Why would a factory like this not be able to produce many other products that require press forged parts like e.g. compressed gas containers (cylinders, similar sizes, also hollow, also closed tops, also painted) or other parts (e.g. forged boat / ship propellers or water turbines).
I totally agree however that 3D printing can’t be an economical viable option for mass producing standardised parts in high numbers.
The final question is, if the metal casing is even still needed if we would instead use similarly sized bombs dropped from drones. The launch forces would be avoided, the shrapnel could be steel balls in a polymer casing with explosives at the core.
Maybe the time of artillery is over when fixed wing autonomous AI-drones (unjammable) are emerging on the battlefield in swarms. That day has either already arrived or is imminent.
You're raising some good questions. I don't think I can definitively answer your question about what else could be made, though there is a big question around optimization. If it's optimized for steel casings for artillery shells, there aren't many other products that require that metallurgy (it's designed to fracture for maximum lethality) or that size. Financially, the business plan for most defense production only closes if the government owns the tools due to how cost plus pricing works. If the government owns the tools, they are less sensitive to someone else making it. That said, the new plant in Mesquite uses a completely different form of metal working known as flow forming, which I understand was previously used for wheels. May be an opportunity there.
As for bombs via drones, again, you're asking the right question. That said, the distinguishing feature of the war in Ukraine is the lack of air superiority. NATO's military strategy is predicated on air superiority to a degree that it's difficult to believe a NATO war would involve relatively static lines seen in Ukraine. That's not to underestimate the impact of drones and so-called loitering munitions—they have proven their place on the modern battlefield. I'm just not sure how all these pieces would work together, and in general, I suspect precise, low-latency artillery will still have a place alongside hunter drones with explosives.
I can of course also just speculate but usually a certain share of the industrial process equipment can be dual used if enough creativity is allowed to enter the planning process. i once visited a factory for gas cylinders that are widely used to hold industrial gases (e.g. for welding) and not the process looked almost identical to the shell production video which is not surprising because it has a very similar shape. The metallurgical detailed specifications for the respective alloys might differ but a hydraulic forging press etc. should be generic enough to handle both - especially if taking it into account in the planning phase already.
Similarly my drone and drone bomb speculation is just that. It seems very plausible to me that autonomous low cost drone swarms will change most doctrines - including achieving classical air superiority with a small number of super sophisticated and expensive jet fighter and air defence systems. They are easily outnumbered and overwhelmed in the future - somewhat comparable to what we saw with the Russian Black Sea navy.
So many what ifs and possibilities. EG what if a plant were established in a country where exports were more flexible the documents were sourced from a country other than the US?
Hi Rob, very interesting reading. Im not an engineer, though well educated, in the manufacturing space in Australia and somewhat of an entrepreneur.
I'm fascinated and intrigued as to how and why the governments of the free world have historically and in today's information world allowed themselves to become so weak they cannot produce/ manufacture enough product to protect themselves.
These governments who are acting more like small school children, are constantly passing the buck. Constantly playing the lazy bully who steals the smaller kids lunch untill the small kids grow up and realize how to say NO MORE.
Our democratic governments have now put themselves in a position that collectively they cannot combat the manufacturing capability of a single country that now poses serious threat to the free world.
That said and rant finished, have you discovered a viable way to obtain the correct machinery, raw resources and manufacturing personnel to increase production of the 155mm M795 if money or permission wasn't a concern? Maybe a cost effective redesign with slightly modified steel as someone had previously mentioned a pineapple or maybe honeycomb design?
My impression is that all the technology involved is actually pretty available. There are some smart things you could do to modernize it, which will be the subject of a piece I'm working on at the moment. Everything just keeps coming back to a lack of will, primarily from Congress.
My understanding is that it is possible but complicated. The DoD would have to release the design files to you. The equipment exists but is also ~$100mm. The parts may be under export control, so without a DoD contract you may not be able to sell the parts. But I don’t think the government would care if you’re making a private stockpile of steel parts.
Thank you for the legwork. Is there room for optimization by trading off alloy fragmentation properties for additional complexity in the explosives filler? Or is yield strength the limiting factor and this is the shortest path?
You would need an ordnance expert to really answer that question, as I'm not sure how much you could optimize the explosives filler. Notably, the Army is backing off their modern explosive and returning to TNT, so it's a non-trivial problem to change anything about the explosive.
The main thing that the research showed is that how the cracks in the metal form is key to getting the highest level of fragmentation and therefore lethality. Notably, the military's new factories continue to use HF-1, so I suspect there is no appetite for trading off lethality. Simpler steel shells may also not be drop-in compatible with the existing weapons systems, which has its own headaches.
