We're Charging Our Cars Wrong

Buried deep in the article, what the clickbait headline alludes to: Car chargers are too complicated and expensive. We can make them simpler and cheaper while keeping them safe, which would let us build many more charging stations.

It would be nice if instead of the fast charging problem the focus would be shifted to standardized battery packs, that can be field replaced. I don't really want to own 50-100kwh battery. I just want to use the charge in it and happy to pay for that.

The argument sounds kind of persuasive to a layman (even one that did a little bit of EE a while back), but my gut says there would be no need to write this article if there weren&#x27;t regulations or other engineering traditions that decided galvanic isolation is necessary for safety.<p>I guess the difference is galvanic isolation is physically (i.e. passively) fail safe, while ground detection is an active safety measure? You can always put in two ground wires in the current system too...<p>Is there some way for the ground wire to have a signal in it even if it were broken (via EMF)? Can the chip fail in an unsafe way?

I can&#x27;t help but wonder while reading the article: why didn&#x27;t this win in the market? As the author mentioned, the tech was there with the prototypes they built over 20 years ago. Why did Tesla and every other electric car manufacturer decide to go in a different direction?

If only we could set up subsidies for this infrastructure in the same way we subsidize ag and oil…

A big issue I see with this is that it would necessitate a new fast charging standard, which (due to the extra ground connection) could not be backwards compatible with any existing fast charging standard - rendering it useless for the 40 million electric cars already on the roads worldwide.

Require a coin-operated 120V outlet with gfci and accessible way to reset the breaker for every two parking spots, everywhere. Many benefits:<p>1. Everywhere you go, you know there&#x27;ll be somewhere to charge overnight.<p>2. It&#x27;s the cheapest per installed spot, by far, allowing way more locations.<p>3. Renters can safely by an electric car and have home charging.<p>4. Coin-op 120V are far more robust than cables with valuable copper.<p>5. It de-incentivizes excessive parking.

You don’t really need two ground wires, CP has a diode with series resistance 2.74K to PE (ground) so it’s trivially possible for the EVSE to just send a push&#x2F;pull waveform over CP and if it reads push but not pull (remember diode) then it knows it’s connected properly.<p>The return path for the pilot current will be over the PE contact, so it can also be detect and doubly verify that PE is low resistance, although ideally you’d want a much smaller series resistance.

So basically, remove isolation transformers, and feed the line voltage (7.2kV or so) through a simple buck converter into the car.<p>Instead of galvanic isolation, use redundant ground connectors that are monitored for continuity.<p>I&#x27;m not sure I agree with this. It puts too much responsibility onto the EVSE side, as it&#x27;ll have to be able to break the fault if it happens. If something like emergency pyro disconnects are mandated, I might be able to trust it a bit more.

&gt; We estimate that the cost of the bill of materials and assembly of a galvanically isolated charging port is about $300 per kilowatt. So a single 300-kW port in a public charging station includes about $90,000 of power electronics, of which about $54,000 is for the isolation link.<p>I would love to see a concrete BOM for a sample build and mouser links to back this up.<p>Seems ridiculously high, like paying some military&#x2F;space grade premiums, or just using very niche parts without economies of scale driving the cost down.

This doesn&#x27;t solve fundamental problems with how America has built its charging infrastructure.<p>We <i>never</i> should have placed hardwired cables on the L2 or L3 chargers. The CCS plugs are broken about 50% of the time? It&#x27;s a joke.<p>Rented an EV in Europe... the cable is in the trunk! You own it. Take it out, plug it into the charger, plug it into your car.<p>Having a bad day? Drive over the plug? Smash it with a hammer? Don&#x27;t know how plugs work so you tried to kick it off your car? No problem, go buy yourself a new one. The charger still works for everyone else.<p>The charger HW can be agnostic. Car manufacturers can make up whatever crazy plug they want for their car, just provide the cable.<p>Tesla wants a 1&quot; cable? No problem!<p>You want a 300&#x27; cable? Do it!

I&#x27;m no electrical engineer but would that really be so much cheaper? A buck converted is essentially the same thing where the transformer is replaced by an inductor. You get rid of a bit of copper but all the power electronics to do the high-frequency switching is still there. Together with most of the components.<p>They &quot;estimate&quot; this would save 60% on galvanic link, and on the rest, you can save another half. So that would mean you are at 20% cost plus some ground wire fault detection HW.

It&#x27;s a pity they didn&#x27;t discuss 2-way charging

To reduce public infra cost we simply need to ERASE the speculation around it. Simply only grid operators are allowed to install chargers and them are allowed to be payed only with a bank card not with apps not with dedicated cards and so on.<p>Cutting this part will makes the very same hw much cheaper.

So the problem is about preventing electric shock. Which is not a new problem at all.<p>Using galvanic isolation is 1 of the solution, but it&#x27;s expensive. How about using something like RCB to detect mismatch between output current and input current. This is well tested after all

We could use smaller external batteries that take seconds to replace more often?<p>I still think road charging would be cool and obviously the fastest if we never have to stop anyways. Couldn’t work everywhere in the US probably but I can dream.<p>Trains and buses do it already where I live.

Related, a recent video intereview with the Peterson Car museum with one of the authors, discussing the EVs he&#x27;s worked on:<p><a href="https:&#x2F;&#x2F;youtu.be&#x2F;Wv9hPIPhR9w" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;Wv9hPIPhR9w</a>

&gt; indicated that 62 percent of EV owners were so anxious about EV range that they had at times curtailed their travel plans.<p>So.. wouldn&#x27;t increasing overall range be the &quot;one thing&quot; you could do to actually improve this?

I&#x27;m not entirely sure about safty in this case. Especially in a system that has to work in a wide range of real-world conditions...

BTW everyone is fond of Norway while having zero idea how it&#x27;s actually over there.<p>Consider this movie<p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;In_Order_of_Disappearance" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;In_Order_of_Disappearance</a><p>The protagonist plows snow for a living, using 200-300 bhp diesel truck for like half a year each year, day after day. This thing just can not be replaced with an electric no matter what you try. And without it there would be zero traffic, electric or no.<p>It often looks like people here mostly do not comprehend what snowfall is.

Lightning strikes...how do EVs currently guard against it?<p>- With galvanic isolation baked directly into onboard charger, I&#x27;m thinking worst case outcome is charging infra and isolation link get fried, but that <i>really</i> expensive EV battery remains safe.<p>- Without galvanic isolation, battery goes...boom? Hopefully without you or your family waiting inside the vehicle under some false pretense of safety.

tl;dr existing galvanic isolation in fast chargers is very costly (~60% of capital cost) and delaying deployment of unsubsidized EV charging infrastructure; the authors think it can be replaced safely and cost-effectively with a second ground wire and some logic to detect&#x2F;verify grounding.