EV Outlet Wiring
During construction had garage pre-wired for EV. 6 AWG Romex and 60 amp breaker. Bought a new car that comes with a dual 120v & 240v charging cable. (Level 2) 240v cable has a NEMA 14-50 plug. I bought a NEMA 14-50 receptacle, but now I’m wondering if breaker needs to be reduced to 50 amps? A NEMA 14-60 receptacle has a different plug arrangement.
If I have to change the breaker, and shut all the power, that’s further complicated by having a Tesla Powerwall. Thanks.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
1) the breaker protects the conductors in the cabling, not the receptacle or the device on the other end of it. It is quite common to have 15A receptacles on a 20A breaker and constitutes no problem. Furthermore, many of the devices plugged into those circuits draw significantly less than 1A!
2) what is the power level of your charger? Circuits (both cabling and breaker) need to be oversized by 20% vs the load. Typically a 60A circuit is used for a 48A (11.5kW) charger so I assume that’s what you have. If it’s lower (40A , 9.6kW) you can use a 50A circuit.
3) What is the length of the cabling run from the panel to the receptacle? Looking at a couple cable sizing tables on google a 60A circuit can only be run with 6AWG if it’s relatively short. One table told me <10’ and another table told me <30’. Longer than that and you need to upsize to 4AWG.
I work in the EV game but I’m not an electrician; hopefully one can chime in for #3 above or you can find one to give you the proper cable sizing for a 60A circuit.
EDIT: See more informed comments below and ignore #1 above.
Breaker needs to match the plug. It is not just about protecting the wiring but also a device that plugs into it. Since a device with a NEMA 14-50 plug would be designed for 50A, you can't connect it to a 60A power feed, if something goes wrong and the wiring on the device can melt before the breaker trips.
Swap out the breaker and you should be good to go, it is always better to have bigger wiring when you have larger continuous loads like charging an EV.
PS. If you really can't swap out that breaker, there are RV outlet boxes with breakers you can get. You can use this instead of a standard receptacle in your garage. Put a 50A breaker and the Nema 14-50 receptacle into the outlet box and you are good to go.
You should swap the breaker for a 50A breaker. You'll need a double pole 50A breaker that matches your panel (same manufacturer). It's a quick and easy swap, and the breakers aren't very expensive.
While it is true that "the breaker protects the wire", there is a little more to it than that. The easiest example is that you are permitted to use 15A receptacles on a 20A, 120V circuit (although I don't recommend doing that), but you are NOT permitted to use 15A or 20A receptacles on a 30A, 120V circuit. The reason is that at 30A, there is a risk that the wire in the cords of connected devices may pose a fire risk that won't cause the breaker to open. The easiest example for that would be so-called "tinsel cord", which is the light gauge wire used for things like christmas lights. Tis' hopefully not the season for house fires, after all :-)
You have the same issue here with the receptacle, since the connected cabling also needs to be protected by the breaker in this case.
BTW, regarding #3 for Frasca's post: the code RECOMMENDS, but does not REQUIRE, that the total voltage drop not exceed 5% for feeders and branch circuits, with 3% being the usual recommendation for branch circuits such as the wiring to an EV receptacle. For a 240 volt circuit, 3% is about 7.2 volts, so if you're using 6 gauge copper cable in NM cable or PVC conduit, with a 50 amp circuit, you're good for about 162 feet. If you run a 60A circuit, then you're good for about 135 feet. If you're running a 120/240v circuit (which would include a neutral), and you want to be conservative for unbalanced loads, then your upper limit for that 60A circuit with #6 copper wire is about 68 feet.
Tables showing really short runs for that cable are either trying to sell you wire, or they are thinking about motor loads with big starting currents, where you may need to be more conservative. Personally, I'd probably think about upsizing to #4 wire around 100 feet here, to be a little more conservative. If I was feeding something like a subpanel in a garage workshop, then I'd use the 2% code recommendation for a "feeder", and assume unbalanced loads (which are 120v loads like a circular saw running on one "leg" of that 240v circuit), and upsize to heavier wire with shorter runs.
Bill
I agree it needs a 50 amp breaker based on my original post. But, I'm still a bit confused if that's OK as the AWG 6 length appears to be about 52 ft. Can't say exactly as some is hidden. On another web site, it says a 50 amp wire loses 20% of voltage every 100 ft. Presumably, I'm losing 10%. The cable is too inaccessible to change. I have to live with what I have.
The 2022 Lincoln Corsair is a PHEV with a 14.4 kWh battery so should be relatively easy charging, even on 120v outlet. See two citations from owners manual, below. Is this saying I should have a 40 amp breaker? Also seems to be saying it only charges at 32 amps max. I also have a 30 amp and a 20 amp garage outlet but that's an even longer run, probably 70 ft and inconvenient from outlet to car. The 20 amp breaker is not marked in panel.
I also read (can't remember where) if using 120v charger, should be dedicated circuit. All my 15 amp garage outlets are on one circuit. The only other regular use on circuit is garage door openers and one or two battery maintainers on other cars. Is that OK? From the owners manual:
WARNING: To reduce the risk of fire, connect only to a circuit provided with 40 amps maximum branch circuit over-current protection installed in accordance with NEC and local electrical code.
Your vehicle comes with a mobile charger that has a low power connector to use with a 120V plug NEMA 5-15 and a high power connector to use with a 240V plug NEMA 14-50. Other connectors, such as NEMA 6-20, are available as service parts at authorized dealers.
Specification Type - Specification Value
Charger Model - SAE
Voltage - 120-240 VAC
Maximum Current - 32 amps max continuous
Frequency - 60Hz
Power draw when idle < 2 Watts
Power draw when charging < 4 Watts
Cable length - Approximately 20 ft (6.1 m)
Weight 8.6 lb (3.9 kg)
A "50" amp charger will have an actual load of 42 amps.
Over a 52' run I calculate less than 1% voltage drop. This install is perfectly OK after you swap the breaker. If the receptacle is outdoors or in a garage that breaker likely needs to be a GFCI type.
You need to derate 20% for a continuous load (like EV charging). This is why you need a 40 amp breaker/recpleptacle for a 32 amp charger.
Longer periods (as in hours) of charging will expose issues in wiring or at panels. I had to address an issue with EV charging (16A @ 240V) where one leg of an older panel feeding a garage was tripping a breaker…but after 30 minutes of charging. It was a poor connection in the main panel feeding the garage sub panel.
Most EVs with a 120V OEM charger option max at 12 amps as this is the legal max for a 15 amp circuit continuous load.
240V is the more efficient option for charging as EVs have a surprising amount of power overhead when charging…in the 300 watt range. So charging less time = fewer hours overhead = higher “efficiency” for a given charge level.
OK, lots of not quite accurate info here that needs to be cleared up.
==========
For post #4:
Wire doesn't "lose" any fixed voltage just because of length. What matters is the CURRENT, in AMPERES passing through that wire. It's simple ohms law, so many amps multiplied by so many ohms of resistance for the wire gives so many volts of volt drop. The code book gives some guidelines for acceptable levels of volt drop as a percentage, and for branch circuits (which includes feeds to EV chargers), it's recommended you not exceed 3% volt drop. With a 240v circuit, that allows 7.2 volts of volt drop. With 6 gauge wire and a 40 amp breaker (which makes it a "40 amp circuit"), you're within code allowable limits up to 162 feet. If your run is only 50-60 feet, you're well within limits and you'll be fine. Most of those "volt drop" tables you see on the internet are for extension cords, and they tend to be *very* conservative, and they don't apply here.
A 14.4kWh battery would take at least 10 hours to fully charge on a typical 120v charger, so not very fun. A Level 2 charge would be much better, which would need the 240v circuit. A typical level 2 charger will use a 40A circuit, and allow up to 32A of charging, due to what is known as "the 80% rule" (which I'll explain later). This charger could charge that same battery in about 2 hours. Note that actual charge times will be different, since I just did a "battery capacity divided by power supply" calculation, which is very simplistic and doesn't allow for fancy things battery chargers do, or for losses in the system.