You mentioned yield strength, which was mostly associated with the need to be able to push the shell as far as possible i.e. range. I'm not sure if that's the limiting factor. There are a number of versions that extend the range with things like base-burners. I suspect the default range was "good enough" but the lethality had to be maximized.
One other important dimension I didn't cover much is that the M795 (the US Army's default 155 mm shell) is noted as 103 lbs filled, which means a single soldier could conceivably move it and two can effectively load a howitzer. It's a major reason the 155 mm became so popular in the first place, it's a good combination of range, lethality, and practicality in logistics.
From what I can find in my very shallow search, the shift toward TNT rather than Octol would be driven by cost and environmental concerns, rather than performance. I'm not qualified to know what tactical advantage it would provide over TNT fires, given that sheer quantity of available fires is the chief concern. (For guided munitions I haven't found anything to suggest a move away from Octol.)
So long as casings are the bottleneck, guidance systems to reduce the number of fires needed are probably more important than increasing the energetic or fragmentation characteristics of the payload. AFAIK it's still an open question as to whether the inferior fragmentation properties of a different, more readily-available and inexpensive steel may be offset by a different loading, such as frangibles set into the explosives, or a different manufacturing process, say an internal "pineapple pattern" cut into the shell or some other form of weakening. Maybe 3D printing offers novel solutions in this direction?
If sheer quantity is more important than lethality, then I'd think producing twice as many rounds at the expense of lethality would be worth it, especially to offset temporary shortages. For whatever my opinion is worth.
A key question I didn’t see in this piece: Could a factory that produces 155 mm shells also produce other munitions that would be more useful in a pacific war? Or could it be economically retrofitted to do so?
Based on the CSIS wargame, essentially no munitions seemed useful in the Pacific theater. Missile range dominates every other factor, and each ship is a huge, concentrated target. With bombers shooting 50+ missiles per aircraft, large munition are a complete non-factor.
Nice article, that got my gears turning. I think that producing casings with 3d printing is the wrong focus, given that 3d printers trade speed for flexibility. What about printing the presses or other tooling, or parts of it? Would it be possible to cut time and cost for a new production line?
I covered it for completeness, not really because I thought it would be a great idea. Notably, the machine tools are what DOD points to as the critical path, such that accelerating getting them in would allegedly accelerate the overall production schedule. However, I have interpreted that to mean the large, heavy equipment required to build a new factory. I didn't see many small, fine parts in the manufacturing line or final product BOM that lend themselves to modern 3D printing.
Not to disparage our industrial base issues (which we definitely have), it's worth noting that that it took YEARS to build up that industrial base for WW2.
Based on this, 2+ years to scale up production capacity seems right in track (esp. for a war we're not actually fighting, so less political oomph at home)
I don't think a liberal democracy will ever be able to be as nimble as an authoritarian state, but we can make moves when the needs arise. Agree that deterrence would be a hell of a lot cheaper!
A very fair point. The elephant in the room here is China, who far outpaces the US in manufacturing capacity. Most analysts say that the US doesn't need to outpace China, rather they can rely on other liberalized, democratic states to provide the manufacturing capacity that would outpace China alone. I believe that, but I worry a lot about what those years would look like and the possible loss of life. China is just at a different scale from Germany or Japan at the time, and they are working now on being wholly self-sufficient in establishing the machine tools and factories required to sustain that nightmarish conflict that never reaches the nuclear threshold. And while war may not be on our doorstep yet and we certainly should not go looking for it, I recognize this is a war of authoritarianism versus the free people of the world. I would have us give the arms needed to the countries that are willing to fight and die for the right to rule themselves, the only legitimate form of governance. We simply have to get out of our own way.
That is not an area I researched much, but I can offer a few thoughts that could guide your own research. For one, they had less or no permitting, you could put a steel mill wherever it was most economical. As a result, they had a much higher industrial base to begin with. Similarly, those sites were likely versatile. SCAAP started its life making train cars and re-tooled into shells. Many of those steel mills during WW1 could have re-tooled relatively quickly to meet the demand from the governments. Considering HF-1 wasn't invented until the 70s, they were using some less precise form of steel. Based on how Bethlehem Steel described the sophistication of HF-1, I would also assume earlier shell steel was far simpler to manufacture. So I would expect they had many more facilities with fewer rules producing many more shells that were slightly worse than what SCAAP produces today.
Very informative. Ukraine is a serious issue. If they win, the Russians and Chinese will back off.
If they lose, Russia will start taking over by force in Europe, China by force in Asia.
Just as Hitler and WWII could have been easily stopped in 1934, Now is the time and the Ukraine is the place to stop Russia and China.
Why would a factory like this not be able to produce many other products that require press forged parts like e.g. compressed gas containers (cylinders, similar sizes, also hollow, also closed tops, also painted) or other parts (e.g. forged boat / ship propellers or water turbines).
