Hacker News new | past | comments | ask | show | jobs | submit login

It's interesting that "cost" is the reason given for why these vehicles aren't electric, but... how could that be? Batteries for electric carmakers are not that expensive any more, maybe $150/kWh.

LLVs are currently horrendously inefficient (fleet average of about 9mpg). They have a ~13 gallon tank for a range of about 120 miles. They drive about 18 miles a day on average (total fleet average, so some are idle and probably the typical route is 45 miles, but 18 miles is good for our purposes) and burn 2 gallons of gas. At $3/gallon, that's about $6 a day in fuel costs (and let's ignore brakes and oil changes). We could put a 75kWh battery in that thing, easily get 250 mile range (double the range of what the LLV is). For an efficiency of about 3.33 miles per kWh, and an average commercial electricity price of 11 cents per kWh, that's 59 cents per day in electricity cost. So every day, the electric one would save $5.40 in fuel costs (net).

Even with that generous 75kWh battery, you're talking just 6 years for it to pay for the electric battery to pay for itself in fuel costs vs the old vehicle. And these things are designed to last for multiple decades, plus this will drastically reduce oil changes and brake servicing costs. Are they really that cash strapped they'll ignore something that can save them that much? The US federal government definitely isn't strapped for cash. Does this really boil down to punitive fiscal requirements for the USPS put on by Congress and DeJoy? Because if so, that's inexcusable.

Or perhaps no one bid a competitively low price for an electric powertrain?




People on HN are technical, so we're discussing technical reasons. However, that's not really how the government works.

USPS leadership had instructions to electrify the fleet. For whatever reason, they decided they didn't want to. Maybe they didn't want to deal with a risky change, maybe they thought it would hinder meeting their other obligations. Doesn't matter. They decided they didn't want to do it.

Their next step is then to write requirements in their rfp such that they can select a contractor that cannot meet the electric mandate. This allows them to legally pass the buck for the electric thing to their contractor, who will get paid big dollars to "develop" electric vehicles that their customer doesn't want.

Congress gets to brag about their environmental contributions, USPS gets the vehicles they want, and the contractor gets a sizable sum to do nothing. Win-win-win.


I suspect they where afraid of push back from the previous administration as well.

I am surprised they didn't look at using LNG instead of petrol.


Even a 75kWh battery would be expensive and heavy, and probably overkill in this case. If 100 miles of range is more than enough for most routes for the day, then something like 30kWh might be more reasonable.

Also one might be able to keep costs down by optimizing for cost rather than energy density. LiFeP04 cells have been comparatively expensive, but it seems price has been dropping lately. It's also a good choice for safety and durability reasons, and because they don't use cobalt which is sort of expensive and sort of rare. Not using cobalt means avoiding a resource that would bottleneck other manufacturers of EVs.

It'd be interesting to know what the actual specs are on the EV version of the USPS vans (battery type, capacity, etc...) and see if my speculation corresponds to what they chose. I don't know if they've released that publicly yet.

A 30kWh battery could easily be 5 or 10 thousand dollars, which may well be worth it to save on gas for a long time; on the other hand, gas engines are well-understood and really cheap, because we have a huge established industry that builds engines by the millions and has been optimizing for cost for over a hundred years.

This sounds a lot like the "Captain Samuel Vimes 'Boots' theory of socioeconomic unfairness" from Terry Pratchett that generally eventually gets quoted in this sorts of discussion.

https://www.goodreads.com/quotes/72745-the-reason-that-the-r...


I think it’s more about timing. Rfp in 2015, the ev landscape was different, and not super clear battery technology would be where it is now.


To some people it was. I have wondered then and I wonder now how it wasn't clear.


The LLV's are not merely old, they're ancient. The EV marketplace still isn't particularly mature: there are competing charging networks, the charging standards are still evolving, and in the US, there are perhaps two mainstream manufacturers of pure BEVs (Tesla and Nissan).

If you're the sort of customer who keeps a vehicle 3 years and then upgrades, none of this is likely a worry for you. Any shakeup in the charging networks or incompatible changes to the standards will probably happen long after you sell the car.

The USPS, on the other hand is in this for the long haul. It seems reasonable to suppose that they'll have their own charging network, so problem solved, right? Wrong. They'll still need parts support every part of this operation for at least 35 years if they keep them as long as the LLVs.

