Live data from Hacker News

Launch HN: Charge Robotics (YC S21) - Robots that build solar farms

news.ycombinator.com

51–60 of 82 posts

Re: Launch HN: Charge Robotics (YC S21) - Robots that build solar farms

#51
post #15

Is labor cost a significant obstacle to building solar farms? What % of initial capital goes to labor?

I'm more interested in what kind of possibilities this unlocks. Sure, labor might not be a huge cost in projects at present moment. But that might be because all projects today are happening in places where labor is cheap. What if this unlocks projects in more remote and harsher terrains?

Re: Launch HN: Charge Robotics (YC S21) - Robots that build solar farms

#52
post #21
post #15

Is labor cost a significant obstacle to building solar farms? What % of initial capital goes to labor?

Reports are somewhat varied here but we find that about 10-20% of the overall site cost goes into labor. The bulk of the rest is component cost for solar modules and tracking systems, which are relatively commoditized and so are a thin margin above the cost of the raw materials to make them (silicon, steel). Those costs won't drop unless the underlying materials get cheaper. We find that beyond the direct cost, the l…

I am a little surprised that silicon is still competitive. I had the impression that cadmium telluride was much cheaper, in bulk. Or is it about reliability/lifetime?

(For home use, I worry about my neighborhood becoming a superfund site in event of a fire. I read claims that fire burning the house under CdTe panels will not release the 8g/m^2 of cadmium on them, but don't know how to evaluate that.)

Re: Launch HN: Charge Robotics (YC S21) - Robots that build solar farms

#53
post #47
post #33

Earlier quoted context omitted.

Definitely, we've experienced workers being sent home for a number of reasons (too cold, too windy, too rainy, materials didn't show up as scheduled). Fortunately with an autonomous system there's no logistical overhead, if some external factor prevents us from working then we aren't stuck covering hundreds of per diems and sending everyone home. Yes, we'll have some equipment depreciation as it's not working, but it…

“Fortunately with an autonomous system there's no logistical overhead” It might be too early to tell. Those telehandlers and skidsteers are almost always rentals, so they get called off and picked up at the site. Moving heavy machinery is much more difficult than moving people, and all of your equipment is unique so it’s glued to that job site until it’s complete. Owning your own equipment presents its own set of cha…

Yes! Sorry I explained poorly above, I meant there isn't additional logistical overhead involved in stopping operations for a day for robots, whereas there's tons when workers are involved.

You're dead on with regards to the logistics involved in operating this equipment on site, we'll absolutely be facing each of the challenges you mentioned.

Re: Launch HN: Charge Robotics (YC S21) - Robots that build solar farms

#54
post #52
post #21

Earlier quoted context omitted.

Reports are somewhat varied here but we find that about 10-20% of the overall site cost goes into labor. The bulk of the rest is component cost for solar modules and tracking systems, which are relatively commoditized and so are a thin margin above the cost of the raw materials to make them (silicon, steel). Those costs won't drop unless the underlying materials get cheaper. We find that beyond the direct cost, the l…

I am a little surprised that silicon is still competitive. I had the impression that cadmium telluride was much cheaper, in bulk. Or is it about reliability/lifetime? (For home use, I worry about my neighborhood becoming a superfund site in event of a fire. I read claims that fire burning the house under CdTe panels will not release the 8g/m^2 of cadmium on them, but don't know how to evaluate that.)

Yes, silicon is still the majority[1], and it really just boils down to all-in cost. In watts per dollar thin-film ends up being reasonably competitive with silicon, but only because it's manufactured in the US which has steep import tariffs on foreign PV modules.

[1]https://www.usa-cdte.org/

Re: Launch HN: Charge Robotics (YC S21) - Robots that build solar farms

#56

This sounds awesome and it’s great that Solar is so cheap. But if it’s only available several hours in the day at best is it really that practical? Or am I just misinformed and we have grid scale storage already?

A dollar spent on renewables is still better spent on generating capacity, most places, even if peak power is given away part of every day.

Storage cost is falling even faster than solar ever did, so waiting until next year gets you more storage for the price. Also, transmission to local storage is low-loss, so you don't strictly need to be responsible for both. Finally, demand for peak generation will only increase as synthesis (H2, NH3, etc.) capacity comes online. Stockpiled H2 and NH3 will prove entirely adequate as bulk storage despite their low round-trip efficiency, given just a few hours of a better alternative, and anyway that efficiency will only ever improve.

The main attraction of H2 and NH3 storage, despite current lossiness, is that the equipment can operate even after the tanks are full, because there will be unlimited demand for both, for a long time. That is, anytime you are not actually drawing down stored energy, the expensive synthesis equipment can be producing to sell, producing revenue. Similarly for liquified air (N2, really).

Re: Launch HN: Charge Robotics (YC S21) - Robots that build solar farms

#57
I wonder to what degree construction of solar farms over pasture could be automated. It seems like an easy competitive advantage over single-use: there is already revenue on the parcel, grass yield is increased because of reduced heat stress, evaporation is reduced, and generation efficiency is higher. Maybe combined with water collection, cutting erosion?

Sharing with row crops seems like the next level, where you probably need to share with irrigation and harvesting equipment. In the near future, production of NH3 on-site for fertilizer and fuel when grid price bottoms might become practical.

Re: Launch HN: Charge Robotics (YC S21) - Robots that build solar farms

#58
post #55

Will your system be robust to changes in solar farm design? Like if the industry goes towards the Erthos model of installing panels on the ground without racking or trackers?

Single-axis tracked solar isn't going anywhere anytime soon, but we are going to have to adapt if and when the industry changes. We're building relatively generic robotic primitives that should be useful for many different types of system.

Re: Launch HN: Charge Robotics (YC S21) - Robots that build solar farms

#59
post #54
post #52

Earlier quoted context omitted.

I am a little surprised that silicon is still competitive. I had the impression that cadmium telluride was much cheaper, in bulk. Or is it about reliability/lifetime? (For home use, I worry about my neighborhood becoming a superfund site in event of a fire. I read claims that fire burning the house under CdTe panels will not release the 8g/m^2 of cadmium on them, but don't know how to evaluate that.)

Yes, silicon is still the majority[1], and it really just boils down to all-in cost. In watts per dollar thin-film ends up being reasonably competitive with silicon, but only because it's manufactured in the US which has steep import tariffs on foreign PV modules. [1] https://www.usa-cdte.org/

Ah, thanks, I missed the tariff bit.

Re: Launch HN: Charge Robotics (YC S21) - Robots that build solar farms

#60
post #57

I wonder to what degree construction of solar farms over pasture could be automated. It seems like an easy competitive advantage over single-use: there is already revenue on the parcel, grass yield is increased because of reduced heat stress, evaporation is reduced, and generation efficiency is higher. Maybe combined with water collection, cutting erosion? Sharing with row crops seems like the next level, where you p…

It seems like if the soil profile and ground is mapped correctly it could be a much easier operation than one may feel.
Post reply on HN