I own more ARs than AKs, and I wasn't talking about DI. I agree with you that DI has more benefits than problems.
It's stuff like e.g. the interface between the receiver and the buffer tube, which is weak and known to break under stress. Or the telescoping stock design, which has a similar problem with the retaining pin - which is why they tell you to never mortar it unless fully collapsed.
There are some issues with the bolt longevity - in an AR, the bolt eventually fails with sheared locking lugs - which is unheard of in AK and not common in other designs. Part of it is due to asymmetric distribution of forces, because of the extractor taking the place of one lug. Modern bolt redesigns try to mitigate this in various ways.
Then there's the enclosed magwell, which is great for quick mag changes, but requires the upper part of the magazine to be straight - which caused troubles for 30-round AR mags historically, until the modern polymer magazine design with internal constant curvature made possible by the manufacturing process. In addition, the magwell is prone to dirt-induced malfunctions - once you get the dirt in, e.g. from the lips of a magazine that fell to the ground, it stays inside, and can block the mag from going all the way in easily. Ice can cause similar issues. The button magazine release can get stuck relatively easily as well. AK well-less paddle system, while slower, is less susceptible to dirt, and the rocking insertion motion and the paddle magazine release provide more leverage even if there is an obstruction.
Or we could look at the antiquated handguard design, which results in precious little rail space on the upper receiver, because the top rail on the handguard, even free-floated, is not as stable (and hence not used for magnifying optics usually, but red dots, lasers etc). Notice how the latest trends in military ARs have been monolithic upper receivers, like in New Zealand's MARS-L, or Canadian MMR?
Of course, you could easily make a rifle based on the Stoner action that has none of these problems - there are more exotic AR variants and accessories that solve one or the other, you'd just need to combine all those solutions. But I don't think it would count as an AR at that point. Even the monolithic upper is already a major difference that breaks compatibility in so many ways. I mean, we don't say that SIG 550 is an AK variant, even though it's also long stroke piston, and the BCG looks very similar to AK. There's more to a gun than its action.
Are there any good commercial designs which stand out as being particularly “modernized” without going full custom? Or are the trade-offs too numerous?
It’s really hard to improve DI systems without compromise, and harder still (probably impossible) to do so without also breaking compatibility with other parts like chamber lugs pattern.
The KAC SR15 solves the bolt longevity issues, but now it’s bolt and barrel extension interface are not compatible with standard bolts.
The LMT MRP platform solves most of the other problems, but not as much the bolt issues. But similarly, the LMT MRP monolithic upper cannot be compatible with standard rails, because that’s exactly the point: to move way from the old brittle way of connecting the hand guard to the upper receiver.
KAC SR15 will last upwards of 50,000 rounds without needing replacement parts, but externally it’s no more physically durable than a standard AR.
LMT MRP is among the most solid and sturdy AR out there, with virtually no way you’re going to break it if you throw it around, drop it, and otherwise abuse it. But internally, it will only last marginally longer than a standard AR before needing replacement parts to keep running reliably.
Civilian guns also have the luxury of using more exotic materials. So e.g. S7 steel or even Aermet 100 for bolts and carriers, 2055 aluminum alloy rather than 7075 for receivers (or titanium, but at that point the cost is through the roof even for the civilian market - it's very expensive to machine something that large), 7075 rather than 6061 for rails, and so on.
The only problem is diminishing returns. This is true in general, of course, but it's especially true with brute force solutions like this. If I remember correctly, Aermet 100 has roughly twice as much tensile strength as Carpenter 158, but costs about 3x for raw material alone, even before you account for how much harder it is to machine - and that's the best price-to-benefit ratio on my list.
(This is probably why Aermet bolts are gradually making their way in via LMT and KAC, but not anything else.)
> Of course, you could easily make a rifle based on the Stoner action that has none of these problems - there are more exotic AR variants and accessories that solve one or the other, you'd just need to combine all those solutions. But I don't think it would count as an AR at that point.
I suppose that’s fair to say, but my mental taxonomy still considers the Knights Armament SR15 and LMT Monolithic Rail Platform to be AR’s. It seems like one could make a pretty strong argument for the SR15 at least, since I believe it was partly designed by Eugene Stoner. Then again, it is also the least compatible out there given the completely proprietary bolt design.
But that SR15’s proprietary bolt design by all accounts solves the bolt longevity issues, and the LMT MRP solves the rail and interfacing fragility issues. But you’re right that I don’t think there’s any single AR that solves both simultaneously. LMT’s BCG is improved but I don’t know that it solves the problem as well as the SR15.
I’m not sure it’s fair to hold the stock mortaring situation against the AR, since this same issue appears in virtually every collapsible polymer stock design. It only is not a problem with fixed stock designs, independent of whether they’re an AR or not. I believe the FN Scar for example received a lot of criticism for its stock being considerably more fragile than even standard M4 stocks.
Also, I’ve never heard of dust in the magwell being an issue as long as standard practices like keeping a mag in place in dusty/dirty environments are used. In fact, more often I’ve heard of issues with polymer AK mags breaking, since the AK mag design essentially offloads the rigidity and durability problems to the mag rather than the rifle. This could be a good or bad thing in some ways, but I think everyone agrees it increases weight too much when you need steel mags. Example: All modern designs use enclosed magwells and polymer mags.
In any case, I’d be inclined to concede to you that ARs don’t last as long as other designs (without replacement parts every 5-10k rounds or so). And it certainly makes sense too when just dry running the action: compare the locking/unlocking cam action on an AR, vs a modern piston design like a Steyr AUG, FN Scar, CZ Bren, etc. and it’s so obvious how much less violent the forces are on the bolt as it locks/unlocks considerably slower and smoother on these more modern designs.
P.S. Fun fact for interested readers: All these “more modern” designs I mentioned are all descendants of Eugene Stoner’s AR18 design (which is itself a descendent if others though, of course). The influence of Stoner’s design in modern rifles is impressively prolific. If you want to learn more fascinating trivia, I recommend the YouTube channel “Forgotten Weapons” which is incredibly educational.
What is and isn't an AR is one of those questions along the lines of angels dancing on the head of a pin.
In practice, I think we mostly go by "I know it when I see it", which usually works because most cases are clear-cut. But then you get designs like, say, ARAK-21 or SIG MCX, where it could be reasonably argued either way - and no matter how you argue it, it just muddies the water further.
Let's take MCX, for example - is it an AR? If it is not, then what exactly pushed it over the line? Piston? But we have piston ARs. Incompatible lower and/or upper? Then CMMG Mutant isn't an AR, either. Recoil spring in the receiver, and no buffer or buffer tube? Extar EXP has the same arrangement, but is otherwise a conventional Stoner action.
Okay, so suppose it's an AR. But nobody calls SCAR an AR; so what's the difference between SCAR and MCX that makes the latter an AR?
I don't really have an answer here, if the difference is meant to be qualitative. On the other hand, if it's not about one particular difference, then we could draw a line somewhere and say that it needs to have that much in common - so one difference doesn't push it over the line, but in aggregate, enough of them can.
It's stuff like e.g. the interface between the receiver and the buffer tube, which is weak and known to break under stress. Or the telescoping stock design, which has a similar problem with the retaining pin - which is why they tell you to never mortar it unless fully collapsed.
There are some issues with the bolt longevity - in an AR, the bolt eventually fails with sheared locking lugs - which is unheard of in AK and not common in other designs. Part of it is due to asymmetric distribution of forces, because of the extractor taking the place of one lug. Modern bolt redesigns try to mitigate this in various ways.
Then there's the enclosed magwell, which is great for quick mag changes, but requires the upper part of the magazine to be straight - which caused troubles for 30-round AR mags historically, until the modern polymer magazine design with internal constant curvature made possible by the manufacturing process. In addition, the magwell is prone to dirt-induced malfunctions - once you get the dirt in, e.g. from the lips of a magazine that fell to the ground, it stays inside, and can block the mag from going all the way in easily. Ice can cause similar issues. The button magazine release can get stuck relatively easily as well. AK well-less paddle system, while slower, is less susceptible to dirt, and the rocking insertion motion and the paddle magazine release provide more leverage even if there is an obstruction.
Or we could look at the antiquated handguard design, which results in precious little rail space on the upper receiver, because the top rail on the handguard, even free-floated, is not as stable (and hence not used for magnifying optics usually, but red dots, lasers etc). Notice how the latest trends in military ARs have been monolithic upper receivers, like in New Zealand's MARS-L, or Canadian MMR?
Of course, you could easily make a rifle based on the Stoner action that has none of these problems - there are more exotic AR variants and accessories that solve one or the other, you'd just need to combine all those solutions. But I don't think it would count as an AR at that point. Even the monolithic upper is already a major difference that breaks compatibility in so many ways. I mean, we don't say that SIG 550 is an AK variant, even though it's also long stroke piston, and the BCG looks very similar to AK. There's more to a gun than its action.