Ideas for Additional "Night Owl" Train Service

[ehbowen]

Budd built 50 twenty one roomette cars, primarily for the PRR, the "Inn" series. Delivery started in 1949, by 1964 all were converted to coaches. Seems like the 38-42 capacity for an all roomette car may be on the high side.

I think it would depend on how the design specifications were written. The Budd cars were built with traditional roomettes which only accommodated one passenger. Recent Amtrak roomettes have been built for two passengers. The big question is how much space is set aside for the attendant, toilets, and showers.

ETA: And the inevitable call for ADA accommodations. If you're going to install one wheelchair-accessible bedroom (and you probably have to), it's not a bad idea to add another pair of revenue bedrooms. So the current Viewliner design really doesn't look so bad....

 

[Anderson]

Budd built 50 twenty one roomette cars, primarily for the PRR, the "Inn" series. Delivery started in 1949, by 1964 all were converted to coaches. Seems like the 38-42 capacity for an all roomette car may be on the high side.

I think it would depend on how the design specifications were written. The Budd cars were built with traditional roomettes which only accommodated one passenger. Recent Amtrak roomettes have been built for two passengers. The big question is how much space is set aside for the attendant, toilets, and showers.

ETA: And the inevitable call for ADA accommodations. If you're going to install one wheelchair-accessible bedroom (and you probably have to), it's not a bad idea to add another pair of revenue bedrooms. So the current Viewliner design really doesn't look so bad....

That's going to depend on some questions such as demand (with and without the roomette toilets) for bedrooms. I'd have to check the schematics, but presuming that the 21 included an attendant room as part of the roomette count, a 21-roomette car would have a capacity of 40 with in-room toilets and no shower or 38 with either no in-room toilets but with a shower or with in-room toilets but no shower. If you need separate toilets and shower rooms that gives you 36. If the 21 roomettes are in addition to the attendant, add two to each of the prior count.

The ADA space requirement has multiple workarounds (TBH as long as Amtrak is running 2-3 normal sleepers and is willing to "take a bath" on the Accessible Room rates, you could probably get away without the "full" ADA room), but depending on demand adding the two bedrooms might or might not make sense. I think that depends on the route, frankly: On the Florida trains a 10-bedroom car might well sell while on other routes a 20-1 car might make more sense. If you're pitching these as part of some sort of "discount" option that increases the chances of being able to make a 20-1 sell.

By the way, since jis raised it for lie-flat seats, now I'm wondering about a possible room count with a one-person roomette plan on an NJT-style car. I can already eyeball some specification changes, actually (having a single door set at one end of the car should stretch the bilevel space by a few feet and/or consolidate the single-level space to one end of the car...I can't find a set of drawings that are detailed/high-res enough quickly to really ferret out the spacing, but presuming that you lose 22 feet to stairs, the single-level portion, and that you place the toilets on the single-level portion you would have space for 9 roomettes on each side of each level...so that's about 36 single rooms at seven feet [1] allocated for each I think this is on par with a Viewliner, but I could be wrong. Call it either 36-1 (36 roomettes, one ADA room) or 32-1 depending on how some of the math shakes out.

FWIW, in theory you could probably flog a little bit more space out if you were to run "married pairs" and have only the top level go through while the bottom level terminated. Removing the lost space for stairs at one end of the car (and the doors, etc.) might get you another two roomettes upstairs (downstairs isn't going through) but I'd have to mess around a little bit with where the wheels go and so on. 20 upstairs and 16 downstairs might be the best to hope for.

Edit: Ok, now I'm pondering this...if you took one of these cars, is there a viable room layout for a "bedroom type" space downstairs? Obviously you can't stack the bunks (the clearance isn't there) but I'm pondering couchette-style fold-out seats (the beds going sideways vis-a-vis the car's motion) with the bunks having a similar width to a roomette bunk instead of the bedroom lower (so, 2'4" each...on a 6'6" space that gives about 2' in between the bunks for the pax to enter/leave at night). That would potentially give you a set of 20-8-1 cars, which isn't too far off of the 11-2-1 we're about to have with the Viewliner IIs. [2]

[1] 6'6" for the roomette and 6" for the walls, etc. Would it be possible to shave 2-3" each off of that "wall space" presumption for a little more slack on wheel space/stairwell requirements?