If your charger says use a 40A breaker, then wire it like that. If it can be plugged into a 50A plug though, you should be fine using a 50 amp breaker for the plug, which is the standard configuration. Either way, the charge will communicate with the car and tell it how much power the charge can deliver, and it will NOT exceed whatever the charger is capable of delivering, regardless of circuit capacity.
Ideally the 120v charger would be on a dedicated circuit, but they are designed to work on a standard 15 amp residential circuit -- although they will use ALL of the available capacity of such a circuit.
==========
For post #5:
The "charger" you install on the wall is really just a contactor (big relay) and GFCI. 120v receptacles installed in garages are required by code to be GFCI protected, but the larger ones aren't (yet -- that may be changing). There is some argument in the industry with code writers wanting to require GFCI breakers to power EV chargers, and the EV charger manfuctures saying that's not necassary, because the EV charger is ALREADY a GFCI. I'm not sure if the code has actually changed yet, but the last time I checked, you could supply an EV charger from a "regular" (non-GFCI) breaker.
==========
For post #6:
You derate by 20%, not 12%, for "continuous" loads. This is known in the trades as "the 80% rule", and it means any circuit running a "continuous" load, which the coad defines as any load of 3 hours or more, can only be loaded to 80% of capacity. This is done for thermal limit reasons, and applies to even very large circuits of hundreds of amps in most cases (the exceptions to this are outside the scope of this thread). You are correct that this is why EV chargers using 40A circuits only actually deliver 32A to the EV.
The efficiency has to do with losses in the system, which are depenent on current (amperes), not voltage. This is the same reason why power lines run very high voltages. I'm not sure what you mean by "power overhead", but more deliverable energy will always result in faster charging times up to the capacity limit of the actual charger.
BTW, what I mean by "actual charger" is that the electronics that are involved in actually charging the EV battery are actually in the EV. The "charger" you mount in your garage doesn't actually "charge" anything -- it only does three things: communicate with the vehicle, to tell the EV how much power is available, close the circuit by closing a large contactor (relay) when the EV says it's ready to charge, and monitor the entire thing for ground faults the whole time to shut things down if dangerous conditions exist (which is the GFCI functionality that is integrated with the EV chargers).
Bill
Bill, I am extremely grateful for you sharing your knowledge with us. Can you clarify one thing. You say either a 40 or 50 amp breaker would be OK. Which would be preferable? Sounds like the 40.
Would the 50 be beneficial for a future EV with higher charger capacity? I see the Ford Lightening can do 80 amps and think Tesla does 60. Is my 50-60 ft run OK for 50 amp breaker? Thanks.
Your 50-60 foot run of #6 copper cable will be fine for a 50 amp circuit. Since the EV charger is a 240v load, you're well within allowable limits at 60 feet here. If your charger only wants a 40A supply, then you'd be better off with the 40A breaker, which will work fine with the wire. If you will be wiring the charger using a 50A receptacle, I would put a label on the receptacle indicating that it is only supplied from a 40A circuit (and I'd also label the receptacle with the panel and circuit number supplying it).
If you use 6 gauge NMB cable, which is the usual nonmetalic sheathed "Romex" style cable, then you're limited to 50A. If you are running 6 gauge conductors in conduit, you can run 60A on the 6 gauge cable.
If you have an "80A" charger, you want to check if that's the supply to the charger, or the current the charger will deliver to the EV. If the supply to the charger is an 80A circuit, you can use 4 gauge copper cable (although I'd probably up this to 3 gauge to help things to run a little cooler). If the charger can actually supply 80A, then the 80% rule comes into play again, and you need a 100 amp circuit which can use 3 gauge wire, although I may up that to 2 gauge depending on the situation. You can usually use SE cable for those circuits, although I prefer to run them in conduit.
Note that if running individual conductors in conduit for those larger circuits, you don't need to run the two hots and the ground all with the same gauge. An 80A or a 100A circuit are both OK using only an 8 gauge wire for a ground. You can save some money using the smaller ground wire this way, and you still meet code.
Bill
Deleted
2020 NEC says if this is a receptacle and in a garage, it needs to be on a GFCI breaker(not every state has adopted 2020 code). If the unit is hardwired this is not required.
Using a 40 or 50 amp breaker depends on the charge rate the user will select(a lot of the units these days have selectable charge rates)
If you want to charge at 32Amps then use a 40Amp breaker. If you want to charge at 40amps then choose a 50amp breaker.
And we needn't even get into the question of where the receptacle is: article 625.60(D) in the 2020 code explicitly says that at receptacle for EV charging must have GFCI protection.
Thanks Bill for pointing out my mistake there...20% is indeed the correct de-rate factor for continuous load.
"Ideally the 120v charger would be on a dedicated circuit, but they are designed to work on a standard 15 amp residential circuit -- although they will use ALL of the available capacity of such a circuit."
I'd add one comment on your statement above. 12 amps is the typical limit of OEM chargers as they are not short term loads, but a 15 amp circuit is designed to deliver 15 amps. There are a few after market EVSE units that can be set to less than 12 amps, and a few I've seen a few (amazon types) that actually run at 16 amps. I'd avoid these unless you have a 20 amp 120 v circuit installed. Very few receptacles out there in the wild are 20 amp capable.
Also:
"The efficiency has to do with losses in the system, which are depenent on current (amperes), not voltage. This is the same reason why power lines run very high voltages. I'm not sure what you mean by "power overhead", but more deliverable energy will always result in faster charging times up to the capacity limit of the actual charger.:
Not really. I'm speaking specifically about charging at 120V vs 240 V (AC). The majority difference in efficiency has more to do with charge times correlated with overhead. If a typical AC-DC converter in an EV uses 300 watts overhead, then halving your charge time reduces the loss due to overhead. The AC to DC inverter has pretty much the same overhead at 120V or 240V.
>"but a 15 amp circuit is designed to deliver 15 amps. There are a few after market EVSE units that can be set to less than 12 amps, and a few I've seen a few (amazon types) that actually run at 16 amps. "
A 15 amp circuit is designed to deliver 15 amps, but must be derated when operating continuous loads, and that means only 80% of the circuit capacity -- 12 amps -- is available for EV chargers on 15 amp circuits. It's true that NON-continuous loads could use the remaining 3 amps, as long as those loads kept to under 3 hours of uninterrupted operation.
A 16 amp charger would be able to run on a 20 amp circuit, since 80% of 20 amps is 16 amps. Note that more recent editions of the electric code have required garages to be supplied with at least one 20 amp circuit, which would be capable of supporting a single 16A EV charger.
Now I understand what you mean by "overhead", I just don't usually see it referred to like that. You're correct though that watt losses are primarily determined by current, in amperes, and not voltage, so the loss in watts can be held reasonably constant in some designs, adding to the efficiency of a higher voltage charger since more power would be delivered to the EV for the same amount of losses in that case.
Bill
Bill, the question of 120V vs 240V charging efficiency has been bounced around a lot and yes, there is quite a bit of misinformation about it. With a look at actual wattage used by various systems during charging (LeafSpy tells you this via the ODBII connection if you're playing with a LEAF), compared to input from the EVSE, a few observant folks have figured out that that that the difference in charging "efficiency" was largely due to the difference running the car systems curing charging. 300 watts is pretty considerable (surprisingly high really) but I would make an educated guess in saying much of that is the loss in running the AC-DC converter during charging. Charge for six hours and you've used 1.8 kWh in charging overhead. Charge for three hours and that drops to .9 kWh. If you compare the actual input to the system in watts vs the gain in kWh capacity to the EV pack, 120V looks less efficient...but the overhead is where the loss is.
Since sorting all that out, we stopped charging our EV at 120V just based on the knowledge that we use less overall power to charge at 240V.
Another thing the manufacturers don't tell you (and this likely varies amoung EVs) but there is limited current available during charging (or off charge) at 120V to precondition the car at low temps, and typically they don't dip in to pack voltage to precondition when plugged in. Plugging a cold EV outside in cold weather (as in -10 C or below) at 120V will likely result in the cabin not being at your set temps for precondition. Plug in to 240 V and that issue goes away :-)
800 volt packs used in the newer Kia/Hyundai, Porsche etc. get their faster charging performance over the current crop of 400 V packs by lowered amperage during charging, and therefore heat...which is a problem for every EV during DC fast charge.
I decided to go with the 40 amp breaker per the owners manual. Home Depot had neither the 40 or 50. Lowes had the 50 and no stores in the metro had the 40. Guess 50 amps it is. Thanks.
See attached picture. Safety question. Can somebody confirm for the circuit breaker exchange, all I need to do is switch off the 200 amp breaker in the Tesla Powerwall Gateway and I can leave my main breaker outside, the PV breakers outside and the Powerwall breaker & power switch on.
Boxes left to right are the Gateway, sub-panel off main panel and main panel. The blue arrow is power from grid & PV system. Orange arrow is power from Powerwall. Green arrow is Gateway to main panel. Yellow is main to sub panel. Thanks.
I would not feel comfortable describing how to safely shut down your panel from the pictures and info you provided. You generally need to shut off ALL potential sources on something like this though, which means shut off the solar system, shut off the powerwall, and shut off the main from the utility, preferably in that order. If you have a generator and transfer switch, make sure to shut off and/or lock that out too.
If you have *ANY* doubt about your ability to safely work on this system, stop, then call an electrician and have them do the work for you. I like to tell customers that "electricity is like an angry little genie, just itching to get out of the bottle, and he doesn't care what, or who, he hurts on his way out". I also don't recommend doing this work alone, be sure someone else is around to call for help in case anything goes wrong.
Bill
As Bill says, be 100% sure it's all off. Then, in addition, measure to make sure, using a good quality meter. Test the meter on with the power on, turn the power off, test again to make sure the power is gone.
And I'll repeat that if you have doubts, hire an electrician.
What state are you in? If your state is on the 2020 electrical code, you are required to have a GFCI breaker if you use a receptacle for EV charging.
Also note that low quality receptacles, such as the Levitons readily available at Home Depot, etc., have been known to overheat in EV applications. They might be OK if torqued properly with a torque screwdriver, but high-quality expensive ones like Hubbell have much better wire terminations that won't have that problem.
By the time you are done buying the GFCI breaker and the high-quality receptacle, you might consider buying a second charger to hard-wire, at which point you don't need either the receptacle or the GFCI, and you can keep your plug-in charger in your car for possible mobile use.
Missouri
Missouri is an outlier: No statewide adoption of electrical code. Some local juridictions are up to 2017 code, but I don't see any on 2020. So likely you get to do what you want, for better or worse. Not a bad idea to follow the latest code anyway.
https://www.iaei.org/page/missouri-electrical-ceus
Note that Leviton doesn't mean low quality. Leviton makes a sort of "builder grade" device, which is the cheap kind that is usually sold loose (no box). They also make "spec grade", and home depot has those too -- they are the more expensive ones that are individually boxed. Leviton also makes a better grade of spec grade receptacle that you have to buy at electrical supply houses, and it has a more robust wire clamp under the screw.
I agree Hubbell also makes high quality devices, but you'll probably have to go to a supply house to buy their stuff.
I would NOT use no-brand stuff off of online vendors for something like an EV charging receptacle though. Long-duration heavy loads, which includes EV chargers, is a sort of worst case scenario for poor quality electrical devices due to the heat that the heavy load can generate in the contact area of those connectors.
Bill
Yup, I didn't mean to denigrate Leviton in general. For regular 5-15 receptacles, going with the higher grade is more important than the brand name (assuming it's a legit brand), and the good Levitons are good. I've been using them in my own house, in fact. Thanks for that clarification.
But Leviton calls their cheap 14-50R, model 279-S00, "industrial grade". That's the one sold at Home Depot for $11.42, that has had problems in EV-charging duty, and they don't make anything higher grade, unless you count the 279-PM which is made for mounting in a panel, not a junction box.
"spec grade" is a defined term, just like "hospital grade". "Industrial grade" doesn't really have a defined meaning. I have found that the large (30+ amp) straight blade receptacles generally tend to not be what I would think of "high quality" regardless of manufacturer. What I typically recommend is to either hardwire the charger (which is always best), or put in a 4" square of 4-11 box that can fit one of the large receptacles, but put a blank cover with a knockout on instead of a receptacle. You can then hardware your charger to the knockout in that blank cover plate using a short piece of sealtite (flexible water resistant conduit) and a 90* fitting. Wiring this way allows for the charger to be swapped out relatively easily, but avoids the weak spot of a plug and receptacle.
Bill
Yes, agreed that "industrial" doesn't necessarily mean anything and you might note that I already recommended hardwiring in my comment above.
I assume that by "spec grade" you mean Federal Specification Grade, i.e., US Federal Specification W-C-596, “General Specification for Electrical Power Connectors," which can be certified by UL with an F-UL-S mark. That's a good idea, to look for that. The Hubbell that is a solid quality device (part no. 9450A) appears to carry that, although it's kind of sloppy--they don't put that on the spec sheet but only at the top of the catalog page, making it seem like it applies, but not as definitely as you might hope.
In any case, hardwiring is the best choice.
I did get the "bad" Leviton 279-S00 receptacle. I'll look for a replacement.
Not sure I understand the circuit breaker brand quality. The breaker I got at Lowes was loose & says Eaton with a made in Dominican Republic sticker. There were no boxed units that I saw. Isn't an "Eaton" made by Eaton and therefore good quality?
Are there third parties, say Hubbell, that makes Eaton fitting breakers? I've also read that I can use SquareD, Siemens, GE, Thomas & Betts and Cutler-Hammer.
Saw another chart online that says while you can use GE, Thomas & Betts, Siemens, Murry and Crouse-Hinds, they are not exactly the same as Eaton. Exactly the same as Eaton are SquareD, Westinghouse & Cutler-Hammer.
Your panel is Eaton? In that case I would use Eaton. There are special cases where there's another manufacturer's breaker that has actually been listed for use in a given panel but figuring that out is a whole can of worms and it's better just to use the same brand.
Deleted
Disregard my last. I called Eaton. All the breakers are the same quality. Looking back on thread, maybe I got confused. Folks were talking about different quality receptacles, not breakers. Sorry.
You have to use the breaker listed for use in your panel, otherwise you void the UL listing for the panel. It doesn't matter if the breaker physically fits, it has to be LISTED too, and very few breakers are listed for use in more than one manufacturer's panel.
Eaton is what used to be Cutler Hammer, as they acquired them years ago. They will typical have a brown handle. Eaton/Cutler Hammer, Siemens, Square D, and GE are all regarded as similar quality. Murray is owned by Siemens, but usually thought of as their "value line", and not as good as their "Siemens" branded stuff.
FPE panels (orange ends on the handles) are considered dangerous and it's usually recommended to replace them. That company hasn't existed for decades now, and the main reason they went out of buisness is that circuit breakers didn't trip reliably and there were liability issues.
Bill
I found an Eaton 1258-SP receptacle. But, how to know its quality?
https://www.eaton.com/us/en-us/skuPage.1258-SP.html
Menards has Legrand
https://www.menards.com/main/electrical/light-switches-dimmers-outlets/electrical-outlets/legrand-reg-pass-seymour-reg-50-amp-125-250-volt-4-wire-black-flush-range-outlet/3894/p-1444451148520-c-9526.htm?tid=-1668908726491717805&ipos=1
Lowes has Utilitech
https://www.lowes.com/pd/Utilitech-Black-50-Amp-Round-Range-Industrial-Range/3775483
Eaton/Cutler Hammer should be fine. Legrand is pretty good too. Utilitech is Lowe's house brand, and I don't know how they are for this type of device.
