  And the people who created that created a software called Keygen, and they launched a website called
[07:40.600 --> 07:47.180]  keygen.co. This website is absolutely amazing. In your type, you select the type of key you're
[07:47.180 --> 07:52.600]  working with. Your outline might be like, if it's a five pin, six pin, seven pin, whatever,
[07:52.600 --> 07:57.160]  master lock, padlock, a four pin. And then your warding type, depending on which key you're using,
[07:57.160 --> 08:01.880]  might have a different warding. And then using the previous slide, we figured out the cuts,
[08:01.880 --> 08:06.480]  like a seven, five, five, four, five. You just simply enter that number combination into the
[08:06.480 --> 08:12.680]  bidding and hit generate. And this generates an STL file. For those of you not familiar with 3D
[08:12.680 --> 08:19.880]  printing, an STL file is the base file you use as your 3D model to generate your 3D prints.
[08:20.680 --> 08:26.520]  So here's a quick video demo of me using the site. This is not the key I showed you previously.
[08:26.520 --> 08:29.840]  It's just a demo real quick, but this is a Kwikset. On Kwikset, they only have the KW1
[08:29.840 --> 08:35.760]  keyway and warding. And this is a bidding I entered for a test key I have for an actual lock I own.
[08:37.020 --> 08:42.440]  It's 25363. I hit generate. It takes a couple seconds here, and then it generates this beautiful
[08:42.440 --> 08:47.060]  3D render of the key. I mean, they even have the end of the key to match what it would actually
[08:47.060 --> 08:52.720]  look like on the key. You hit download STL. It downloads the STL file to your computer.
[08:54.040 --> 08:59.720]  And then I opened this up real quick in Windows 3D file viewer. They call it print 3D. And this
[08:59.720 --> 09:06.580]  is just a basic idea of how it imports. It's an absolute perfect replication of a Kwikset key.
[09:07.320 --> 09:12.560]  At the time, I was printing this on an Ultimaker 2 Plus, and the slicer software I used for that
[09:12.560 --> 09:16.540]  is called Cura. If you don't know what a slicer software is, it's a software that
[09:17.660 --> 09:26.140]  3D printers print on layers. This slices the STL file into layers so that the 3D printer can print.
[09:26.140 --> 09:30.100]  Now, I don't print vertically like this because it makes it very unstable and weak,
[09:30.100 --> 09:34.660]  and it wobbles as it prints. So the first thing I do is I rotate it flat down to the bed.
[09:35.360 --> 09:38.940]  And in this preview, you're going to see this big ring around the outside.
[09:38.960 --> 09:42.980]  That's called a brim. It just helps bed adhesion. When I actually go to print it,
[09:42.980 --> 09:48.300]  I don't use that brim. I do print with supports depending on the warding,
[09:48.300 --> 09:54.820]  but for the most point, you don't need to use even supports on that. So here's just a quick
[09:54.820 --> 09:57.600]  time lapse. I'll probably skip ahead. I'm not going to watch the whole thing on this
[09:58.360 --> 10:01.640]  of my Ultimaker printing that exact key we just showed and modeled.
[10:04.610 --> 10:08.510]  So if you're not familiar with 3D printing, this is just a quick overview of what that looks like.
[10:13.810 --> 10:17.750]  The gap you see on the bottom is part of the warding. I didn't do supports on this one
[10:17.750 --> 10:25.840]  because I didn't need them. And you can see towards the top there, there's a little bit
[10:25.840 --> 10:29.540]  of stringing between the cut. That's simple. When I'm done, I usually take an X-Acto knife
[10:29.540 --> 10:34.280]  and just clean that up. It doesn't really affect a quick little cut with an X-Acto knife. It cleans
[10:34.280 --> 10:43.490]  up any stringing or weird little print artifacts that you may receive on the print. So this gives
[10:43.490 --> 10:47.350]  you the general idea. We don't need to watch the whole print. On average, this print takes between
[10:47.350 --> 10:51.610]  10 to 15 minutes, depending on your printer, but it takes about 10 to 15 minutes to knock
[10:51.610 --> 10:59.350]  out an entire key print. So this is the Kwikset lock I modeled, or that I modeled the key off of.
[10:59.710 --> 11:04.490]  This is the key I printed. You can see I stopped it early. That's why it's cut off on top. I insert
[11:04.490 --> 11:10.290]  it in and it turns the key, or it turns the lock, opens the lock. One thing to keep in mind here
[11:10.290 --> 11:17.310]  is that this is a deadbolt and this deadbolt is not attached to the actual bolt, right?
[11:17.890 --> 11:22.650]  So a 3D printed key will twist this open because there's nothing, there's no resistance from the
[11:22.650 --> 11:27.070]  bolt having to be slid forward and backwards. One thing I could suggest if you're worried
[11:27.070 --> 11:31.750]  about snapping the key off in a lock, is you can either use a tension wrench at the bottom,
[11:31.750 --> 11:35.310]  trim a little bit off the bottom of the 3D printed key with an X-Acto knife, and slide
[11:35.550 --> 11:42.830]  a thin tension wrench in, or a small screwdriver to assist the turn. Because when you actually
[11:42.830 --> 11:46.510]  throw a bolt, you're going to have a lot more weight behind it and it's going to be a little
[11:46.510 --> 11:51.030]  more difficult to turn, and you don't want to bend or snap a 3D plastic key off inside the
[11:51.030 --> 11:58.230]  deadbolt lock. So that's just something to keep in mind. So I had this key, but what now? I have a
[11:58.230 --> 12:05.630]  3D printed key. 3D printed keys are not durable. They will not last. So I was walking through my
[12:05.630 --> 12:10.850]  local hardware store, Mine's Menards, and I saw the KeyMe machine. And I'm not going to get into
[12:10.850 --> 12:17.050]  how much I hate this company and the ridiculous security options they offer.
[12:17.890 --> 12:22.650]  And by ridiculous, I mean they don't have them. So I printed two keys on this machine. And what
[12:22.650 --> 12:27.090]  this is, is if you have a key, it'll duplicate your key for you unmanned. And you can see I'm
[12:27.090 --> 12:31.430]  entering... this is a Schlage. It's a totally different key. I was testing it another time.
