2017 Investor Roundtable:General Discussion

[mmd]

Looks like another big TE job could be incoming:

View attachment 217884

Looking a bit desperate with this sort of promises. Shorts will start smelling blood
Tesla already had billions in order from April 2015 and is sold out. No desperation should be needed.

But why so much focus on South Africa all of a sudden? Their power issue has been there since few years.
Speaking of speed, nothing beats the Powerships. These actually generate power; not just recycle power with 10% loss as the batteries do, besides being easily un-deployed when they are no longer needed.

Powership - Wikipedia

Here is Karadeniz with 2 GW production in 4 ships, in SA back in 2015.
Keeping South Africa’s power problems at bay

Come to think of it, that'll be ~9 GWh of batteries needed for just 4 hours @ 2GW. With current GF cell production at roughly 100-150 MWh/month, cell production alone will take 7 years. The cost will be humongous too. If SA could pay so much, they could build better and more power plants already.

 

[MP3Mike]

But why so much focus on South Africa all of a sudden?

I have no idea what you are talking about. Tesla/Elon were talking about getting battery storage setup in South Australia.

Tesla in talks with Australian utilities to fix country’s power problems by deploying large-scale energy storage

 

[30seconds]

Nobody seem to be interested to do a napkin math exercise on Tesla's bet of installing up to 300MWh worth of batteries in Australia. Here is my shot at it. Assuming four shipments 75MWh each with the following schedule would require current production rate of 37.5MWh/week or 1.8GWh/year assuming 48 production weeks.

0 weeks - sign Contract
2 weeks - ship first 75MWh batch of PowerPacks (#1)
4 weeks - ship 75MWh batch of PowerPacks (#2)
5 weeks - deliver batch #1 to SA Site
6 weeks - ship 75MWh batch of PowerPacks (#3)
7 weeks - deliver batch #2 to SA Site
8 weeks - ship 75MWh batch of PowerPacks (#4)
9 weeks - deliver batch #3 to SA Site
11 weeks - deliver batch #4 to SA Site

Why wait the two weeks for the first shipment? Contacts take awhile and you get pretty clear on intention towards late stage. They could probably have 20% ready to go on day 1.

 

[erthquake]

Strictly speaking, Pruitt is right. We don't know enough about natural warming and cooling cycles to know whether CO2 is the primary contributor to the 0.8C rise in global temperatures since 1880. Most likely, it is. But there is a small chance that a natural warming cycle has been the primary contributor, and our emissions are a very large secondary contributor (say, 49%). Not very likely, but possible.

Of course, it's also possible that our emissions have caused more than 100% of the observed global warming. If we are in a natural cooling cycle, our emissions may have caused more warming (say, 1.0C or even 1.5C), but nature negated some of it. In that case, we're causing global warming at a faster rate than we are capable of observing.

At any rate, Pruitt's nit-picky, lawyerly point should not be the basis for policy.

Strictly speaking, Pruitt is flat out wrong.

Pruitt is either woefully ignorant, or intentionally deceitful. He will attempt to savage the EPA, but it will have little impact on Tesla's bottom line. Australia is clamoring for Powerpacks and Powerwalls, not because of the environment, but because they are the fastest and least expensive solution to their current grid problems. And Tesla's disruption of the auto industry is due more to the fact they are making an awesome car. That Tesla's are better for the environment is more of an afterthought for most buyers.

 

[30seconds]

Looking a bit desperate with this sort of promises. Shorts will start smelling blood
Tesla already had billions in order from April 2015 and is sold out. No desperation should be needed.

But why so much focus on South Africa all of a sudden? Their power issue has been there since few years.
Speaking of speed, nothing beats the Powerships. These actually generate power; not just recycle power with 10% loss as the batteries do, besides being easily un-deployed when they are no longer needed.

Powership - Wikipedia

Here is Karadeniz with 2 GW production in 4 ships, in SA back in 2015.
Keeping South Africa’s power problems at bay

View attachment 217895

Come to think of it, that'll be ~9 GWh of batteries needed for just 4 hours @ 2GW. With current GF cell production at roughly 100-150 MWh/month, cell production alone will take 7 years. The cost will be humongous too. If SA could pay so much, they could build better and more power plants already.

Yep, that is just the long term solution that ultility engineers should love. That way they can take care of a ship and a power plant AT THE SAME TIME!!

Brilliant

 

[hockeythug]

Looking a bit desperate with this sort of promises. Shorts will start smelling blood
Tesla already had billions in order from April 2015 and is sold out. No desperation should be needed.

