No, Your 3D Printer Doesn’t Have a Fingerprint

Hackers and makers see the desktop 3D printer as something close to a dream come true, a device that enables automated small-scale manufacturing for a few hundred dollars. But it’s not unreasonable to say that most of us are idealists; we see the rise of 3D printing as a positive development because we have positive intentions for the technology. But what of those who would use 3D printers to produce objects of more questionable intent?

We’ve already seen 3D printed credit card skimmers in the wild, and if you have a clear enough picture of a key its been demonstrated that you can print a functional copy. Following this logic, it’s reasonable to conclude that the forensic identification of 3D printed objects could one day become a valuable tool for law enforcement. If a printed credit card skimmer is recovered by authorities, being able to tell how and when it was printed could provide valuable clues as to who put it there.

This precise line of thinking is how the paper “PrinTracker: Fingerprinting 3D Printers using Commodity Scanners” (PDF link) came to be. This research, led by the University at Buffalo, aims to develop a system which would allow investigators to scan a 3D printed object recovered from a crime scene and identify which printer was used to produce it. The document claims that microscopic inconsistencies in the object are distinctive enough that they’re analogous to the human fingerprint.

But like many of you, I had considerable doubts about this proposal when it was recently featured here on Hackaday. Those of us who use 3D printers on a regular basis know how many variables are involved in getting consistent prints, and how introducing even the smallest change can have a huge impact on the final product. The idea that a visual inspection could make any useful identification with all of these parameters in play was exceptionally difficult to believe.

In light of my own doubts, and some of the excellent points brought up by reader comments, I thought a closer examination of the PrinTracker concept was in order. How exactly is this identification system supposed to work? How well does it adapt to the highly dynamic nature of 3D printing? But perhaps most importantly, could these techniques really be trusted in a criminal investigation?

The Elephant In the Room is a Red Herring

We can certainly debate about how common it is for a criminal to employ 3D printed objects, but we can’t deny it’s possible. But interestingly, when we hear about the nefarious use of 3D printers in the media, CC skimmers and 3D-printed keys are rarely the examples given. It seems all anyone wants to talk about is printing untraceable “Ghost Guns”.

Hackaday printed a gun in 2013, and it sucked

Yes, it’s possible to create a rudimentary firearm on a $200 3D printer, but a far more reliable weapon could be built using $20 in parts from the hardware store. In the other extreme, a CNC mill can be used to produce a gun that’s not only untraceable, but legitimately practical. Yet we don’t often see calls for more regulations on the products sold by Haas or Bridgeport. A 3D printer is arguably the worst possible way an individual could produce a firearm, but thanks to their low cost and high availability, some would have you believe a Monoprice Mini is little more than a personal weapons factory.

One thing is abundantly clear, 3D printing isn’t going anywhere. If today we already have a handful of verified examples of 3D printers being used to commit criminal acts, it would be naive to think it won’t happen in the future as the technology gets better. But for the purposes of this article I’m not considering firearms as they are are not a viable 3D printable item in the foreseeable future. This 3D printed gun was crappy six years ago, it’ll still be crappy if printed today, and all signs point to it being crappy in another six years. So fix your mind on present-day crimes being committed with 3D printers, and credit card skimmers are the gold standard in real-world examples.

Matching 3D Prints to 3D Printers: The Burden of Proof

PrinTracker is the marketing-speak for the research paper outlining a way of forensically identifying what type of 3D printer was used to produce a given print. At first glance, the PrinTracker method sounds plausible enough: by closely examining a printed sample, visible manifestations of the printer’s hardware and software configuration such as extrusion rate, nozzle diameter, temperature stability, and acceleration can be observed and cataloged. With a large enough database of these observations, the lineage of a printed object can be determined by comparing its surface irregularities to that of previously identified printers. In principle, this is similar to how a fired bullet or ejected cartridge can be matched to a given firearm.

Examining the “fingerprint” of a gun

But as you look deeper into the report, it quickly becomes clear that there are some serious issues that would prevent this technology from being useful in the real world. For one, the authors note that you would need to build and maintain a database of “fingerprints” for every possible permutation of hardware and software. In the same way a human fingerprint found at a crime scene needs to be matched to one already on record, PrinTracker would only work if there was already a cataloged example that the print under investigation could be matched to.

Attempting to build such a database of even just the commercially available printers would be a monumental undertaking, and that wouldn’t even take into account custom designed printers or those built from kits that might exhibit different behavior from pre-built models. There are simply too many machines, controlled by too many software packages, to hope to catalog them all.

To that end, the paper suggests that 3D printer owners could someday be required to not only register their printers, but provide regularly updated samples of printed objects to be added to the database. Without such an Orwellian system, the paper concludes that PrinTracker could not determine which specific printer actually produced an object, but at best only the make and model which was used.

In light of these facts, it seems conclusions drawn by PrinTracker would be circumstantial at best. Not only is there a high probability that the system won’t find an existing match in the database, but even if does, it cannot say conclusively that the print under examination was made by a specific printer. To return to the ballistics example, it could not identify the particular gun, it could only verify that it was the same model of gun used to commit the crime. This might be a corroborating data point, but it’s not enough to condemn the suspect.

An Exceptionally Narrow Scope

Be that as it may, every journey must start with the first step. PrinTracker may not be a perfect solution, but it’s certainly worth researching. So if we ignore the logistical issues of implementing the system, does the method itself hold up to scrutiny? Unfortunately, that’s not terribly clear either.

According to the paper, testing was limited to just five models of FDM printers. But even with so few printers examined, one would think that variations in software configuration should have been enough to make identification more difficult. After all, manipulating various parameters to produce noticeable changes on the final print quality is how these machines are fine-tuned. Yet according to the paper, the system was tested with only the most minute adjustment to print settings:

PrinTracker is resistant to the variation in printing parameters within 120 – 100 mm/s nozzle speed and 0.06 – 0.15 mm layer thickness in different materials. Out-of-range configuration might pose a risk of spoofing our solution, but the products will suffer from severe deformation and poor quality, which compromises the usability.

