Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register
This site may earn commission on affiliate links.
Normally, yeah, but if the do have mechanisms that raises the high side, they can get more power out of the side that is facing the side. So start with the East panels tilted toward the sunrise. Then at noon they are at the flattest, toward the end of the day it raises them again to boost the West facing panels.
So a lower cost, lower profile, limited range tilt mechanism. Not saying they are doing that, but it seemed remotely possible after staring at the photo too long...
If that was the design wouldn't the panel of a pair closest to the sun be flat at its most un-optimized angle ?
 
I'll guess at least 600 GWh, perhaps more if the original estimate included a lot of wind energy that has a higher capacity factor than PV. Wind easily produces 3 kWh a year per rated watt, so a 100% wind source of 300 MW = 900 GWh annually.

I agree with the earlier math estimating 200 * 10^6 kWh = 200 GWh generation annually from the entire roof, so if the energy requirements are correct the roof can supply in the neighborhood of 2 - 3 parts of 9 required.

That actually amazes me considering how energy intensive this type of factory must be. Consider that residential homes can easily cover 1/4 of the roof with PV just for everyday living.

Industrial energy use is often far more concentrated in KWH/sf than living space. The GigaFactory is probably more concentrated than the typical industrial building. I'm not sure they would need to charge the batteries after construction. If anything they probably need to drain off some charge.

The materials of the anode and cathode would be completely separate when the cell is put together. The cell discharges as lithium ions migrate to the other side. Recharging moves the lithium ions back to where they started.

It takes a lot of energy to make a cell though. The ingredients need to be extracted from the ore, then processed into the might mix for the components of the cell. The parts of the cell need to be fabricated and cured, then the whole thing needs to be assembled. The actual energy expended making the cell is probably more than the cell can hold.
 
  • Like
Reactions: T3SLA3 and neroden
Industrial energy use is often far more concentrated in KWH/sf than living space. The GigaFactory is probably more concentrated than the typical industrial building. I'm not sure they would need to charge the batteries after construction. If anything they probably need to drain off some charge.

The materials of the anode and cathode would be completely separate when the cell is put together. The cell discharges as lithium ions migrate to the other side. Recharging moves the lithium ions back to where they started.

It takes a lot of energy to make a cell though. The ingredients need to be extracted from the ore, then processed into the might mix for the components of the cell. The parts of the cell need to be fabricated and cured, then the whole thing needs to be assembled. The actual energy expended making the cell is probably more than the cell can hold.

The discharge process involves moving Li and electrons from the graphite (typically) anode to the Li (NMC/NCA) cathode. A new cell has no Lithium at the anode, so does not start in a charged state. A new lithium ion cell also requires a formation charge to create the correct interface on the anode before real use.

Thread discussing this:
Cell production and charging (out of MA)

A rechargeable Li-ion cell is more like a capacitor than a one use zinc carbon cell.

Agree the 3 floor high density design of GF1 will have a considerable energy usage per land area. They did integrate energy recovery techniques (contra flow material preheating, solvent recapture) that will bring the energy down somewhat. Having no combustion based heating processes helps with that.
 
Industrial energy use is often far more concentrated in KWH/sf than living space. The GigaFactory is probably more concentrated than the typical industrial building. I'm not sure they would need to charge the batteries after construction. If anything they probably need to drain off some charge.
I may have been unclear, I was saying that the the Gigafactory roof supplies a higher fraction of the building's requirements than I would have guessed. Tesla must be doing an outstanding job at energy conservation. Perhaps there is also recycling.
 
OK - full collection of all sunlight on this one roof is a big number. 100 million kWh/yr +.

The initial statement regarding GF1's solar roof put its nameplate capacity at 70 MW:

Tesla Gigafactory 1 new drone flyover shows latest massive solar array progress

- and it seems this has not changed (although it could go up a bit, with increasing panel efficiency).

The capacity factor should not exceed 0.28 by much (i.e. around 2500 full-load hours annually), compared to e.g.
Agua Caliente Solar Project - Wikipedia

So the annual production, assuming they get to 70 MW could be on the order of 2.5kh * 70 MW = 175 GWh.

PS. At some point solar panels will be cheap enough that GF1 employees can park in their shade. That will naturally increase the production significantly.
 
Last edited:
  • Like
Reactions: mongo
The initial statement regarding GF1's solar roof put its nameplate capacity at 70 MW:

Tesla Gigafactory 1 new drone flyover shows latest massive solar array progress

- and it seems this has not changed (although it could go up a bit, with increasing panel efficiency).