[2] It also seems possible to configure the last room on the bottom a little bit differently for a few inches of space-saving, etc. by having the door at one end. I might move some stuff around and pitch it as a "premium room", stuffing any "saved space" in the design downstairs into it (but potentially selling it within the last week or two as a "regular" bedroom).

 

[jis]

I have some more details from when I looked at this, back home. Right now I am in Kolkata, India and get back mid next week.

The bottom line as I recall was that it is advantageous to go with MLV configuration only if you are using airline style lie flat seats. When you use railroad style seats and berths in general single level with high ceiling is advantageous. In the extreme if you go to a three tier six per compartment European Couchette style layout you can maximize the capacity. But of course in the US all that capacity may not be effectively usable. For the same reason, in the US a multi-level airline style lie flat style seating may be the most efficient effective capacity than double roomettes in a single level layout; because the second berth in a third or more of the roomettes would not get used.

 

[Seaboard92]

What about the odd duplex bedroom cars that DBAG used to use for the City Night Line. I want to say ÖBB has bought those. But that design.

 

[ehbowen]

The ADA space requirement has multiple workarounds (TBH as long as Amtrak is running 2-3 normal sleepers and is willing to "take a bath" on the Accessible Room rates, you could probably get away without the "full" ADA room), but depending on demand adding the two bedrooms might or might not make sense. I think that depends on the route, frankly: On the Florida trains a 10-bedroom car might well sell while on other routes a 20-1 car might make more sense. If you're pitching these as part of some sort of "discount" option that increases the chances of being able to make a 20-1 sell.

To my understanding, the Viewliners were designed for interchangeable room modules. So, if you had enough of the right kind of modules, you could convert an existing Viewliner to an all-Bedroom car for the Florida trade or configure others in an all-Roomette layout. If you had enough of the right kind of modules...and enough extra car shells to play with.

If I were appointed Amtrak czar and had enough funding to make a difference, one of the first things I would do would be to change the car building policy completely. I'm sick and tired of Amtrak placing One Big Order, putting all its eggs in that one basket, dealing with companies who have no recent experience building long-distance passenger cars, fussing at the inevitable delays and cost overruns...and then, once the assembly line finally gets clicking and the builder starts to "get it", declaring the order complete and then ordering nada for another 20 years until another equipment shortage throws everyone into crisis mode yet again. Instead, I would issue orders for ten to twenty car bodies per year, every year, on an ongoing basis...a mix of Superliners and Viewliners. Keep the production lines hot, the skilled workers employed, and the companies and suppliers familiar with and ready to handle the unusual needs that long-distance passenger cars demand. And, incidentally, build up that inventory of Viewliner shells and modules....

 

[Anderson]

I have some more details from when I looked at this, back home. Right now I am in Kolkata, India and get back mid next week.

The bottom line as I recall was that it is advantageous to go with MLV configuration only if you are using airline style lie flat seats. When you use railroad style seats and berths in general single level with high ceiling is advantageous. In the extreme if you go to a three tier six per compartment European Couchette style layout you can maximize the capacity. But of course in the US all that capacity may not be effectively usable. For the same reason, in the US a multi-level airline style lie flat style seating may be the most efficient effective capacity than double roomettes in a single level layout; because the second berth in a third or more of the roomettes would not get used.