Remember the lengths one dedicated person had to go to to resurrect the DE-19 connector[0]. You're looking at that problem multiplied by thousands of parts.

I've always coveted the original Honda Insight (the two seater), and one popped up for sale near me last year so I did some research. First party support is pretty well done. Third party support exists, but it's not a volume business, and it's priced accordingly. And that's a car that's only 15 years old.

The mail truck fleet will be bigger, but the price of parts will still be high if there are big shifts in the market and nobody's making anything similar. The market has matured and stabilized since Honda quit making the original Insight, but probably not enough to inexpensively support a vehicle for 3 decades.

[0] https://www.bigmessowires.com/2016/06/04/db-19-resurrecting-...


Except this is a PERFECT application of battery electric drivetrain. I literally can’t think of a better one. If this isn’t “good enough,” then precisely what is? What WON’T be sunk by short-sighted risk averseness (actually nothing risky here) or plain indifference if this can’t be electric?

And now they’ll have forced themselves into expensive conversions if they ever want to go electric in the next 3 decades.

It’s legitimately depressing.


You're not wrong, but as somebody who buys cars used and hangs on to them for a while, I'm sitting out going electric for at least another 5 years in the hope that things shake out somewhat on the charging side.

That, and for all of the reasons the USPS is sitting this out. The TCO has some pretty big unknowns over a multi-decade timescale at this point. I do a lot of my own work on my cars, and long-term parts availability is a real struggle at times.


I think J1772 has basically won in the U.S. for AC charging. It's hard to imagine something replacing it at this point. The competitors are Tesla's charge port and the other DC charging standards. DC chargers though require a lot of power and are expensive, so it's not realistic to expect them to be ubiquitous. J1772 is not much more than a 110 or 220 outlet with a fancy plug and electronic handshake protocol, and can be installed just about anywhere that a regular 110 or 220 outlet is, assuming adequate power is available at the site.

(A typical house might have dozens of outlets, but no one expects to be able to run a space heater off of every single plug all at once. Someone installing dozens of chargers, on the other hand, probably intends to use them simultaneously.)


Between all the people looking at a problem you'll probably get every possible solution and one (or more) of them could be correct. Hindsight reinforces that opinion and makes it look more certain than it was at the time. But would you have bet the farm for your assumption then? It's easy to "see the future" when you have no risks to consider.


Also, I think that view only looks at the vehicle itself and neglects the onerous funding model the USPS is saddled with and the investment required on the supply chain/logistics side.

The fleet itself is a massive capital investment. But changing the fuel would be sizable as well. The USPS already has a supply chain network and facilities for gasoline. Replacing that will be huge in itself.

Municipalities are a better first-step and have had some success with moving bus fleets and maintenance vehicles to NG or EV. But they usually just have a handful of fueling stations and suppliers.

Fleet renewal is a pretty well understood budgeting process in any large scale logistics network (airlines, military, delivery..). You reach a point where replacing the fleet is a better investment than maintaining the current one.

To migrate to EVs, the USPS would likely wait out success in the private sector (like FedEx’s announcement) to demonstrate ROI, and retrofit this design or phase in a 2nd or 3rd iteration on it as a separate budget/RFP process.


That's essentially what the German Post did. They developed their own all electric van the streetscooter. Unfortunately they could not turn it into a viable business and sell enough units to others, so they shuttered the operation last year. Quite unfortunate, because I believe the project was well ahead of its time.


iirc German Post didn't really develop the Streetscoter [1] themselves, instead they acquired the company that did. It was a start up lead by a professor from university Aachen and from what I can remember, they had quite a lot of problems besides the ones you mentioned. There were a lot of recalls and other problems, especially towards the end of the company [2]

The post employees also often complained about the relatively high loading edge for the storage box, which makes it super hard to load and unload heavy packages.

Regardless of the problems, I think it was a really good idea and a rather cool looking vehicle. And when it comes to urban streets in Germany, the street scooter probably was the perfect vehicle when it comes to space requirements.

[1] https://en.wikipedia.org/wiki/StreetScooter [2] (German) https://www.businessinsider.de/gruenderszene/allgemein/stree... [3] (German) https://www.elektroauto-news.net/elektroautos/streetscooter-...


and they developed it it "by themselves" i.e. by a company they acquired for that purpose, because neither the volkswagen group, nor mercedes/daimler were able or willing to just shipmelectric local delivery mini vans.

last year deutsche post lost two streetscooter CEOs for "diverging opinions on the future of streetscooter". deutsche post is now trying to sell their subsidiary.


