Ivanpah Solar Thermal Units Shutting Down, as Tech Shifts (opens in new tab)

(enr.com)

17 pointsLMSolar0y ago4 comments

4 comments

I wonder if it would be worth retrofitting the mirror trackers as tracking panel racks? It seems like a ready-made solar panel facility, with land, exposure, trackers, and power egress?

Are there any plans for a retrofit?

LMSolarOP11mo ago

I have no inside info re Ivanpah, but we often get folks asking if our H1 heliostat [1] can extend the generation window of an existing PV array, or if they can mount PV on our H1 to extend the working day.

While both 'work' technically, you're still paying to buy and maintain the electro-mechanical system of the heliostat to get an improvement in PV output [2]

So, the active-tracking investment is in competition with the brute-force approach: just buy and install more super-affordable PV panels. e.g. put some oriented for morning light, some for afternoon, etc.

I've heard first-hand that SV VC investors were shocked how fast PV got cheap vs more-complex / active systems like Ivanpah.

If you're optimizing a single PV asset for max output (constrained PV panel supply), active tracking pays off - but in most non-maritime or space applications, it's better just to install more fixed-position PV panels because they are cheap and because supporting a mechanical system can be expensive.

All that said, we think there are huge applications for non-CSP use of heliostats, e.g. for industrial and agricultural heat. [3]

But CSP's usefullness for electricity production is uncertain at best, IMO.

[1] http://www.lm.solar/heliostats/ [2] https://www.sciencedirect.com/science/article/abs/pii/S22131... [3] https://lm.solar/solar-plastic-molding/

K0balt11mo ago

Yeah, more panels almost always FTW. I manage a microgrid on Hispaniola that serves a farm and a small community. The bottom line almost always falls on”more panels instead” no matter what improvements we are contemplating. Sometimes it’s “better batteries” but usually it’s “more panels”.

pingou0y ago

Are the maintenance costs so high that it's better to just kill it? Why is that? I thought the costly thing would be to build it (and indeed it was, at more than $2 billions).

j / k navigate · click thread line to collapse