Dismissing print speeds outside of the tested 20 mm/s range or layer heights higher than 0.15 mm because they would cause “severe deformation” is disingenuous to say the least, and is a claim which holds absolutely no weight to anyone who owns a 3D printer. For rapid prototyping one might use a layer height as thick as 0.3 mm and run at high speed, but for high detail work, switch to a “low and slow” approach. The reality is that wild variations in speed and layer height are perfectly normal.

Of course, the software is only one half of the equation. What about when modifications are made to the printer’s hardware? According to the paper, when hotends were swapped between different printers, PrinTracker was unable to identify any of the test objects printed. Which is precisely what you’d expect, given the fact that it would completely invalidate the original observations.

But rather than acknowledge this deficiency, the paper frames this as evidence that the system cannot be fooled into producing a false positive:

Upon testing the scanned images using PrinTracker, we observe that the entire set of these images is refused and classified as an alien device. The reason is that the fingerprint not only generates from the manufacture variations but also from the complex integrated effect of mechanical components. Thus, PrinTracker would remain unaffected.

Both of these explanations appear to be clear examples of confirmation bias. Modification of the software and hardware of the printer is common, perhaps even expected. But here modified printers are dismissed as inconsequential outliers. If all it takes to evade detection is a new hotend, or altering print settings, what good would this system be?

Active Countermeasures

We must also consider the possibility that a criminal would actively try to obscure the identity of their printer, in the same way serial numbers may be filed off of a gun. The paper touches on this subject as well, but here again the testing method doesn’t appear to be rigorous enough to make any clear determination.

For example, they describe a “Scribbling Attack”; wherein a criminal would apply something to the surface of the print to make it harder to examine visually. In the paper, this is simulated by dabbing a marker onto the surface of the print. But the testing procedure takes something of an unexpected turn:

We observe that alcohol can be used to clean the ink from the key’s surface effortlessly. We test the scanned images of the cleaned surface’s texture with results showing an accuracy of 100%, implying the PrinTracker has a high tolerance to the scribbling attack.

As difficult as it is to believe, rather than actually testing if PrinTracker could identify prints which had been obfuscated in this manner, they simply removed the ink before scanning them and claimed the test to be a success; completely ignoring the actual question being posed.

Even if we grant the somewhat questionable determination that any ink or paint applied to the print could simply be removed before scanning, that doesn’t take into account physical alterations to the surface. What if the attacker took a sander to their printed object, obliterating all surface detail?

This is considered in the paper as a “Scratching Attack”, but here again the legitimate question seems largely ignored. In this test, only a small section of the object’s surface area was smoothed out with a file before testing. Why would a criminal who is actively trying to avoid detection only obfuscate a tiny fraction of the surface detail? The level of abrasion in this test looks closer to normal wear and tear, and is likely something that should have been considered in the baseline performance of the system.

Adding Fuel to the Fire

It seems obvious that the testing described in the paper was not nearly rigorous enough to take into account the vast number of variables that impact the appearance of a 3D printed object; unintentionally or otherwise. But this is also first of its kind research, so there’s perhaps a little leeway to be given. It’s not perfect, but at least it poses an interesting question.

Unfortunately, what could have been the start of a useful dialog is mired by the same dubious claims against 3D printing that we’ve seen time and time again. Rather than focus on plausible scenarios such as printed credit card skimmers or duplicated keys, the paper references objects which can only be called the products of fantasy:

After the attack is conducted, he leaves the crime scene without leaving any personal marks such as body hair or fingerprint. Instead, he unintentionally or intentionally abandons the tool at the crime scene (e.g., credit card skimmer, cartridge case or magazine) or is unable to retrieve the broken object due to certain circumstances (e.g., grenade debris or broken key in the lock cylinder).

They double down on these examples as part of their official testing procedure, as seen in Figure 20:

This paper is insinuating, either through an intent to deceive or a woeful ignorance, the existence of functional 3D printed grenades and ammunition. These are ideas which belong to the realm of science fiction, and their inclusion in a scholarly paper is completely inappropriate. Printing out objects which look like a hand grenade or bullet without clarification that they’re completely inert is a distraction from the real point of the paper, and further strains the legitimacy of the already questionable conclusions the document draws.

A Missed Opportunity

There are several issues with “PrinTracker: Fingerprinting 3D Printers using Commodity Scanners” that stand out to anyone with experience in 3D printing. Not enough printers were tested, crucial variables are either ignored or underplayed, and the choice of reference objects play into unfair tropes about the technology. This is really a shame, as there was a chance to perform some valuable research here.

A truthful examination of visual print identification, which took into account a realistic array of software and hardware variables and avoided unproductive security theater would still have been first of its kind research. Alternately, the team could have looked at the problem from the other side, and investigated current or near-future methods of “watermarking” prints, such as 3D steganography. This is already something we’ve been seeing in more thoughtful media portrayals of 3D printing, and research into the practicality of these methods would have likely garnered considerable interest.

Instead, we are left with only hints at what could have been. While it wasn’t the intent, this paper seems destined to spread more fear, uncertainty and doubt about 3D printed firearms or other criminal implements. Before long we’re going to need to address the bad actors who want to turn additive manufacturing against society, but it will take something a lot more comprehensive than PrinTracker to do it.

103 thoughts on “No, Your 3D Printer Doesn’t Have a Fingerprint

      1. The paper actually does mention the idea of removing the outer layer of a print to study the internal geometry in such a scenario, but it doesn’t explain how one would accomplish this non-destructively.

        1. Must it be non-destructive?

          I would assume that if it is a large print then only a small part would be cut in order to minimize the damage. Also no cut would be made before thoroughly imaging and measuring the item.