The capacity factor should not exceed 0.28 by much (i.e. around 2500 full-load hours annually), compared to e.g.
Agua Caliente Solar Project - Wikipedia

So the annual production, assuming they get to 70 MW could be on the order of 2.5kh * 70 MW = 175 GWh.

PS. At some point solar panels will be cheap enough that GF1 employees can park in their shade. That will naturally increase the production significantly.

Why do the panels need to get cheaper to do that? Kettleman SC for instance. Will GF1 employees get free charging?

Need to correct a previous post of mine. GF1 is targeted to produce 150GWh of pack, but only 105 GWh of cells. So the power needed to send the cells/ packs out mostly charged is ~70% of my previous value. (105GWh * 80% SOC = 84 GWh, ignoring processing)
 
Why do the panels need to get cheaper to do that? Kettleman SC for instance. Will GF1 employees get free charging?

OK, in general the principle is that the more PV-electricity you produce, the smaller the fraction of that increased production you are able to consume yourself. That in turn means that the incremental cost of increased nameplate capacity goes up, because you will at times have a (peak) PV-production that you cannot utilize (e.g. during summer), for the benefit of having more power available during non-peak production times (e.g. during winter).

So the panels would need to get cheaper, if the GF1 roof-top installation is (or will be) anywhere close to covering their own consumption.

GF1 however has a _lot_ of battery capacity (e.g. from being able to do burn in tests with their production), which changes things.

If we assume for a moment that GF1 has unlimited battery capacity, then they can use all their PV-production as long as it does not actually exceed their electricity consumption, on up to an annual basis.

I think their battery capacity is somewhere below that (ie. that there would be prohibitive costs incurred by keeping battery packs at the factory to store electricity between seasons).

So if at some point (during the summer months) their PV-production starts to exceed their battery capacity for enough weeks, that they cannot store all the electricity and if during the winter months they cannot cover their entire electricity consumption with PV + batteries, then they can opt to increase their nameplate capacity.

This will allow them to be self-supplying to a larger extent also during the winter, but during the summer they will have excess production that they will have to let go to waste. This in turn means that the panels will need to be cheaper for the PV-extension to be profitable, because the added capacity produces at a lower utilization.

Whether this reasoning makes sense depends on the monthly (winter and summer) roof-top production relative to the monthly electricity consumption at GF1, and on the battery capacity available for storing excess production, even between seasons.

As far as I can understand.

PS edit. Net metering is equivalent to unlimited battery capacity, but I believe GF1 does not have that.
 
Last edited:
  • Helpful
Reactions: mongo
OK, in general the principle is that the more PV-electricity you produce, the smaller the fraction of that increased production you are able to consume yourself. That in turn means that the incremental cost of increased nameplate capacity goes up, because you will at times have a (peak) PV-production that you cannot utilize (e.g. during summer), for the benefit of having more power available during non-peak production times (e.g. during winter).

So the panels would need to get cheaper, if the GF1 roof-top installation is (or will be) anywhere close to covering their own consumption.

GF1 however has a _lot_ of battery capacity (e.g. from being able to do burn in tests with their production), which changes things.

If we assume for a moment that GF1 has unlimited battery capacity, then they can use all their PV-production as long as it does not actually exceed their electricity consumption, on up to an annual basis.

I think their battery capacity is somewhere below that (ie. that there would be prohibitive costs incurred by keeping battery packs at the factory to store electricity between seasons).

So if at some point (during the summer months) their PV-production starts to exceed their battery capacity for enough weeks, that they cannot store all the electricity and if during the winter months they cannot cover their entire electricity consumption with PV + batteries, then they can opt to increase their nameplate capacity.

This will allow them to be self-supplying to a larger extent also during the winter, but during the summer they will have excess production that they will have to let go to waste. This in turn means that the panels will need to be cheaper for the PV-extension to be profitable, because the added capacity produces at a lower utilization.

Whether this reasoning makes sense depends on the monthly (winter and summer) roof-top production relative to the monthly electricity consumption at GF1, and on the battery capacity available for storing excess production, even between seasons.

As far as I can understand.

PS edit. Net metering is equivalent to unlimited battery capacity, but I believe GF1 does not have that.

Sounds like the most cost-effective solution is solar PV (enough to cover summer + fall consumption) + battery + grid (for the winter + spring months). Net-metering wouldn't matter, since they would be a net consumer of electricity in all months. Unless wind provides more reliable winter season power?
 