Th Norwegian Post Office uses electric vehicles. I saw one in the street in Drammen this morning. See https://elbil.no/posten-elbil/


Local postal company started using Swiss electric tricycles, though slightly customized. I imagine they aren't cheap to buy, but maybe total cost of ownership is reasonable? Also I assume they can tolerate winters well compared to two wheeled vehicles, yet can navigate narrow spaces.

https://www.post.ch/en/about-us/responsibility/our-prioritie...


It seems like you are comparing a theoretical modern EV to the terrible efficiency of the old LLV. They didn’t choose to continue buying the old LLV, they are buying a vehicle with a new ICE. So the comparison makes little sense.


Repair and maintenance are probably far more expensive than the gas costs, especially as a government contract. Gas prices will also increase as demand falls of a cliff over the coming decade.

Electric wins on all fronts in cost, simplicity, and city health


There is no reason an efficient ICE can't have cheaper maintenance than an EV.

In this day and age, reliable ICE drivetrains are pretty much a solved issue if you're not trying to make tons of horsepower (which they're not)

Most of the maintenance with a modern ICE has nothing to do with the drivetrain. Tires, brakes, suspension components, filters, all things that EVs still have (brakes get a respite from regen, but the extra weight takes the savings back for the tires).

Oil changes with modern synthetics and low performance demands can be 15k+ apart.

Then there's the fact you can get an ICE drivetrain repaired anywhere and fuel up anywhere.

The cost of building up a repair and charging infrastructure alone would probably have killed any savings.

This is not an anti-EV post, but it is an anti-antiquated-ideas-about-ICE post.


> brakes get a respite from regen

The brake cylinders might seize instead. The pads and disks are quite cheap to replace.

How many private cars actually change brake systems because of braking too much and not environment attrition.


I doubt maintenance & repair costs be more expensive for an ICE engine. The actual engine part requires little maintenance: occasional oil changes each year, spark plugs every five years, and maybe a belt in 10 years? I guess brake pads might be more common, but those are still a once every few years thing and inexpensive.

All other common repair items would the same for an EV: suspension, various ball and CV joints, tires, AC, body damage, sensors, bushings, etc


I sometimes try to find an article I’m sure I read here at hacker news. It said something to the effect that the top ten wearables on a ICE car, measured by revenue in the auto shop market, don’t even exist on an electrical vehicle.


Well that’s hard to believe, unless you pedantically itemize every tiny little thing. For example, a single oil change typically requires a new oil filter, a new crush washer on the drain plug, oil, possible additives, and sometimes an air filter swap. So that’s five items right there, but we’re still talking only 20 minutes of work and $50 of maintenance.


Ye ... my experience with old cars is that the ICE is not the main time or money sink.

Thinks like drive axis, suspension parts, wheel bearings, brakes etc is common with EVs - and those fall apart.

The timing and generator belt headache is replaced by two selfdriven motors/pumps on EVs for the oil or water cooling system and servo. Hardly less mentainance and more expensive parts.

The main thing is that most EVs are new so most have not had their exposed systems beaten out of spec and turned into rust yet, I assume.


Smooth delivery of power has surprising downstream effects. The wear on ball joints and tyres for EVs can be significantly less due to not having as much jerk (derivative of acceleration) from their motor.


Wear on tires and ball joints are more due to increased weight and high torque from zero RPM.

https://www.fleetowner.com/equipment/tires/article/21147010/...

Not sure why you thought otherwise.


Cited article is about _potential_ torque loads and the need to design for it, and does not address light duty typical use case for mail delivery.

Jerk causes chatter (slippage) between parts and can significantly reduce designed life. Electric motors are amazing at smoothing out discontinuities. Nokian has stated EVs are easier on tyres for this reason.

In this case for short-distance batteries I estimate battery pack mass of 150kg more than ICE equipment. Increased mass would not have a major impact on ball joints and tyres, these loads will be designed in.


There are literally hundreds of sources you could find for how EVs have worse wear.

I have no idea what you're on about with "chatter" without specifying any specific part of a drivetrain...