          Still, even if they just sawed the thing in half, so what? So long as the cut is made cleanly and with a fine kerf the prosecutor or defender could simply hold the two pieces up, push them together and say “it looked like this before the lab did their work on it”. Is that going to confuse a jury?

    1. Besides the already mentioned acetone process, one could also have replacement extruder nozzles. It’s like matching a fired round to a firearm by comparing the striation marks left by rifling (or ejector and firing pin marks). So, have a few spares stashed? Plan to do something questionable? Swap in a new set, use it, then discard it far from home and reinstall your standard ones (which will match up to all the abortive prints in your workshop).

      Not suggesting anyone DO this, but I personally find it hilarious just how much forensics relies on someone not taking a minimally active effort to change a _part_ of the equation that will render much of the rest of the parts you have seemingly unrelated.

      1. You have a point there. Anyone 3D printing the equipment for their crime is investing a lot more premeditation in the process than many shooters do.
        You don’t design, 3D print and use a gun to shoot someone as a spur-of-the-moment decision.

  1. So, it seems that you can’t find a printer from an object.
    But maybe you could prove that an object that was found was made on the printer of your suspect, once you have all three items (suspect, printer and item) in your custody.
    So, more like matching the fingerprint of a suspect to the crime scene than using the fingerprint to find the suspect.

    1. If I’m interpreting this correctly, introducing random micro-variations in layer thickness, print speed etc. would make even that problematic.

      The bottom line, as those American chaps would say, is that the old ‘cop-show’ trope of “Find Fingerprint –> Match Fingerprint to Criminal Records Database Entry –> Arrest Suspect” very rarely happens in real life.

      Even before shows such as CSI, etc. All but the most stupid of criminals knew to wear gloves.

      1. No.

        TFA said that changing those parameters didn’t fool their system. Ok, they probably meant printing whole items with different settings but if that doesn’t work why would changing settings within an item work.

    2. This was my thought, too. No, it may not be possible to maintain a database of all printers, and so this would not be of much use to police trying to solve a crime. But it may be very possible to match a print to a printer, making this a potentially useful tool for prosecutors looking for more evidence to build a case.

      The discussion of obfuscation seems a red herring, too; sure, criminals could try to obscure their prints. That doesn’t make this a useless tool, just one that won’t be useful in some circumstances. Likewise, the discussion of new hotends and other printer modifications: Yes, they may change the fingerprint of a printer beyond recognition, but if you find the printer and the hotend hasn’t been changed, you could still connect a printer to an object (and therefore a crime).

      Still, being able to resolve between five printers is a long way from being able to definitively connect a single print to a single printer. In a way, their point about changed hotends demonstrating the test’s resilience does have some validity; if you consider a printer with a new hotend to be effectively a different printer, then doing this has increased their sample size from five printers to more than five printers. But I’m thinking a study of thousands, maybe hundreds of thousands of printers is needed to actually show that this test could be reliably used for evidence.

      1. Another more likely use – if a rash of crimes involving 3D printed card scanners or IEDs with 3D printed components occurred, such analysis would be useful to see if the separate crimes were tied to the same printer. A negative result might not be too useful – the criminals may have simply tinkered with their software settings – but a positive would rule out that there were unrelated criminal organizations who had merely found a file for something on the Internet.

        But a lot of the sort of 3D printed parts I’d expect to see in crimes are something that criminals would want to use some kind of smoothing, paint, or other tricks already – not to disguise what printer the object was made on, but to hide that it was 3D printed at all. An obviously 3D printed object stuck to an ATM is bound to arouse suspicion.

    3. “So, more like matching the fingerprint of a suspect to the crime scene than using the fingerprint to find the suspect.”

      Well first you have to prove that the so called fingerprint is unique and then you have to match the two. Both of those are non starters here, the point is that the variables that affect prints are largely environmental. That means that if you take the exact same printer and print the exact same thing in 6 different locations then you will have 6 different prints, that is enough to prove that a printer does not have a unique fingerprint and provides reasonable doubt. Given the small sample size of the paper referenced i would say that they have a long way to go before they can prove that each individual printer has a unique “fingerprint” (that transcends all environmental variables) because until that is proven there is always the possibility that someone else out there could have printed it.

  2. Just go back to the old Reprap frames that were made of threaded rods. Print your crime tool. Pick your printer up and move it across the room. There you go, new fingerprint now! Even better if some detective throws it in the back of his/her car and drives it to the station. Hah!

  3. I think very few people have any interest in printing weapons. If for some reason I did (why?), I would add a lathe and a mill to my shop along with the CNC, and skip the printer. If we were all talking about replicators, this would be a very different conversation. But … reality.

    The practicality of this system is non-existent. I have to imagine a pretty large portion of 3D printers are custom built, and my Anycubic i3 has been modified significantly since I got it in November. I don’t even think that a printer having printed a part even has any legal relevance – sure it ties someone to the crime peripherally, but its not actionable in regards to proving that person did anything illegal with it.

    1. So, in your mind it goes something like this right?

      Suspect: Yes, the knives found sticking out of the victims torso were printed on my personal 3d printer but honest, I have no idea how they got there!

      Jury: Not Guilty!

      1. Suspect: Yes, The knives found sticking out of the victims torso were from my personal kitchen drawer but honest, I have no idea how they got there!

        Jury: Not Guilty!

        Yes, actually. It *has* gone something like that many times. Being a 3D printer doesn’t change anything.

  4. Looking purely at the mechanical properties of the prints seems like a waste of time if you are trying to connect printers.

    Better investigative angle – look at the trace chemical composition of a print – the trace elements from using a variety of filaments along with the specific lubricants that are used on that printer should make for some unique chemistry in a given print. connecting a 3d print back to a printer seems more reliable this way – like linking the chemicals used to start a fire back to the chemicals on someones clothing.

