Hacker News new | ask | show | jobs
by kevinventullo 1572 days ago
I would love for these trucks to be EV’s as much as anyone, but I can see where USPS is coming from. Some selected quotes from the article:

The USPS did say that it plans to put 5000 electric delivery trucks into service starting in 2023 and claims that there is room for more EVs to be added to the mix "should additional funding become available."

“While we can understand why some who are not responsible for the financial sustainability of the Postal Service might prefer that the Postal Service acquire more electric vehicles, the law requires the Postal Service to be self-sufficient,” a USPS spokesperson told the Post in a statement.

Armchair take: If the EPA really wants these changes, they should be lobbying higher up the food chain in order to subsidize these EV’s in some way.

6 comments

Its a BS quote, the deck was financially stacked against EV procurement from the start. For example:

1. EV proposals had to account for a 100% EV fleet. The small percentage of rural routes where this was difficult created a long tail of costs. There's no reason why this couldn't have been scaled down to a more reasonable 90% of routes (or whatever the breakdown happens to be).

2. TCO calculations were inadequate. The real benefit of EVs come from operating costs (much lower electricity costs compared to gas, much lower maintenance). Interest rates are so low right now, there would have been some way to capitalize the initial purchases, if the USPS leadership had appetite.

Yes, there's no way that the EV version should cost $30k more than the ICE version. It's possible to buy a whole EV for that price. It's not like you need a large battery to cover a 45 mile rural route.
weird, rural route carriers are independent contractors who provide their own vehicles, or at least that is how it used to work.
still works that way even in the north end of the phoenix metro area in the foothills of the mountains, where we live. and this is not like the boonies, just the edge of suburbia abutted with mountains.
Could be just longer routes in rural areas that caused the effect in the analysis rather than necessarily actual designated "rural routes."
I don’t now why USPS what evaluate this on any other yard stick other than cost. It’s really hard to know exactly what their numbers look like but I have a hard time believing that their leadership is backwards etc like the epa rep would want everyone to believe. On the other hand I guess this could be a gambit for federal money. Either way it’d be great to see public thorough analysis.
This might be the lobbying. If so, just cut them a check for the EV cost delta and installing EV charging stations at USPS facilities where vehicles are parked.

Assuming outrageous costs (~$50k per usps branch) and 6000 USPS facilities, that’s $300M to install EV chargers. Say you do it over five years (slowly working from highest to lowest utilization geographies), that’s only $60M a year.

USPS spends $300M a year on fuel, and vehicle fuel accounts for half of their energy use. Per the GAO, USPS spends $2B a year maintaining their bespoke Grumman LLVs. The costs savings are blindingly obvious.

https://www.vox.com/energy-and-environment/2020/4/22/2122913...

That's interesting, since electric vehicles should be cheaper in the long run -- especially if they can arrange lower prices for overnight charging.

Perhaps the up-front costs of an electric vehicle are prohibitive for their constrained financial situation. If so, that is an unfortunate bind, if it causes more financial costs in the long run.

> since electric vehicles should be cheaper in the long run

Do you have a reference for this? I'm curious to see where the costs went. For someone like USPS, who purchases mostly one model, in bulk, with long term maintenance as part of the plan, the math might be a little different.

My numbers may be a bit out of date, but I feel are still somewhat in the ballpark. A battery that gets you 300 miles on a charge (90 kWh?) can be charged 1500 times, and costs $20,000, so that is $13 of wear per charge, plus about $12 worth of electricity (probably more, just basing off battery capacity not counting charge / discharge losses). So that is $25 per 300 miles. A car that gets 30 miles/gallon will cost around $30 per 300 miles. So electric saves you ballpark about 16% in fueling costs.

I'm sure though that the battery packs will come down in price by the time they need replacing, on the other hand the charging costs may be higher due to thermal losses.

The average rural USPS route is 45 miles. Non-rural routes are closer to 20 miles. Low average speed (lots of stop and go) means they would likely beat what we are used to for EV energy usage per mile.

A 30kWh battery could probably handle most use cases even accounting for range loss in cold weather.

Edit: Also the USPS estimates the new gas-powered trucks will get ~8 miles per gallon with the A/C on, and ~14 miles per gallon with it off.

That would make your estimate closer to $16 per 300mi for the EV and $64-$112 per 300mi for the gas truck.

> In actual use by the USPS, which includes extensive stop-and-go driving for residential delivery, average fuel economy is about 10 miles per US gallon (24 L/100 km).[10]

From: https://en.wikipedia.org/wiki/Grumman_LLV

30MPG is a pipe dream for these vehicles. Going electric would save a lot more than 16%.

The charging infrastructure is the real killer, you need 1:1 charger to vehicle or some kind of smart charger that can plug into X vehicles and charge each up overnight. There's also the power draw requirements that will take, so it may not actually be able to install sufficient charging on existing sites. There's also the maintenance on the chargers, which is admittedly small. Switching to EV means for the same budget you get say 20% less vehicles (assuming same cost), just because you have to spin up the charging.
Yes, this is a bigger policy and legislative failure that let the previous gen trucks get so old and the new contract go through without a mandate and funding for reduced emissions.
EV or not, why is it so fuel inefficient?
Because they spend the vast majority of their operating time stopping, idling, and accelerating at low speeds, . They are pretty much a worst case scenario for ICE efficiency.
But then the cars do not need to be electric! A hybrid with a small battery will suite this as well and, given the current price of batteries, will be cheaper.
... because the test driving cycle is based on 8 miles city, 6 miles freeway, and then 11 miles stop/go with 700 stops; and if I remember correctly is based on all accessories (i.e. air conditioning) running for the entire time.

This is not your EPA passenger vehicle test cycle.

They spend lots of time accelerating, braking, and idling.
Constant stopping and starting. This is where most car energy is expended. Cars are most fuel efficient when they're rolling about 50MPH.
> Cars are most fuel efficient when they're rolling about 50MPH.

They're even more efficient rolling at 40 MPH or 30 MPH or even 20 MPH, as long as it's consistent (not stop-start). Air resistance scales with something like the cube of velocity and begins to be noticeable around 20 MPH.

Efficiency is easy to ball park, but if you want to get precise then every vehicle is going to have it's own moving target. I would argue that no vehicle is especially efficient at 20mph for long durations. Sure you have no air resistance, but you also are spinning a 6 speed transmission when you really only need one. 90% of the car is being wasted. You might as well just drive a self propelled lawn mower.

Even two vehicles of the same make will be different due to differences in operating conditions.

An ICE engine, especially gasoline powered ones, operate with widely varying levels of efficiency. Lets think of a car in 4th gear at 50mph with 15% throttle applied. Now imagine climbing a steep hill. The car can stay in 4th gear and climb the hill with 65% throttle, or downshift to 3rd and climb the hill with 45% throttle. Speed alone doesn't determine fuel efficiency. Intake vacuum (or boost in certain applications) determines fuel efficiency. The trick is to try to go as fast as you can while consuming the least amount of fuel. There is nothing "efficient" about driving a 250hp, 3.0L V6 at 20mph.

Every vehicle has a different efficiency curve. I was into hyper milling for awhile and the vehicle I drove had a sweet spot at 62mph.
I’ve heard this many times, but I’ve never actually seen it cited.
It's the reason why vehicles have a separate "city" and a "highway" fuel economy rating, since the former takes into account the expected stopping and starting you do in city traffic which you would not on an uncongested highway. It also makes intuitive sense if you understand physics: in a non-hybrid ICE vehicle, you impart energy from the engine into the wheels to accelerate the vehicle and you impart energy from the wheels into the brakes to decelerate, effectively wasting the original energy from the engine as heat. In hybrids and EVs some of that energy can be recovered by using the motor as a generator to recapture it in a useful form.
It's just basic newtonian physics, it takes energy to start and stop moving and a lot less to keep doing what you are doing.
Wouldn't it be more-so that with ICE vehicles you loose 100% of the kinetic energy as heat when stopping. With an EV you convert a portion back to the battery packs with regenerative breaking.
Correct, though many times costs are not just fuel - total ROI includes service cost, availability of replacement parts, insurance costs, etc - a NG car could cost more in fuel and still be cheaper for effectively forever depending on how the books land.
Here's a couple of good ones

- https://sci-hub.se/https://doi.org/10.4271/2017-01-9379.

- https://sci-hub.se/https://doi.org/10.4271/2013-01-1113

You'll want the second just to compare different speeds. Nice pretty graph to follow.

Not trying to be rude, but it's because it's intro Physics (F=m*a, with F coming from fuel, battery, or brakes), with an understanding that the process that generated that force has an efficiency that's less than 100%, mixed with a dash of intro Calculus (integral of force).
Respectfully, intro physics mentions nothing about ICE operating most efficiently at 50 MPH.
Unsure why USPS wouldn't look up to another company that is all based on logistics and delivery. If Pepsico is already buying Tesla Semi trucks af of last Q of '21 [1] then they clearly did the homework and decided electric trucks makes more economical sense than other means. After all, Pepsico as a publicly traded company, is all about squeezing the last cent out of the cost of freight.

https://www.cnbc.com/2021/11/08/pepsico-ceo-says-he-expects-...

I'm all for electrifying most of the USPS fleet but I don't know that a Tesla Semi running regional highway routes is a good reference point.