Regional Rail (RUR) & North-South Rail Link (NSRL)

[Joel N. Weber II]

Hey Data,

Were hydrogen powered EMUs something you looked at for lines that aren't currently electrified? Is there not enough data to guage their cost performance vs electrification with straight EMUs?

Hydrogen is totally impractical in many ways. The compressors are probably 10x-100x more expensive than DC fast charge stations for a given amount of throughput; if you make hydrogen from electrolysis of water it's less than half as energy efficient as just using batteries (although you can make hydrogen from crude oil or fracked natural gas, too, which is what tends to happen in practice).

Worcester's Proterra battery powered buses seem to get an awful lot more revenue service than the T's hydrogen powered bus that I read was going to be tested in revenue service but somehow don't recall ever seeing a report of anyone riding...

 

[Joel N. Weber II]

We put everything we had into proven, global industry-standard electric technology.

Given that https://reneweconomy.com.au/the-stunning-numbers-behind-success-of-tesla-big-battery-63917/ says ``The Tesla big battery in South Australia has already taken a 55 per cent share in the state’s frequency and ancillary services market'', is the Tesla PowerPack technology sufficiently proven at this point that it would make sense for you to look carefully at whether installing it at substations feeding overhead power would save money on demand charges?

 

[Joel N. Weber II]

We're focused on improving MDBF:

If we're trying to optimize 2030 reliability, one thing to think about is that recovering from fallen trees is faster when you don't have overhead wire. It's probably a minor thing, though.

And bringing down operating costs:

We really should be focused on minimizing operating costs + interest and principal payments on capital expenses and not just operating costs. SEPTA already happens to have the overhead wire everywhere; on most lines, we don't.

 

[dwash59]

Have you bothered to look at Tesla's latest quarterly statement?

?? Many of the things you argue are debatable, but the fact that Tesla's quarterly statement shows they lost around $700 million is not. Depending on what accounting method you want to use, that range could be $650 million to $785 million, but the fact is, they aren't making a profit.

Their gross profit on automobile sales was ~$540 million. That means, they could double auto sales next quarter and still lose money. Especially with this concerning statement: "Model 3 gross margin remained negative in Q1", so the model 3 production lost money last quarter on a gross basis, not just net. Do they expect to have a positive gross margin on Model 3 next quarter? Seems so, and it would be damn helpful to their financials if they did.

 

[ceo]

Putting Tesla Semi parts into railroad trucks ought to be less effort than the Type 9 design project, since the battery project just involves the trucks and not building the entire car body.

The trucks on commuter rail coaches are not powered trucks, and adding motors to trucks that were not designed to be powered doesn't strike me as remotely low-effort. You'd be much better served to start with existing powered trucks and put the batteries under the carbody. I don't know if you could use an off-the-shelf battery management system here or would have to custom-design one. (Li-ion batteries require very careful management to keep them from catching on fire.) Then there's the issue of the different crash strength and inspection requirements for powered rail vehicles; I don't remember what the rules are here.

I'm not sure whether battery power for commuter rail is a viable idea or not, but I do know that hacking batteries and motors into existing coaches is not the way to go about it. If it does turn out to be viable, you can be sure Alstom, Siemens et. al. will be all over it.

 

[Mongo]

Looks like there will be a bit more data on Hydrogen in the near future. The UK is starting to convert trains and is looking at using Hydrogen to replace diesel trains on the 50% of its' network that isn't electrified.

 

[Arlington]

Joel Weber, I love ya, but please start your own thread for battery electrics.

TransitMatters Phase 1 (PVD and allied lines) has the overhead wires. Don't let the perfect (battery-electric) be the enemy of the (vastly) better (off the shelf electrics)

TransitMatters has earned our unanimous support for it's PVD-Fairmont-Stoughton plans. Don't confuse matters with "but it could be Awesomer"

 

[Joel N. Weber II]

The trucks on commuter rail coaches are not powered trucks, and adding motors to trucks that were not designed to be powered doesn't strike me as remotely low-effort. You'd be much better served to start with existing powered trucks and put the batteries under the carbody. I don't know if you could use an off-the-shelf battery management system here or would have to custom-design one. (Li-ion batteries require very careful management to keep them from catching on fire.) Then there's the issue of the different crash strength and inspection requirements for powered rail vehicles; I don't remember what the rules are here.

That's not what I've been suggesting. Instead, build brand new powered trucks where the batteries are built into the trucks, lift the existing coaches off their unpowered trucks, and insert the new trucks that have the motors and batteries under the coaches and set the coaches back down onto those trucks. By putting the batteries in the trucks, the existing coach bodies don't have to bear the extra weight of the batteries, and if the existing crashworthiness is a matter of the car bodies and not the trucks hopefully you can avoid having to deal with that. And I've been suggesting using the battery management systems and motors that are common to the Tesla Model 3 (which is already the most produced EV in 2018) and the Tesla Semi. This should largely be a matter of designing a new or revised suspension for the truck and maybe adjusting the gearing from the motors to the wheels and dealing with extending the MU control cables to the trucks and getting the Tesla motors to react to train MU cables to control acceleration.

I'm not sure whether battery power for commuter rail is a viable idea or not, but I do know that hacking batteries and motors into existing coaches is not the way to go about it. If it does turn out to be viable, you can be sure Alstom, Siemens et. al. will be all over it.

Elon Musk figured that GM and friends would be all over building battery powered cars for years before he realized the truth and decided to found Tesla.

 

[DominusNovus]

Except companies like Siemens actually do have electric power transmission systems, and they seem to consider that some sort of remote generation is best.

 

[Joel N. Weber II]

?? Many of the things you argue are debatable, but the fact that Tesla's quarterly statement shows they lost around $700 million is not. Depending on what accounting method you want to use, that range could be $650 million to $785 million, but the fact is, they aren't making a profit.

Their gross profit on automobile sales was ~$540 million. That means, they could double auto sales next quarter and still lose money. Especially with this concerning statement: "Model 3 gross margin remained negative in Q1", so the model 3 production lost money last quarter on a gross basis, not just net. Do they expect to have a positive gross margin on Model 3 next quarter? Seems so, and it would be damn helpful to their financials if they did.

http://ir.tesla.com/secfiling.cfm?filingID=1564590-18-10306&CIK=1318605 is focused on an expectation of profitability in Q3 and Q4:

Quarterly non-GAAP operating expenses should grow sequentially at approximately the same rate as in the past four quarters, with our gross profit expected to grow much faster than our operating expenses. Thus, provided that we hit the 5,000 unit milestone in our projected timeframe and execute to the rest of our plan, we will at least be profitable in Q3 and Q4 excluding non-cash stock based compensation and we expect to achieve full GAAP profitability in each of those quarters as well. Also, considering our capex targets,
we expect to generate positive cash in Q3 and Q4, including the inflow of cash that we receive in the normal course of our business from financing activities on leased vehicle and solar products.

We have significantly cut back our capex projections by focusing on the critical near-term needs that benefit us primarily in the next couple of years. At this stage, we are expecting total 2018 capex to be slightly below $3 billion, which is below the total 2017 level of $3.4 billion. Ultimately, our capex guidance will develop in line with Model 3 production and profitability. We will be able to adjust our capital expenditures significantly depending on our operating cash generation.

I'm curious how the payments to CRRC for the Red and Orange Line cars are structured. Has CRRC been losing money so far on that order because they've built the factory in Springfield but not collected any payments that the contract might specify they don't get until the train cars are delivered? If so, are the Tesla skeptics here thinking the T made a mistake in placing the order with CRRC because of that? Or is the state paying CRRC for R&D if the state paid for the Springfield CRRC factory up front?

 

[Joel N. Weber II]

What percentage of the MARC Penn Line trains are using overhead power vs diesel these days, and how easy has MARC found Amtrak to be to work with at getting reasonable access to the overhead power at a reasonable price?

 

[ceo]

Problem with putting the batteries in the trucks is that a battery pack that can power a commuter rail coach any reasonable distance is going to be kind of big, and powered trucks don't have a whole lot of extra space.

 

[Arlington]

A bilevel/MLV EMU is the vehicle of the near future and it doesn't have space for batteries.

 

[datadyne007]

A bilevel/MLV EMU is the vehicle of the near future and it doesn't have space for batteries.

We actually aren't recommending bilevels. Bilevels increase dwell time dramatically. We are currently working on explaining how we can have enough capacity with flats only in a capacity appendix.

 

[Arlington]

We actually aren't recommending bilevels. Bilevels increase dwell time dramatically. We are currently working on explaining how we can have enough capacity with flats only in a capacity appendix.

flats offer more and wider doors? (and, once moving faster, doing more turns per shift can offer same capacity) is that it?

 

[datadyne007]

flats offer more and wider doors? (and, once moving faster, doing more turns per shift can offer same capacity) is that it?

For dwell time, yes. For general capacity, we believe we can run trains frequently enough to boost capacity on the lines even with eliminating bilevels.

 

[jass]

We actually aren't recommending bilevels. Bilevels increase dwell time dramatically. We are currently working on explaining how we can have enough capacity with flats only in a capacity appendix.

I think it depends. The current MBTA bi-levels arent great because they have poor circulation.

The ones RER uses in Paris are much better. Doors are twice as wide, and there are twice the staircases.

 

[datadyne007]

I think it depends. The current MBTA bi-levels arent great because they have poor circulation.

The ones RER uses in Paris are much better. Doors are twice as wide, and there are twice the staircases.

Paris RER dwell times are actually far above ideal standards.

 

[Joel N. Weber II]

Problem with putting the batteries in the trucks is that a battery pack that can power a commuter rail coach any reasonable distance is going to be kind of big, and powered trucks don't have a whole lot of extra space.

At https://teslamotorsclub.com/tmc/threads/railroad-trucks-with-batteries.115796/ I explore this in detail; 6' long x 4' wide x 30" tall inside the wheels, plus 6' long x 1' wide x 30" tall on the outside on each side for each train axle would have about the same total volume as 4 Tesla Model S battery packs capable of storing 4 x 100kWh = 400kWh, and I think with a compact suspension design a railroad truck probably could be made with space for that.

Also, while for some reason my memory was convinced I'd seen 3 MPH/s as the rate of acceleration for some Metro North cars, the best reference I could find today was http://www.railroad.net/forums/viewtopic.php?f=16&t=158567 which claims 2 MPH/s. Does anyone have other data that might help to confirm how well the best North American commuter trains currently accelerate? Can TransitMatters cite a source for the rate of acceleration of their preferred overhead power option? (If 2 MPH/s turns out to be the best Metro North can do, it looks quite possible that a fully loaded Tesla Semi might be able to accelerate faster than a Metro North EMU train using overhead power.)

 

[ceo]

You're still forgetting the space taken up by the motors, which aren't small. If you've ever looked at a powered truck without a railcar on it, there is no space left.