    1. Ummm So you grease your filament? No inuendo intended.

      I think that if your lube ends up in your print you are doing it wrong. Sure, you should use some on your threaded rods and maybe on the extruder gears if you are using a Wades or something like that. But if you use it on your filament I think you might be doing it wrong!

      1. If you use it on your printer it’ll likely be on your filament in trace amounts. Whether those trace amounts stay there or outgas quickly after printing is a different matter. I think finding the balance in something like a mass spectrometer would be difficult at best though

        1. I suppose if you touch a greasy or oily part of the printer and then touch your filament it’s going to end up in your hotend. How well will it survive the heating though? Also, how often do you touch those things. I try to avoid it myself! Any grease that ever got in that way, if it doesn’t burn off has still probably long since worked it’s way through and out into some past prints.

          Besides, I don’t know about everyone else but I chose my lubricants based on reading a bunch of Reprap forums that were high up in the Google search results. How unique could they possibly be?

    2. I think this would end up the same place bullet lead analysis did. In the garbage heap after convicting people on false positives.
      Then there is this, I very SERIOUSLY doubt the the contamination of a printed part by its lubrication would be consistent enough to be tracked. This assumes that all consumer grade lubricates are stable at the temperatures used? If not and they do degrade, is it always in same fashion?

      Using this, you MIGHT be able to narrow it down to group composed of printers, controllers, raw materials and users. But I see that group as being too disjoint to be useful.

    3. I think this is a more reasonable approach to associating printer with print. I just don’t think the association is necessarily actionable. People print things for other people, and they discard prints they made (or give them away). Plausible deniability.

  5. In the scenario where printers would have to be registered: only law abiding people will register their printers and submit print samples to be added to database. Criminals by definition don’t follow the law. They will use a unregistered printer made of randomly sourced parts to print their “tools“.

    This has the same issue as gun bans and gun-free-zones: criminals don’t care about what is and what is not allowed.

  6. The only thing I don’t like about 3D printing is that it can potentially create lots of plastic waste. I can’t wait till we can reliably create good filaments from plastic and other recyclable materials.

    1. What? What kind of non-plastic filament are you using?

      I’d love to see recycle bots become more affordable and more commonplace. They are certainly on my “some day” list. Meanwhile I’d be thrilled if our curbside recycling accepted 3d printing scrap. They could even require that we box it sorted by type.

      1. PLA is only compostable at about 60C/140F. I doubt it’s composting much in landfills. The legal limit for a landfill in the US is 50C with typical being from 35-45C. I assume other countries have similar laws?

  7. Use your brains. If your gonna print something that can get you into trouble throw the printer out when done, and get a new one. They are cheap enough now a days. Simple Solution.

  8. It’s a common practice to seize printers (e.g. dot matrix, laser, inkjet) in a criminal investigation. You can’t possibly find a particular printer if you don’t know where to look for, but you can determine a “certainty” level of whether a particular printer was used for the crime. Hobby level 3D printers would be the worst case scenario, as they mutate in multiple ways, but I imagine that they can even stablish whether a particular frame size/mass was used to print the piece (by the vibration armonics left in the build). It’s certainly not fail proof.

    1. Right, but the paper is specifically talking about situations in which the printed object was the only evidence left at the scene.

      So you’d need to use their proposed printer database method, since you wouldn’t have anything else to go on. If you had a suspect and enough evidence against them to sieze their possessions, youre probably past the point where this system would be necessary.

      It’s a classic solution looking for a problem.

      1. Yes, that does sound kind of dumb. Whose to say that given this tool they would follow that suggestion and only use it when no other evidence is available. It seems to me that they would collect every bit of evidence that they legally can (and often more). If that evidence includes a printed part and a printer fingerprinting method exists why wouldn’t they use it, even if they did already have other evidence (which they needed before seizing the printer)?

  9. General comments about science and technology in the hands of law enforcement and the U.S. court system. Have dealt with many lawyers and court systems for both personal issues and for my employers.

    1. My anecdotal statistic is that approximately one in 43 lawyers (sitting judges included in this count) have the intelligence, education, and willingness to analyze and understand science and engineering data.

    2. For both criminal and civil trials, lawyers, judges, and police officers do not care about the truth of physical reality. Their concern is whether ‘evidence’ can be presented in such a manner to prevail in a hearing or trial.

    3. The seven juries that I have been exposed to (sat on one, witness in front of six) were all ill-suited for the technical nature of the claims and evidence. This was by design. Lawyers tend to disallow scientists technicians, and engineers from juries. Most juries tend to be too stupid to know that they are stupid. Once heard a prosecuting attorney ask an English teacher if she had taken any ‘advanced’ math – yes, she had taken algebra. “Do you remember any of this math?” – no, did not do well in the course and does not remember any ‘advanced’ math. The defense and prosecuting attorneys both had agreed to ask this question of all panel members that were college educated, and empaneled only those without ‘advance’ math.

    4. Have been empaneled once and only for a single day. Requested relief from the judge after the prosecuting attorney dismissed physics and said stuff that was not possible in the opening statements. The judge granted my dismissal from the jury.

    5. Have had two moving violations dismissed where the police officer claimed an act that violated Newtonian mechanics. But it was difficult explaining simple algebra and high school physics to the idiot traffic commissioners (judge wannabes and lawyers). In the last incident, the officer commented that “this mumbo-jumbo should not be allowed”.

      1. Privatize courts? Is this a real thing? Do people actually advocate for such clearly harmful things? Look how terrible privatized incarceration is, with economic pressures ending up with judges being found guilty of taking bribes to send kids to their facility in Pennsylvania and lobbyists lobbying for more reasons to lock people up.

        1. No, you agree to accept judgements from a certain court, who are competing to provide the best justice in the marketplace, and charge fees to both parties for the judgement, which is quality controlled by assessing body the court submits to.

          1. So, what you want to do is throw a (bigger) profit motive into the justice system? What makes you think that a commercial court system would be any better than a public one?

  10. Or maybe they force all printers to insert an identifying set of parameters into the infill. Then MP Select Mini SN# 123459876 would put “123459876” someplace in the infill, repetitively so that it could be forensically found. Just like HP does with the color copiers.

  11. Live well to the west of this abstract and meaningless political boundary. Live and work in Southern California.

    Sorry to cast doubt on racist prejudices, but the ‘better’ jury where my testimony was presented, was found in the Macon-Bibb superior court.

    1. Ha Ha Ha! So Cal? Well that explains it!

      You had better fix that comment of yours before it gets you into trouble. Ah, nevermind. I am feeling nice. I’ll do it for you.

      — This comment has been found by the state of California to cause cancer and birth defects —

  12. It’s great to see hackaday fearlessly taking on science again, with the rigor we have all come to expect.

    “In the same way a human fingerprint found at a crime scene needs to be matched to one already on record, PrinTracker would only work if there was already a cataloged example that the print under investigation could be matched to.” Ah yes, this method is compromised because it has fundamental flaws which, oh wait, current state of the art techniques which are widely used also have.

    “the paper concludes that PrinTracker could not determine which specific printer actually produced an object, but at best only the make and model which was used. In light of these facts, it seems conclusions drawn by PrinTracker would be circumstantial at best… it cannot say conclusively that the print under examination was made by a specific printer. ” Yes, you’re very good at repeating what the paper has said.

    “For example, they describe a “Scribbling Attack”; wherein a criminal would apply something to the surface of the print to make it harder to examine visually. In the paper, this is simulated by dabbing a marker onto the surface of the print. But the testing procedure takes something of an unexpected turn:”
    Forget everything I just said. They deserve mocking. public mocking. Scientists showed that alcohol can be used to remove marker ink from plastic. It’s groundbreaking, I tell you. What’s next? Could you remove pencil from paper? Why, there’s untapped commercial potential in such an invention.

    As others have said, it would be interesting to see actual smoothing done with acetone or some other chemical. There’s still structure inside the print, but I’d imagine that would make identification more difficult. On the other hand, it might be that the internal structure is more identifiable.

    Seriously though, the scenario I see here that could be useful is police find a card skimmer, then when they executa a search warrant and happen to find a 3D printer forensics could be useful to determine if it created the prints. In practice, it’s probably enough to check if the filament matches and the criminal has the file ‘card_skimmer.stl’ on their computer.

  13. I’d be interested just how sensitive this method is. Swapping hot-ends already detects as a different printer, but what about simply removing and then replacing the same hot-end/nozzle (cleaning, clearing jams, tinkering, swapping for a bigger nozzle, etc)? If basic maintenance on a printer already throws off the results then I think this method is pretty much useless.

    1. Not necessarily. This assumes that maintenance has been done in between the crime and catching a suspect, which may or may not be the case. If the criminal does this deliberately to prevent this line of investigation, then sure – but there are plenty of ways to deliberately prevent the fingerprinting idea from working of the criminal thinks to do so. CSI techniques can all fail or be fooled. The trick is to find the one thing the perpetrator *didn’t* account for. This could be that thing. If it works in 1/100 of potential cases, it will be a good method – probably on par with may other CSI techniques (though I have no idea what realistic success rates are).

  14. This article raises some valid concerns, but by the end I was less a skeptic of the fingerprinting idea then before I read it. If those are the only complaints you can come up with, then they’re on to something.

    1) The most repeated attack in this article was a completely irrelevant red herring. So they used unrealistic examples of nefarious 3d printed objects – who cares? That’s not the point.

    2) So the technique could be fooled. Yeah, and so can every other CSI technique in existence. CSI is more about finding evidence that was left by mistake then developing foolproof methods that work regardless of what the criminal does – that does not exist. The criminal who thinks of everything can’t be caught. The more realistic criminal who missed 1 out of 1000 things might be. Every additional tool helps.

    3) The database idea would be easy to implement if volunteers sent in a few thousand objects. This could help identify and unknown printed down to make and model, which would still be useful information. Once a suspect is found, the printer could then be matched, if the criminal hasn’t made modifications (see point 2).

    4) The basic concept could be extended to inside an object (i.e. carefully saw in half, examine infill texture, etc.)

    5) Back to point 1 a little. Even though it’s irrelevant, I think this article way overstated the case against 3d printed guns. Yes, there are better or cheaper ways to make a gun. However, guns capable of murder have been demonstrated, so it wouldn’t surprise me if someone used one in a crime at some point. Criminals aren’t intelligent creative hackers/engineers topically. They are likely to do something that’s easier/more accessible over something that is technically superior. 3D printers are much more used friendly now days than CNC machines. A complete idiot could download a file to a 3d printer and press start. Working with metals is a bit trickier.

          1. Actually, his point was that I don’t have experience in 3D printing, therefore my arguments are wrong, which is a silly argument. If they’re wrong, show me how. Not to mention that neither of you have explained what I’ve said that gives you insight into my experience with 3D printing. The only thing I said about 3D printing is that functional guns can, indeed, be made. That is a fact you can easily verify, so it hardly demonstrates an understanding or lack of understanding – just something you’d be aware of if you were into 3D printing and followed related news.

            Among other things, this line of attack is weird since the majority of my post wasn’t even about 3D printing.

            I will admit that I use a stratasys at work, so I don’t have as much experience with cheap printers. I have printed a few things on makerbots, etc. though and it never seemed to be all that different.

    1. “However, guns capable of murder have been demonstrated”

      Only when printed using expensive industrial-grade printers that are far beyond what is available to typical users. Even then the result is a gun that has a very high probability of blowing up in the perpetrators face/hand.

      “so it wouldn’t surprise me if someone used one in a crime at some point”

      With 8 billion potential perpetrators on the planet very little would truly surprise me either. What is your point. For the foreseeable future there will probably be more murders committed using strawberry jam than 3d printed guns.

      “Criminals aren’t intelligent creative hackers/engineers topically.”

      citation needed.

      “3D printers are much more used[sic] friendly now days than CNC machines.”

      3D printers ARE CNC machines. No doubt you mean CNC mills. Yes, yes they are. So what? It still isn’t easy.

      “A complete idiot could download a file to a 3d printer and press start. ”

      Sure. Then the not-so-well-tuned printer will spit out a mess of plastic. Imagine your “complete idiot” leveling a bed, dealing with a clogged hotend, figuring out the best temperature/speed settings or how best to use supports.

      “They are likely to do something that’s easier/more accessible over something that is technically superior.”

      All it takes is a steel pipe of the right size a hammer and some ammo. Add a cap and a nail for slightly better odds of not maiming ones self. That would be far easier than printing anything! Wake me when a 3d printer can produce ammo.

      “Working with metals is a bit trickier.”
      To produce something that looks like a commercially produced gun yes it is.. Just to make something that can fire a bullet only requires the ability to saw through a pipe and maybe drill a hole.

      1. “Only when printed using expensive industrial-grade printers that are far beyond what is available to typical users.”

        This was done with a pretty typical type of printer available to users at libraries, universities, and maker spaces (in the US), not to mention that the printer costs less than a macbook (i.e is well within the range that many people in the US could afford).

        Look, I’m not arguing that 3d printing guns is a smart way to go, or anything to fear. I just said his point was overstated. Functional weapons can be 3d printed (fact), and so will probably be used, to some degree. I agree that murders by first, hammer, knifes, etc will be a much larger category though (barring some major change in accessibility of metal deposition techniques, a novel filament, or some other unforeseen breakthrough).

        “Then the not-so-well-tuned printer will spit out a mess of plastic.”

        Maybe. Depends on the printer, some require much less tuning out of box than others. More importantly, though, already tuned and functional printers are also available to most people who go looking (see point above).

        “Criminals aren’t intelligent creative hackers/engineers [typically].”
        “citation needed.”
        Really? I mean c’mon. This is kinda off-topic, but in case you really think that your average criminal is a highly intelligent engineer (a scary thought), here you go:

        Keeping in mind the average IQ of an egineer/hacker/etc.:

        As for the rest – yes, there are other better ways to make a weapon. I already agreed with that in my post.

        1. Check out that funny setup they used to fire that gun. It’s attached to a piece of aluminum extrusion and the trigger is tie-wrapped to a cable. Why do you think they did that? Has anyone ever held a gun which was 3d printed on a commonly available printer in their hands, lined the sites up by eye and fired? Would you? If not then why?

          As for the IQs of criminals.. meh. People with more selfishness than sympathy for their fellow human beings turn to crime regardless of IQ. IQ just determines what kind of crime. Smart people go for the crimes where they can gain the most for the least personal risk. Dumb people take a more smash and grab approach. That’s right, smart criminals go into business and politics! We have a culture that vastly overlooks white collar crime and focuses on the lower end so we mostly only perceive the dumber criminals. Honestly though, who has killed more people, whoever holds the liquor store hold-up record, the world’s most prolific mugger, Ted Bundy, a former or current leader of Monsanto or Dick Cheney?

          Obviously there are exceptions to the rule. Surely there is some genius petty thief out there somewhere to counter that mentally impaired sociopath that just keeps failing up and whose name rhymes with Dump.

  15. “There are simply too many machines, controlled my too many software packages, to hope to catalog them all.”


    Tom, have you considered cutting back, just for your own sanity’s sake?

    1. Yah. I kind of felt that was similar to saying there are too many finger prints or too many gene variants out there to catalog them all. It doesn’t really stop finger printing and dna collecting from being useful does it?

  16. Their time, and the taxpayers, money would be better spent conducting a study to determine that water is wet. People worrying about 3D printed firearms have absolutely no idea how firearms work, how 3D printers work, or how easy it is to build a firearm from hardware store parts.

    1. “Their time, and the taxpayers, money would be better spent conducting a study to determine that water is wet. ”

      Sure. Understanding the movement of water molecules and how that affects viscosity no doubt has many positive applications. That doesn’t mean that potentially catching people that harm others isn’t valuable too though.

      “People worrying about 3D printed firearms have absolutely no idea how firearms work, how 3D printers work, or how easy it is to build a firearm from hardware store parts.”

      I entirely agree. But, did the article say anything about this only being applicable to guns? How about 3d printed knives? That sounds much more practical to me anyway. Unlike a 3d printed gun it will not explode maiming the bearer. Unlike a regular knife it will not be found by metal detectors.

      Why the weapons fixation though. Even if the killer/rapist/thief just accidentally dropped some random object that happened to be 3d printed if you can finger print it and have enough reason to suspect the person that you can get a warrant to enter their home and seize their printer then that could be evidence.

    1. You’d find something. “Yep, it came back from the lab. It’s ABS printed with an FDM, at a later height of about 0.2mm.” “anything else?” “the overhangs string a bit, about 2-3mm strings, and it’s warped by 3 mm along the longest axis. The bottom is really smooth like it was against glass” “There can’t be too many printers out there that do that. We got the bastard!”

  17. This study is will be used as basis for inserting finger print into all 3D print object like color photocopier have micro dots that can identify it. With the primitive ABS 3D printers this is not needed, but for high quality industrial printer with laser on metal particle, this is doable.

    1. This is a study about how to identify a printer WITHOUT doing that. How would such a study be a tool for someone trying to add micro-dots or an equivalent? That makes no sense!

      Further, what if “they” passed a law requiring such signatures? Are all the home hobbyists going to implement them? I don’t think so! They might as well legislate that the sun be dark.

      1. Every hotend nozzle would have to be factory fingerprinted with an extrusion sample, and every nozzle would need a serial number filed with the sample.

