NEMA 14-50 home charging advice

[TexLou]

So, I installed the Hubbell. Yup, it makes a difference. It is only 54 degrees right now, but the plug has been pulling 5kw continuously for 3 hours now and the outlet is in a brick wall with a cover so I doubt the mild outside temp plays much of a role. I unplugged it and held the prongs in my hand and could not perceive any warmth. Probably it was warmer than ambient, but with the Leviton it would have been uncomfortably hot to hold onto.

I do have one of those IR thermometers, but it was not handy at the time and I already un-installed the old one.

 

[TexLou]

Just to clarify, 25 amps is 5kw here. This is 208 service.

 

[mr10012]

This is a typical and very incorrect assumption. The wire must be rated at the full 60A. You see, there is something called the 80% rule. On a continuous load circuit the load cannot exceed 80% of the wire rating. So then on a 55A wire the maximum is 44A, which is not acceptable. Here is a chart with options you can pick from:

Cerrowire Resources - Ampacity Charts

Ampacity is the maximum current that a conductor can carry continuously under the conditions of use without exceeding its temperature rating. Cerrowire's ampacity chart helps calculate the load requirement for a circuit.

www.cerrowire.com

BTW, I am pretty sure that anything other than the NM-B options must be in a conduit.

Update: I ended having the electrician run #6 copper wire in a conduit (2 phases plus neutral) from panel in basement to disconnect (indoors) and then to 14-50 outlet (outdoors in a metal enclosure). I changed the cheapo outlet that came with the enclosure to a high-quality Hubell (required enlarging the hole for the outlet, but works fine). I have 3-phase service to my house, so this should get 208V nominal to the outlet.

I asked the electrician to install a high-quality 50A GFCI breaker at the panel, but the work was done while I was away and what I see there is a no-name 50A breaker, so a bit concerned about that. Is there a way to tell if it is GFCI? Even though there is a disconnect before the 14-50 outlet, I could be charging for hours outside in the rain, so I would feel better with a GFCI breaker.

Another concern: With no load, voltage between neutral and either phase is 119.5V. Voltage between the phases is 205V which I guess is close enough to the 207 I would expect. However, as I draw more power the voltage drops. When I charge with my gen-1 mobile connector, at 10A, the car reports 205V. At 20A it reports 202V. At 30A it reports 200V and at 40A it reports 198V. Is that normal for a house with 600A service (3 phases x 200A each)? By the way, after an hour or so of 40A charging, the 50A circuit breaker is quite hot to the touch.

TIA for any insights...

 

[jcanoe]

I would not be concerned by the voltage drop while charging at 40A. I routinely charge at a public Level 2 charging station (30 amp) and the 3-phase voltage varies between 193V and 200V (but never higher.)

A GFCI circuit breaker will have a test/reset button just like the button on a GFCI receptacle.

I thought that the service disconnect must be located within line of sight of the equipment. (This way the service technician can have visual and control of the power being switched off to the equipment during service.)

 

[exxxviii]

Another concern: With no load, voltage between neutral and either phase is 119.5V. Voltage between the phases is 205V which I guess is close enough to the 207 I would expect. However, as I draw power the voltage drops. When I charge with my gen-1 mobile connector, at 10A, the car reports 205V. At 20A it reports 202V. At 30A it reports 200V and at 40A it reports 198V. Is that normal? By the way, after an hour or so of 40A charging, the 50A circuit breaker is quite hot to the touch.

I’m an EE and I spent some time working for power utilities years ago in my career. But I cannot remember the specifics to answer with authority. In general, I think most of this is normal…

Stated voltages are more like guidelines, like the pirate’s code. Actual voltages vary by power company and system load. Yours look fine. And the fact they decrease under load is probably fine.

But, I don’t think the breaker should be hot. That indicates an internal resistive load, and I don’t think that should happen in a breaker. (And an internal resistive load in the breaker would decrease voltage.)

 

[mr10012]

I would not be concerned by the voltage drop while charging at 40A. I routinely charge at a public Level 2 charging station (30 amp) and the 3-phase voltage varies between 193V and 200V (but never higher.)

A GFCI circuit breaker will have a test/reset button just like the button on a GFCI receptacle.

I thought that the service disconnect must be located within line of sight of the equipment. (This way the service technician can have visual and control of the power being switched off to the equipment during service.)

Thanks! The breaker doesn't have a test/reset button, at least not one visible without removing the panel cover. I think I am going to ask for replacement with a better quality GFCI one. Do this or this look good?

