Perfectlogic
Member
Efficiency is not what matters though, cost is.
@jhm
Tesla will definately install the panels before 30% ITC expiration, regardless of how far along they are with the rest of the factory.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Tesla Gigafactory Investor Thread
- Thread starter teddyg
- Start date
-
- Tags
- Gigafactory TSLA
Efficiency is not what matters though, cost is.
@jhm
Tesla will definately install the panels before 30% ITC expiration, regardless of how far along they are with the rest of the factory.
Off Shore
Off Topic Member
I'd like to see some analysis for this. It must be something like the amount of heat, speed of heat, level of heat, efficiency of heat transfer, heat retention, heat format, etc.. Am I getting warm? Pun only incidental.
I mean, if you take mirrors and point them at a manufacturing process, the sun could bring any focus point to pretty much any desired temperature, given enough mirrors. Add in tracking and/or holographic tracking, and it becomes a question of tracking and mirror cost vs. other heat source cost.
Are some of the needed chemical reactions in PV manufacturing actually requiring certain types of fuel that cannot be done with focused sunlight? You could turn sunlight into lasers and heat up pretty much any point to any level, without even using electricity at all.
So, ok, electricity itself may be inefficient. Fair enough. I'd still like to know the details.
Last edited:
I believe the idea is that electricity is a high value energy source, while heat is a very low value energy source. But since all processes produce heat as a waste product, turning any energy source into heat is a very efficient conversion. As energy can't go anywhere, where should the energy go? As transporting eletricity is fairly efficient over short distances i.e. a few thousand yards. The only loss is there. Norway gets about 60 or 70% of it's heat from electric heating. The only reason other countries don't do that is because they have cheaper sources of energy, like CH4.
Cobos
Cobos
jhm
Well-Known Member
1 kWh = 0.003414 MMBTU. So NG at around $3/MMBTU is about 1 c/kWh. This is only puting a price on heat energy, but that of course is only part of the story. Consider the difference between arc welding and welding with gas. In arc welding the use of energy is quite focused heating up metal first before waste heat is lost to the environment. In welding with gas a lot of heat is lost to the environment before it heats the metal. So when you look at the efficiency of actually heating metal it is plausible that electricity is more cost effective than gas owing to differences in electrical conductivity verses heat diffusion.
I do not know the engineering well enough to quantify this, so I am only arguing for plausibility. Certainly aluminum plants run cost efficiently off of hydroelectric power so there must be process efficiencies that make this work. The GF will need to heat a lot of metal and the waste heat can be captured for other uses. Natural gas is all waste heat used both to heat metal and everything else. Focused energy is worth more than waste heat.
- - - Updated - - -
Solar efficiency is not an issue for ground mount systems. There they may just go with the lowest cost per Watt which they can source from China.
Anybody got an estimate on the total MW solar the GF will install?
I do not know the engineering well enough to quantify this, so I am only arguing for plausibility. Certainly aluminum plants run cost efficiently off of hydroelectric power so there must be process efficiencies that make this work. The GF will need to heat a lot of metal and the waste heat can be captured for other uses. Natural gas is all waste heat used both to heat metal and everything else. Focused energy is worth more than waste heat.
- - - Updated - - -
Yeah, the question this leaves for me is whether Riverbend can scale up production fast enough. The GF roof will likely want the high 22% efficiency panels Silevo is producing. The roof may support about 100 MW which is the annual output of the piliot plant in the Bay Area. SolarCity certainly has other customers who need high efficiency panels too. So they really need to fire up Riverbend ASAP.
Solar efficiency is not an issue for ground mount systems. There they may just go with the lowest cost per Watt which they can source from China.
Anybody got an estimate on the total MW solar the GF will install?
Bangor Bob
Member
Electricity is not a PRICE efficient way to produce heat. Mostly because carbon pollution is free. But electric resistance heating is nearly 100% efficient. Heat pumps can be 3-4x that depending on how big of a temperature differential you're trying to achieve.
Best answer is to redesign your manufacturing processes so they don't need a lot of heat in the first place.
jhm
Well-Known Member
That's exactly right. Minimize and reuse waste heat through better process design. Also the GF is to be carbon emissions free just like the car. This is is important to customers, and it is also critical demonstration of the concept of moving to a carbon neutral economy. You can use batteries to move electricity toward carbon neutrality, and if you feed cabon neutral energy into the Gigafactory to make those batteries, you've got a closed loop. Moreover those batteries need to support cabon neutral transportation, mining and all other supply chain operations. Of course, it will take along time to clean up all upstream carbon, but it is critical that GF itself not be a net carbon emitter just like the Tesla cars. And yes, Tesla needs to clean up auto manufacturing as well.