        And it would be utterly useless, just as firearm cartridge ‘fingerprinting’ was found to be because any normal wear alters any imprints left by enough to make the sample useless as a means of identification.

  18. There are so many variables that are very easy to change. Print nozzle, speed, filament brand, software used, belts, steppers. I don’t see any usefulness here really.

    1. Nah, they addressed that. Perhaps they didn’t address it satisfactorily but they did address it so if you want to use that argument then go back, RTFA and explain how they didn’t address it sufficiently.

  19. Fast forward a short time later…Referencing the published paper…blah, blah, blah…We propose the following study at a cost of several nice retirements for the researchers and follow on contracts for concept development. Get a hysterical Nj politician or 4 behind the whole thing and next thing you know, DHS contracts for everyone.

      1. He didn’t say that they receive the cost of several retirements. He said that they request it. Of course they did! What fraction of what is requested gets granted? They probably had to request the GDP of a small island nation just to get their parking validated!

      2. Promotion in academic institutions depends almost entirely on the ability to bring in grant funding (the Universities skim a massive chunk off the grant up front). If you don’t get the grants, you don’t get promoted, and you don’t get the $200k + per year salary. Does this lead to corruption and BS science? Who knows? Maybe scientists are all really honest and incorruptible, and pigs can fly.

  20. So much written, so much of it irrelevant.

    Not only weapons might need to be tracked back to the maker. It might be a camera bracket on a quad shot down at an airport.

    Nobody needs to fool with the print parameters to get something identifiable. Few beds are mathematically flat. The imperfections on any bed are probably unique, and are imprinted on the bottom of each piece printed, unless you took time to sand them off. (Do you always?)

    Finding which bit of bed the article was printed on would require some rough pattern matching, but you wouldn’t need a perfect match for that. Once you identify the spot, there are probably thousands of unique, microscopic details to compare.

    1. No way is this a serious comment. So you think there’s enough surface variation in a piece of glass in an area of a few cm that it’s not only detectable, but could be matched after the fact?

      1. I see: All 3D printers (except every single one I have seen) have perfect, optically flat glass beds? My mistake, then.

        Lots do have that bumpy blue tape on their (perfect, optically flat glass) beds, I guess to keep prints from sticking, or not sticking enough, or something. Many have that Kapton adhesive plastic sheeting, with scratches from previous prints, or from getting previous prints unstuck. But I guess that wouldn’t count, because… Remind me why that wouldn’t count?

        1. Blue tape and Kapton are very rarely used on modern printers, though even when those materials were commonly used, they would not have been used with a glass bed.

          With respect, I think you’re out if your depth or operating on outdated information.

          1. Indeed, bumpy enamel surfaces are quite common. Perfect optically flat surfaces are vanishingly rare. Rare enough, in fact, that something printed on one could be easily shown not to have been printed on any of the others.

          2. Nope. Now instead of blue or kapton tape it’s that stupid build tac stuff.

            I’ve been printing with a piece of glass on a heated bed for a couple of years now and for the life of me I can’t figure out why anyone would consider doing it any other way. It is so much easier and better!

            A friend bought a printer with some sort of build tac like surface. It’s so hard to use! Depending on how close the nozzle is to the bed it either doesn’t want to stick or it sticks so tight you can barely get it off. Maybe there is a sweet spot in-between but if so it sure is a small one and it’s going to take a lot more time than I have to sit in his house for me to find it.

            PLA sticks pretty well to hot clean glass. Then if you let it cool it often pops itself off all on it’s own! Just add a little glue stick and ABS is about the same.

        2. It wouldn’t count because it changes after every few prints, or more realistically after every print. At least I replace the bed surface tape after every print. And I assume you’d definitely do it after printing something that you wouldn’t want identified in that manner!

        3. I don’t think there will be enough data in the small variations in the glass plate. While it’s never really flat, the actual details are very smooth. And thus contain no real fingerprinting information.
          Also, the glass changes shape for about 30 minutes after it reaches the final temperature. Surprise!

          (I work at Ultimaker. We had our fair share of issues with bed correction and done more measurements and tests then I’m willing to admit)

          1. Hi Daid303,

            I wouldn’t expect variations in the overall shape of the plate to tell us much, but microscopic pits and scratches should. It all turns on how closely the melt conforms to the plate. I can’t imagine anything that would introduce systematic differences in these details, excepting that articles are often placed differently to make room for others printed in a batch, and that sometimes tape or soluble goo might be laid on before a print and stripped or washed off after.

            The most likely circumstance where the question would arise would be the question: “Was this article printed on this gadget?” Lots of circumstances would make the answer “I don’t know”, but sometimes you could say, “definitely yes”.

            No need to for attention to changes in material, temperature, nozzle, layer thickness, infill pattern, or slew rate. None of those make any difference.

  21. An episode of “Elementary” addressed 3D printed guns by having the murderer use acetone to dissolve the white ABS Liberator pistol he used.

    Sherlock figured it out by noticing the bottle of “milk” in the refrigerator, the dead wife was vegetarian, her killer husband was lactose intolerant.

    Nevermind the anachronism of a glass bottle of milk or that the acetone would smell, or that the husband could’ve simply poured the dissolved pistol down the toilet, Why would he even think to hide it as a bottle of milk when he wouldn’t be accustomed to having milk in the fridge?

    1. “Why would he even think to hide it as a bottle of milk”
      It hadn’t fully dissolved, but then, would the cooler temperature of the fridge slow down the process further?
      (I’m just guessing, I didn’t see the show)
      Would acetone dissolve an HDPE milk bottle?

  22. Excellent and very thorough article even for hackaday, I applaud the author.

    I have been wondering about this myself since that paper came out and it seemed like there were many flaws in a lot of parts of it and a lot of assumptions made.

    Nice to see that someone else noticed and addressed the idea.

  23. They’d have a much better chance of implanting a code of invisible markers in the filament, like a barcoded serial number of IR ink, than getting everyone to register their print. This code would be repeated in the print but stretched out. But even this is pretty much pointless because a criminal could simply make their own filament or soak it in something that looks like the marker. It would be about as hopeless as trying to print serial numbers on bullets. Not saying it’s a good idea, just that even as a bad idea it’s many times better than this idea.

    1. IIRC, plastic explosive mfgr’s (legitimate ones anyway) implant tiny “paint” chips in their product that will survive the blast, leaving investigators a clue as to who mfr’d it.

      1. Gunpowder manufacturers have to include such ‘taggant’ particles in their product so forensic examiners can trace where the powder residue came from if it’s used in a bomb.

  24. Think this just another grab for taxpayer money. Just how often are 3d printed parts actually used in crimes? How many places can you get a part printed? Heck, I’d probably print a few things, if paid, and probably wouldn’t worry or wonder if the part is just a prop, or actually going to be used in some crime. Hell, maybe it was intended as a prop, but someone got hold of it, and used it as a weapon. They really over hype a non-problem. Criminal activity will continue, the tools they use, aren’t the problem. There are so many things to change, just to get a decent part, little adjustment do to the daily environmental conditions, different brand filaments, part characteristics. Basically, this is fishing for life time government grant money. Doesn’t really seem to matter is you can match a part, to one printer, you’d still have to prove who used the part, to commit the crime, at which point, it doesn’t really matter where he got the part, or how.

    Skimmers are more than just 3d printed parts, people using them, probably aren’t making their own, but buying them online, even if the own a 3d printer… Suppose the printer doesn’t match the print, does that make the suspect innocent? Works both ways, even DNA isn’t perfect, this might help in a very rare few cases, but not enough to prove guilt, only open other ways to weasel out.

  25. not going to read their documents, but i wonder if they even bothered trying to identify a printer by scanning a print other than the one they used to create its fingerprint with…

  26. It’s about power and control. The people who come up with this baloney (same as ‘fingerprinting’ firearms by marks left on bullet casings) want to sell governments on the idea that it’s a viable way to have more control over what people are allowed to have and what they’re allowed to do with it. If these ‘researchers’ can sell the idea to some foolish government control freaks, they’ll be set for life.

    As for gun ‘fingerprinting’ AFAIK in all the time any government has required it, it’s been used to help solve few, if any, crimes, at a cost of many millions of dollars. Maryland quit it after 15 years with no crimes solved using the data.

    The problem is that firearms are made of many interchangeable parts, and wear on those parts changes the marks left on shell casings. Unless someone uses a gun in a crime after no or very few rounds have been fired since it was ‘fingerprinted’, the data is useless. There’s also the problem of tooling marks being very close to identical across batches of parts machined with the same tooling. It’s not possible to tell if slight differences are due to the wear of the machine tooling or wear caused by using the gun.

    3D printers are even worse for this. Change the filament, change the nozzle, change any of a dozen controlling parameters, change the temperature and humidity in the room. Such a large number of factors that will cause even two prints made with the same settings and filament come out different enough they could’ve been made by two different printers of the same model despite being printed one after the other on the same printer.

    So how do we put a stop to governments even considering listening to this foolish notion? By educating legislators on how 3D printers work, and how they *don’t*, with emphasis on how the output is so highly variable at the micro level to the point where outside of such carefully controlled experiments it’s simply impossible to tell which specific printer was used to produce an item.

  27. The source paper, “PrinTracker: Fingerprinting 3D Printers using Commodity Scanners” was produced under support of an NSF grant. So, this paper is the result of some academics likely having a deadline to continue their “research”. Without showing some level of success, they likely can not get further grants, hence the excessive amount of positive spin in their paper about the success they have had. When you have to show positive results to get future grants, even failure will be shown as success!

    1. I am astonished by the level of contempt shown for researchers and academics in this thread. If there is anywhere where one might anticipate a positive view of academic research it is here at Hackaday. I am terrified to consider how low the opinion of the general population must be for Science.
      And it is not at all uncommon to publish that something can’t be done (or, more likely, that it is more difficult than expected). My own PhD thesis basically said “the phenomenon studied in this research appears not to exist, largely as previously assumed”

      1. My reading is that, while some of those posting were simply appalled at the unsophisticated process being presented as legitimate research, many imagine themselves at risk from investigations, and fervently hope that home-manufactured objects would not be traceable to them. In other words, if the research had been presented as a way to absolve innocent printer owners from suspicion, the response might have been different.

        As it is, the effect of most of the posts, were they persuasive, would be that all printer owners must be considered suspect of producing any printed object used in a crime, because any printer at all could have produced it, with no way to rule out any.

        The most likely response to this post is personal attacks on me for having said so.

  28. Conceivably every tool could/would leave a unique mark on the material doing it’s work, even extrusion dies. Most likely only usable until other evidence leads to a suspect. Self fear mongering must lead some to turn to 3 gun . At moment in Kansas its very easy to purchase a hand gun. After pruchase,put om glove tp clean it well, and load it. While wearing gloves go do your criminal act, when you’re done toss the gun in river or pond.

  29. Surely genetic/environmental material that gets incorporated into prints produced by FDM methods would be a more reliable method of identification – because of the way they work, material must get incorporated into objects in the air gaps between filament, infill and in finger grease from handling the printed object and the filament. You could also do a chemical analysis on the filament reel and print scraps such as brims and rafts, and print bed residue, found at the site of the (presumably) confiscated 3D printer of the suspect? If you can match the DNA from a skin fragment found in a 3d printed object to an individual in possession of a 3d printer with evidence of the same filament composition, you might have a link, right?

    This would probably also cover resin and sintering printers and hybrid methods – not many people can afford to operate their equipment in a clean-room and reliably manage to avoid all physical skin-to-object contact.

    And what about good old fingerprints – you know, the human kind?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.