I totally agree however that 3D printing can’t be an economical viable option for mass producing standardised parts in high numbers.
The final question is, if the metal casing is even still needed if we would instead use similarly sized bombs dropped from drones. The launch forces would be avoided, the shrapnel could be steel balls in a polymer casing with explosives at the core.
Maybe the time of artillery is over when fixed wing autonomous AI-drones (unjammable) are emerging on the battlefield in swarms. That day has either already arrived or is imminent.
You're raising some good questions. I don't think I can definitively answer your question about what else could be made, though there is a big question around optimization. If it's optimized for steel casings for artillery shells, there aren't many other products that require that metallurgy (it's designed to fracture for maximum lethality) or that size. Financially, the business plan for most defense production only closes if the government owns the tools due to how cost plus pricing works. If the government owns the tools, they are less sensitive to someone else making it. That said, the new plant in Mesquite uses a completely different form of metal working known as flow forming, which I understand was previously used for wheels. May be an opportunity there.
As for bombs via drones, again, you're asking the right question. That said, the distinguishing feature of the war in Ukraine is the lack of air superiority. NATO's military strategy is predicated on air superiority to a degree that it's difficult to believe a NATO war would involve relatively static lines seen in Ukraine. That's not to underestimate the impact of drones and so-called loitering munitions—they have proven their place on the modern battlefield. I'm just not sure how all these pieces would work together, and in general, I suspect precise, low-latency artillery will still have a place alongside hunter drones with explosives.
I can of course also just speculate but usually a certain share of the industrial process equipment can be dual used if enough creativity is allowed to enter the planning process. i once visited a factory for gas cylinders that are widely used to hold industrial gases (e.g. for welding) and not the process looked almost identical to the shell production video which is not surprising because it has a very similar shape. The metallurgical detailed specifications for the respective alloys might differ but a hydraulic forging press etc. should be generic enough to handle both - especially if taking it into account in the planning phase already.
Similarly my drone and drone bomb speculation is just that. It seems very plausible to me that autonomous low cost drone swarms will change most doctrines - including achieving classical air superiority with a small number of super sophisticated and expensive jet fighter and air defence systems. They are easily outnumbered and overwhelmed in the future - somewhat comparable to what we saw with the Russian Black Sea navy.
"It treats time like it’s free. It does not respond quickly. It does not build or innovate around current problems"
This is also how the FDA behaves: https://jakeseliger.com/2024/01/29/the-dead-and-dying-at-the-gates-of-oncology-clinical-trials/. No one at the FDA seems to consider opportunity costs or what can be done with money apart from investing in pharmaceuticals. The result is the retarding of the pharma market and a lot of people, like me, dying prematurely.
So many what ifs and possibilities. EG what if a plant were established in a country where exports were more flexible the documents were sourced from a country other than the US?
Hi Rob, very interesting reading. Im not an engineer, though well educated, in the manufacturing space in Australia and somewhat of an entrepreneur.
I'm fascinated and intrigued as to how and why the governments of the free world have historically and in today's information world allowed themselves to become so weak they cannot produce/ manufacture enough product to protect themselves.
These governments who are acting more like small school children, are constantly passing the buck. Constantly playing the lazy bully who steals the smaller kids lunch untill the small kids grow up and realize how to say NO MORE.
Our democratic governments have now put themselves in a position that collectively they cannot combat the manufacturing capability of a single country that now poses serious threat to the free world.
That said and rant finished, have you discovered a viable way to obtain the correct machinery, raw resources and manufacturing personnel to increase production of the 155mm M795 if money or permission wasn't a concern? Maybe a cost effective redesign with slightly modified steel as someone had previously mentioned a pineapple or maybe honeycomb design?
My impression is that all the technology involved is actually pretty available. There are some smart things you could do to modernize it, which will be the subject of a piece I'm working on at the moment. Everything just keeps coming back to a lack of will, primarily from Congress.
What if it was fully funded by private institutions?
My understanding is that it is possible but complicated. The DoD would have to release the design files to you. The equipment exists but is also ~$100mm. The parts may be under export control, so without a DoD contract you may not be able to sell the parts. But I don’t think the government would care if you’re making a private stockpile of steel parts.
Thank you for the legwork. Is there room for optimization by trading off alloy fragmentation properties for additional complexity in the explosives filler? Or is yield strength the limiting factor and this is the shortest path?
You would need an ordnance expert to really answer that question, as I'm not sure how much you could optimize the explosives filler. Notably, the Army is backing off their modern explosive and returning to TNT, so it's a non-trivial problem to change anything about the explosive.
The main thing that the research showed is that how the cracks in the metal form is key to getting the highest level of fragmentation and therefore lethality. Notably, the military's new factories continue to use HF-1, so I suspect there is no appetite for trading off lethality. Simpler steel shells may also not be drop-in compatible with the existing weapons systems, which has its own headaches.
You mentioned yield strength, which was mostly associated with the need to be able to push the shell as far as possible i.e. range. I'm not sure if that's the limiting factor. There are a number of versions that extend the range with things like base-burners. I suspect the default range was "good enough" but the lethality had to be maximized.
One other important dimension I didn't cover much is that the M795 (the US Army's default 155 mm shell) is noted as 103 lbs filled, which means a single soldier could conceivably move it and two can effectively load a howitzer. It's a major reason the 155 mm became so popular in the first place, it's a good combination of range, lethality, and practicality in logistics.
From what I can find in my very shallow search, the shift toward TNT rather than Octol would be driven by cost and environmental concerns, rather than performance. I'm not qualified to know what tactical advantage it would provide over TNT fires, given that sheer quantity of available fires is the chief concern. (For guided munitions I haven't found anything to suggest a move away from Octol.)
So long as casings are the bottleneck, guidance systems to reduce the number of fires needed are probably more important than increasing the energetic or fragmentation characteristics of the payload. AFAIK it's still an open question as to whether the inferior fragmentation properties of a different, more readily-available and inexpensive steel may be offset by a different loading, such as frangibles set into the explosives, or a different manufacturing process, say an internal "pineapple pattern" cut into the shell or some other form of weakening. Maybe 3D printing offers novel solutions in this direction?
If sheer quantity is more important than lethality, then I'd think producing twice as many rounds at the expense of lethality would be worth it, especially to offset temporary shortages. For whatever my opinion is worth.
A key question I didn’t see in this piece: Could a factory that produces 155 mm shells also produce other munitions that would be more useful in a pacific war? Or could it be economically retrofitted to do so?
Based on the CSIS wargame, essentially no munitions seemed useful in the Pacific theater. Missile range dominates every other factor, and each ship is a huge, concentrated target. With bombers shooting 50+ missiles per aircraft, large munition are a complete non-factor.
Nice article, that got my gears turning. I think that producing casings with 3d printing is the wrong focus, given that 3d printers trade speed for flexibility. What about printing the presses or other tooling, or parts of it? Would it be possible to cut time and cost for a new production line?
I covered it for completeness, not really because I thought it would be a great idea. Notably, the machine tools are what DOD points to as the critical path, such that accelerating getting them in would allegedly accelerate the overall production schedule. However, I have interpreted that to mean the large, heavy equipment required to build a new factory. I didn't see many small, fine parts in the manufacturing line or final product BOM that lend themselves to modern 3D printing.
Not to disparage our industrial base issues (which we definitely have), it's worth noting that that it took YEARS to build up that industrial base for WW2.
https://www.ibiblio.org/hyperwar/USA/BigL/img/BigL-p59.jpg
Based on this, 2+ years to scale up production capacity seems right in track (esp. for a war we're not actually fighting, so less political oomph at home)
I don't think a liberal democracy will ever be able to be as nimble as an authoritarian state, but we can make moves when the needs arise. Agree that deterrence would be a hell of a lot cheaper!
A very fair point. The elephant in the room here is China, who far outpaces the US in manufacturing capacity. Most analysts say that the US doesn't need to outpace China, rather they can rely on other liberalized, democratic states to provide the manufacturing capacity that would outpace China alone. I believe that, but I worry a lot about what those years would look like and the possible loss of life. China is just at a different scale from Germany or Japan at the time, and they are working now on being wholly self-sufficient in establishing the machine tools and factories required to sustain that nightmarish conflict that never reaches the nuclear threshold. And while war may not be on our doorstep yet and we certainly should not go looking for it, I recognize this is a war of authoritarianism versus the free people of the world. I would have us give the arms needed to the countries that are willing to fight and die for the right to rule themselves, the only legitimate form of governance. We simply have to get out of our own way.
Thank you. It is very informative.
My question is how all sides participated in WW1 succeded producing around 1.5 billions of shells in short period of time?
That is not an area I researched much, but I can offer a few thoughts that could guide your own research. For one, they had less or no permitting, you could put a steel mill wherever it was most economical. As a result, they had a much higher industrial base to begin with. Similarly, those sites were likely versatile. SCAAP started its life making train cars and re-tooled into shells. Many of those steel mills during WW1 could have re-tooled relatively quickly to meet the demand from the governments. Considering HF-1 wasn't invented until the 70s, they were using some less precise form of steel. Based on how Bethlehem Steel described the sophistication of HF-1, I would also assume earlier shell steel was far simpler to manufacture. So I would expect they had many more facilities with fewer rules producing many more shells that were slightly worse than what SCAAP produces today.