Hubbell makes reliable devices, as does Eaton/Cutler Hammer. Legrand has some variants just like Leviton does, but I'm not sure how much difference there really is between their box store stuff and their supply house stuff.
If you want a quality receptacle, you have a better chance of getting something good at an electrical supply house than you do at a box store. That doesn't mean all box store stuff is bad, it's just that the supply houses lose customers when things they sell break and electricians have to go back to fix something at no charge (the dreaded "call back" as it's known in the trades). Commercial electrical contractors tend to try to avoid cheap stuff and the supply houses know this, and usually stock better quality devices to keep their customers happy.
Bill
Based on this:
https://teslamotorsclub.com/tmc/threads/master-thread-definitive-14-50-nema-outlet-guide.140694/
I would be inclined to order a Bryant 14-50R from Zoro. But then this report makes it sound like the Bryant can have a ground issue that doesn't affect the apparently identical and much more expensive Hubbell.
https://www.reddit.com/r/evcharging/comments/z8702t/leviton_works_bryant_1450_doesnt_update_defective/
I installed the Eaton 1254, it's sibling without the neutral, about a week ago. (Not for an EVSE; that's getting hard wired.). I tightened the terminals with a torque screwdriver. Every day since, I have been able to further tighten the terminals. This hasn't happened to me before. When I check torque the next day normally no adjustment is required.
(This receptacle won't see much use, so I wasn't concerned about buying a better quality device.)
I would never normally recommend this check, but based on my experience, when charging an EV on a new circuit, I would 100% do it.
1. Charge your EV for 30-45 minutes at the highest level possible given your setup.
2. Check the receptacle, panel, and if in a garage with it's own panel, the feed breaker from the main panel for heat. Use a FLIR or just a point and shoot infrared temp gun.
If there are any points of higher resistance, you'll see them. I had to pull a main feed breaker and redo connections to the sub panel to resolve a bizarre issue with one leg overheating during longer charges and popping the main feed to my shop. There were zero issues over 20 years until I started charging an EV. In a way it was a good thing the EV charging load on that circuit exposed a potential fire hazard that I could correct.
This is why I insist on a hardwired install if possible(not all chargers allow for it but most do).
And at this point I would recommend the OP hires a reputable licensed electrician. DIY is great but judging by some of the questions it would be irresponsible to lead the how-to much further.
Here's a final update. I tried to hardwire over the outlet which would require drilling a hole on the back of the charger but was dissuaded that hole would be on the wrong side of the stud and not sure the electrician left enough wire. Seems best to avoid splicing 6 AWG if possible.
Since the car has a max charge of 32 amps/40 amp breaker and actually automatically charges at only 240v 15 amps, and I could find a Bryant (by Hubbell) 14-50 receptacle at Grainger for $45, I went that route.
One last obstacle. Because the Hubbell wires come in from the bottom (with ground pin up), there wasn't enough room to fit the neutral (white) at the very bottom. Considered leaving it disconnected as not needed by EVSE charger but want outlet to be fully functional in case ever used for something else.
I found the pictured extension box at Home Depot, which made for a much easier fit and doesn't look bad. If I later get a BEV instead of a PHEV, I can worry about hard-wiring at that point to get 48 amps instead of 32 amps max. Thanks for all the help along the way.
One more question. The Bryant/Hubbell instructions say torque wires to 75 in-lbs. My small wrench maxes out at 65, but was too hard to turn any tighter anyway. Leviton 14-50 receptacle goes for 25 in-lbs. Does 75 seem right?
Looks fine to me. You do need to connect the neutral though -- you're not supposed to not wire all the connections on a receptacle. Not wiring the receptacle's neutral would technically mean you miswired it.
It's very unsual for anyone to actually torque those screws down using a torque wrench. Nearly everyone just uses what we jokingly call the "german method" -- we tighten the lugs down "gutentight" (good'n tight :-). 75 inch pounds is probably fine, and if that's what it says to use and you want to measure the torque, then I'd go with what the receptacle's manufacturer says to use as the proper torque value.
Bill