Like your whole comment reads like someone trying to apply extremely general theoretical concern while handwaving away actually connecting it to practical application...

And where are you getting the idea that:

a) the weight difference will only be 150kg. These things can end up doing 50 miles with a heavily loaded truck in high heat. They'll need the same cooling systems the ICE does, they'll need battery packs over-provisioned to meet their 25 year life span mandate.

b) even 150kg wouldn't increase weight. I don't know what you think "designed in" means. It means either make the parts beefier so that they can withstand greater wear before replacement, or it means replace parts more... neither option erases the problem.

ICE or EV obviously they're going to design the suspension to handle the curb weight + occupants and cargo but adding to the curb weight will have the result of making something more expensive. Doesn't really matter if you're paying for it upfront or later, you're paying for it.

My Spec Miata's suspension alignment was designed with my weight in mind and literally every pound counts. Even the weight of the gas in the tank opposite me can have a noticeable effect on how the suspension sits, which in turn affects tires wear.

The same thing plays out on a "normal" car in much slower motion


My comment only addressed wear on components, and specifically mentioned tyres and ball joints. I am not making a statement on upfront costs. I’m also not considering behavioural changes of users when switching to electric motors.


Decreased demand would tend to result in reduced prices, not increased prices.


The laws of supply and demand only work that way if supply is limited, which is very often not the case. A lot of production benefits from economies of scale, where fixed overheads are dwarfed by unit costs. In such cases, decreased demand results in reduced production, which results in the loss of economies of scale as production overheads remain the same or similar, thereby increasing in magnitude relative to unit costs.

Maybe it costs ten gajillion dollars to produce one litre of petrol, but twenty gajillion dollars to produce one bazillion litres of petrol.


We are thousands of times over capturing economies of scale in gas production, we support thousands of independent drilling and refining sites.


Only instantaneously. In the long term supply decreases to meet decreased (now niche) demand which decreases duty cycle on operations, hence increasing per unit cost.


Refined petroleum products will certainly not be niche within a decade. So much of global transportation simply can't swap out ICE with electric now, or in many cases ever. Huge swaths of the globe lack the electrical grid infrastructure to entertain even small moves in that direction. And huge numbers of drivers in countries with adequate infrastructure can't plug in at home due to living in apartment complexes, urban street parking, etc. Electric transport is a great growth area but it will not make gasoline niche any time soon.

More likely the decreasing cost of electric transportation options will have downward competitive price pressure on gasoline.


I have seen one concern expressed as potential imbalance in demand for different refined components. For example, if demand for diesel were to hold steady, but for gasoline diminish, to some degree gasoline potentially becomes a leftover pollutant.


It seems to me that a parcel-laden EV that starts and stops repeatedly will still be pretty lossy -- ie, the effective fuel efficiency will be quite low compared to typical EV city driving because regen doesn't work quite so well for precise stops. You're still going to engage the brake pads for a full stop in front of every mailbox.

Also, maintenance would probably be more expensive if you have to train people to work on the electric powertrain.

But presumably they could always modify the powertrain in a future iteration.


And somebody think about the extra load on the power grid!

It's not that these aren't valid concerns, they have to be addressed, they're just so minute they don't change the calculation even slightly.

EVs at that speed may lose a bit more (already more efficiently generated and used) power in regen - vs. a guaranteed 100% loss on a gas vehicle.


We already know there's a guaranteed loss on ICE braking. If that were the only difference between ICE and EV then I would totally agree with you...BUT

Cost/benefit analysis comparing ICE and EV changes wildly if the effective range of the vehicle is lower and the comparable fuel efficiency is not enough to justify the additional costs of EV.

These are fleet vehicles that get deployed all across the country, and they have to be serviced, moved, etc...I doubt the choice not to go EV was an oversight, considering that fleet has been one of the beachhead markets for alternative-energy vehicles.


It’s too bad, a massive fleet built to spec, with central fueling and maintenance locations is a perfect use case for electric now even though we are potentially a long way from charging stations as convenient as gas stations.


from the other linked article:

> While some will be EVs, some of the NGDVs will come with modern, fuel-efficient internal combustion engines. As electric vehicle infrastructure improves, these NGDVs will be able to be retrofitted with the battery-electric powertrain later on.


That doesn't make any sense. All the "electrical infrastructure" needed to drive the typical 45 mile route is a regular old 120V exterior outlet. The retrofit cost is going to probably be double the marginal cost it would've been to just make it electric in the first place.

Whoever is making these decisions either wasn't given competitive bids or they're poorly informed about electric vehicles.


How long will it take to recharge these hundreds of vehicles, and who will be around to take them out of their charging stations and move in a new vehicle?

Electric vehicles are great but at scale, the issues around charging them daily are very real. It works for a single homeowner, but if a post office has 100(?) trucks, and it takes 2 hours to charge, someone needs to be there to keep track of them, unless you expect 100 charging stations.


I think if a parking area has many vehicles in one place it might be easier to move the charging infrastructure around them. I’d assume that can be done easier than moving refuelling infrastructure. So maybe 100 vehicles could have 10 charging widgets on wheels with cords to their base point, 2 people to hook up and move to the next one that is done.


Just put a power outlet - only minimally more beefy than a normal wall socket - at each parking lot. Over night, they are easily charged no need to move them around.


You don’t need a “charging station,” you just need an outlet. And you don’t need someone to “charge” them except the person who parked it. It charges at night after use. An exterior outlet costs almost nothing compared to the fully burdened cost of a parking space.

Of all the “very real” problems with electric vehicles, you picked two or three things that aren’t.


If you have more than one vehicle that you need to charge, you need more than one outlet. Worse, you cannot just get connect all of them to one outlet through daisy chained extension cords: you need separate 120V circuit for each vehicle. Even if you are lucky enough to have multiple outlets around where you park your fleet, these will likely all be on the same circuit.

This problem is of course solvable, you need to build a charging station. The point is that you need more than "just an outlet".


Nope. You only need a 15A branch per vehicle. You can install a 30A branch, too, if you like, and two poles of those are enough for 4 vehicles. For slow charging purposes, 120V exterior outlets are more than good enough and have commonly been used for government electric vehicles for decades.


Slow charging may not be fast enough. Around here, I'll sometimes see them making (mostly) parcel deliverys as late as 9PM, and then they're out in the morning at 4 or 5AM.


Those areas should splurge for fast charging then. Plenty of areas could get by with slow charging though.


At most post offices the trucks are parked outside without electrical outlets nearby. Sure those can be built but it will take a while. In many cases the circuit supplying the facility will have to be upgraded.


Any parking lot with lamp posts has electricity. Takes way less time to install an exterior outlet than it does to build a vehicle.


Those lamp posts aren't close to every space.

Are you going to run extension cords for the 150+ feet to each outlying vehicle? No?

Then you have to tear up asphalt to lay down new lines. That's not cheap, especially with a USPS that's so strapped for money right now (for political reasons).


1. Do they park at the same place every night?

2. Do they have to install electrical charging to all possible parking space?

3. Are these installation paid by government or the USPS?

4. Are those installation done on government property? Or on private property? Do the owner agree to do so?

5. Miles and Maintenance / TCO flavour Electric in these usage?

6. Insurance?


Are you implying that circuit driving the lamp posts can handle multiple 15A load?


You'd be surprised how much load runs through a chain of streetlights. 3phase/400V and they pull heavy duty conductors through the conduit. Could definitely handle multiple 120v/15A equivalents (but it'd be way cheaper and more effective to pull dedicated conduit + conductor)


also: In my city (in UK) the lamp posts date from a long time ago. Only a few years ago all the sodium lamps were replaced by lower power LEDs, which presumably means that much of the cabling is now massively redundant.


Don't underestimate the efficiency of old-fashioned sodium lamps, they're pretty close to that of LEDs from what I understand. I think part of the reason places are so keen to install LED streetlights is that on paper they have a much longer lifespan than the bulbs in the sodium lamps they're replacing. Also, they're more modern and everyone knows how efficient they are.


Low pressure sodium vapor lamps are also very yellow [1], whereas a modern LED is much more broad-spectrum and thus looks nicer.

Unfortunately, it does mean that LED lamps are a greater hazard for light pollution.

[1] Low-pressure sodium vapor lamp spectrum: https://en.wikipedia.org/wiki/Sodium-vapor_lamp#/media/File:...


interesting, perhaps the new design for the lamp covers is part of it but in my direct experience from a boat there is massively less light pollution with the LEDs which shine downwards only and show in a town on the side of the hill as a set of twinkling fairy lights and in a city over the hill as not much at all whereas the sodium lights produce an orange glow which can be seen for 20 miles or more at night. However, that doesn't say anything about the frequency which could of course be more damaging locally even if it scatters less.


Correct, which is why they're installing lamp post EV chargers


There's quite a bit more charging overhead for 110 vs 220.

Also, there are 30,000 post offices. I'd imagine it will be a long process, even if they just put in 110. one evse requires large amounts of continuous current, let alone with a parking lot full of vehicles.


I suspect there is a high level of variance in both the length of different routes and the availability of power in parking areas for mail trucks.


As I pointed out in another thread about these vehicals, the USPS still operates mule trains, hovercraft and helicopters for some deliveries. That's the most extreme examples, and I imagine there are many journeys that are somewhere between in terms of range and difficulty. People have a very narrow view of the world and forget that not everywhere is urban or suburban, or even just farming country levels of density. The USPS has to service them all.


Grumman LLVs only have a 120 mile range. This is already double. And sure, maybe you’ll want an RV style outlet (also pretty cheap) in case of 5% of vehicles where 120V wouldn’t be enough.


The LLV can refuel in 5 minutes at over 160,000 public locations.


It sure can!

That's also much much more flexibility than it needs. For a low-mileage fleet vehicle it's deep into diminishing returns.


You're assuming that vehicles are charged disparately, rather than having several vehicles charging in tandem in the lot behind the post office.


120V is tremendously inefficient when converting to DC. 220V is a little less so but it's better to have other means of charging.


> They drive about 18 miles a day on average

The LLVs don't drive for an hour a day and sit in the parking lot turned off for 23 hours a day. Many routes might be 6 hours or more. You need vehicles that have AC or heat and the ability to run for an entire day. The differential cost for such a vehicle is not the cost of a battery. I'm not saying the current purchasing process is waste-free, but you need to use realistic numbers.

The longest route is about 185 miles in Phillipsburg, KS. There are averages, medians, and modes, none of which are the right numbers. The average isn't as important as probable mileage.

For example, say a bus carries a maximum of 80 people. When 10 people get on your bus, your costs aren't 12%.


Carriers on many rural routes use their own vehicles. My parent's carrier drives a Subaru Outback


I hadn't even contemplated that. Thank you. That's an important use case.


Aaron Gordon at Vice gives some explanations - most important cost-related bit seems to be they weight upfront cost a lot more than amortized cost, and non-electric does better on that front.

https://www.vice.com/en/article/3an4k8/dont-expect-the-usps-...


Weight-per-wheel also figures into wear inflicted on local roads.


Everyone hates OpEx, even Feddy'Gov.


ITT people arguing that a country with $7 trillion of spending budget can't install a few power outlets in its post offices.


The US is massive and its grid of infrastructure can't always keep up. But you're right that cities should have electrical USPS vehicles. Dominoes here have electrical bikes in city locations, why can't the post office have electrical cars?


I don't think that spending really goes to the post office tho.


They would also need to provide infrastructure for charging the fleet, possibly multiple times a day depending on what the routes look like, which is a significant cost if they want to do more than one battery worth of a route in a day, even if they did have short enough routes that they could do overnight charging for everything that's still a lot of power draw and would need expensive electrical upgrades. Sticking with gas there's no further infrastructure investment required.


Maybe the charging infrastructure isn't available?


A literal 120V exterior outlet? 83% of these vehicles drive in the city. They drive on average 18 miles a day (typical routes being about 45 miles). You don't need a 150kW Supercharger station, you just need a 1.2kW regular old 120V exterior outlet used for trimming the bushes. Even a cheap RV hookup would be overkill.

If a federal government agency in the most powerful national on the planet can't take the most obvious step toward electrification in a case which is the best possible situation for an electric vehicle (very stop and go traffic, slow speeds, short routes, long total miles on odometer over life of vehicle), what hope does the rest of the world have in taking action on climate?

Greta Thunberg is right. People don't care about climate. Literally everything else is a priority.


Yes, there are probably very few outlets where mail trucks currently park. These things are parked in the middle of parking lots, not in garages. And you need more than one outlet, you need one for every truck that parks there. And I am sure no current USPS office has power service hooked up for several hundred amps of draw that was never anticipated to be used.

