Nominal voltages in different installations

Nominal voltages in different installations

110V and 220V are outdated nominal voltages. Many Model S owners may be seeing the measured voltage at their home (as measured by the car) for the first time. It grates on me when I continue to see references to 110V or 220V. Maybe I’m the Brian H of the voltage world.

If you actually measure 110V or 220V you have a system with a problem. Neither of those voltages should ever occur in the US. US residential home voltage is nominal 120V +/- 5%. The national average delivered is about 118V. The higher voltage is 240V if you’re in a two-wire feed system or 208V if you’re in a three-phase system also +/-5%. Most higher-voltage devices will be labeled something like 208-240V because they need to work on both voltage systems. For an electric motor that is striving to maintain a fixed RPM, if the voltage is the lower one, it just tries to pull more amps to get the same power and generally succeeds, up to a point. The Tesla charger tries to get the nominal amperage, let’s say 40A for a standard single charger on a NEMA 14-50. If this is connected to a 240V system it will yield a nominal 9.6 kW. If it’s on a 208 system, it will yield a nominal 8.3 kW, 14% less power and 14% less miles per hour of charge. It will not act like a motor and will not try to pull the 46A necessary to achieve the same 9.6 kW while on a 208V system. In reality with voltage drops that WILL occur from the main panel to the car, voltages may be down and additional few percent from the nominal when run at 40A - 80% of rated amperage.

Anecdotal stories:
In my own case, my garage has a subpanel fed from a 50A breaker in the house. I installed a NEMA 14-50 fed from the subpanel on a 50A breaker as specified. Because of either other trivial loads in the garage panel that I think are turned off or a weak, 30 year-old feed breaker in the house, if I pull 40A to the car it will trip the breaker in the house that feeds the garage in about an hour. I have no problem with the car set to 35 amps and I always get a charge overnight. My measured voltage with no load in the car is 241V. At 40A charge it is 233V. I think that’s a fairly normal drop considering it’s from a sub-panel and running through a UMC cabled that gets warmer than it really should (see the thermal photos posted on another string; they all do that). Someday maybe I’ll measure the voltage at all the intermediate points but for now, 35A works fine.

Second story:
The power company substations have to deal with loads that change during the day and the resulting voltage drop from inherent resistance in their distribution system. When loads are higher, their transformers have “tap changers” that adjust the voltage up or down to compensate for loses in their distribution system. Sometimes they don’t work. I had a case a few years ago at the university where I was responsible for computer power, about 30% of the office UPSs were reporting being online to lower the voltage. The supplied voltage was 128-129 volts. After several emails to the campus power folks, the reply came back that they had discovered that one of the substation transformers had been incorrectly set in “Manual” for voltage control. So much for nominal voltage provision.

If you’re having issues, you need to measure the voltage at no-load and at full-load. If it’s out of spec (+/- 5% for the respective system voltage) you need to contact an electrician or your power company.

sule | 6 juillet 2014

8V drop at 40A means you have a 320W loss somewhere, or 0.2Ω resistance somewhere. If spread along a length of the cable not an issue, but 320W at a single point of contact could produce significant heat. Do check your connection points - this is where of most of such resistance occurs.

My HPWC (connected to main panel, via some 60 feet of cable) has 240-241V at no load, drops at most to 239V (have never seen less) at full 80A.

zwede | 7 juillet 2014

I have 250V at no load, about 246V @40A. My outlet is very close to the breaker panel. The "110V" outlets measure 125V.

johncrab | 7 juillet 2014

Yep, you are bang-on here. The numbers 110 and 220 are theoretical and just talking points. I put a recorder on my house at various positions and it's interesting to look at a full day of data. It's up and down like the Assyrian Empire. I typically see 122-124 on either leg of my single phase panel but I have seen it as low as 118 and as high as 126. That's actually pretty tight.

A home's panel voltage depends upon what is feeding it nearby. Delta or Wye will make a difference. Wye is rare out West but common in the Northeast.

277 or "wild phase" is typically used for commercial lighting and that would be great for charging my Tesla but I'm on a 220 nominal single phase panel. 480 cannot be brought into any residential dwelling in any jurisdiction of which I am aware.

One thing nobody talks about is Power Factor. I'm really curious as to what kind of PF a Model S has and since Tesla does not reveal any techs about their chargers, I can't even speculate but I'm reasonably sure they took this into account in the design. Thinking of PF as allegorical to ignition timing is a simple way to understand it. It can be unity, lagging or leading. Resistive loads always operate at unity PF and inductive loads generally lead or lag. Then we get into the netherworld of VARs and the relationship of apparent power to actual power.

There is also the concept of balance between or among legs and phases. Europe uses 220 (nom) for everything because it creates an inherent leg to leg balance. In North America, we use the neutral to divide the voltage so we have to add some circuits to one leg and other circuits to the other to keep in balance. Large loads like ovens and stoves go leg to leg to avoid dropping a big load onto one leg and creating an imbalance at the transformer with the resulting loss to VARs.

In other words, it's not as simple as sticking a plug into a wall outlet.

AndyO | 7 juillet 2014

@zwede: 250V is on the high side but within spec. You might actually get a higher miles per hour charge than most folk depending on the limitations of the built-in car charger.

AndyO | 7 juillet 2014

@johncrab: Thanks for the corroboration and I agree with your additional comments. The ignition timing is a good analogy to PF which I hadn't heard before. I had assumed that the Tesla car charger was near to PF 1 but I have no basis for that.

In my post I was trying to explain nominal voltage to noobies regarding the difference in some of the voltage numbers. I know a large portion, possibly majority of posters on this forum know all of this stuff. I admit that I have a pet peeve about people talking about 110 & 220 V circuits. I read somewhere that 80 years ago 110 and 220 were the correct nominal voltages.

As @sule mentioned, I do need to determine whether my 8V loss is distributed (which I had assumed) or in some particular poor connection. I'll check into that some time after my wife's knee surgery tomorrow.

diegoPasadena | 7 juillet 2014

I have a dedicated circuit running the 50 feet from my panel into the garage. The panel is from 1963, but for some reason it is a 200 A panel, which I understand to be unusual for single family homes of that era. I don't know if total rating affects anything, but I get a pretty much rock solid 240V 40A charge. Occasionally, it will vacillate between 239V and 240V for the first couple of minutes after starting the charge, but it settles in at 240 very quickly.

bonaire | 8 juillet 2014

I have the same problem. It is 120V or 240V here in the USA. Voltage drops below that may mean "really far from the transformer" or "really hot out today". I always get 121V on my Kill-a-Watt device. We are about 150' from the transformer and very few homes clustered off the transformer.