That's fair and I could see that. To be fair, I'm explicitly looking at not putting two people in the roomettes (and a different layout in the bedrooms) but rather going back to the Budd single-occupancy roomettes with some variation (probably having the seat slide out rather than fold out, for example, to avoid a height/length crunch). I agree that couchettes probably wouldn't sell well in the US (if we can't make berths fly, even with potentially doing something like "pairing" uppers and lowers on a single-gender or single-reservation basis, couchettes are probably hopeless) though but for legal concerns I wish an attempt could be made to try them.

The key with what I was looking at is that you'd have, give or take, 16-20 roomettes upstairs...but you'd be selling them as 16-20 singles and so you'd actually get a practical capacity of the whole sum (rather than losing about 1/4 to 1/3 to singles in a double room). Add in the downstairs, either as bedrooms or roomettes and you end up with a car with a practical capacity of 35-37 versus a Viewliner II's 22.5.

What's worth asking is how a private room variant of the car with a de facto 1-1 seating arrangement would compare with a lie-flat car with a mostly 2-1 arrangement. Some of that will be down to the mechanics of the seats, among other things, but I also wonder how much folks are prepared to pay for the presence of a door and the ability to turn off their lights (the latter complaint was probably the biggest issue with the Spirit of Queensland's service).

For the record, I'd also be tempted to look at whether some more capacity could be "shaken out" of a different room arrangement (e.g. running the hallway down one side of the car and arranging roomette beds the same as bedroom beds): Keeping the dimensions close to the same you could probably get close to 20 roomettes upstairs (3'6" wide roomettes would give you 20 in 70' plus the walls; if you go with the bedroom depth, you'd actually be able to add some net storage space at one end or something similar, and you might be able to shave a few inches off the width at the same time to dispose of the "wall problem"). Seat design becomes the real question (e.g. Is a 3' by 7'6" compartment a sellable proposition if you offer a 3' wide bed? It would beat the dimensions of most lie-flats out there in both dimensions but you'd have trouble looking out the window unless you did something funky with folding the bed down [1])...and you might be able to jam upstairs capacity up a bit more.

Downstairs/with a bedroom, it's time for a fun-but-serious question: Would two fold-out beds from the size that form a queen bed (60"x80" would compare favorably with the 78"x90" dimensions of an existing bedroom...18"/bedroom is a ton of space in context) sell better or worse than two bunks? Presuming that you kept a wash-stand but chucked the separate shower and presuming that the 78" dimension is from the window to the door, knocking 18" off a room is a lot (it's the difference between 10 bedrooms and 12 bedrooms in a dedicated single-level car at the price of moving a toilet opposite the attendant's room); even bumping off 12" (allowing for a wider bed and perhaps some sort of divider at pax request) would still give you an extra room and some extra space to work with.

[1] I can see some ideas here: If the bed folds from one side of the length-wise compartment you'd still have a 27" seat, room for overhead luggage storage, and possibly a place to hang a coat by the door.

 

[railiner]

The Pullman Company tried many different car plans thru the years...one of the types they tried was a car containing some "Duplex Single Rooms", which in their heirarchy, were between Roomettes and Double Bedrooms.

They were crosswise oriented beds, and staggered alternatively, floor level, or up a couple of steps, in the manner of Duplex Roomettes, or Single Slumbercoach Rooms.

 

[WoodyinNYC]

... the Viewliners were designed for interchangeable room modules. So, if you had enough of the right kind of modules, you could convert an existing Viewliner to an all-Bedroom car for the Florida trade or configure others in an all-Roomette layout. If you had enough of the right kind of modules...and enough extra car shells to play with.

If I were appointed Amtrak czar and had enough funding, one of the first things I'd do would be to change the car building policy completely. I'm sick and tired of Amtrak placing One Big Order, putting all its eggs in that one basket, dealing with companies who have no recent experience building long-distance passenger cars, fussing at the inevitable delays and cost overruns...and then, once the assembly line finally gets clicking and the builder starts to "get it", declaring the order complete and then ordering nada for another 20 years until another equipment shortage throws everyone into crisis mode yet again. Instead, I would issue orders for ten to twenty car bodies per year, every year, on an ongoing basis...a mix of Superliners and Viewliners. Keep the production lines hot, the skilled workers employed, and the companies and suppliers familiar with and ready to handle the unusual needs that long-distance passenger cars demand. And, incidentally, build up that inventory of Viewliner shells and modules....

In one of its fleet plans, Amtrak went on and on about this question, noting that the on-again-off-off-off-again practice had wiped out the domestic supplier base.

The fleet plan said the company had talked to several likely bidders, who told them than an order needed to be about 100 units a year to gain the better prices from economies of scale. The fleet plan went on with Amtrak planning to order 100 single-level cars for iirc 6 or 7 years, and starting a year later, 100 bi-level cars for 6 years or so. The orders were going to be strictly 1:1 replacements, no expansion included.

Larger orders -- to include equipment needed for future extension of service -- would run for more years, keeping the lines open for possible still further orders from Amtrak, or from some states. A sweet possibility is that VIA could add its own forthcoming order for new single-level trains, stretching that run by another year, or two.

I don't think changing Viewliner modules would be all so cheap. The sleepers would need to add plumbing to the modules for the wash basins, and the toilets and shower rooms. Anyway, the least costly to modify would be bag cars and bag-dorms.

Sadly, the CAF Viewliner order is shaping up as the perfect example of your complaint. It's on, but will soon be off again. From the order for 130 cars, Amtrak got 70 baggage cars, and now we wait for the 60 other cars. Of course, originally Amtrak wanted 200 cars, not 130, but had to cut back without enuff funding from Congress.

In a normal world, with 70 cars in hand, and 60 in the bush, Congress might put in a few more pennies now to order more sleepers, more bag-dorms, and more baggage cars.

More sleepers should make more money, later if not sooner. Looks from here like Amtrak has just barely enuff baggage cars, even with adding modules to them. More bag-dorms would give more flexibility when adding or adjusting capacity. And it could happen that Congress won't want to have two big orders, one for bi-level coaches and one for single-levels. In that case, Amtrak could switch the Capitol Limited, and perhaps the City of New Orleans]/i], to single-level cars and use the freed-up Superliner equipment to fill out Western trains with another coach and sleeper where demand warrants it (like every train that has a lot of sold-out runs, like most of them).

 

But Amtrak is being strangled by its on-going capacity crunch. So any serious expansion -- the Broadway Ltd, the fabled 'day train' to Atlanta, or simply taking the 3/7 trains to daily 7/7 service -- seems out of the question. That may be just what too many in Congress want to see.

 

[railiner]

It's too bad that Amtrak,and the continent's commuter railroads couldn't share a basic single (or dual) level car body, and outfit them to order, to achieve that "economy of scale". Something like the Horizon car body, perhaps with a little more height, could probably work...

 

[ehbowen]

In one of its fleet plans, Amtrak went on and on about this question, noting that the on-again-off-off-off-again practice had wiped out the domestic supplier base.

The fleet plan said the company had talked to several likely bidders, who told them than an order needed to be about 100 units a year to gain the better prices from economies of scale. The fleet plan went on with Amtrak planning to order 100 single-level cars for iirc 6 or 7 years, and starting a year later, 100 bi-level cars for 6 years or so. The orders were going to be strictly 1:1 replacements, no expansion included.

...

In a normal world, with 70 cars in hand, and 60 in the bush, Congress might put in a few more pennies now to order more sleepers, more bag-dorms, and more baggage cars.

I agree that economies of scale are desirable, and in a perfect world there would be enough orders from enough customers (more than just Amtrak) to achieve them. In the current environment I'd rather spend, say, 20% more per unit on my fifteen a year to get reliable equipment delivered on time by an experienced supplier with skilled employees than the current Viewliner charlie-foxtrot. But, as you correctly point out, as long as Amtrak's equipment budget is wholly dependent upon Congressional largesse, that may be difficult to achieve [my scenario did specify that I was 'Amtrak czar', with the sufficient funding to make a difference]. We may have to go back to the days of Lyndon Johnson spreading Apollo program patronage around all 50 states....

 

[Anderson]

In one of its fleet plans, Amtrak went on and on about this question, noting that the on-again-off-off-off-again practice had wiped out the domestic supplier base.

The fleet plan said the company had talked to several likely bidders, who told them than an order needed to be about 100 units a year to gain the better prices from economies of scale. The fleet plan went on with Amtrak planning to order 100 single-level cars for iirc 6 or 7 years, and starting a year later, 100 bi-level cars for 6 years or so. The orders were going to be strictly 1:1 replacements, no expansion included.

...

In a normal world, with 70 cars in hand, and 60 in the bush, Congress might put in a few more pennies now to order more sleepers, more bag-dorms, and more baggage cars.

I agree that economies of scale are desirable, and in a perfect world there would be enough orders from enough customers (more than just Amtrak) to achieve them. In the current environment I'd rather spend, say, 20% more per unit on my fifteen a year to get reliable equipment delivered on time by an experienced supplier with skilled employees than the current Viewliner charlie-foxtrot. But, as you correctly point out, as long as Amtrak's equipment budget is wholly dependent upon Congressional largesse, that may be difficult to achieve [my scenario did specify that I was 'Amtrak czar', with the sufficient funding to make a difference]. We may have to go back to the days of Lyndon Johnson spreading Apollo program patronage around all 50 states....

I don't think the CAF order (or the N-S order) would do much for Congress at this point. Both have been rather humiliating messes: CAF is running on seven years with only a batch of baggage cars (and a single diner) delivered while the N-S order failed crash testing (if only by a hair, per some rumors).

Additionally, my understanding is not that Amtrak intended to/planned to/wanted to buy 200 cars: They had options on another 70 (IIRC it was 35 more bags and 15 more each of diners, bag-dorms, and sleepers). However while I could speculate on the utility of more sleepers and bag-dorms, I don't think anyone here could think of a use for more than a small number of those diners (let alone 35 baggage cars). I think it was jis who noted that the option was a planning tool that might get used to shake a few spare cars out if we got lucky; as it was, it was used to switch some bags and bag-dorms within the order as things dragged on (I suspect to finish replacing some of the Heritage bags sooner seeing as I don't think there's a bag-dorm in sight).

Moreover I would argue that CAF is to blame for that option being unusable: IIRC the option expired years ago (it had already expired when I approached VHSR about talking to the VA DRPT about exercising some slots on it) and CAF hasn't even delivered a single sleeper. If they had been able to churn out the sleepers first there's a chance that Amtrak could have "counted beans" and gotten an RRIF loan to thrown a few sleepers or bag-dorms on the back end (while the baggage cars were coming down the pike) but as it stands that's not likely.

Hindsight being 20/20, the Alstom-Bombardier bid was probably the smarter option...but my understanding is that their bid was simply too expensive by comparison for Amtrak to justify. If more money were to come available, I'd be mighty inclined to go with Siemens at this point (seeing as a Siemens-based order from AAF has gone from conception to delivery and, knock on wood, start of service entirely within the timeframe it's taken CAF to deliver a single sleeper).

Edit: To be clear, Amtrak's latest fleet strategy plans have all presumed something of a boom-bust scenario in this respect. Part of the problem, IMHO, is the "split fleet" situation: Amtrak basically has three fleets (Superliner/MSBL, Amfleet/LDSL, and Acela). The Acela split is somewhat unavoidable due to the nature of that stuff (orders for that will always be in batches, and the new federal standards don't require Amtrak to reinvent the wheel on new equipment there) but the single-level/bilevel split is a very real problem for a bunch of reasons (not least being that a new railcar either "belongs" to the extended NEC/NYP-based portion of the network or it doesn't, but also because Amtrak can't just keep pumping out orders for cars in single body types with different, particularly modular, interiors).