But why so much focus on South Africa all of a sudden? Their power issue has been there since few years.
Speaking of speed, nothing beats the Powerships. These actually generate power; not just recycle power with 10% loss as the batteries do, besides being easily un-deployed when they are no longer needed.

*snip*

Wait, are you serious about the "powership"?

 

[vgrinshpun]

Why wait the two weeks for the first shipment? Contacts take awhile and you get pretty clear on intention towards late stage. They could probably have 20% ready to go on day 1.

They need time to manufacture first batch, hence 2 weeks for first shipment. With the backlog of PowerWalls that was delayed because of PowerPack projects, they definitely do not have 75MWH worth of PowerPacks sitting on factory floor, waiting for an Australian order to start shipping immediately.

 

[mmd]

I have no idea what you are talking about. Tesla/Elon were talking about getting battery storage setup in South Australia.

Tesla in talks with Australian utilities to fix country’s power problems by deploying large-scale energy storage

OK, my bad. Haven't seen that article and only been seeing SA mentioned here, and thought we are talking about some power emergency in South Africa
But the observations still stand. Battery storage remains expensive compared to other options of combined cycle plants and peaker plants that can be built in a year to 3 years.
At $500/KWh for Powerpack (does it include everything?), assuming 1000 cycles, that is 50 cents to just recycle some base load electricity. Total cost per KWh is $0.50 + 1.1 * (base load kwh price).

The GE Combined cycle plants are at 7-8 cents/kwh. Peakers (second to second variation) are at 12-15 cents/kwh.
A user's guide to natural gas power plants

I guess batteries will also have some place. How much is the big question.

 

[dennis]

At this point I don't think throwing money at advancing the M3 will have enough value therefore the cap raise isn't in 16Q4 or 17Q1. When proof of concept/beta run is completed and production begins there will be a big need for higher capex. Once M3 production starts it is at Elon's discretion on how fast the ramp up will be. Advance sustainable transportation exponentially or in controllable pace. I bet he and his team is trying to figure out how to balance all of this. Elon wants all this done yesterday.

The problem is not advancing the M3, it is protecting against the downside if M3 ramp is delayed. In Q3 Tesla will be at maximum cash burn with all of the staff hired to produce 5000/week, capital equipment deployed and payments due, supplier parts in house, etc. The time to raise insurance capital is before this situation can arise, I.e before Q3.

 

[Yggdrasill]

OK, my bad. Haven't seen that article and only been seeing SA mentioned here, and thought we are talking about South Africa
But the observations still stand. Battery storage remains expensive compared to other options of combined cycle plants and peaker plants that can be built in a year to 3 years.
At $500/KWh for Powerpack (does it include everything?), assuming 1000 cycles, that is 55 cents to just recycle some base load electricity. Total cost per KWh is $0.55 + (base load kwh price).

The GE Combined cycle plants are at 7-8 cents/kwh. Peakers (second to second variation) are at 12-15 cents/kwh.
A user's guide to natural gas power plants

I guess batteries will also have some place. How much is the big question.

Why are you assuming 1000 cycles? 5,000-10,000 is realistic. That drops the price down to 5-10 cents for the infrastructure.

It's also important to note that a battery can work both ways when balancing the grid. A 100 MWh Powerpack installation can either absorb around 28 MW, or produce around 25 MW. A 25 MW peaker can only operate in the range 0-25 MW. You get more than twice as much balancing power for the same MW-rating.

Edit: https://www.eia.gov/outlooks/capitalcost/pdf/updated_capcost.pdf

A natural gas plant costs around 1000 USD/kW, which is exactly as much as the Powerpack costs (500 USD for 1 kWh or 500W balancing power.)

 

[GoTslaGo]

?? What? Retire, W2?

What @Lump was referring to was the interview. The current state of regulation requires that for the small bank to give a loan to a hypothetical retiree for anything (a house, a car, etc...) that retiree would need to pony up a W2. A W2 in the US is a form that documents yearly earned income from work, which a retiree would not have. Basically the current rules are written so meticulously that it makes the activities of smaller banks much more onerous.

There is a great podcast by "Planet Money" on NPR regarding repealing Dodd-Frank and the various actors involved. It covers a lot of the issues quite well:

Planet Money

See #757, Strong Feelings about Dodd-Frank

 

[techmaven]

But the observations still stand. Battery storage remains expensive compared to other options of combined cycle plants and peaker plants that can be built in a year to 3 years.
At $500/KWh for Powerpack (does it include everything?), assuming 1000 cycles, that is 55 cents to just recycle some base load electricity. Total cost per KWh is $0.55 + (base load kwh price).

The GE Combined cycle plants are at 7-8 cents/kwh. Peakers (second to second variation) are at 12-15 cents/kwh.

Ah, more like 4,000+ cycles. And it's $50 million USD per 100 MWh, all in, so $500/kWh is right. It's about $0.11-0.12/kWh factoring in degradation.

 

[vgrinshpun]

OK, my bad. Haven't seen that article and only been seeing SA mentioned here, and thought we are talking about South Africa
But the observations still stand. Battery storage remains expensive compared to other options of combined cycle plants and peaker plants that can be built in a year to 3 years.
At $500/KWh for Powerpack (does it include everything?), assuming 1000 cycles, that is 55 cents to just recycle some base load electricity. Total cost per KWh is $0.55 + (base load kwh price).

The GE Combined cycle plants are at 7-8 cents/kwh. Peakers (second to second variation) are at 12-15 cents/kwh.
A user's guide to natural gas power plants

I guess batteries will also have some place. How much is the big question.

You've got quite a smoothie here - talking about mixing apple oranges and bananas and loading everything into a blender - delicious!

The Powepacks designed for 5000 cycles, not 1000.

The combined cycle plant cost you quoted is capital cost, not cost of electricity produced.

The peaker cost you quoted is back from 2013. The electricity costs kept going up. The current cost of electricity produced by a peaker plant is 16-19 cents per kWh.

 

[mmd]

Why are you assuming 1000 cycles? 5,000-10,000 is realistic. That drops the price down to 5-10 cents for the infrastructure.

It's also important to note that a battery can work both ways when balancing the grid. A 100 MWh Powerpack installation can either absorb around 28 MW, or produce around 25 MW. A 25 MW peaker can only operate in the range 0-25 MW. You get more than twice as much balancing power for the same MW-rating.

Edit: https://www.eia.gov/outlooks/capitalcost/pdf/updated_capcost.pdf

A natural gas plant costs around 1000 USD/kW, which is exactly as much as the Powerpack costs (500 USD for 1 kWh or 500W balancing power.)

Is the powerpack spec on cycles out somewhere? Please link to it, so we can verify the 5000-10000 cycles warranty. I vaguely rememebr LiFePO4 has 4000-6000 cycle lifespan. Will be curious to see what is warranted number of cycles for Powerpack and Powerwall.

Second para: Agree in concept, disaggree on the positive spin you put on "more than twice". So, if it sucked up 100 MWh to produce 25 MWh of energy when needed, you will spin it as "more than 5 times" ? That's ludicrous!
Consuming electricity when demand is low is not an issue. You just turn off some of the CC plants, or lower price enough so people run their pool pumps or play loud music for no particular reason. No need to pay for that. But if the grid has a lot of baseload generation and very few combined cycles and peaker plants, then it may be a minor issue. Such situation will be rare, and a poor grid design to begin with. The combined cycle electricity cost I quoted earlier assumes only 30%-40% operation.

Your last para makes zero sense. You are comparing initial capital outlay for natural gas plants per kw of capacity with KWh costs of Powerpacks. Here is a chart of how long the natural gas plants work. 51% of the plants are 30 years or older. Let's say, they last only 30 years. Then, it is $1000/(30% of number of hours in 30 years) per kWh produced, which is roughly $0.012/KWh, using your $1000/kw initial cost.
You should just use the number I provided earlier from the utilitydive.com article. That is 7-8 cents/KWh including cost of natural gas.

So, the cost comparison is really between 1.2c/kwh for NG plant vs. $500/(number of cycles) per kwh for Powerpack, if we exclude the fuel cost from both to simplify the comparison.

Age of electric power generators varies widely - Today in Energy - U.S. Energy Information Administration (EIA)

About 530 GW, or 51% of all generating capacity, were at least 30 years old at the end of 2010 (see chart below). Most gas-fired capacity is less than 10 years old, while 73% of all coal-fired capacity was 30 years or older at the end of 2010.

 

[Turing]

 

[Turing]

From CNBC: Trump hosted a meeting with bankers today and directed his cabinet staff to deregulate the banks within 6 months. CNBC featured a banker (didn't get her name), that implicitly called for the return of stated income loans and claimed that 'professional bankers don't need to be regulated to be compelled to ensure they loan money to people who will be able to repay the loan'. For those of us that have followed Tesla from the early days, the banking crisis of 08 was darn near fatal (or so I was told) - Anyone have a feel for this direction, will it fly? Can we really trust that bankers will perform the proper diligence and if not, how long until this blows-up the economy again? Lastly, if the economy has a major contraction, how will it affect Tesla?

It probably wouldn't be as bad this time around, there just isn't the same level of institutional money willing to buy endless amounts of MBS on the premise that residential housing can never dip.

 

[Turing]

I see a cap raise 'soon'. The other scenario would be IF EM is waiting to see if delivery news for Q1 is great OR waiting to see if Q1 ER has the potential to be great.

IIRC the last cap raise they waited too long (IMO). We were all hoping for a cap raise at a healthy SP and it dropped, then the cap raise came.

Why would they admit to a likely cap raise incoming in this last CC, take the short term stock price hit if they weren't going to do it fairly soon? If they were planning on waiting until Q2 or Q3 couldn't they have said no cap raise planned for the short term and then bring it up at Q1 or Q2, ect?

 

[Runarbt]

This really blows my mind.:-O

Everything, pack cost and speed of installation. Not to mention what this opens of other opportunities. The world will scream for deals like this.

I Understand why Elon said the TE market will match/outgrow(dwarf?) Tesla motors. Might happen sooner rather than later. What will this do to SP? When will we see revenue from this? I guess we have to see hard profit before most analysts will feel forced to acknowledge TE and rise sp targets.

I Googled Cannon-brooks, and this guy is seriously capable.:-D this will happen in a few weeks(?) +100 days.

 

[Runarbt]

Why would they admit to a likely cap raise incoming in this last CC, take the short term stock price hit if they weren't going to do it fairly soon? If they were planning on waiting until Q2 or Q3 couldn't they have said no cap raise planned for the short term and then bring it up at Q1 or Q2, ect?

Now it is "out". Let the fact settle, and closer to m3 release, when sp have recovered, then do the raise at new ath.

 

[Yggdrasill]

Is the powerpack spec on cycles out somewhere? Please link to it, so we can verify the 5000-10000 cycles warranty. I vaguely rememebr LiFePO4 has 4000-6000 cycle lifespan. Will be curious to see what is warranted number of cycles for Powerpack and Powerwall.

The powerwall comes with a ten year warranty, and is intended for daily cycle, which works out to 3650 cycles. The warranty can reasonably be expected to cover substantially fewer cycles than the actual projected lifespan. I can't buy a car with a 15 year warranty, even if it will last that long.

The exact lifespan remains to be seen, but 5,000-10,000 cycles is a good bet.

Second para: Agree in concept, disaggree on the positive spin you put on "more than twice". So, if it sucked up 100 MWh to produce 25 MWh of energy when needed, you will spin it as "more than 5 times" ? That's ludicrous!

The hydrogen proponents argue for the exact thing you describe - "efficiency doesn't matter if the energy is free"-type argumentation. But okay, I'll moderate my argument to say that you get twice as much balancing power for the same MW-rating with batteries vs natural gas peakers. Not "more than" twice as much.

Your last para makes zero sense. You are comparing initial capital outlay for natural gas plants per kw of capacity with KWh costs of Powerpacks. Here is a chart of how long the natural gas plants work. 51% of the plants are 30 years or older. Let's say, they last only 30 years. Then, it is $1000/(30% of number of hours in 30 years) per kWh produced, which is roughly $0.012/KWh, using your $1000/kw initial cost.

A natural gas power plant doesn't really last 30+ years. *Major* maintenance is required for that, where they replace almost everything. Same thing with batteries - the Mira Loma Powerpack installation may still be operating in 50 years - they just need to replace all the batteries.

A powerpack installation can be expected to operate 13-27 years (5,000-10,000 cycles), which is more or less the same as a natural gas peaker.

You should just use the number I provided earlier from the utilitydive.com article. That is 7-8 cents/KWh including cost of natural gas.

kWh doesn't describe the balancing power of the installation, you need to look at the power. Let's look at a hypothetical situation, where you have a grid that has an average consumption of 2400 MWh in 24 hours, where the peak consumption is 125 MW and the lowest consumption is 75 MW, and the consumption climbs and falls linearly.

Using batteries, you could even this out by having 100 MW of base power generation combined with a 150 MWh/25 MW battery installation that you charge when demand is low and discharge when demand is high.

Using a natural gas peaker, you can even this out by having 75 MW of base power generation, and 50 MW of natural gas peaking. You have to generate 600 MWh of electricity from natural gas.

Using natural gas, you have to output four times as much electricity as you have to with batteries, and natural gas has to be one quarter as costly as batteries per kWh cycled to be competitive.