Regarding the disconnect, that was all that was feasible at this installation... technically it is in line of sight if the garage door is open while charging.

 

[Art84]

The only assumption I have about your 205V is that you actually have true two phases for two hots, instead of "split phase" type of connection, which is the most typical for residential usage.
"Split phase" is when a single phase transformed into two hot wires with a 180-degree phase, so the phase for each wire is the opposite, which gives a math 120V+120V=240V
In a three-phase type of connection, each phase is at 120-degree to the other, so the voltage is lower

208 Volts Is a Weird Voltage. Where Did It Come From? - Thermal Corporation

208 volts is a weird voltage. Where did it come from? View the Thermal Corporation Blog to find out! More posts can be found on our Blog Archive!

www.thermalcorporation.com

 

[jcanoe]

Regarding the disconnect, that was all that was feasible at this installation... technically it is in line of sight if the garage door is open while charging.

Got it.

The 240V/50A circuit breaker must fit your specific make,model service panel. Your electrician should be able to select the correct GFCI breaker.

 

[mr10012]

The only assumption I have about your 205V is that you actually have true two phases for two hots, instead of "split phase" type of connection, which is the most typical for residential usage.
"Split phase" is when a single phase transformed into two hot wires with a 180-degree phase, so the phase for each wire is the opposite, which gives a math 120V+120V=240V
In a three-phase type of connection, each phase is at 120-degree to the other, so the voltage is lower

208 Volts Is a Weird Voltage. Where Did It Come From? - Thermal Corporation

208 volts is a weird voltage. Where did it come from? View the Thermal Corporation Blog to find out! More posts can be found on our Blog Archive!

www.thermalcorporation.com

Correct. As I said above I have three phase 3 x 200A service, and 205V is "close enough to the 207 I would expect". My concern was that drawing 40A would bring it from 205 down to 194-198 range. To my surprise, the sentiment seems to be that I shouldn't worry about it as it probably just reflects a drop in the supply voltage to the house. I expected better from Con Edison but I will confirm by testing the incoming house voltage next time I charge before and during charging.

 

[mr10012]

The 240V/50A circuit breaker must fit your specific make,model service panel. Your electrician should be able to select the correct GFCI breaker.

Of course. The two breakers I linked to fit the rails in my panel and are made by brands I've heard about (Siemens and Square D).

 

[jcanoe]

Of course. The two breakers I linked to fit the rails in my panel and are made by brands I've heard about (Siemens and Square D).

Your electrician can choose the breaker based on their experience with the quality, including reliability of the product. The intangibles include how easy is it to locally obtain a replacement GFCI if required. This is more of a concern with the ongoing supply chain issues especially for anything manufactured outside of the US.

Example: In 2021 I replaced an ancient furnace and AC unit. The new HVAC system, Lennox, was recommended by the company I have been using to service the old equipment. Within the first year the contactor (relay) for the AC compressor failed. The service technician discovered this while checking the AC system at the beginning of the AC season. The technician had the part (probably cost ~$50) on their truck. Changing the bad contactor took ~30 minutes. If this failed part had been manufactured, sold under a different label I might have had to wait an unknown number of days for a replacement to be available.

 

[Art84]

Correct. As I said above I have three phase 3 x 200A service, and 205V is "close enough to the 207 I would expect". My concern was that drawing 40A would bring it from 205 down to 194-198 range. To my surprise, the sentiment seems to be that I shouldn't worry about it as it probably just reflects a drop in the supply voltage to the house. I expected better from Con Edison but I will confirm by testing the incoming house voltage next time I charge before and during charging.

Yes, voltage drop is totally fine.

 

[mr10012]

Yes, voltage drop is totally fine.

I think you are right. FWIW, the voltage drop is in the charging circuit (circuit breaker, wire and gen-1 mobile connector). With voltage at panel of 203V and drawing 40A, the car reports 196V, for a drop of 7V, or a resistive load of 0.175 Ohm. That is the resistance of about 500 feet of 6 awg wire, and my run is less than 50 feet (including the mobile connector).

Of course it is more complicated as the breaker, the disconnect and the mobile connector electronics add some resistance, and this is AC current--I show phase shift/power factor of 0.9L (inductive) for one of the phases and 0.87C (capacitive) for the other but I have long forgotten how to calculate the implications for apparent/effective voltage. At my beach house (with a hardwired Tesla wall charger) the car reports 240V, but I seem to recall seeing 122-123V at an outlet, which would mean 244-246V at the panel, so there could be a similar drop...

Anyway, after having the electrician replace the breaker with a GCFI one, I am declaring victory and moving on.