This will allow them to be self-supplying to a larger extent also during the winter, but during the summer they will have excess production that they will have to let go to waste. This in turn means that the panels will need to be cheaper for the PV-extension to be profitable, because the added capacity produces at a lower utilization.
If there is excess production, it does not need to go to waste.
One (or more) of those wind turbines that were in the early artists renditions of the gigafactory could be pointed at the gigafactory and use up some excess blowing a breeze to keep things cooler........a really big fan.........

upload_2018-9-20_12-34-24.png
 
  • Like
Reactions: ggies07
Net-metering wouldn't matter, since they would be a net consumer of electricity in all months. Unless wind provides more reliable winter season power?

If available, net metering would matter because it would effectively allow GF1 to consume 100% of their own production, also if that production were to exceed their consumption in the summer half-year (by up to the amount missing in covering their own consumption in the winter half-year).

So with Net Metering GF1 could just look at their projected annual electricity consumption and install PV-panels so their nameplate capacity times the projected annual capacity factor (CF) would match that.

I am unsure about the CF for wind turbines around GF1, except that some landscape features like a ridge or pass can act as a funnel, resulting in a CF of up to about 50%. Land turbines usually lack behind sea turbines, that routinely reach 40%.

But in general, a combination of PV and wind gives a production that better matches the consumption, when averaging over seasons.

However, GF1 is a kind of a special case, because they have so much battery capacity (just from burn-in testing) - and presumably also access to PV-panels at competitive prices (from e.g. GF2). And price performance of PV is dropping exponentially - and faster than it is for wind turbines. PV is just solid state physics, no moving parts, no need for a maintenance person with an oil can, for 30 years...

So maybe someone at Tesla made a 5-year projection and concluded that PV + batteries alone is cheaper than wind turbines. I guess we will see as work at GF1 progresses. I will be keeping an eye on that.
 
If available, net metering would matter because it would effectively allow GF1 to consume 100% of their own production, also if that production were to exceed their consumption in the summer half-year (by up to the amount missing in covering their own consumption in the winter half-year).

So with Net Metering GF1 could just look at their projected annual electricity consumption and install PV-panels so their nameplate capacity times the projected annual capacity factor (CF) would match that.

I am unsure about the CF for wind turbines around GF1, except that some landscape features like a ridge or pass can act as a funnel, resulting in a CF of up to about 50%. Land turbines usually lack behind sea turbines, that routinely reach 40%.

But in general, a combination of PV and wind gives a production that better matches the consumption, when averaging over seasons.

However, GF1 is a kind of a special case, because they have so much battery capacity (just from burn-in testing) - and presumably also access to PV-panels at competitive prices (from e.g. GF2). And price performance of PV is dropping exponentially - and faster than it is for wind turbines. PV is just solid state physics, no moving parts, no need for a maintenance person with an oil can, for 30 years...

So maybe someone at Tesla made a 5-year projection and concluded that PV + batteries alone is cheaper than wind turbines. I guess we will see as work at GF1 progresses. I will be keeping an eye on that.

Perhaps I'm mis-understanding how net-metering works. Isn't that credited on a daily basis? Nevada's net-metering credits solar panel production at the wholesale price, so Tesla would effectively need to pump 3kwh of electricity into the grid (during the summer) for every 1kwh of electricity it pulls (in the winter) in order to break even (from a utility-bill perspective).

Or is the net-metering calculated in aggregate at the end of the year?

Because if I'm right, the cost savings from the difference between the net-metering credit and the industrial electricity rates wouldn't justify the capex of installing the additional panels needed to cover the electricity consumed in the winter/spring times.

e.g:
assuming industrial non-peak electricity is billed at 12c/kwh, while wholesale value is 4c/kwh. 100KW of additional panels would produce 500kwh of electricity per day that the grid would pay 4c/kwh for during the summer. Then during the winter, Tesla uses 500kwh of electricity less per day than they would've otherwise had to buy from the grid (thus saving them 12c/kwh). That gives them a savings of 500 * 4cents + 500 * 12cents per 2 days (1 summer + 1 winter). But it costs them 100,000W * $1/Watt to buy/install those panels.

If net-metering was NOT factored, then it's strictly about the 500 * 12 cents per day in electricity savings.

That's what I meant about it being more cost-effective to NOT factor in net-metering.
 
Perhaps I'm mis-understanding how net-metering works. Isn't that credited on a daily basis? Nevada's net-metering credits solar panel production at the wholesale price, so Tesla would effectively need to pump 3kwh of electricity into the grid (during the summer) for every 1kwh of electricity it pulls (in the winter) in order to break even (from a utility-bill perspective).

Or is the net-metering calculated in aggregate at the end of the year?

Because if I'm right, the cost savings from the difference between the net-metering credit and the industrial electricity rates wouldn't justify the capex of installing the additional panels needed to cover the electricity consumed in the winter/spring times.

e.g:
assuming industrial non-peak electricity is billed at 12c/kwh, while wholesale value is 4c/kwh. 100KW of additional panels would produce 500kwh of electricity per day that the grid would pay 4c/kwh for during the summer. Then during the winter, Tesla uses 500kwh of electricity less per day than they would've otherwise had to buy from the grid (thus saving them 12c/kwh). That gives them a savings of 500 * 4cents + 500 * 12cents per 2 days (1 summer + 1 winter). But it costs them 100,000W * $1/Watt to buy/install those panels.

If net-metering was NOT factored, then it's strictly about the 500 * 12 cents per day in electricity savings.

That's what I meant about it being more cost-effective to NOT factor in net-metering.
Isn't the claim net zero carbon footprint? The price they pay for electricity drawn doesn't affect the fact that they are still producing as many kwh.
 
The discharge process involves moving Li and electrons from the graphite (typically) anode to the Li (NMC/NCA) cathode. A new cell has no Lithium at the anode, so does not start in a charged state. A new lithium ion cell also requires a formation charge to create the correct interface on the anode before real use.

Thread discussing this:
Cell production and charging (out of MA)

A rechargeable Li-ion cell is more like a capacitor than a one use zinc carbon cell.

Agree the 3 floor high density design of GF1 will have a considerable energy usage per land area. They did integrate energy recovery techniques (contra flow material preheating, solvent recapture) that will bring the energy down somewhat. Having no combustion based heating processes helps with that.

Sorry I was shooting from the hip as far as the condition of newly made cells. Yes, they do need to go through a special cycle to initialize the cell after manufacturing.

Actually I would say rechargeable batteries are much more similar to single use batteries than capacitors. The only thing rechargeable batteries and capacitors have in common is the ability to charge them, but the internal processes are very different. All batteries do their thing through a chemical reaction, it's just that with some chemistries once the reaction is done, it can't be reversed and others can be reversed by applying electricity. The charging process reverses the chemical reaction that happened when the battery discharged.

Capacitors do have a lot of research that goes into the electrolyte between the two charged plates and the electrolyte determines how much charge a capacitor can hold, but the process of charging and discharging does not involve a chemical process like batteries do. Neither the plates nor the electrolyte undergo any chemical changes during charge and discharge.
 
Perhaps I'm mis-understanding how net-metering works. Isn't that credited on a daily basis? Nevada's net-metering credits solar panel production at the wholesale price, so Tesla would effectively need to pump 3kwh of electricity into the grid (during the summer) for every 1kwh of electricity it pulls (in the winter) in order to break even (from a utility-bill perspective).

Or is the net-metering calculated in aggregate at the end of the year?

Because if I'm right, the cost savings from the difference between the net-metering credit and the industrial electricity rates wouldn't justify the capex of installing the additional panels needed to cover the electricity consumed in the winter/spring times.

e.g:
assuming industrial non-peak electricity is billed at 12c/kwh, while wholesale value is 4c/kwh. 100KW of additional panels would produce 500kwh of electricity per day that the grid would pay 4c/kwh for during the summer. Then during the winter, Tesla uses 500kwh of electricity less per day than they would've otherwise had to buy from the grid (thus saving them 12c/kwh). That gives them a savings of 500 * 4cents + 500 * 12cents per 2 days (1 summer + 1 winter). But it costs them 100,000W * $1/Watt to buy/install those panels.

If net-metering was NOT factored, then it's strictly about the 500 * 12 cents per day in electricity savings.

That's what I meant about it being more cost-effective to NOT factor in net-metering.
In MY net metering case....I produce kW during the sunny day from my solar array, and those kW above my consumption go into the grid. I pull from the grid at night, and will probably pull from the grid in the winter. In October, we balance the excess minus the pull and I get paid for the excess, or pay for consumption, all at the same rate of $0.12/kWh. This contract is in place for 20 years.
The ability to obtain this net-metering contract expired Sept 1, 2018. The new deal is not as good. New guys are required to sign up on a time-of-use deal, where evening rates are higher than mid day, so you "sell" at mid day rates, and buy at evening rates, but still settle in Oct.. If I was a larger customer, they even put in Demand charges - incentives to keep max power below a certain level by load shifting. A battery would make sense at load shifting but not makes sense under net-metering.