“So critical. Important minerals. Very wow.” Excellent overview of the fundamental rate-limitations and what is causing them. Spamming the re-stack button ;)
Love it. Thanks for leaping over the soundbite wall.
Agreed that policy is just part of the supply chain solve. But my concern is that current US admin lacks minimum viable policy vision. The ultimate bottleneck.
Thanks though for articulating the need with such clarity. Well done.
To what extent could "batteries everywhere" ameliorate the grid situation? While batteries suffer from the same need for electronic control parts as detailed here for the grid and wind turbines, as long as China is selling them, why shouldn't they be installed at scale to at least shift some of the grid problems along the timeline.
Grid-scale battery farms can store off-peak production for peak transmission; 1-4MWh 20-foot containers can be sited at substations and reportedly delay a need to upgrade $5,000,000 transformers and also provide peak-shaving. Local siting as in Minnesota's distributed utility-owned battery program ( https://www.volts.wtf/p/sooner-than-you-think-electricity ) can alleviate local distribution problems. A program to provide utility-owned 10kWh batteries at the meter for households could provide free blackout resilience for residents and VPP-type peak power shaving for the utility.
None of that solves the supply problems which the analysis reveals, but could provide some short-term relief and long-term flexibility while larger supply chain dependencies are addressed. At least China is not (yet) threatening to cut off end-user products like batteries at all scales, solar panels, and EVs.
Solid state transformers would also be good for robustness to solar flares right? BTW, instead of wondering about China, a low-hanging fruit would be using the L1 satellites 15 min warning to reduce risk on transformers - low cost, but you need to fight institutional inertia. Which actually is probably the real issue behind much of the "Chinese threat".
the transformer bottleneck is real for US hyperscalers. the part that does not get said is that it is not symmetric. china three largest transformer manufacturers, TBEA, SINAI, and CHINT, account for something like 40% of global large-transformer capacity. bytedance and alibaba are not waiting 18-24 months for substation equipment, they are ordering from the same domestic suppliers who just spent a decade scaling for grid modernization. the supply problem you are diagnosing is partly a china-built its own supply chain story running in the background.
You are correct. China has a built in advantage because its modernization is recent and its grid has been growing rapidly. For the US, the grid was built 75 years ago. Recent demand has been for replacement and incremental growth. Arguably the US has not invested to modernize but that may have been Special interests. Now the advent of ai has shifted the demand curve radically higher and the US has no inherent production capacity.
the 75-year gap shows up in transformer lead times. TBEA, SINAI, CHINT can cover roughly 40% of global large-transformer capacity, which is why ByteDance and Alibaba are not sitting in the same 18-24 month queue as AWS. the asymmetry is not just historical, it is getting wider.
Yes, when it comes to most aspects of core infrastructure, China has a distinct set of advantages. If the US can expedite SMR nuclear, it can minimize ST demands. Necessity sometimes helps force progress!
Overall, great stuff. My only pushback is on the grid-forming side. Given Spain's blackouts, most countries seem to be aggressively adopting mandatory grid forming from stationary storage through DERs and utility-scale BESS. The future will be much more resilient, adaptive grids, but we need all those power electronics to get there.
I’m continually impressed with the depth of research and quality of the analysis you all put out. This was very informative and interesting.
“So critical. Important minerals. Very wow.” Excellent overview of the fundamental rate-limitations and what is causing them. Spamming the re-stack button ;)
Love it. Thanks for leaping over the soundbite wall.
Agreed that policy is just part of the supply chain solve. But my concern is that current US admin lacks minimum viable policy vision. The ultimate bottleneck.
Thanks though for articulating the need with such clarity. Well done.
Very interesting analysis. So many problems.
To what extent could "batteries everywhere" ameliorate the grid situation? While batteries suffer from the same need for electronic control parts as detailed here for the grid and wind turbines, as long as China is selling them, why shouldn't they be installed at scale to at least shift some of the grid problems along the timeline.
Grid-scale battery farms can store off-peak production for peak transmission; 1-4MWh 20-foot containers can be sited at substations and reportedly delay a need to upgrade $5,000,000 transformers and also provide peak-shaving. Local siting as in Minnesota's distributed utility-owned battery program ( https://www.volts.wtf/p/sooner-than-you-think-electricity ) can alleviate local distribution problems. A program to provide utility-owned 10kWh batteries at the meter for households could provide free blackout resilience for residents and VPP-type peak power shaving for the utility.
None of that solves the supply problems which the analysis reveals, but could provide some short-term relief and long-term flexibility while larger supply chain dependencies are addressed. At least China is not (yet) threatening to cut off end-user products like batteries at all scales, solar panels, and EVs.
A really terrific article on solar and battery potential worldwide, published June 19:
https://www.electrotech-revolution.com/p/harnessing-the-sun-the-solar-battery
Fantastic overview, thank you.
There is a similar situation in *high-frequency* electronics too: GaAs is used for radars, missiles, EW, etc. Without Gallium, no GaAs chips.
Look up MACOM, Skyworks, Triquint/RF Microdevices/Qorvo, etc.
Amazing work.
Solid state transformers would also be good for robustness to solar flares right? BTW, instead of wondering about China, a low-hanging fruit would be using the L1 satellites 15 min warning to reduce risk on transformers - low cost, but you need to fight institutional inertia. Which actually is probably the real issue behind much of the "Chinese threat".
hey did you mean anywhere from China or anywhere but China?
the transformer bottleneck is real for US hyperscalers. the part that does not get said is that it is not symmetric. china three largest transformer manufacturers, TBEA, SINAI, and CHINT, account for something like 40% of global large-transformer capacity. bytedance and alibaba are not waiting 18-24 months for substation equipment, they are ordering from the same domestic suppliers who just spent a decade scaling for grid modernization. the supply problem you are diagnosing is partly a china-built its own supply chain story running in the background.
You are correct. China has a built in advantage because its modernization is recent and its grid has been growing rapidly. For the US, the grid was built 75 years ago. Recent demand has been for replacement and incremental growth. Arguably the US has not invested to modernize but that may have been Special interests. Now the advent of ai has shifted the demand curve radically higher and the US has no inherent production capacity.
the 75-year gap shows up in transformer lead times. TBEA, SINAI, CHINT can cover roughly 40% of global large-transformer capacity, which is why ByteDance and Alibaba are not sitting in the same 18-24 month queue as AWS. the asymmetry is not just historical, it is getting wider.
Yes, when it comes to most aspects of core infrastructure, China has a distinct set of advantages. If the US can expedite SMR nuclear, it can minimize ST demands. Necessity sometimes helps force progress!
Overall, great stuff. My only pushback is on the grid-forming side. Given Spain's blackouts, most countries seem to be aggressively adopting mandatory grid forming from stationary storage through DERs and utility-scale BESS. The future will be much more resilient, adaptive grids, but we need all those power electronics to get there.