There's no way to do a project this big without phasing it in. The USPS has many tens of thousands of buildings.


These vehicles ARE being phased in slowly. And installing 100-200 Amps of commercial service is cheap, and I actually think would largely be unnecessary as they can be set to charge at night when office demand is super low.


Does the electric code permit overallocation in that case? That's not my area of expertise but I doubt it does.

And yes, installing service is "cheap" but it takes time.


It takes way less time than building these vehicles will and can be done cheaply with local electricians. It’s strange how common magical thinking about EVs is, like electricity is this newfangled thing that we don’t understand and need massively new, poorly understood infrastructure. We don’t. And that’s a huge advantage.


Many of these buildings are not owned by the USPS. Many of these buildings do not have adequate electrical right now. USPS has 30k buildings.

Paying someone to build vehicles is much different than retrofitting buildings, each in a different jurisdiction, in many cases that you do not own, to allow plugging in a huge number of vehicles in all sorts of weather conditions - in the middle of often uncovered parking.


It's just not that hard. I charge in uncovered driveway. With an exterior outlet. In the winter or summer, doesn't matter. It's simply not that hard.


Do you also run a national mail delivery business?


I am not saying and have not said electricity is the hard part. The logistics of steering the world's highest volume postal operation (with financial issues as it stands) is the hard part. The USPS handles 43% of the world's mail. There are no projects of this size that aren't phased in, which is apparently what they are doing. Some of these vehicles will be electric, so it stands to reason that they will also be building out the equipment and processes required to roll them out.


Except these vehicles are not going to be immediately dumped onto all USPS parking lots. They are going to be built over a period of years. Adding an outlet for each vehicle would cost on the order of 1% of the vehicle cost and would be paid for in fuel savings within a couple months.

Look, I don't know where this idea came from that electricity is this hard problem. It's not. As it is now, many of these LLVs have to be fueled sometimes multiple times per week. Simply plugging into an external outlet not only is cheap but will actually save time.

People who do not have electric cars keep looking for an inventing some kind of "gotcha," but they really are this easy. You literally can just use a normal external outlet to charge them, and even if you want faster charging (totally unnecessary and probably undesirable for this application), it costs maybe $1200 to get an RV hookup (equivalent to L2). I charge from an old dryer outlet, although sometimes use an exterior 120V. It really is that easy. In fact, I don't know if big USPS locations have refueling depots like bus depots do, but if they went electric, they'd be unnecessary and could be removed.


Parking spaces in the US are, by and large, not commonly populated with dryer outlets or RV hookups. This is a common issue that affects EV adoption, particularly when people lease the property where they park their car.

https://www.nytimes.com/2020/04/16/business/electric-cars-ci...

You are entirely correct that the concept of an electrical outlet is not difficult. Commercial real estate is a bit harder, however.

Bear in mind that this contract is for a company to build trucks, and what you describe would be work that is out of scope anyway. The current fleet is way past its replacement date, and needed to be replaced about 10 years ago already. The USPS has funding issues at the moment and it is entirely possible that they cannot afford to do everything at once.

While you claim that it is "1% of the vehicle cost", I think some quick math would indicate that there's no way this is right. 4.8 million dollars to retrofit 40,000 locations for EV charging? That's $120 per station. You won't even be able to pull a work permit for $120.


I still don't understand your point. For your argument to be valid, you'd have to assert that a factory can build new mail trucks faster than you could add charging points.

There will be one factory pumping out cars on a specialised, serial production line. It will be competing with 30 thousand locations operating in parallel.


It’s an issue of distribution.

LLVs are past their expiration date everywhere and likely need replaced all over the country. But, it doesn’t make sense to add charging capability ad hoc for a single parking space in a lot when a new vehicle is sent out - the marginal cost of each space is not linear — you’d retrofit an entire parking lot at one time.

If 10 locations have 10 existing vehicles each, and all of them need to replace one failing LLV, going 100% electric for new vehicles means you need 10 vehicles but a retrofit for 100 parking spaces. It does not make sense to tear up a parking lot every year for one new vehicle over the next 10 years.


You wouldn’t replace mail trucks individually, you would fully transition each location as a unit.