Once you have electricity, that´s right. But starting with radiation coming from the sun, the more efficient way is going directly to heat, because photovoltaics has a efficiency of twenty-something percent at the best AFAIK. If you need a lot of electricity for other means than heating which is likely, it might be more simple and flexible to just employ photovoltaics though.
From the Q3 shareholder letter:
Since I don't think they will produce a large number of (or maybe even any, the Model3 won't go on sale for quite some time) EV cells before making stationary cells in 2016: What we now know is that there is no cell production in Nevada until late 2016 at the earliest.
Summary: Tesla will continue to import current cells from Panasonic Japan in 2016, but assemble them in Nevada (instead of their car plant in CA) going forward.
Since I don't think they will produce a large number of (or maybe even any, the Model3 won't go on sale for quite some time) EV cells before making stationary cells in 2016: What we now know is that there is no cell production in Nevada until late 2016 at the earliest.
Summary: Tesla will continue to import current cells from Panasonic Japan in 2016, but assemble them in Nevada (instead of their car plant in CA) going forward.
chickensevil
Active Member
Since Tesla just confirmed there won't be EV cell production in Nevada until 2017 (first stationary energy use cell production by the end of 2016) I wonder how Tesla can state with confidence on the latest CC that the rest of the industry is far off ("no one is even close to us") in terms of battery pricing.
As a reminder (cross-posted from the Q3 2015 CC thread) here's GM pricing for 2016 from supplier LG Chem:
LG Chem of course has supply contracts with many large car makers for both PHEVs and EVs. They can achieve high economies of scale long-term given their customer list.
PS: And yes, this price is on the cell level as outlined in the quote. The analyst caller from GS on the Q3 2015 CC (around 21-min mark) was unsure about this.
As a reminder (cross-posted from the Q3 2015 CC thread) here's GM pricing for 2016 from supplier LG Chem:
LG Chem of course has supply contracts with many large car makers for both PHEVs and EVs. They can achieve high economies of scale long-term given their customer list.
PS: And yes, this price is on the cell level as outlined in the quote. The analyst caller from GS on the Q3 2015 CC (around 21-min mark) was unsure about this.
Last edited:
chickensevil
Active Member
Maybe I missed something, but where did they say the EV cell production wouldn't happen until 2017? All they confirmed that I heard was that stationary production wouldn't happen until end of 2016. Maybe I missed something?
Also, it might be a minor distinction but their 145$ per kWh is stated at the cell level and all that people have been estimating for Tesla with reasonable confidence is at the pack level. Depending on how big GM is planning on making their packs vs how small Tesla is willing to drop their pack sizes this could be a reasonable distinction. What I am implying here is that maybe at the cell level Tesla is already seeing at or around that price with the expectation to drop it by 30% when the gigafactory comes online? Many numbers have been thrown around over the years, so it is a little tough to keep it all straight, but I seem to recall that their last statements on cost reduction was basically implying that the 30% was their minimum at this point (as opposed to some optimistic number) and that it would come simply because of scale... so the other price reductions (5-6% every year) should be safe to add on top of that 30%... just a guess here.
So if Elon is seeing that GM has put their cards on the table and is still laughing at the 4 of a kind they laid out, then maybe it is because they know they are sitting on the Royal Flush and it doesn't matter what GM throws down. That or he is bluffing hard core on a 2 pair hoping that GM is somehow lying about their cards? But that doesn't really make any sense to me.
One final point of conjecture, 145$/kwh doesn't state how they are fairing in the energy density part. I could tell you I had 145$/kWh on some lead acids... doesn't mean it would *EVER* make sense to put them in the car...
dc_h
Active Member
Just to be clear, the earnings letters states that stationary products have already begun production at the Gigafactory. Tesla will also begin cell production before the end of 2016 and all goals, unlike the Model X, have moved to the left. The GF will not be cash flow positive, but it will be profitable and the GF will be more and more self funding moving into 2017, freeing cash flow from S & X for the Model 3 ramp up.
