Working on automag twistlock detent improvements.

[Calvin]

I am nearing the end of my roll of TPU filament. I was finishing up making improved Tippmann Carbine/98 detents and was marvelling at how rock solid the positioning of the improved detent was. The back of the ball was on the bolt. The front of the ball was on the detent finger, and the ball was centered in the feed tube.

The printed Automag detents are very functional. In fact, I have only had one ball break while using one, and am pretty sure it was my fault. Nevertheless, I always marveled at how much ball motion forward the detent allowed. After shoveling what felt like 4000 cubic meters of snow, I decided to spend some time with my calipers examining the situation. First thing I noticed is that the fulcrum of the finger was way down in the slot. After I built up the structure to raise the base of the finger, it was clear that I also needed the finger to be forward 2 mm or more to hold the ball in the right position. It took 10 or so variations before I got a workable geometry that also let the bolt ride past easily, but I needed more spring action on the finger. A curved spring section on the back gave it that proper springiness, and folds in on itself to allow the finger back into the slot.
20260125_202823.jpg
These have one side of the finger sized properly for slot clearance and the other goes down to the build plate for proper support while printing. One needs to trim that side of the finger with an Xacto before installing the finished detent. The test detent holds the ball solidly on the front of a long nosed lvl 7 bolt with no forward movement.
20260125_195438.jpg This places the ball directly in the center of the feed tube.
20260125_195444.jpgI fired a pod of paint using it with no issues. OCD assuaged.

Going to print up a bunch to finish off this roll and try them out for real next time I can get back out on the field.
20260125_200920.jpg

[Reflex0283]

Any plans to release the files when they're done? Or will they only be available as physical products like the BW ones

[Calvin]

I'm going to see how they work for me first, and give some away to see if they work for others. Maybe I will use them for trades. I'm not a vendor.

[NONOBLITUS]

I would be down for some to try either by purchase or test subject.

[Calvin]

Just spent some time researching my next roll of flexible filament. There is a fairly new type called PEBA. Its similar to TPU in most ways, but has way higher energy return (springiness). People have been printing balls and other high energy return things with it. Sounds like the perfect material for detents! Just ordered a roll in orange 90a.

[Calvin]

I learned stuff today. Printed a full plate of these last night and went to start installing them in my barrel collection. I already knew that some companies cut the detent slot a little thinner, and some significantly thinner. I scaled the original models thinner for BOA and J&J barrels. Smart Parts was a smidge small, but you could roll them in carefully. What I didn't anticipate is that there is also deviation brand to brand in where the detent slot cut through sits relative to where the barrel locks up on the pin. with 2.5mm of forward slop in the original detent, this went unnoticed. As soon as you try to make one precise, you notice it. My three Dye barrels are good to go, since they were used making the model. I think I will need to go through my collection brand by brand and have tweaked models for these manufacturing variations. I have a few days before the new filament gets here, and still have enough TPU left for this process, so there is nothing for it but to do it.

[BLachance75]

I use a Dye twistlock. If you are looking for some testing I'd be willing to buy some and give them a workout.

[Calvin]

I will be happy to send you some when I am more confident in them. I think I'm going to spend this arctic blast interlude to experiment more before doing anything definitive. I now understand why the guy that made the original model did what he did. It makes a less awesome detent, but works across a wide range of barrels from many manufacturers. When I first started using twistlocks, I thought it was pretty cool that the different brands bought into the standardized spec. Now I know that is only up to a point. What I'm hoping to find is that I can do 2 or 3 styles of finger position and a couple width variations to accomodate all the brands. Thats pretty easy to do. I will play with designs and then finalize them in the new PEBA filament and see how things turn out.

[pneumagger]

Doesn’t bearded works already sell a tpu printed detent just like this?

[superman]

Yes I do and they are available on my website.

[Meglomachinist]

Yeah the bearded ones work great I've got them in most of my bigger barrels need to order some more here soon.

