Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

For a counterweight let me present this interview[0] with the CEO of "the largest privately held solar contracting company in America", near the end of which he says several disparaging things about Tesla's roof, including,

> When I saw the demo he did at Universal Studios... What I saw was a piece of glass that looked like it had a cell in it. The challenges he’s going to have is, how are you going to wire it? Every one of those shingles has to be wired.

> Roofs have valleys and they have hips and they have pipes. … How are you going to work around that? How are you going to cut that glass? Are you going to cut right through the cell?

The latter question is perhaps answered by the posted article, "Solar Roof uses two types of tiles—solar and non-solar." So Petersen's question is moot, the glass/solar tiles don't have to be cut to fit in a hip or around a flue, that will be done to the non-solar tiles that look the same.

The question of wiring is open: imagine the grid of wires that have to underly that roof, and getting them all put down without a break or a short, by big guys with nail guns (if you've ever watched roofers at work -- it isn't a precision operation).

Then Petersen goes on to say,

> So I would say for the record ... it’ll be cost-prohibitive. ... For $55,000 I can give you a brand-new roof that will last forever — 50 years — and I can give you all the solar you can handle. ... (Musk’s) product is going to be north of $100,000.

The graph in the posted article does not directly address total up-front installed cost, but rather tries to combine cost with some anticipated lifetime energy return -- a procedure with a LOT of variables and assumptions. I would like to see real numbers for a Tesla roof, $/sq.yd installed.

[0] http://www.mercurynews.com/2017/05/04/from-summer-job-to-sol...



You can use their calculator to edit any of the assumptions, and the math is fairly transparent.

The question of installation is a straightforward challenge that's easily testable, presumably they have a solution that works or they wouldn't announce yet. It's pretty obvious to see if it's going to work or not: just send a team out to put on a test roof and see how much work it is.

[0] https://www.tesla.com/solarroof


OK I just put my address into the calculator and I think the thing's bogus. Without asking anything about my house size or my solar exposure, based solely on my address, it gives me a hard figure ($53,100) for the roof cost and a 30-year energy income of $76,100.

I call bullshit. First, I do not believe they actually know the size of my house or the orientation of the (numerous) roof pitches to the sun. Spoiler: it's only 1300 sq feet and the longest roof ridge is oriented NW-SE, far from optimal for exposure.

Then there are the three very large oak trees that shadow it from three sides.

Then there is the fact that my utility is a local one, not PG&E, and I doubt they took the different rate structure into account. And if they did, how the F do they know what Palo Alto utilities will charging in 10, 20 or 30 years?

They don't know. The whole calculator is a bogus fantasy toy.


That's actually exactly the data they're getting from Google's project Sunroof. I'm not sure how correct that data is, but I haven't seen many complaints.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: