That’s a good suggestion. I’ll put “maintenance tips” on the todo list.
The short version is that Glocks are designed to run practically dry. One drop of oil on the inside top of the slide in front of the ejection port, one drop on each slide rail channel, and you’re good to go.
Anti-walk (really, anti-rotation) pins are of dubious value for a milspec receiver. The idea is that you want the hammer and trigger to rotate on their pins, but friction can cause the whole pin to rotate. That puts wear on the pin holes in the receiver, which really can’t be fixed without going to oversize pins. But this is really a tens of thousands of rounds with no lube kind of problem, so not relevant for most recreational shooters.
But for a printed receiver, the stamped steel “dogbone” plate takes some of the stress off the receiver pin holes. The hammer spring is constantly pushing the hammer pin away from the trigger pin. Tying them together with a metal plate helps prevent creep (which is the result of a constant stress over an extended period of time).
None of the Amazon offerings are particularly high quality. But most 3D printed guns aren’t duty-grade or precision competition guns. I’ve used an assortment of Chinese anti-walk pins and they’ve all been “good enough”.
The headspace measurement method I describe will work with anything – even an empty case. Please don’t do the headspace measurement with a live round. It’s bad practice to be muzzling yourself and putting things down the barrel with a live round in the chamber. It’s really more of a no-go indicator than a precision measurement, so a slightly out of spec dummy round isn’t really a deal-breaker.
The reason to pull the barrel forward is because you’re trying to determine the maximum space between the bolt face and the case head. That means you want the chambered cartridge as far from the bolt as it can get. Tugging on the barrel to take out any front/rear “slop” gets you the maximum measurement.
If time is money, then something like the SUNLU E2 may be a good investment. It’s expensive, but will do the job with the least amount of fuss. An air fryer is probably the best cost/effort compromise (note that the less expensive models may need to be manually restarted every couple hours). If you have a basic understanding of electrical wiring and you don’t mind investing a little labor in modifications, a cheap toaster oven and a PID controller is the least expensive option (well under $50).
Perhaps not thicker wire, just wire made from a higher quality steel. DIY springs are rarely as good as commercial because most people don’t have the ability to anneal and heat treat them like commercial factories do. However, the fact that you do have a clear firing pin imprint on the primer leads me to think your hammer and hammer spring are probably ok.
Play in the barrel retainer could certainly account for this problem. It’s possible to check headspace on a completed build. Remove the bolt and place a dummy round in the chamber. Hold the dummy round firmly seated in the chamber and place a wood dowel in the barrel from the muzzle end until it contacts the round. Scribe the dowel at the end of the muzzle. Now reassemble the firearm and chamber a dummy round using the bolt. Put the marked dowel in the barrel again making sure to press the round against the bolt face and pull the barrel forward to take up any slack in the retainer. If the previous mark on the dowel is now more than 0.15mm short of the end of the barrel, you have a headspace problem.
Weak hammer spring or hammer binding and loosing energy are possibilities. Too much headspace (either due to a build problem or the bolt not going all the way into battery) can also cause this.
I think I know what my first contribution to the Guncad Wiki needs to be.
You haven’t even told us which frame(s) you printed (there are probably 50 different remixes of two core designs), what slide you’re using, which rails you bought, where you got your LPK, etc. This is a little bit like calling an auto mechanic on the phone and saying only “my Toyota doesn’t start, how do I fix it?”
Since none of us can actually examine your build, you need to give us all the build info. You also need to develop enough knowledge of how it’s supposed to work to be able to describe what is/isn’t happening.
First, put away all live ammo. You should never test with live ammo until you’ve verified basic functions with dummy rounds. Don’t have any dummy rounds? Print some: https://odysee.com/@spaceboundtactical:9/SnapCaps-Dummy-Rounds:9
Watch this video until you know the parts in the gun are supposed to function: https://youtu.be/y2fMRqFEEZI
Print an armorer’s backplate so you can actually see what’s going on with the trigger mechanism: https://www.thingiverse.com/thing:3591065
Install the armorers backplate in your slide, load the mag with dummy rounds, then do a dry-fire functional test. Watch what’s happening compared to the video. When you can describe that, then we may be able to offer suggestions as to the root cause.
The Reddit ban did you a favor :) The old fosscad Discord is a much better place to report GCI connectivity problems: https://discord.gg/WHesE5g4
Creep test results I’ve seen show little difference between Sunlu PA6 and PA12 (both are quite good). I don’t have any trusted creep data for Polymaker PA612 (but I haven’t looked that hard).
But you’re really overthinking it. If you’re getting good prints with Sunlu PA6-CF, use it - it’s more than good enough. Printed lowers aren’t going to last decades or 10s of thousands of rounds no matter what you do. There’s this “dream” that with the right material you can get a premium-quality lower. The reality is that the absolute best you can do is still miles below what you’d get with a $39 Anderson “poverty pony” (yeah, they’re $59 now that Anderson sold to Ruger - still cheaper than 1kg of Polymaker PA612).
My recommendation is print your first one in PLA Pro and run it till it fails. When it does, reprint in something better (if you loved it), or print a different design in PLA Pro to see what that one is like. To mitigate creep skip the Hoffman bushings and get a set of anti-walk pins with the “dogbone” connectors (https://www.amazon.com/GLKTRS-Prevent-Rotation-Precision-Motorcycle/dp/B0D9XBCTQG). Or use a drop-in cartridge trigger (cheap milspec triggers suck and you’ll eventually want to upgrade anyway). Don’t be afraid to iterate. That’s the true joy of 3D2A … being to try out lots of interesting stuff without a huge expense.