 

[Vines]

I think you are right. FWIW, the voltage drop is in the charging circuit (circuit breaker, wire and gen-1 mobile connector). With voltage at panel of 203V and drawing 40A, the car reports 196V, for a drop of 7V, or a resistive load of 0.175 Ohm. That is the resistance of about 500 feet of 6 awg wire, and my run is less than 50 feet (including the mobile connector).

Of course it is more complicated as the breaker, the disconnect and the mobile connector electronics add some resistance, and this is AC current--I show phase shift/power factor of 0.9L (inductive) for one of the phases and 0.87C (capacitive) for the other but I have long forgotten how to calculate the implications for apparent/effective voltage. At my beach house (with a hardwired Tesla wall charger) the car reports 240V, but I seem to recall seeing 122-123V at an outlet, which would mean 244-246V at the panel, so there could be a similar drop...

Anyway, after having the electrician replace the breaker with a GCFI one, I am declaring victory and moving on.

It will be interesting to hear how bad if any is your nuisance tripping with the GFCI breaker. Let us know.

 

[TexLou]

Correct. As I said above I have three phase 3 x 200A service, and 205V is "close enough to the 207 I would expect". My concern was that drawing 40A would bring it from 205 down to 194-198 range. To my surprise, the sentiment seems to be that I shouldn't worry about it as it probably just reflects a drop in the supply voltage to the house. I expected better from Con Edison but I will confirm by testing the incoming house voltage next time I charge before and during charging.

Your voltage drop could be concerning. It depends on if it falls into the expected range. There are tables for any given gauge of wire (need to spec Al vs Cu) at any given amp draw, per foot of run. If the voltage drop exceeds expected, then something is consuming power on the run and whatever is consuming that power could be getting very hot. ie, a bad connection. You mentioned it could be the utility, but I doubt it. While continuously monitoring your voltage at the panel, turn the Tesla charging on and watch for changes. My panel voltage at both my office (3 phase 208) and at home (split phase 240) does not budge when I charge at 7-8kw. A 10 volt drop like that could damage some motors, but I'm pretty sure the Tesla doesn't care. A 10v drop is kind of a lot and to me would mean you either have a very long run, or you have an issue. At home I only see maybe a 1v drop at 8kw. At the office I usually charge at 5kw and only see about 3-4v drop. The office has about a 60 or 70 foot run to the outlet, at home it is right next to the panel.

 

[mr10012]

Your voltage drop could be concerning. It depends on if it falls into the expected range. There are tables for any given gauge of wire (need to spec Al vs Cu) at any given amp draw, per foot of run. If the voltage drop exceeds expected, then something is consuming power on the run and whatever is consuming that power could be getting very hot. ie, a bad connection. You mentioned it could be the utility, but I doubt it. While continuously monitoring your voltage at the panel, turn the Tesla charging on and watch for changes. My panel voltage at both my office (3 phase 208) and at home (split phase 240) does not budge when I charge at 7-8kw. A 10 volt drop like that could damage some motors, but I'm pretty sure the Tesla doesn't care. A 10v drop is kind of a lot and to me would mean you either have a very long run, or you have an issue. At home I only see maybe a 1v drop at 8kw. At the office I usually charge at 5kw and only see about 3-4v drop. The office has about a 60 or 70 foot run to the outlet, at home it is right next to the panel.

Thanks, this is helpful. Late Sunday night my meter monitor (coils on the phase wires at point of entry before the panels) was reporting 119.9V and 120V on the two phases providing power to the Tesla, so the panel voltage should be close to 208V. The car was reporting 199-200V at 40A, charging at 8KW. The total draw on these two phases was about 43A, as there were other house loads on these phases in addition to the car. The run from panel to disconnect and then Hubell outlet is about 20 feet, with #6 copper wire in conduit. The conduit felt a bit warm but not hot. The breaker was hot to the touch, but not too hot to keep the finger on for minutes. Disconnect, outlet and UMC were all at ambient temperature. Typical power factor (cos theta) while charging was 0.87 inductive on one phase and 0.91 capacitive on the other. Right after charging completed, power factor went to 1.0 and 0.86 capacitive, drawing 2.5 and 3.5A respectively at 120V. Not sure if power factor is relevant here, as I think is primarily relevant for motor loads.

So under the above conditions, I had about 8V drop, which seems high but not unreasonable. The one thing I suspect can be improved is the circuit breaker, which is a no-name generic one. I have asked the electrician to replace with a brand-name GFCI one as this is required by code. TIA for any comments/suggestions.