[12:32.390 --> 12:39.410]  It was a white filament. And we assume they use optics to scan this key for depths.
[12:39.410 --> 12:43.550]  And it misread the key. You'll see here in a second that it misread the first key,
[12:43.550 --> 12:49.790]  thinking it was something that it's not. And it failed. I had also printed this in a black or a
[12:49.790 --> 12:54.850]  dark, dark gray PLA as well. And I'll show you that here in a second and how that works. So you
[12:54.850 --> 12:58.970]  can see it says here, this is a less common key. It thinks it's some weird Schlage, which it
[12:58.970 --> 13:04.950]  absolutely is not. So at this point, I was confused why it wasn't working. So I canceled out of this
[13:04.950 --> 13:14.880]  and I decided I'm going to try again. And at this point, I went ahead and pulled the white key out
[13:14.880 --> 13:18.940]  and I put the darker key in. And it definitely helps. So I would suggest if you're going to take
[13:18.940 --> 13:23.900]  this attack vector to the point where you want to try to get a real key made from it,
[13:23.900 --> 13:28.100]  that you use a darker PLA. Don't use white. Something that's going to be super reflective.
[13:28.100 --> 13:34.080]  So here's the darker, dark gray key. I'm being very gentle shoving it in because it's got spring
[13:34.080 --> 13:40.720]  loaded shutters on it and I didn't want to snap it off inside there. And yes, you did see that
[13:40.720 --> 13:45.640]  correct. That is an NFC reader, RFID reader on one side and it's got a fingerprint scanner on the
[13:45.640 --> 13:53.580]  other side. The fingerprint scanner's purpose, as horrible as this is, is that you can save all your
[13:53.580 --> 13:58.960]  key profiles in the cloud. So if you don't have your key with you, you can always come back and
[13:58.960 --> 14:05.980]  later. I don't want to get into the security implications of that. So you see that it's $1.99
[14:06.880 --> 14:12.440]  to get a key duplicated here. Standard brass. If you'd like, you can get your Mickey Mouse or Darth
[14:12.440 --> 14:18.520]  Vader key, but this is for testing purposes. So after tax, it was $2.12. This is the part
[14:18.520 --> 14:22.920]  where I turn the camera away as I insert my credit card to pay for this transaction.
[14:28.090 --> 14:31.110]  Sorry, this video is a bit of a lengthy video because it took me a minute.
[14:31.730 --> 14:35.830]  Before they had this KeyMe, they used to have a machine called MinuteKey. And I liked the MinuteKey.
[14:35.830 --> 14:39.930]  It worked well. It was very simple. There wasn't all this extra save your data, all this different
[14:39.930 --> 14:44.990]  crap. They're trying to do an upsell here. I get a secondary key here for a few bucks cheaper.
[14:45.410 --> 14:48.270]  I'm like, no, I'm good. I just need to test to make sure this works. And then they want your
[14:48.270 --> 14:54.650]  email address so they can email your receipt, aka mark it to you. So of course, you know me,
[14:54.650 --> 15:09.530]  I enter my real email address and they want to know if you want to save your key for later.
[15:09.530 --> 15:16.770]  That's where the fingerprint comes in. Hell no. Moving on. So it starts going down this path
[15:16.770 --> 15:20.690]  where it starts updating you. So here's the concept. You have a 3D printed key that is not
[15:20.690 --> 15:28.330]  durable. However, I can take a 3D printed key, go to an unmanned key duplication system in the very
[15:28.330 --> 15:34.930]  back of a hardware store that is in my local town. And I can duplicate this key. Now you saw it does
[15:34.930 --> 15:40.990]  have the ability to do higher security keys than your standard Schlage or Kwikset or less common
[15:40.990 --> 15:45.890]  keys, we'll call it. To do that though, they do ask for a mailing address because they cut it and
[15:45.890 --> 15:49.910]  then they mail it to you. So that's something to keep in mind as well. But if you have a standard
[15:49.910 --> 15:56.150]  Schlage, a standard Kwikset or other master lock standard keys that are what this company considers
[15:56.150 --> 16:02.730]  common, it's going to cut this key in front of you and then dispense it out from the machine.
[16:02.730 --> 16:10.790]  And again, I have to definitely apologize for the time of this video.
[16:12.230 --> 16:16.170]  The other thing is it says that it does have an Android and iOS app that it can scan your
[16:16.170 --> 16:20.650]  keys in home. I have not tested this, but I'm planning on it where I'm wondering if I'm able
[16:20.650 --> 16:27.050]  to just set my key down on a desk and the app will scan it. So I'm going to be testing that in the
[16:27.050 --> 16:32.130]  because if you could do that, if you had access to a physical key, it'd be set.
[16:32.130 --> 16:37.070]  Okay, so the key's done cutting. You can see it does an absolutely horrid job. You can see little
[16:37.070 --> 16:44.670]  pieces of flashing sticking off where it cut it. But nonetheless, you now have a real physical
[16:44.670 --> 16:50.810]  metal duplicated brass key of something you just 3D printed, which is awesome.
[16:50.810 --> 16:58.250]  So that concludes the 3D printing and cutting a key. Like I said, it's very simple. The hardest
[16:58.250 --> 17:03.590]  part is... well, let me rephrase that. If you have access to the key, this is dead simple.
[17:03.610 --> 17:08.830]  You get the measurements with the key gauge, you type it into keygen.co, you spit out an STL file
[17:08.830 --> 17:12.550]  and you print it and you're done. You have a 3D printed key. You can take that to an unmanned
[17:12.550 --> 17:17.990]  key duplication because if you go to a hardware store and ask a real person to cut this key,
[17:17.990 --> 17:21.630]  you might get a weird look. I'm considering trying that just to see what happens if they'll
[17:21.630 --> 17:28.730]  even do it. But you'll have a physical key if you can go to an unmanned key duplication center.
[17:29.410 --> 17:32.410]  So now we're going to go into molding and casting a key. And this is where you start
[17:32.410 --> 17:37.810]  to feel a little more like James Bond or a spy in this case, because you're starting to do a
[17:37.810 --> 17:42.210]  little more covert, see it in the movies kind of stuff. And the cool part is you get to play
[17:42.210 --> 17:47.470]  with fire. So where did I get the idea to try this? So I've been wanting to try key duplication
[17:47.470 --> 17:54.370]  for a while now. But I'd always seen every video, you need a clamshell, you need something to put
[17:54.370 --> 18:00.310]  the clay in to mold this key in. And they're always milled or CNC'd out of like an aluminum
[18:00.310 --> 18:05.530]  or some kind of metal. And I didn't have access to that. I live in a very small city, like population
[18:05.530 --> 18:11.950]  10,000 people. And I didn't know anyone. And if you were to send this off to be done, it would
[18:11.950 --> 18:16.870]  be kind of a high cost. And then Devian Olof put out a video on his YouTube channel, which if you're
[18:16.870 --> 18:22.730]  not subscribing, I don't know why you absolutely should be subscribing to his videos. He put
[18:22.730 --> 18:29.790]  out a video and he talked about how you could absolutely 3D print this. So I went out and
[18:29.790 --> 18:35.430]  searched Thingiverse for a 3D printable clamshell. And lo and behold, I found one. And I'll get into
[18:35.430 --> 18:41.950]  who that's by here in a second. So once I saw that, Deviant talks about every little thing you need
[18:41.950 --> 18:47.550]  to cast a key. He goes step-by-step process on how to do this. So I watched his video,
[18:47.550 --> 18:51.530]  took inspiration from it and tested it myself. And here's the link at the bottom.
[18:51.530 --> 18:55.070]  If you're seeing that to his original key molding and casting video.
[18:56.730 --> 19:01.310]  If you don't know who Devian Olof is, there's this beautiful mug right there.
[19:01.310 --> 19:07.150]  His website is deviating.net, Devian Olof at Twitter. Check him out on Twitter for sure.
[19:08.290 --> 19:13.330]  So we're going to go over everything you need. There's a much higher bill of materials on this.
[19:13.530 --> 19:16.890]  So let's go through them one by one. I try to give you an idea of where you can get them off Amazon
[19:16.890 --> 19:24.530]  pretty cheap and or the prices you can expect to pay. So in this case, you are going to need
[19:26.310 --> 19:31.810]  physical access to the key you're wanting to duplicate. If you prepare ahead of time,
[19:31.810 --> 19:37.290]  I'll talk about in a second, you don't need much time. You maybe need like
[19:37.970 --> 19:43.010]  anywhere between 15 to 30 seconds hands on with that key.
[19:44.830 --> 19:51.550]  Depending on what you're allowed to do on a job or your story for your job,
[19:51.550 --> 19:57.370]  you could get access to this key for like requesting access to a server room or wherever
[19:57.370 --> 20:03.270]  you're trying to access specifically. Had you need that key, you may have just enough time to
[20:03.270 --> 20:10.610]  get the mold of that key. And again, I did this all with a Kwikset KW1 key. Then you're going to
[20:10.610 --> 20:18.230]  need a low melting point metal. So I do this demonstration at a lot of cons, and some of
[20:18.230 --> 20:23.710]  these cons have kids. One metal that Deviant talks about, and if you talk to other people,
[20:23.710 --> 20:27.990]  they suggest is called woods metal. The problem with woods metal is that it has
[20:27.990 --> 20:32.850]  lead in it and other toxic chemicals that when burnt, the fumes rise up and you're breathing
[20:32.850 --> 20:39.570]  these in. Woods metal is relatively cheap. You can get a lot of it for a very low cost.
[20:39.750 --> 20:43.510]  However, because I was going to have kids, A, touching the metal after they're done,
[20:43.510 --> 20:47.150]  melting it themselves, all this different stuff, I didn't feel comfortable using a woods metal
[20:47.150 --> 20:53.030]  because of the toxins. And talking to some people in the tool slack, they suggested fields metal.
[20:53.730 --> 21:00.690]  Fields metal is much more expensive by weight. I was able to get a small little brick of it,
[21:00.690 --> 21:07.730]  probably about two and a half to three inches by three inches by quarter of an inch thick for about
[21:07.730 --> 21:13.910]  35 bucks off eBay. That's the exact link to the one I buy. This is a reputable buyer. He's got
[21:13.970 --> 21:19.190]  a hundred percent positive feedback as of this presentation, and he's always sent me the products
[21:19.190 --> 21:28.790]  I've needed. I did a con in Omaha, Nebraska, and at that con, I ordered five different bricks,
[21:28.790 --> 21:31.810]  and he got them to me, and he was willing to work with me on price, which I thought was awesome.
[21:32.010 --> 21:36.050]  So this is where I get my metal. It's fields metal. It's very low point metal. Like the
[21:36.050 --> 21:42.470]  picture shows, it melts at about 144 degrees Fahrenheit, and it doesn't have the toxins in
[21:42.470 --> 21:47.510]  it. And then you're going to need something to melt that brick you get down so you can make
[21:47.510 --> 21:53.770]  ingots. Deviant has some really cool big melting pot that he bought. I looked it up on Amazon. I
[21:53.770 --> 21:57.350]  think it was like 35 or 40 bucks. He might've gotten it cheaper somewhere else. But I was
[21:57.350 --> 22:01.790]  walking through Walmart one day in the camping section, and I saw this cast iron skillet,
[22:01.790 --> 22:06.090]  and it was like five bucks. So I was willing to drop five bucks on a cast iron skillet,
[22:06.090 --> 22:09.330]  which I bought. Obviously, I'll only use it for this purpose moving forward.
[22:10.370 --> 22:20.370]  And then Deviant had these perfectly square, small ingots of his metal. And it just so happened that this
[22:20.370 --> 22:27.650]  ingot was exactly the right amount of metal to mold and cast one single key. So I reached out to him
[22:27.650 --> 22:32.850]  on, I can't remember if it was Twitter or Slack, and talked to him. And he's like, oh dude, it's
[22:32.850 --> 22:38.350]  awesome. And there are these ice cube trays. So these are these silicone rubber ice cube trays
[22:38.350 --> 22:41.770]  that are meant to make tiny ice cubes. You can drop them in like in the mouth of a bottle of
[22:41.770 --> 22:47.990]  water or soda. And they're super tiny. You can get them off Amazon and you get two trays, which
[22:47.990 --> 22:53.070]  make a ton of ice cubes or ingots in this case. And they're 10 bucks for both of them. Again,
[22:53.070 --> 22:57.470]  these prices might be different now, especially with COVID going on. They could have completely
[22:57.470 --> 23:02.190]  changed because all prices and availability on Amazon are off the charts compared to what they
[23:02.190 --> 23:07.010]  used to be. But I got two of them for 10 bucks. So you can either use them both or use one for
[23:07.010 --> 23:12.110]  key casting and one for your ice cubes. Then you're going to need the clay. And the clay is
[23:12.110 --> 23:20.910]  what you actually mold the shape and the indent of the key into. And Sculpey is something that
[23:20.910 --> 23:28.330]  Deviant recommended. One tip, when you search Sculpey clay on Amazon, select all the colors
[23:28.330 --> 23:34.750]  available. The reason I say this is that the price varies based on the color you select.