That Hoffman video was a huge setback to 3D firearm material knowledge. He’s a smart guy with excellent design skills, but when you see statements like “PLA Pro … is not known to have any significant levels of creep …” which are provably incorrect, you have to be very skeptical of everything else he claims. The paper he wrote to accompany the article “looks” like an academic research paper but would be laughed out of a real academic peer review.
The AR lower failure he demonstrates was caused by buffer tower flex, which leads to the buffer tube becoming misaligned with respect to the upper. When that happens the BCG binds up as it travels rearward. Imaging trying to push a straight bar through a bent tube. In his video, you can actually see the BCG getting stuck rearward. Hoffman does a piss-poor job of explaining that in the video.
When it comes to creep, you can see huge differences across different brands of the same polymer. For example, Polymaker PA12-CF is known to be very high creep, while Sunlu PA12-CF exhibits even less creep than most PA6-CF brands. There are factors other than base polymer chemistry (in this case rate of crystallization) that can dramatically affect performance. For reasons unknown, Hoffman doesn’t get into any of that and that’s one more reason why his results should probably not be informing your material selection.
Pins loosen due to creep, which is a result of how the polymer crystalizes and has nothing to do with moisture content. Polymaker’s PA6 has additives that slow crystalization resulting in very low creep. Some brands crystalize faster which leads to more creep and may require annealing.
Pretty sure my wrist will fail long before a decent quality M5. A grade 8 M5 is rated for over 2500 lbs of tensile force.
ar15discounts.com often runs some pretty good LPK deals. Their shipping is slow, but usually cheaper than PSA.
People really need to stop worrying so much about PA6-CF moisture absorption. Unless you’re going to store your build underwater, it’s just not a problem. At 5.3% moisture content (which you’re unlikely to reach even if you live on the gulf coast and store everything outdoors) the bending modulus of Fiberon PA6-CF is still 2286 MPa. That’s only 18% less than Polylite PLA Pro and 16% more than eSun PLA+.
PA612-CF, PA12-CF and PPA-CF all have more than enough impact strength for an AR lower and will get you a little more stiffness. with PA612-CF and PA12-CF you may need to anneal to mitigate creep issues. PPA-CF needs higher print temps (up to 320°C) to achieve optimal strength which rules it out for some people. Builds in any of these materials will be slightly more susceptible to breakage due to drops or rough handling vs. moisture conditioned PA6-CF (though if that’s a serious concern, a $40 Anderson lower – if you can still find one – is probably a better choice than printing).
Find the nut on https://www.mcmaster.com/ (for example https://www.mcmaster.com/products/nuts/hex-nuts-5~/hex-nuts-2~~/system-of-measurement~metric/thread-size~m5/). Click on the part number to bring up the detail window and you’ll see a CAD Download button. Select “3D-STEP No Threads” and click “Download”. You can load the downloaded STEP file directly into your slicer as a “Negative Part”. Position it where you want it and you’ll get a recess the correct size and shape.
If the fit is a little tight when you install the nut (this method gives you no extra tolerancing), just heat the nut with a soldering iron like you would a threaded insert and it should easily press into place. Alternatively, you can scale the negative part up a few percent in the slicer to create some tolerance. Test by adding the scaled negative to a simple cube and iterating on that until you get the fit where you want it.
I just put the length of the printed part. If you want to be really precise, add a quarter inch for the thread adapter. But I can assure you the ATF doesn’t actually care. I’m pretty sure there has never been a case of someone being charged with a violation because their suppressor missed the approved length by tenths of an inch.
You really don’t need to provide much. In fact, too much info can actually slow things down since it’s more stuff for them to review. For suppressors, this section is very antiquated and didn’t anticipate 3D printing as a viable construction method. If you show you know what you’re making (there’s an actual design) and what you’re making it from, that’s really good enough.
If the model is already in the selection list, that means the design as already been through “research” and approved on another Form 1. For my FTN.5 Rimfire that I submitted Jan 1, I just said I was building the design as released and identified the brand and type of filament. I uploaded an image of the filament from Siraya’s website as “documentation”. Approved in 10 days.
When I did a Form 1 last year for a suppressor design not in the previously approved list, I loaded the STL into a viewer, turned up transparency so it showed the internal design, and saved that image. I annotated it to show how/where I would be applying the S/N and other NFA markings and submitted that as my documentation (along with filament info). That one went to “pending research” status and was approved in about two weeks.
This will shoot to the top of my “must print” list as soon as it drops. CZ was practically giving away the 20 round mags at one point last year so I stocked up.
Gen 3 slide locks (SP00301) generally work fine with Gen 5 springs. The only difference is the notch where the spring sits is a little narrower. You can use a Gen 5 slide lock (SP33706) if you want, but I doubt it will address your problem.
Not every frame design gets the spring pocket exactly to spec. You might need a slightly longer spring than stock. The spring doesn’t have any critical specs. Anything that fits and reliably holds the slide lock up is good enough. I’ve heard some disposable lighters have springs that work well. Alternatively, a drop of glue in the bottom of the spring pocket allowed to harden will add tension to the OEM spring. Don’t go crazy – reducing depth by 1mm will make a significant difference on a spring this small.
Trigger not resetting is likely unrelated. For testing/troubleshooting see: https://wiki.guncadindex.com/index.php/Fix_My_Glock