[Calvin]

BW prints out detents that I believe to be from a publically available model. This is in no way disparaging on him, as he prints them very well. I always send people his way when they ask about twistlock detents. When I began using mags, I purchased some of his, and then began printing my own to compare. This project is to see if I can address some of the deficiencies I see in the original design, which is workmanlike, but not what my OCD deems as perfect. I am doing this from the point of view of an enthusiast who wants to get the most out of my mags.

[superman]

Mine were my own design from scratch and my model is not publicly available. What deficiencies?

[Calvin]

My apologies. The last time I had looked at them side by side was early October. I didn't notice that the finger on his design was slightly thinner. Otherwise they are startlingly identical.
20260126_205114.jpg
In this case, I think he copied you since his was posted well after your release. I should have noticed that in October. I was wrong. It happens all the time.

20260126_205029.jpg
Nothing is wrong with your design. My reasons for trying different designs are listed clearly in the post. I still think your offering is what most people need. I am not most people.

[Calvin]

In light of the suspect nature of the model available on Thingiverse, which I was using for the base strip, I made my own model from the ground up. I concentrated on filling the slot till it was flush with the barrel ID and followed the slope of the feed ramp. Once that was done, I had a base to work with. I switched from trying to have a curved surface following the contour of the ball to having a point interface like a spyder detent. That made it easier to match ball positioning. After multiple designs, I like these two the best.
20260128_095257.jpgI printed up enough of the curved one on the right and got perfect or near perfect centering of the ball on all my barrels except the chrome one piece SP AA. It wanted the finger back about 1.25mm. I also only needed to make them wider for the J&J and BOA slots. Thats pretty minimal variations, which is what I was looking for. They work just fine in the guns. Now I'm going to wait for the PEBA to be delivered on Friday and try to dial in the settings for it. Reading about that material makes me think it will be perfect for such a small and flexy part. I will pick up the experimenting once I have that roll on the printer.

[Calvin]

PEBA roll arrived today. I already had the setting in the slicer, so I could just load up. First two detents printed great. A plate of 12 is going now. Printing by object, not layer to avoid reported stringiness. This means you have to spread the objects apart on the build plate with resultingly less objects. I will try by layer with a full plate later and see the results.

The material itself is.......incredible. TPU is plenty springy, but can be a bit lazy to go back to original shape. This stuff snaps back instantly. I cant wait to make my 68 Carbine bumper from it. The bolt has no problem moving past it, yet it is stiff to hold the ball and returns to upright the instant the bolt clears. Very very very cool stuff. Will be playing with it more tonight.

[StaticShocked]

PEBA roll arrived today. I already had the setting in the slicer, so I could just load up. First two detents printed great. A plate of 12 is going now. Printing by object, not layer to avoid reported stringiness. This means you have to spread the objects apart on the build plate with resultingly less objects. I will try by layer with a full plate later and see the results.

The material itself is.......incredible. TPU is plenty springy, but can be a bit lazy to go back to original shape. This stuff snaps back instantly. I cant wait to make my 68 Carbine bumper from it. The bolt has no problem moving past it, yet it is stiff to hold the ball and returns to upright the instant the bolt clears. Very very very cool stuff. Will be playing with it morr tonight.

Is a different nozzle or other special components necessary to run this type of material?

[Calvin]

So far, not for me. Given how supple the filament is, I think direct drive is mandatory. The material is new enough that my slicer didnt have a profile. Reading about the experiences of others indicated that I could start with a TPU profile, so I did. I bumped up the nozzle temp to 240 and the bed temp up to 60. Fans are off first 6 layers, and on low after that, just like TPU. My biggest worry was bed adhesion. I had been printing TPU with the z-axis adjustment up pretty high in order to eliminate elephant foot on the parts. Turns out PEBA adheres just fine. I also lowered my print speed to 30mm p/second at all times. The slow speed is a non issue on such small parts. If it proves that I can do by-layer printing, as normal, without stringing, this will be just like using TPU.