[23:34.750 --> 23:39.590]  I don't know. I don't get it. I've gotten yellow and green so far because they're usually between
[23:39.590 --> 23:46.830]  two to $3. But if you select black or white or orange, it might be $7 to $10 for the same size,
[23:46.830 --> 23:51.690]  the two ounce size I buy here. This two ounce size, you can see from the picture,
[23:51.690 --> 23:57.930]  there's indents along each of the thing. And pretty much it's got four logs of clay.
[23:57.930 --> 24:06.970]  One log of clay is exactly what you need to cast a key. So you get a minimum of four castings.
[24:06.970 --> 24:13.510]  I found that you can reuse this clay over and over again. It does dry out over time.
[24:13.610 --> 24:18.810]  It is heat set. It's oven baked clay. So heat sets it and makes it hardened. So you're not
[24:18.810 --> 24:22.990]  going to get 100% reusability over and over again. And your results might vary. But I've
[24:22.990 --> 24:28.150]  gotten four or five keys out of a single piece of clay. But again, it all depends on how hot
[24:28.150 --> 24:33.990]  you heat your metal to before you pour it in. And then you're gonna need some kind of talcum
[24:33.990 --> 24:39.510]  powder. I bought mine at Walmart. I bought the little baby powder. It's a little size about this
[24:39.510 --> 24:43.510]  big. It's a travel size baby powder. And it's perfect. And this just stops the clay from
[24:43.510 --> 24:47.990]  sticking to itself. So when you cast the two key or take the two pieces of the clamshell and stick
[24:47.990 --> 24:51.650]  them together, when you pull them apart, they separate without the clay sticking together and
[24:51.650 --> 24:58.870]  ruining your whole mold. Then you're gonna need the clamshell. I 3D printed mine from Thingiverse.
[24:59.670 --> 25:06.030]  ZBZZN Boris is the person who created this. There's a bitly link I created to link straight
[25:06.030 --> 25:13.410]  to the Thingiverse link for that. It works absolutely perfect. It's three pieces. You're
[25:13.410 --> 25:18.250]  gonna have a left and a right to your clamshell. And then a base. And this base is awesome. In
[25:18.250 --> 25:21.750]  Deviant's video, he had an aluminum one, I believe. And then he used like a C-clamp
[25:22.650 --> 25:27.330]  to clamp it to hold it together and like kind of stabilize it on the table.
[25:27.330 --> 25:32.050]  This base you print for it works absolutely perfect. I love this clamshell. I've used it over
[25:32.250 --> 25:38.570]  a hundred times. And there's no deformation. It absolutely does not deform. It holds up to the
[25:38.570 --> 25:46.230]  heat. You have to realize that when you print PLA, you print at a minimum usually of 190 degrees
[25:46.230 --> 25:55.010]  Celsius to 220 degrees Celsius. And we're only hitting around 140 to 160 degrees Fahrenheit.
[25:55.170 --> 25:59.790]  Now, unfortunately, I use American units. I wish we'd switched to metric and Celsius,
[25:59.790 --> 26:03.910]  but I don't understand it off the top of my head. So you could do the conversion. But the point is
[26:03.910 --> 26:08.550]  you never get close to that melting temperature or the glass temperature of the filament. So
[26:08.550 --> 26:12.870]  you're gonna be absolutely fine. Then you're gonna need a melting ladle. This thing was like
[26:12.870 --> 26:18.350]  10 bucks off Amazon. It's for melting lead, you know, casting ammunition, things along those lines.
[26:18.450 --> 26:23.150]  It was 10 bucks on Amazon. It fits a perfect ingot inside of it. And you can hold a lighter
[26:23.150 --> 26:32.550]  underneath it to melt that down. Then you're gonna need a lighter. Deviant suggested a torch lighter.
[26:32.550 --> 26:37.030]  I do as well. I think it puts out a little more heat than just a standard lighter. But I've tested
[26:37.030 --> 26:41.450]  with fields metal. You can use like a tea candle or a standard Bic lighter or a Zippo lighter. It
[26:41.450 --> 26:47.070]  will get it hot enough over time. It's just not as fast. If you're like me and you like a cigar
[26:47.070 --> 26:51.790]  renown again, I already own a cigar lighter. I bought this one off Amazon for 15 bucks like
[26:52.230 --> 26:57.130]  years ago for the purpose of actually having cigars. It just so happened to work perfectly
[26:57.130 --> 27:02.850]  for this situation as well. So this is a video of me trying not to upset my wife.
[27:03.610 --> 27:09.410]  This is the five dollar cast iron skillet I bought from Walmart with the brick of metal of
[27:09.410 --> 27:15.050]  fields metal I bought off of eBay. I'm slowly just melting it down inside this pan.
[27:15.810 --> 27:22.310]  I make a very noob mistake here. Do not make my mistake. I took a like a kebab skewer
[27:22.310 --> 27:27.050]  and I was making sure it was fully melted and kind of stirring it around to make sure there
[27:27.050 --> 27:31.810]  was no more chunks of metal that it was 100% completely melted on my stove. And then to make
[27:31.810 --> 27:37.390]  sure I, like an idiot, swirl it around in the pan like you would with butter if you're about to fry
[27:37.390 --> 27:43.910]  steak. And what I do is I just coat my entire pan with fields metal. So I lost a bit of metal
[27:43.910 --> 27:50.670]  because of that. But like I said, I'm not using this pan for anything but melting fields metal
[27:50.670 --> 27:55.930]  in the future. So it works perfectly. I put the wax paper underneath the ice cube tray in case I
[27:55.930 --> 28:00.310]  spilt. That's the only person I have that there. If I did an overspill, it would land on the wax
[28:00.310 --> 28:05.210]  paper and not my stove and my wife would still love me. So I quickly just pour this in here.
[28:05.750 --> 28:10.690]  And you can see that I got a whole bunch of full cube pours and then a whole bunch of random ones.
[28:11.530 --> 28:15.150]  And then you can see the cubes it makes. I think I got about 16 cubes out of that one piece of
[28:15.150 --> 28:21.070]  metal. So that's enough for 16 keys. One note, when I poured into this ice cube tray, it doesn't
[28:21.070 --> 28:26.030]  fully fill all the cubes. So what I did is I popped them all out. Anything that was a full cube,
[28:26.030 --> 28:31.790]  I set aside. Anything that was a partial cube, I took into my melting ladle. I melted it down,
[28:31.790 --> 28:36.270]  all the little fragments, little pieces, and I slowly poured them in cube by cube
[28:36.270 --> 28:40.570]  until I got a full ingot for each one. So that's just a little pointer there.
[28:41.950 --> 28:49.010]  So now we're going to switch to demo time, aka time to play with fire. Don't try this at home
[28:49.550 --> 28:55.430]  or if you do, just don't blame me if you get hurt. I highly suggest you try this at home.
[28:56.310 --> 29:02.390]  Full disclosure, I've had this demo pre-recorded for a time because it's one of those things that
[29:02.390 --> 29:10.130]  with COVID going on, I've done a lot of virtual conferences. And with a virtual conference,
[29:10.130 --> 29:14.530]  it's just difficult to plan ahead. So I pre-recorded the actual demonstration of this video.
[29:14.990 --> 29:19.370]  So you're going to see here, the background is going to change. My hat won't, you won't see my
[29:19.370 --> 29:23.330]  face. It's going to be all on table kind of thing, but you're going to see the whole demonstration.
[29:23.330 --> 29:27.710]  So we're going to do that demonstration and then I'll come back right after that. Enjoy.
[29:28.390 --> 29:31.650]  Okay, so I'm going to get started with my demo here. I'm going to go over what I have here on
[29:31.650 --> 29:38.170]  the table. So some of the things you're going to need. The first thing is a clamshell. This is
[29:38.170 --> 29:43.290]  what we're going to use to actually mold the key. This is actually 3D printed. It's fully available
[29:43.290 --> 29:53.310]  for free download off thingiverse.com. And I'll try to make sure the links are available. They're
[29:53.310 --> 29:58.210]  on the base. The base is pretty much so you don't have to hold this as you're pouring molten metal
[29:58.210 --> 30:04.650]  into it. It just lets you freehand it. Speaking of pouring molten metal, the next thing we have
[30:04.650 --> 30:09.890]  here is a ladle. And this is a lead ladle. You could use this for pouring lead, but it works
[30:09.890 --> 30:15.830]  perfectly for what we're doing. So we'll be pouring the metal in to the clamshell like that.
[30:16.550 --> 30:21.950]  The next thing we're going to need is some sculpting clay. This is Sculpty. It's like
[30:22.230 --> 30:28.650]  a few bucks on Amazon for this package. This package is pre-split into four different bars.
[30:29.290 --> 30:34.550]  And just one bar is exactly perfect to fill this mold. You rip that one bar in half and you fill
[30:34.550 --> 30:39.810]  each half of the mold. It works perfect. You're going to need some baby powder, some talcum
[30:39.810 --> 30:45.290]  powder. I buy this nice little jar or nice little bottle from Walmart. It's like a buck something,
[30:45.290 --> 30:53.090]  just because it's small and convenient. And I travel with all this stuff in this like pencil
[30:53.090 --> 30:58.670]  bag I found one time at Walmart during school shopping time. It's like a dollar or two dollars
[30:58.670 --> 31:05.590]  for this zipper bag and it holds everything I need. You're going to need a lighter. I suggest
[31:05.930 --> 31:10.470]  a torture cigar lighter. This is a cigar lighter. You can't really see the flame on camera I'm
[31:10.470 --> 31:17.190]  guessing, but that's what that is. And then I also have the metal. So there's different kinds of
[31:17.190 --> 31:23.670]  metals you can use for this. A very common popular one is called woods metal. The problem with woods
[31:23.670 --> 31:29.490]  metal is that it contains lead, so it's toxic. The fumes it puts off are toxic. So I would highly
[31:29.490 --> 31:32.930]  suggest against it if you're going to be doing this around kids or in the public. For your own
[31:32.930 --> 31:38.070]  use, that's up to you. You decide if that's okay with you. This is Fields metal. Fields metal is
[31:38.070 --> 31:44.830]  non-toxic. I get mine off Amazon and it's like $35 for a brick. And one brick makes pretty much
[31:44.830 --> 31:50.590]  enough to fill this little sleeve here. So about 16 little cubes, which you saw in the previous video
[31:50.590 --> 31:57.670]  that I made melting it into a pan and then pouring it in this mini ice cube tray. So let's get into
[31:57.670 --> 32:04.310]  molding an actual key. So I have this lock. It's a Kwikset lock and here's the key for that lock.
[32:05.430 --> 32:09.130]  It works perfect. So this is the key we're going to mold.
[32:10.470 --> 32:13.070]  So to get started, I'm going to move some stuff out of the way here.
[32:14.390 --> 32:18.990]  First thing I'm going to do is get my clamshell opened up and laid flat in front of me and I'm
[32:18.990 --> 32:26.820]  going to open my clay. Now I try to open this nicely so that I can kind of wrap it back in
[32:26.820 --> 32:31.280]  the clay later. That's only because I'm doing it for demo purposes. This doesn't really matter.
[32:31.280 --> 32:36.720]  I've actually 3D printed a holder for this clay just because I like to be organized with my stuff,
[32:36.720 --> 32:41.260]  especially in 3D printing. So if you look at this again, you can see that there's,
[32:41.260 --> 32:44.640]  it's hard, maybe hard to see, but you can see there are these lines that pre-cuts into these
[32:44.640 --> 32:49.840]  bars. And this one bar is exactly the right shape. So if I take this bar and break it in half
[32:51.140 --> 32:57.920]  or relatively in half, like so, this is exactly perfect for each side. So just take one and
[32:58.460 --> 33:05.400]  mash it up, just like play-doh, soften it up just a little bit. And then what you're going to do,
[33:05.400 --> 33:08.520]  and it's up to you, some people like to take talcum powder and put this in so the clay comes
[33:08.520 --> 33:13.640]  out easier. For demo, I don't really care. If you needed to do multiple back-to-back,
[33:13.640 --> 33:18.120]  maybe that would make sense. Otherwise it doesn't matter. Anyways, I get into about a shape like
[33:18.120 --> 33:25.060]  that and then I just push it down into the clamshell and try to get as flat as possible.
[33:26.080 --> 33:32.140]  So I'm pushing it down in, back-filling this big gap at the top, the big void at the top.
[33:36.830 --> 33:43.110]  And you want it to be as level with the top of this as possible. You don't want a big
[33:43.110 --> 33:48.150]  bulge of clay sticking up because it won't give you a good mold. It'll wrap around the key, which
[33:48.150 --> 33:59.370]  you do not want. So get this as flat as flat as you can. Again, it doesn't have to be absolutely
[33:59.370 --> 34:03.930]  perfect. When you squish these two together, you're probably going to get some leak out. That's fine.
[34:03.930 --> 34:11.710]  Just get it as best as you can. So you can see I have it. It's pretty level. It's hard to see on
[34:11.710 --> 34:17.090]  camera, but it's pretty level. Now this part, you're going to want to depress this down in,
[34:17.090 --> 34:22.610]  like so, to create a well because you need some place to pour the molten metal into. So I'm
[34:22.610 --> 34:29.070]  creating like this lip that I can bend backwards. And it's just this well because the top of the
[34:29.070 --> 34:34.010]  doesn't need to be molded. That's not important. So pretty much you do this again and just repeat
[34:34.010 --> 34:41.490]  it on the other side. So I'm going to take the clay and again soften it up by squishing it.
[34:54.100 --> 34:59.400]  And you want to make sure that you go above the border here. You want that lump and then you want
[34:59.400 --> 35:04.380]  to press it down to flatten it out. Otherwise, if you try to skim it in, you might get an air
[35:04.380 --> 35:08.040]  pocket below. And when you squish the key in, all the clay moves down and you won't get a good
[35:08.040 --> 35:15.560]  impression. Okay, so I've pushed it all down in and I still have a bulge. So now I'm going to work
[35:15.560 --> 35:22.980]  on getting that bulge down so it's not above the plane of the top of the clamshell. And again,
[35:22.980 --> 35:34.340]  I'm going to fill this void in on the top. People have asked me if the color of clay matters. It
[35:34.340 --> 35:39.020]  doesn't. Literally, if you go on Amazon, one day one color is cheaper than the other color.
[35:39.820 --> 35:43.740]  So sometimes you'll see this in yellow. Sometimes you'll see it in green, black, red, blue. It
[35:43.740 --> 35:51.380]  doesn't matter. It is literally just whatever's the cheapest I buy. Because why pay $8 or $5 for
[35:51.380 --> 36:00.280]  red when green is $1.99. So okay, so I have that pretty well in there. I'm going to check
[36:00.280 --> 36:06.780]  to make sure I don't have it too too low. I think I'm good. Now I'm just gonna again create that
[36:06.780 --> 36:11.940]  well on the top of my thumb, squishing it back. So I create a lip and when I put these two together,
[36:11.940 --> 36:15.840]  it'll create a pore spout kind of like this that the metal can go down in.
[36:17.160 --> 36:22.120]  So now the next important part is to cover these two pieces or these two sides in talcum powder,
[36:22.120 --> 36:26.340]  so that when you push the two clays together, and you separate them that they separate and
[36:26.340 --> 36:30.720]  they don't just stick into one big glob of clay. So you just take a little talcum powder and
[36:30.720 --> 36:38.400]  sprinkle it on the sides here. And then just take your finger and rub that talcum powder across the
[36:38.400 --> 36:57.430]  top of all the clay. Just like that. Check this, it seems a little bulgy. Now that I've pressed
[36:57.430 --> 37:03.590]  it down, I'm going to put a little more powder on. Okay, so we have these both sides coated.
[37:03.750 --> 37:09.590]  They're good to go. So I'm done with the talcum powder. I'm done with the clay. So the next thing
[37:09.590 --> 37:13.250]  we're going to need here is we're going to need the key. Now it's important that when you press
[37:13.250 --> 37:19.770]  this down that you think about this key as having two halves. Have one, have two over here. You don't
[37:19.770 --> 37:25.690]  want to shove the whole key down 100% into one side. You want to push it down about halfway on
[37:25.690 --> 37:30.770]  one side so that when you put the other clamshell on top, half the keys on one clamshell and half
[37:30.770 --> 37:36.250]  the keys on the other side of the clamshell. So to start with, I'm just going to place this in here
[37:36.250 --> 37:42.210]  so that the bow of the key, which is right here,
[37:42.210 --> 37:48.950]  is make sure that that's in thick clay, not in the well. So I want it down far enough that we're
[37:48.950 --> 37:53.990]  good there. Okay, and then I'm going to press down and rock it back and forth, just trying to get it
[37:53.990 --> 38:01.070]  down about halfway in. And just lean in and check on it. Make sure you're about halfway in all the
[38:01.070 --> 38:07.350]  around. So now you can see the key stuck in halfway there. Now the next thing to do is take
[38:07.350 --> 38:14.550]  this clamshell, place it on top, and then squeeze these two pieces together. Now it's suggested that
[38:14.550 --> 38:18.130]  you stand on this, which I'm not going to do because you're not going to see it anyways.
[38:18.250 --> 38:25.050]  So I'm just going to place my hands on the table and press. You want to get a nice firm pressure
[38:25.050 --> 38:29.830]  on this and just stand on it. Put some weight on it. Give it a second for the clay to move out of
[38:29.830 --> 38:35.090]  the way for the key to come in. Because remember, you have the warding and all the parts of the key
[38:35.090 --> 38:42.110]  that you need to get up in. Okay, that should be good. Now you want to carefully open this,
[38:42.110 --> 38:45.490]  separate it apart. Don't rock it back and forth because you're gonna mess up the clay. Just try
[38:45.490 --> 38:51.430]  to get it apart as cleanly as possible, like so. And we have a pretty decent impression over here.
[38:51.430 --> 38:56.070]  It looks like the key sunk in a lot on this side, so I'm hoping this all turns out well.
[38:56.870 --> 39:01.230]  But you saw how the clay came apart without sticking. That's because we used the talcum powder.
[39:01.310 --> 39:07.410]  So I'm going to bend this lip back. I'm going to use the true method of flicking the key out.
[39:07.830 --> 39:11.570]  The reason we do this is we want to get it out as fast as possible because if you lift it up,
[39:11.570 --> 39:15.990]  it can cause the tip of the key to dig into it. So we want to get it out as fast as possible.
[39:17.550 --> 39:24.430]  So there, it's out. It shot across the room, but that's fine. So where do we go from here?
[39:24.430 --> 39:28.290]  So the next step to do is we're going to go ahead and take these two pieces of clamshell. We're
[39:28.290 --> 39:33.590]  going to put them back together. And now we have a void. Now I'm going to open up these lips on the
[39:33.590 --> 39:38.790]  top to make sure that I can see down into where we're about to pour, which I can see fine. So I'm
[39:38.790 --> 39:44.090]  going to place that into this holder. And this does two things. One, it keeps these two pieces
[39:44.090 --> 39:49.110]  tight together. And two, it makes it so I don't have to hold this in my hands as I pour molten
[39:49.110 --> 39:53.590]  metal over it. And if I bump it or miss, I'm not going to end up with melted metal all over my
[39:53.590 --> 40:01.790]  hands. So the next step I'm going to do is I'm going to take out one cube of my Fields metal.
[40:01.790 --> 40:08.050]  And again, this metal is meant to melt at very low melting points. So a candle or a lighter works
[40:08.050 --> 40:14.750]  fine. And one cube is almost absolutely perfect for your average key. So like a standard Schlage
[40:14.750 --> 40:22.690]  key or Kwikset key. So I'm going to drop this in the end of the ladle, like so. And I'm going to
[40:22.690 --> 40:27.410]  try to do my best here to let you see this melt and how fast it goes. But it's going to
[40:27.410 --> 40:35.770]  be kind of hard to do. So we'll do our best here. So I can already see it moving. So it's already
[40:35.770 --> 40:41.450]  starting to melt. And as soon as the metal below it starts melting, you get a little bit of liquid
[40:41.450 --> 40:46.270]  metal. That metal heats up the rest of the cube and it melts really quickly. So there it goes.
[40:47.050 --> 40:52.070]  So now I've got to kind of look at it here to verify. So it just hit completely liquid. I'm
[40:52.070 --> 40:55.070]  going to go a little bit longer to make sure it doesn't start solidifying on me quite yet.
[40:56.110 --> 40:59.710]  There we go. We're good to go. And now we just pour this metal
[41:04.790 --> 41:08.450]  down in. And I'm kind of doing this because I want it to get down in and I want any air
[41:08.450 --> 41:15.910]  bubbles to get out. But that should be good. And now we just let it sit. Now,
[41:15.910 --> 41:21.410]  depending on how long you heat this for... let me bring this in. This is silicone,
[41:21.410 --> 41:27.270]  so I can put something hot on it and it won't melt. Depending on how long you heat the metal
[41:27.270 --> 41:32.450]  for, it's going to take longer to cool. Now this takes quite a bit longer to cool because it's
[41:32.450 --> 41:37.890]  such a low melting point. If you use woods metal, the toxic stuff, it doesn't take as long to
[41:38.870 --> 41:45.050]  do. If you watch any of Deviant's videos, you'll see that he uses... his clamshell is actually...
[41:45.050 --> 41:49.830]  I think it's milled out of aluminum, which if you think about what aluminum is, it's a heat sink.
[41:50.310 --> 41:54.990]  So that would help awesome to cool the key faster. Where this is plastic, this isn't going
[41:54.990 --> 42:00.650]  as a heat sink. So if you can get a metal clamshell, great. Again, it's not needed. This is
[42:00.650 --> 42:05.830]  super simple to make. It's probably like a few dollars worth of plastic in this. So it works
[42:05.830 --> 42:09.430]  really well. You don't have to have metal if you've ever seen it. When I first started, that's
[42:09.430 --> 42:12.690]  why, again, that's what I thought. I thought, oh, I didn't have a metal clamshell. How am I supposed
[42:12.690 --> 42:19.870]  to try this? And when I saw the 3D printed one, it was fantastic. So I'm watching the top. I'm kind
[42:19.870 --> 42:25.110]  of shaking it, and I can see that the center is starting to solidify. But the sides are still
[42:25.110 --> 42:30.290]  jiggly. And I wish I could tip this to show you, but if I do, that metal might run out. So it doesn't
[42:30.290 --> 42:35.870]  have to be cooled down to take it out, but it does need to be solidified. So we're going to finish
[42:35.870 --> 42:40.490]  letting this go here. I'm just tapping it. And pretty much what I'm looking for is if the metal
[42:40.490 --> 42:45.410]  is still kind of jello-y on top, or you can see liquid still. And as soon as I don't see that
[42:45.410 --> 42:49.830]  anymore, then I'm pretty sure we're good to go. So there's just a few little spots that are still
[42:49.830 --> 42:59.830]  like that down in here. We're very close now. Some very, very small spots still jiggling, which means
[42:59.830 --> 43:04.410]  they're cooling off. And if you open this too early, and it's not solidified down, you're going
[43:04.410 --> 43:27.520]  to have molten metal just running out. So better to be safe than sorry. Okay, it looks pretty
[43:27.520 --> 43:33.460]  solidified. I'm going to give it a couple, like one more minute here, just to cool off a little bit
[43:33.460 --> 43:40.480]  to guarantee it's good all the way down. And when we pull this out, you may or may not see some gaps,
[43:40.480 --> 43:45.380]  some air gaps, that's fine. The key can still work. The important part, obviously, is the
[43:45.380 --> 43:50.340]  bidding of the key and the warding. But if there's holes in the warding, we're fine. If there's holes
[43:50.340 --> 43:54.320]  in the bidding, then we have problems because there might make a dip where it won't lift the
[43:54.320 --> 43:59.540]  pins to the proper height. Okay, I think we're good. So I'm gonna go ahead and take the base off this.
[44:00.980 --> 44:03.120]  Okay, and then we're going to split this open.
[44:05.120 --> 44:10.140]  And it looks like we have a good casting. So let's see, well, I've got to keep it in the light so you can
[44:10.140 --> 44:16.180]  see it here. So there's the key. And you can see, we just, we don't have the entire top of the head.
[44:16.220 --> 44:20.500]  We have just enough, though, that we can grab onto it and turn it. And we can see all the key in here,
[44:20.500 --> 44:26.940]  we can see the warding. So let's go ahead and peel this out. So we've got a little clay on the tip,
[44:26.940 --> 44:31.620]  that's not a big deal. Just pull it off. So you can see there's a lot of flashing around here and
[44:31.620 --> 44:35.760]  stuff like that. But again, none of this matters. And if that bothers you, it's so thin and soft.
[44:35.760 --> 44:43.060]  Just bend it back. Okay, now we have a little bit of flashing, it looks like, around this bidding area.
[44:44.340 --> 44:49.960]  But again, I'm not overly concerned with that because of the way the cut is. I can just kind
[44:49.960 --> 44:54.820]  of bend that flashing out of the way. So I think we're completely good to go here.
[44:55.220 --> 44:59.100]  So now it's the moment of truth. Because, you know, demos fail.
[45:00.180 --> 45:06.100]  What's new? So let's place it into the lock. It goes into the lock. That's good news. Now we'll
[45:06.100 --> 45:11.700]  open the lock. There it goes. So there was that little bit of flashing. So when I first started
[45:11.700 --> 45:15.940]  turning it, it caught. But as soon as I turned that flashing bent out of the way, we're good to go.
[45:15.940 --> 45:22.440]  So now we have a fully working duplicate of that key. So like when we talked about with 3D printing
[45:22.440 --> 45:28.100]  of this key, this key is very bendable. It's very soft. But I can take a pair of calipers or again,
[45:28.100 --> 45:34.700]  I can take the key depth measurement guide here. Let's see, quick set. Where'd you go?
[45:35.320 --> 45:40.960]  There. And I can take this now. I can get the measurements from this on my own time because
[45:40.960 --> 45:46.640]  I have a duplicate of this key now. Take my own time to get my measurements. I can 3D print this
[45:46.640 --> 45:52.720]  key and take this to a key me or a key duplication, unmanned key duplication machine and have a
[45:52.720 --> 45:56.960]  permanent copy of this key forever. If you only need to get them once or twice, this will work.
[45:56.960 --> 46:05.240]  Again, this will last for a while. So I'm making a mess, but like that, it worked perfectly.
[46:05.240 --> 46:10.540]  However, I want to show that this key is soft. So I'm going to bend it here just to show you.
[46:11.960 --> 46:17.200]  That doesn't take a lot of power to bend that. You can see it's completely bent key now.
[46:19.040 --> 46:25.620]  But if I lay it on a table, I flatten it back out. We're still, we should be good to go. Let
[46:25.620 --> 46:35.580]  me just test it. Yep, we still get an open. So that is duplicating key with casting a key and
[46:35.580 --> 46:44.080]  melting metal and casting it in clay. So let me flip back to my other screen here and we'll
[46:44.080 --> 46:51.660]  continue on. Okay. So time to get some credit here. So first of all, we have the keygen.co.
[46:51.660 --> 46:57.580]  There's the GitHub repository for that. And below that is the talk I specifically mentioned earlier
[46:58.280 --> 47:05.560]  where keygen.co came from. It was from DEF CON 21, key decoding and duplication attacks for the
[47:05.560 --> 47:11.160]  YouTube video of that talk. And that was done by David Lawrence, Eric Van Albert, and Robert
[47:11.160 --> 47:16.860]  Johnson. Big shout out to them. I believe the tool is also available as just a Python script you can
[47:16.860 --> 47:20.660]  download and run locally. So you don't need to rely on their website. The website is just super
[47:20.660 --> 47:27.320]  quick and easy. Obviously, I have to give a massive shout out to Devin Olaf. I didn't think I
[47:27.320 --> 47:32.200]  could do key duplication before I saw his video. After watching that video, he gave an in-depth,
[47:32.200 --> 47:36.720]  very, very detailed way of how to do key duplication. And the fact that he mentioned
[47:36.720 --> 47:41.260]  you could use 3D printed clamshells is what even sent me down this path. So check out his website
[47:41.260 --> 47:48.220]  and check out his Twitter. And then the link below that is the original key molding and casting video.
[47:48.220 --> 47:53.880]  And again, a big shout out to Boris for the 3D printed clamshell that he makes freely available
[47:53.880 --> 47:59.580]  to the world on Thingiverse. If you go back in the slides, you will definitely see the link to that.
[47:59.580 --> 48:04.760]  So I want to thank you guys for letting me talk to you today about key duplication.
[48:05.080 --> 48:09.820]  If you have a 3D printer, please give this a try. It's super simple. Just enter a random
[48:09.820 --> 48:13.860]  bidding code, if you will, just to test it out and print a 3D key just so you can see how easy
[48:13.860 --> 48:17.500]  it is. If you have your own key, go ahead and duplicate your own key just to test it out and
[48:17.500 --> 48:22.320]  see how well it works. It's super simple using thekeygen.co. If you have any questions, any
[48:22.320 --> 48:25.860]  comments, or anything else you want to talk about, again, hit me up at Twitter,
[48:25.860 --> 48:31.560]  at Vareli. It's the easiest way to find me. And then questions. If you have questions,
[48:31.560 --> 48:34.540]  obviously I'll be sticking around. You can find me in the Discord channel.
[48:34.540 --> 48:38.900]  I'll be available to answer any questions you might have, details about printing,
[48:38.900 --> 48:43.480]  molding and casting, anything like that. So again, I want to thank you for taking the time to watch
[48:43.480 --> 48:48.360]  this presentation. I hope you have a great DEF CON safe mode and stick around at the
[48:48.360 --> 48:53.480]  Lockpicking Village with Tool for all the other amazing talks that are coming up. Thanks a lot.