"In early Q4, we relocated production from Fremont to an automated assembly line at the Gigafactory.This positions us for strong growth in 2016, but the Gigafactory pull-ahead will push some Tesla Energy Q4 production anddeliveries into Q1."
"In early Q4, we relocated production from Fremont to an automated assembly line at the Gigafactory.This positions us for strong growth in 2016, but the Gigafactory pull-ahead will push some Tesla Energy Q4 production anddeliveries into Q1."
On just a currency basis, the current cells supplied by Panasonic to Tesla may already be below $145/kWh. Of course, the exact cell pricing has not been released, but given various statements in 2012 and 2013, we had expected cell pricing to be significantly under $200/kWh at that point in time. Tesla re-negotiated the supplier agreement with Panasonic in the fall of 2013 and from Tesla's 10-Q, we know that the cell prices are paid in yen. Assuming $180/kWh in the fall of 2013, that's already $144 today. There was a SAE report that Tesla was paying $160/kWh, so at that price, Tesla is already paying $128/kWh.
Further, there is integration costs... and Tesla may know that the final pricing at the pack level may be far lower than GM's.
Finally, there are technological improvements... Tesla's current cell is already much better than the chemistry that LG can supply in 2017. Tesla has already hinted several times that they expect to improve their current cell at least once, if not twice before the Model 3 ships. That also lowers the per kWh price.
JRP3
Hyperactive Member
They "confirmed" no such thing, they merely stated that stationary use cell production would happen by the end of 2016. That does not mean vehicle cell production could not start sooner.
You seem to conveniently forget how LG freaked out when GM announced that price, probably because it's a loss leader and none of their other customers are getting that pricing.
How about listening to the actual Q3 CC around the 50-51 minute mark?
Tesla will only start EV cell production by 2017 in Nevada.
This answer comes straight from the CTO (Straubel).
dhanson865
Well-Known Member
+1 listen to the call guys.
I had to play that section 3 times to get this rough transcription, it still has errors but it's pretty close considering both guys stutter/stumble with unrehearsed Q&A.
49:33 Our next call...
Q - Colin Rush, can you talk about acceleration of ramp of the Gigafactory? What are you seeing... 2016, when will we see cells [from there] going into cars
A - JB we've accelerated storage products ... we do expect to start product of cells in second half next year*, which is again ahead of plan. Some of those cells will be initially allocated for Tesla Energy products. And we are still targeting around 2017 for the first cell production that would be for going into vehicles and going into the Model 3 so that remains on target.
*I interpret that as 2H 2016.
That leaves ambiguity for if cells from the gigafactory will be used for S/X before 3 but says no car cell use for 2016.
Last edited:
Yuri_G
Member
"We do expect that cell production will start up in the second half of next year." 50:15
Not the end of 2016.
dc_h
Active Member
According to the shareholder letter, which the CFO would have signed off on, "We have also accelerated plans to begin cellproduction for Tesla Energy products at theGigafactory by the end of 2016. This is severalquarters ahead of our initial plan."
We can quibble over the details, but Tesla is ahead of schedule at the GF. They are moving battery production out of Freemont to free up resources to build more cars and prepare for the Model 3. They are building product and generating cash flow at the GF and automating the battery pack production, increasing productivity to drive down costs. You are entitled to your own narrative, but not your own facts. Personally I was hedged and concerned about the bearish model, but based on this solid report, Tesla is positioned for a long run of positive news. Making delivery numbers for 2015, which the market discounted. Ramping up Model X in 2015, which the market discounted. Going live in the GF in 2015, which the market did not anticipate, all leading up to the Model 3 pre-reveal in March, which many expected to be delayed. All things considered, being sold out for the GF through 2017, seems more important than producing cells in December 2016 or March 2017.
We can quibble over the details, but Tesla is ahead of schedule at the GF. They are moving battery production out of Freemont to free up resources to build more cars and prepare for the Model 3. They are building product and generating cash flow at the GF and automating the battery pack production, increasing productivity to drive down costs. You are entitled to your own narrative, but not your own facts. Personally I was hedged and concerned about the bearish model, but based on this solid report, Tesla is positioned for a long run of positive news. Making delivery numbers for 2015, which the market discounted. Ramping up Model X in 2015, which the market discounted. Going live in the GF in 2015, which the market did not anticipate, all leading up to the Model 3 pre-reveal in March, which many expected to be delayed. All things considered, being sold out for the GF through 2017, seems more important than producing cells in December 2016 or March 2017.
Similar threads