Also, you overestimate the complexity of installing charging points. In many cases the parking will be directly against the post office building. In many cases you would only have to dig a small trench. In some cases you would string up a new overhead wire. When you’re doing 10 at a time, this isn’t complicated or expensive on a per vehicle basis.


$482 million is just the first payment. They're not making the vehicles for $12k each.

And while 1% is too optimistic, you'd presumably install multiple outlets at a time.


1% of the 6 billion total estimate sounds like it's in the ball park.

But that's an expected expenditure over 10 years, and you have to have the ability to charge before you can use the vehicles. That's probably another important factor, because LLVs are past their expiration date now. You're right that it makes sense to upgrade an entire facility's electricals at once. That's almost certainly what will happen over the course of the rollout, but the need for replacing LLVs is likely evenly distributed among locations.


It seems like the right way to charge a fleet of vehicles if your site has limited power is to have some central device that coordinates charging rates, so that together they only pull as much current as is available. You could also set priorities. That way you can install as many charge ports as you want, and just have to worry about whether you have enough power to charge all your vehicles every day.

I don't know if that sort of setup is commonly done or is allowed by code.


A 100 vehicle fleet would require not 100-200 Amps service, but 1500 Amp service, and that's a whole different problem.


Outlier locations will have outlier problems. That's not necessarily something to be worried about.

The average post office has six delivery vehicles if I did the math right.


I am in a small town, with one post office for 40k residents. We have approximately 20 mail trucks.


Well, you've probably got 14 to 16 hours of time to phase that charge across your fleet. You're still going to need a lot of outlets, but you only need to feed power to 20-25% of them at a time.


In many locations the charging window is going to be a lot smaller than you think with longer routes and later deliveries pushing parts of the fleet into sub-10 hour charging windows. Now you have a need to shuffle power and charging across this local fleet in the middle of the night. Who exactly is going to be doing this? If you are automating it then you are making some very large bets on the charging infrastructure being 100% reliable.

It is a problem that sounds simple at first but becomes more complex when you actually dig into the details. A fleet that is ICE-ready on day one but which can be retrofitted to EV means that you are able to deploy something that works now and pick locations to test EV and all of the details that are involved in the process by doing partial upgrades as various locations. Overall, sounds like current plan is the smart one.


A) The USPS is not technically a government agency. B) They're pretty strapped for funding right now.


I think the USPS actually is a government agency—from https://en.wikipedia.org/wiki/United_States_Postal_Service#G...:

> The USPS is often mistaken for a state-owned enterprise or government-owned corporation (e.g., Amtrak) because it operates much like a business. It is, however, an "establishment of the executive branch of the Government of the United States", (39 U.S.C. § 201) as it is controlled by presidential appointees and the postmaster general.


One outlet is certainly fine, 100 outsets, not so much.


Just putting this out there.."Adapting a lamp post into a charging point for electric vehicles can cost as little as £2,000" Seems like a good infrastructure project ...


2000 pounds? Try $200 maybe. And these vehicles cost on the order of $40k. Penny wise and pound foolish to spend thousands of dollars on a combustion engine and drivetrain just because you're afraid of a $200 exterior outlet.


$200 won’t even cover the cost of the electrician’s labor. And beyond just the cost of materials and labor to actually do the thing, consider the cost of the consultation on what is actually possible for a given location, the time and cost of permitting, the time involved in all the paperwork such as the drafting and filing of the plans, etc. This isn’t and informal process like some exterior facing outlet you decide to add to your home as a weekend project after looking up a YouTube video and doing a Home Depot run.

Then there’s all the aggregate costs associated with switching. Maintenance of an EV requires different skills and different tooling. Establishing new or amending existing maintenance contracts, training new mechanics or retraining old ones, rewriting guidelines, purchasing new tools and equipment, etc. Many hundreds of thousands of dollars of costs per region.

And you’re going to do that across far-flung areas of the US, many which have a current EV install base near or at zero?

They have to work to transition this across tens of thousands of locations, while not jeopardizing the delivery of hundreds of millions of pieces of mail a day. And you’re going to suggest that they NOT take a slow phased approach? To transition one of the largest fleets of vehicles in the world?

Please. Consider for a moment you may be wrong. That things are not as simple as you think. That when every single person is trying to explain why, you shouldn’t automatically attempt to find ways to dismiss or invalidate what they have to say.




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: