The comment section from the last post is about to overflow, so here is an open thread to keep the discussions going. Some noteworthy developments fished out of the previous comments:
- Caltrain's corridor-wide grade separation strategy continues to evolve towards its final form, with a prioritized list of grade separations due for adoption this summer. Of note, grade separations are now allowed 2% grades without a design exception, which removes the need to design an entire 1% project before seeking the exception. Steeper grades are a good start for shorter grade separations, but we also need to reduce freight train speeds to 45 mph throughout the corridor to tighten up vertical curve radii. A freight train requires the same vertical curve radius at 45 mph as a passenger train at 110 mph, such that freight speed limits any higher than 45 mph result in freight-driven designs that are longer and more expensive to build. An important but neglected part of the grade separation strategy should be to reduce the freight train speed limit from 50 mph to 45 mph.
- Speaking of grade separations, costs continue to rise out of control, with the Broadway project in Burlingame (see agenda item 11 of the March 3rd city council meeting) flirting with $900 million. It's not just inflation. A grade separation cost model discussed a few years ago predicted that, after adjusting for inflation, the Broadway project should cost $120M all-up, including the "soft cost" category that today forms a metastasizing cancer on these projects. The city is now considering deleting the Broadway station (not a bad decision, due to proximity to Bvrlingame) to bring costs down to the still eye-watering sum of $600M. This is a prime example of the transportation industrial complex's capture of a project designated as safety-critical, where cost becomes no object because you just can't put a price on safety. In this abject fleecing, the city and Caltrain are just along for the ride.
- Development of the BEMU continues, in spite of the looming "fiscal cliff" where one of the most logical cost cutting moves will be to suspend Gilroy service and dispose of the diesel fleet and its attendant operating & maintenance expenses. The BEMU has fewer seats (280) and more batteries (2.3 megawatt-hours) than previously understood, making it even more of a turkey for the $80M pricetag.
- On the bright side: Caltrain's EMU service is holding up nicely! Well done to all involved. Ridership should continue to increase as the freeways rapidly return to pre-pandemic levels of congestion misery. The way this works: 101 overflows onto 280, which overflows onto Caltrain. 280 is starting to get congested again, which augurs well for Caltrain.
Posts
Re: BEMU scam:
ReplyDelete$9.64 million "Project management and administration costs".
Ten million dollars of the purest fattiest lard for contractors and agency do-nothings heaped on top of the $60.98 + $7.47 (because "contingency" on a fraudulent porkfest always means "spend it all") $68.5 million dollars on one single 280-seat train designed to make one round trip a day carrying, optimistically, one hundred humans on one 60 mile round trip on five days per week.
Oh, and another cool million dollars of public cash for unspecifiable "wayside". Maybe they're constructing a marble viewing pavilion in Morgan Hill from which to observe the gold-plated once-daily regal procession as it glides by?
Meanwhile, a normal fully-functional, zero-million-dollar "Project management and administration costs" 200+ seat Stadler FLIRT (like the ones running in Dallas) costs about $15 million, all in, including "Buy American" ~40% overhead. (Not that Gilroy service can possibly justuify any new trains, ever. Or any trains at all.)
Meanwhile, a normal fully-functional zero-million-dollar "project management and administration costs" over the road bus costs under a million dollars. Go ahead, splurge a bit in case ridership shatters expectations ... buy two buses! So indulgent.
TEN MILLION DOLLARS of overhead. SEVEN MILLION DOLLARS of "contingency" (against what? Bank robbery? Inside job!)
You see why these things have a life of their own.
You see why the country and the planet are doomed.
It's one banana, Michael. What could it cost? Ten dollars?
Criminals. Fucking criminals.
Redwood City’s Jefferson Avenue grade separation (fully-depressed street underpass built using a shoofly) only cost around $10m in 1999/2000.
ReplyDeleteThe preferred Broadway grade separation design that fully elevates the tracks over the street and includes a new center-boarding station is clearly inherently more costly … but (admittedly unfairly ignoring inflation) $889m is like 89 Jefferson grade seps and well over one-third the $2.44b cost of the entire 7-year systemwide electrification and fleet replacement.
All planned grade seps and bike/ped underpasses (e.g. Menlo Park @ Middle, Sunnyvale @ Bernardo, and the deferred-due-to-insanely-5-story-high-over-catenary-design-and-hence-cost NFO overpass) are being hit & hobbled by similar cost increases.
Something is seriously wrong with this picture.
You're right, these things are insane. I suspect a lot of this has to do with nonstandard designs, and varying levels of competence among governments up and down the corridor. On top of that, there's an over reliance on consultants because there's no in-house competency which just increases costs 20%+ at each and every stage. See Alon Levy's project on construction costs for best practices that align with this.
DeleteI think there are two solutions, one short term and one long term that need to happen:
1. Short term: We need a project authority that is backed by counties, state gov, Caltrain, and transit agencies that is dedicated solely to grade separations. This authority should have a standardized design kit and stable of engineers that can do the design work, the environmental work, permitting, etc. They also can do oversight, meaning that the only things that would go out to contract would be the construction aspect of the grade separations. From there, this authority expertise and staffing can be folded into MTC or even the state level.
2. Long term: At the state level, we need a good rail department like the highway department that has standardized design kits, standardized contracts, a group of engineers and professionals that do the design, clearance, and management work in-house up and down the state. Ideally, a lot of CAHSR staff with their expertise can go to this department and serve not just CAHSR but the entire state in managing, designing, and delivering these rail projects like the highway department already does today, IMO.
@Anonymous-S,
DeletePoint #1, that's needed for grade separations on arterial roadways, too, not only to separate rail and road (bike-pedestrian) traffic.
Point #2, it's a reminder that rather than thinking of the high-speed project by itself, what's really needed for the state is a system, which would include Bay Area-Sacramento regional service through an improved Altamont Pass, with connections to Central Valley and L.A. service, and the same type of service along the Central Valley as Bay Area-Sacramento regional, serving the smaller cities, with high-speed trains, if they ever materialized, serving only the largest cities between the Bay Area or Sacramento and Los Angeles, with a common, and transfer, station in the vicinity of Manteca. It's a shame the thinking of so many is poor, but what may be needed to induce some to think more systematically about rail service in the state is to have a seriously dedicated rail transportation department under Caltrans with highways a separate department under it, same with aviation, etc.
The RWC Jefferson grade sep cost $14.2M in YOE, and was completed in 1999. If you take 1998 as the mid-year of construction, then inflate to today's dollars, you get $28M.
DeleteThe Broadway project is an interesting test for the question "does money grow on trees near the edge of a fiscal cliff?" We're about to find out!
RE: Anon from 20:38
DeleteAgreed, and good point about arterials, highways, etc. The construction and design standards are very similar for roads and for rail.
I saw a recent draft of the CAHSR plan that outlines as such for point 2 to what you are saying - see this link for more: https://hsr.ca.gov/about/project-update-reports/2025-project-update-report/. Page 27 of the PDF or Page 14 on the document shows an interesting map of a broader southwest HSR network, that also breaks the system down into additional segments. For example we see Segment D, which is Gilroy - Merced; or Bakersfield - Palmdale as Segment F, and the High Desert Corridor as Segment G. I think it's a lot more realistic to break it down into these pieces as funding comes online, and encourages third parties like Metrolink/LA Metro and Caltrain to invest in their segments over time.
RE: Clem, fair point. I do think that is why a lot of this design/permitting/contracting/etc need to be taken from cities and standardized across the state. If cities want a fancier alternative, they can pay for it, but creating standardized approaches and design help deflate these costs completely, IMO. One option, while less palatable, would be simply to build viaducts directly over the existing tracks and electric catenary. Once the viaducts are complete, close the lower tracks, and use that space as parks, trails, and station space. While this may be more expensive in some areas, it may also be logistically easier and more straightforward, especially if you standardize the design and construction like you do for highways.
@Clem: Does the $600 million no-station option still include 1% maximum grade, no vertical curvature on bridges, and other unwarranted, megadollar-adding design rules?
DeleteDo you have any estimates of (Caltrain-style) costs after fixing those ... stupid ... new design constraints? Deliberately not including U-shaped bridges, as that would require the transport-industrial mafiosi to think outside their box, rather than just going back to design constraints broadly similar to, oh, maybe San Carlos or Belmont grade separations? (Without sinking the overpassed road.) Not holding those up as a good example; just less horrifically expensive than current Broadway plans.
@Clem: thanks for the corrected ($14.2m YOE) Jefferson grade separation cost. I regret my error.
DeleteAt cost of nearly $1billion, might as well expand the Broadway grade separation project into full grade separation within Burlingame, or even including San Mateo.
DeleteIs the CMGC method not working to reduce time and cost? If Caltrain doesn't have in-house design/project management expertise, can it outsource to other public agencies that do, MTC, BART, Caltrans, etc...?
Watch beginning at the 2:54:26 mark of last night’s meeting video as Caltrain staffers get berated by Burlingame’s city council members after a detailed presentation of the recent Broadway grade separation project’s fraught history and the pros & cons of the latest round of “value engineering” options it seeks council guidance on, and more millions just to proceed to finish designing — none of which anybody has any realistic idea how to possibly fund before the presented costs spiral even further into the stratosphere.
ReplyDeleteYes, as Caltrain’s chief if construction Robert Barnard explains, they’ve used the newly-allowed steeper grades of up to 2%, and have even figured out ways to shift stuff toward parallel Carolan Ave. to avoid building a double-track shoofly and to minimize impacts on the claimed seven!? creeks, eliminated the nice bike/ped passages punched through the berm, etc., etc.
In their supposedly extensive value engineering exercise, I will bet you very good money that they still have not:
Delete1) allowed an uninterrupted vertical curve over the bridge deck
2) reduced freight speed limit to 45 mph to tighten vertical radius (no impact to HSR, still at 110 mph)
3) aggressively minimized structure depth, using a through girder or similar with minimal span lengths
4) reduced track center spacing to 15 feet (if no station)
5) reduced vertical clearance under the bridge from a luxurious 16'6" to a merely comfortable 16' with the requirement that paving must be ground down before future replacement
They can't and won't explain why in the "good old days" (the San Bruno grade separation completed in 2014) they could buy 0.8 miles of elevated track, 3 bridges, 2 pedestrian tunnels and an elevated station for $165M in YOE (equivalent to $225M today). Costs have not tripled after adjusting for inflation.
All this "post-pandemic" "fragile supply chain" "geotechnical risk" yammering is worthy of the world's tiniest violin playing a sad tune to lament the fact that in ten short years, Caltrain has utterly and completely lost the ability to plan and manage a grade separation project.
Wait, what? 3x in 10 years is ~12% year-on-year (real) cost growth. That is damn impressive, actually, in a way.
DeleteClass 1s itch and moan about 2% maximum grades and hill climbs raising their costs while ignoring that math is taught to every child in America and we can easily see that if you raise a trackbed 20' over a quarter mile its a 1.5% grade and for many grades the track height is already superior to the gross ground level of the road it intersects. a 1.5% grade becomes an issue when that grade is carried on for 3-30+ miles, as long as the requisite AAR guidelines of vertical radius are followed, eliminating a substantial number of our nations grade crossings is simply a matter of tightwads being loosened, rather than the engineers being brought in to do miracle work.
ReplyDeleteResponding to the (apparently deleted) @Anonymous post bemoaning the unavailability of Caltrain Burlingame Broadway Project “value engineering” Update slide deck, someone I know saved a copy HERE for posterity.
ReplyDeleteIt will likely be presented again tonight during agenda item 10b of tonight’s 5 p.m. SMCo. TA Board Meeting.
Completely off topic but can anyone here tell me what’s the purpose of the multiple giant hallways of electrical boxes that look like something of out Space Mountain on the new Caltrain cars? I’ve ridden similar Stadler trains in Europe and I don’t recall seeing such use of seating space.
ReplyDeleteWith only 6 unpowered wheel trucks (i.e. “bogies”) out of 14, Caltrain’s over-motored/powered(*) 7-car KISS sets have more interior-space-robbing traction power electrical cabinets than probably most other of the world’s KISS sets.
DeleteOnly the 2 bike cars and 1 truck on each cab are unpowered.
(*) Caltrain’s COO has said that after observing excessively strong acceleration during early testing, they decided to cap or limit peak acceleration for rider comfort and safety, particularly for ones that aren’t sitting and may fall.
Thanks for the details on this.
DeleteBonus question: since the 7-car set just barely fit on many existing platforms, there is never a normal revenue service scenario in which the fancy hands-free Euro-style “Schaku” couplers will be used to couple/uncouple sets, and because they need to use one of their janky US AAR/Schaku coupler adapters anytime a set needs to be rescued or towed anywhere, why in the heck didn’t they just have Stadler fit & deliver them with cheaper and less collision-damage-prone US AAR couplers?
ReplyDeleteI had the same question when the Caltrain KISS was first spotted. However, if BEMU sets were realized, then it is possible BEMU will couple with regular sets to make trips north to SF, instead of shuttling between SJ and Gilroy.
DeleteNope, the 7-car sets just barely fit on many platforms … so coupling the 4-car BEMU proof-of-concept demonstration train to them for an 11-car train wouldn’t fit on most mid-line platforms and would block numerous crossings during various station stops.
DeletePlans are for the BEMU (with only 3 passenger cars) to be used just as the diesel Gilroy South County Connector shuttle trains are for unelectrified UP territory, and never normally be run north of Diridon where they would also have insufficient bike & rider capacity.
This makes me wonder about a cheap and easy way to increase capacity without needing increased service - lengthened platforms and lengthened trains, from 7 cars to 10 or 11 cars. Since the trains are somewhat overpowered, they could add in 2-3 unpowered cars pretty easily, and even allow for an accessible transition car while level boarding is implemented (like Clem's older post on this https://caltrain-hsr.blogspot.com/2024/10/another-path-to-level-boarding.html).
DeleteI assume the constraints are that extending platforms at stations are probably pretty expensive, and would also require reworking of the CEMOF (which really should be relocated to the quarry at CP Lick)...?
I was never a fan of island platforms like Caltrain and others are, but damn... For Broadway, a center-boarding platform commands a $150 million premium over side-platforms. Glad that this should put an end to island platforms.
ReplyDeleteWhy on earth would you have such an obviously wrong preference? You LIKE missing trains by being on the wrong platform? You LIKE longer access times? You LIKE the extra cost of duplicating all the station services and amenities?
DeleteJust bizarre.
Island platforms are always (when feasible) ALWAYS the correct station design, just as fast-slow-slow-fast is ALWAYS (when feasible, which it is everywhere on Caltrain, ignoring isolated Lawrence mistake) the correct track configuration.
Surprise! Projected cost of Mtn. View’s Rengstorff Ave. planned grade sep just shot up by $118m or 35% from $335m in January of 2024 to $453m in January of 2025.
ReplyDeleteSo city staff tersely reported at last Tuesday’s Council Transportation Committee meeting that since there’s now a $159m funding shortfall that they’ll be “value engineering” the cost down and report back to council with their results by this Fall.
It was only $120m in 2014 and $262m in 2022.
Rengstorff grade separation is a terrible, terrible, horrible disgusting hellscape project: an absolute ORGY of road widening, road excavation and endless utility replacement masquerading as a safety project masquerading as anything at all to do with Caltrain, resulting in an even more hopelessly shit human pedestrian environment than they already have.
DeleteIf it were a rail project (it isn't) it could be constructed — as every Caltrain grade separation should be, with the soleexception of 16th Street in SF — as a rail over road bridge. With 2% grades (actually can only get 1.45% on the south side, for different crazed Caltrain AREMA-cut-and-paste "design standards" requiring constant grade between vertical curves), 45mph freight speed (110mph passenger) design speed the project wouldn't even touch San Antonio station, and would be just under 1.2km end to end and on the Caltrain right of way. If Caltrain can't build that for under $450 million they shouldn't be building anything, anywhere, at any cost — just stick a fork in it, put them out of our misery.
If Mountain View wants to turns its traffic sewer stroads into subterranean traffic sewers, that's on them. The only point where a rail grade separation should even touch the streets is exactly at the street crossing, and then only to build any needed temporary tracks. Anything else, NOT A RAIL PROJECT.
My accidentally-"anonymous" comment, obviously.
DeleteSpeaking of totally out of control costs (as we always are), Caltrain's San Mateo Replacement Parking Track project, comprising one (1) turnout, one (1) derail, construction of about 300m of non-revenue track on existing flat, bridge-free right of way, and without planned electrification, is budgeted at $10 million. Oh yes, for reasons there's 1300 feet of 10 foot high "sound wall" (one side of the ROW only, supported by 16 foot deep piles) thrown in.
ReplyDeleteNo, I didn't add an extra zero to the dollar figure. They did.
They simply can't build anything any more; certainly nothing that needs building.
I wonder how much of this massive cost increase is due to Caltrain's revised rules for grade separation post-electrification. These rules were just developed last year and probably weren't part of the original design criteria for Broadway.
ReplyDeleteLast year, Palo Alto asked for Caltrain input on their (interminable) deliberations on grade separations at Churchill and Meadow/Charlston. They were quite shocked by the results. The following are a samples of new rules:
1) 16.5ft clearance required over roadways (up from 15.5ft previously). This clearance must extend over the entire width of Caltrain ROW, even if bridge is not full width (to allow bridge widening later for 4-tracking without lowering roadway). It seems that 15.5ft might be allowed in some cases with a variance and sacrificial beam.
2) 20-25ft (!) clearance from track center for any construction work adjacent to an active electrified line. This makes shoo-flies need to be really far from viaduct/berm construction. If this clearance cannot be met then work can only happen overnight which would dramatically increase the costs.
3) Rail bridges must be wide enough to include a 10ft roadway adjacent to the tracks for OCS inspection / maintenance etc.
For details, see the presentations and reports at the Jan 2024 and March 2024 Palo Alto Rail Committee.
Write your own "standards", write your own blank checks.
Delete"Oh sure, they might be able to do "that" (pick anything, anything at all rail related) in places like Switzerland with [checks notes] famously cheap wages and famously poor safety records, but we can't ... because standards. Our hands are tied! Please whatever you do don't throw us into the briar patch!"
Caltrain is 100% captured by the construction-industrial consultancy mafia; like VTA and like BART before it, it is a barely-alive brain-dead husked-out host body that exists solely to provide sweet, sweet nutrients to the thriving parasitic chest-bursters.
The Daily Post story on the huge Rengstorff cost increase also mentions that the simple-seeming Middle Avenue bike/ped tunnel under the tracks in Menlo Park recently doubled in one year’s time to $62m. From what I read elsewhere this was largely due to Caltrain imposing the new requirements @jpk122s mentioned on the project (e.g. tunnel must now be the full width of the ROW for possible future quad-tracking).
ReplyDeleteTo be clear, I think Caltrain is 100% correct in not allowing anything in their ROW that would restrict them from 4-tracking in the future. Any realistic 8 Caltrain / 4 Pacheco HSR plans are going to need 4 tracking (or at least 3-tracking) in this area.
DeleteDue to even Caltrain not including Menlo Park in the areas it has identified where quad tracking may someday be needed to accommodate overtakes (primarily by/for HSR, if it ever gets here), it seems Caltrain could have offered Menlo Park the option of a binding agreement to later fund whatever modifications may be needed to their bike/ped underpass for quad-tracking, if & when ever actually needed (vs. requiring them to needlessly, speculatively bear those costs today).
DeleteI believe at one point Caltrain did say that maybe part of the tunnel inside Caltrain's ROW could remain uncovered so long as MP agreed to pay to cover it up at some future point if needed. Not sure if Menlo Park took them up on this offer. It really does not make sense to build ramps inside the Caltrain ROW when there is plenty of space outside it. Would you let your neighbor build a structure partially on your property so long as he promised to rebuild it if you ever really needed him to?
Delete"Any realistic 8 Caltrain / 4 Pacheco HSR plans are going to need 4 tracking (or at least 3-tracking) in this area [Menlo Park—Palo Alto]"
DeleteUnlikely.
Not that 8tph Caltrain south of Redwood City makes any sense at all, or is in any way "realistic".
As I've been saying for decades, look at BART, which doesn't and never has provided that level of service, ever, on any of its branching lines to far-off suburban nowheresvilles, on lines with vastly less parallel freeway capacity and decades-ahead suburban "development", and never will.
8tph SF—SJ is simply insane. It's not going to happen, ever, and certainly not before the soon-rising seas and soon-uninhabitable equatorial zones make all of this a sick joke.
But regardless, "realistic" 4tph Pacheco (Pacheco HSR itself both insane and Never Going To Happen) needn't involve multi-tracking in Menlo Park: it looks something like this. Caltrain service (the stuff actual humans might actually use, here in the shitty real world) wants quadruple tracking Belmont—San Carlos—Redwood City; 4tph imaginary Altamont HSR wants quadrupling Bayshore—South San Francisco; delusional imaginary Pacheco HSR wants Bayshore—South San Francisco and Palo Alto—California Avenue—San Antonio and triple-tracking San Jose Cahill—San Jose Tamien.
Anyway, all angels on the head of a pin.
Caltrain can't build anything. Not for less than $500 million/km. Which means nothing can possibly be justified ... unless you're actively on the take. Stick a fork in it. Kill it with fire.
Given the current state of affairs and out of control costs it's hard to imagine any of this happening, but I do think that you have to maintain the ROW assuming that someday things might improve.
DeleteThe middle avenue undercrossing is just north of PA station and it's not hard to imagine 4-tracking PA station having impacts that extend that far north. Not to mention the rebuilding of the decrepit San Francisquito Creek bridge that needs to happen urgently. Both of these projects could be made more expensive and complicated by needlessly narrowing the ROW to save MP a few bucks on their pedestrian undercrossing.
Also, when considering realistic 8/4 Pacheco plans I wasn't really thinking about those with a HSR stop in RWC. I was assuming 45-50 minutes SF-SJ times with stops at Mission Bay and Millbrae as is currently proposed.