Thoughts on using generator to charge

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,173
Reaction score
2,283
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
Saltwater? Bottled or Natural :rolleyes:
Frank always taught me.
"Watch out where the huskies go, and don't you eat that yellow snow"
Well I did think of that but as you can reuse the 4 footers...





Advertisement

 

Bill906

Well-known member
Joined
Mar 21, 2020
Messages
93
Reaction score
123
Location
Wisconsin
Vehicles
Jeep
Country flag
Most portable generators today are designed to not require a ground rod. If you are connecting it to your house, then yes you will most likely require a ground rod to meet code. If you’re plugging your car charging cord directly into the generator, and the generator meets some basic bonding requirements you will not need to drive a rod into the earth. Nor will you need water- salt or fresh. Check your generator documentation to determine if it requires a ground rod.

I highly recommend using a pure (true) sinewave generator. The first bridge in the charging circuit of the Tesla chargers uses active components (Transistors). I find it unlikely the control Tesla uses for the firing of these transistors would be designed to adapt to voltage waveforms other than sine. There would be no need to as all utility power produces sine wave voltage waveforms.

The Tesla does look to see if neutral is bonded to ground. This has been demonstrated in many posts and YouTube videos. Your generator may have that built in, or you may need to acquire an adapter that will do it. If you understand what you are doing, you could bond it yourself.

If you do require a ground rod, one thing we did in the military was drive them in at an angle. When we needed to pull them out we just tried the end of the rod to the back of a truck, and slowly drove forward.
 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,173
Reaction score
2,283
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
Most portable generators today are designed to not require a ground rod.
The only "design" element for a ground rod is a bolt in the frame to which one can connect the wire from a grounding rod. Obviously every generator has a bolt in the frame that can be used for earthing but as awareness of such things as requirement for grounding advances some are provided with a conveniently located one that serves no other purpose and is marked with the ground symbol. If no bolt is so designated any frame bolt will do.

If you are connecting it to your house, then yes you will most likely require a ground rod to meet code.
No "most likely" about it. If it is connected into your house wiring it must be properly grounded and unless your transfer switch switches neutral as well as the hots or your transfer switch is the "service entrance" the generator may not be connected to the house grounding system and must, therefore, have its own ground.

If you’re plugging your car charging cord directly into the generator, and the generator meets some basic bonding requirements you will not need to drive a rod into the earth.
You do not have to ground the car to charge it from a generator. You only need to fool it into thinking it is grounded. The "basic bonding requirements" referred to here is connecting neutral to the frame. Is that safe? Well no it isn't and that's why if you hadn't fooled the car into thinking it is grounded it wouldn't allow you to charge. Are you going to get killed? Probably with liklihood about the same as being struck by a meteorite. But people have been struck by meteorites. It's up to you to decide whether you want to go to the trouble of properly grounding the generator or not.


Nor will you need water- salt or fresh. Check your generator documentation to determine if it requires a ground rod.
No generator requires a ground rod.


I highly recommend using a pure (true) sinewave generator. The first bridge in the charging circuit of the Tesla chargers uses active components (Transistors). I find it unlikely the control Tesla uses for the firing of these transistors would be designed to adapt to voltage waveforms other than sine.
It has been proven that the first bridge has diodes. It has been proven that it has transistors. Both right here on the internet. There are many reasons why the input bridge could use active devices but the most likely one in this context "power factor" correction. That's a misnomer as in the case of the Tesla the true power factor is 1.0 but it has come to refer to correction of distortion of the current waveform. To do this, of course, the system must be able to fire correctly when the waveform is distorted as that is what it is designed to do.



There would be no need to as all utility power produces sine wave voltage waveforms.
The utility is highly motivated to produce harmonic free waveforms but that is impossible. More to the point in today's world where so many power supplies are "switchers" lots of harmonics are injected into the network to the point where requirements for "power factor" correction have been levied. If the Tesla has an active front end it is to meet these requirements.

I get my power here from Quebec Hydro which, obviously, serves Quebec but also much of New York. They deliver a voltage waveform with 1.7 % THD. The current waveform to my charging model X exhibits 2.0% THD. Given what I have been able to deduce about the power supply architecture in the car I don't think the bridge is active - but it could be. In any case it doesn't corrupt the utility much. On generator the current THD goes up to about 2.9% (relying on memory for this one - the utility numbers I wrote down so they are solid).

When I expressed dismay in No. 4 at having this topic come up again so soon it was because I knew someone would come out of the woodwork with the pure sine inverter nonsense again. And Voila! You do not need a sine wave inverter generator! Were I to buy a portable generator I would probably get one as they are advantageous in other ways. But the power they produce is better than your utility's.



The Tesla does look to see if neutral is bonded to ground.
The Tesla has no way to tell whether your generator's neutral is bonded to earth. But it knows that in a properly wired (passed inspection) building it will be. It, therefore, looks for 120V between hot and earth and if it sees it, allows charging. The company further protects itself by plainly advising that one should not charge from a privately owned generator.



This has been demonstrated in many posts and YouTube videos. Your generator may have that built in, or you may need to acquire an adapter that will do it. If you understand what you are doing, you could bond it yourself.
An older generator may have earth and neutral pins connected but a newer one will probably not and may even have a sticker saying "Ground and Neutral Isolated" on it.


If you do require a ground rod, one thing we did in the military was drive them in at an angle. When we needed to pull them out we just tried the end of the rod to the back of a truck, and slowly drove forward.
Now that's a great tip!
 

tjordan

Member
First Name
Tristan
Joined
Jun 15, 2020
Messages
10
Reaction score
10
Location
Sproat Lake
Vehicles
Cybertuck (on order - dual engine model)
Occupation
Forester
Country flag
Can you not just plug into 110V with the Cybertruck? There are lots of charging stations through BC and the number is growing rapidly. But 110V is literally everywhere that has a town or gas station (or house).
I've had my EV for 3 years and mostly charge with my 110V at home. Obviously on a long road trip that's not ideal - but if you plan accordingly it should be a better option than hauling a generator and fuel for the entire trip.
 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,173
Reaction score
2,283
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
A 120V outlet provide about 1440 W of charging power. Using such an outlet to charge a CT for 24 hours would add 78 miles to its range. So yes, in desperation in order to take on a few miles to get to a more suitable charging location is doable. But one should really look for a gas station with a welder outlet, a welding shop or someplace that has 240 and might be willing to let you plug in.
 

Crissa

Well-known member
First Name
Crissa
Joined
Jul 8, 2020
Messages
3,037
Reaction score
3,205
Location
Santa Cruz
Vehicles
2014 Zero S, 2013 Mazda 3
Country flag
Well, it wouldn't be desperation if you weren't going more than an average of 78 miles a day, would it?

The 'granny cable' does serve a purpose, and charging overnight is a good idea. That pre-heats (or cools) the cabin, maintains the battery, and adds a few miles, too.

-Crissa
 

Red61224

Well-known member
Joined
Jul 10, 2020
Messages
54
Reaction score
61
Location
Florida
Vehicles
No Thanks
Occupation
Amature Lizard Wrestler
Country flag
A 120V outlet provide about 1440 W of charging power. Using such an outlet to charge a CT for 24 hours would add 78 miles to its range. So yes, in desperation in order to take on a few miles to get to a more suitable charging location is doable. But one should really look for a gas station with a welder outlet, a welding shop or someplace that has 240 and might be willing to let you plug in.
Or lest we forget the RV park 14-50, hurray.
 

Bill906

Well-known member
Joined
Mar 21, 2020
Messages
93
Reaction score
123
Location
Wisconsin
Vehicles
Jeep
Country flag
I really had a hard time understanding why someone critiqued my post with such thoroughness. It was a post on a public forum, not a lab report. I had an even harder time trying to figure out why it claims so much of what I said was wrong. After reading the critique several times I realized there were some bad assumptions made along with possibly some misconceptions about electricity. Let me try to bring these to light and explain them.

I did some google searches to verify what I thought to be true and found many sites that confirmed what I said. If you want to see all of them simply google "portable generator grounding". I chose two that presented the data well and I consider trusted sources - OSHA and Generac. OSHA stands for Occupational Safety and Health Administration. A government agency that has safety right in its name. Generac is an established generator manufacturer. Here are the links to the websites.

https://www.osha.gov/OshDoc/data_Hurricane_Facts/grounding_port_generator.html

https://www.generac.com/service-support/faqs/portable-generator-faqs

The only "design" element for a ground rod is a bolt in the frame to which one can connect the wire from a grounding rod. Obviously every generator has a bolt in the frame that can be used for earthing but as awareness of such things as requirement for grounding advances some are provided with a conveniently located one that serves no other purpose and is marked with the ground symbol. If no bolt is so designated any frame bolt will do.
The design element I was referring to was that all non-current carrying metal parts need to be bonded to the generator frame. As specified in many websites I found including the OSHA website referenced above.


No "most likely" about it. If it is connected into your house wiring it must be properly grounded and unless your transfer switch switches neutral as well as the hots or your transfer switch is the "service entrance" the generator may not be connected to the house grounding system and must, therefore, have its own ground.
There are a lot of jokes about engineers know everything and how they are always right. I, unlike some, am not that kind of engineer. I said "most likely" because I do not pretend I know every possible situation. I do not know the electrical code in all countries, I do not know all situations. Maybe some electrical codes provide other options in case a ground rod is not feasible. I really do not know why this criticism was deemed necessary.

The Tesla has no way to tell whether your generator's neutral is bonded to earth. But it knows that in a properly wired (passed inspection) building it will be. It, therefore, looks for 120V between hot and earth and if it sees it, allows charging. The company further protects itself by plainly advising that one should not charge from a privately owned generator.
I said " The Tesla does look to see if neutral is bonded to ground." At first I could not understand the reply. First it says the Tesla does not do it, immediately following that the reply explains how it does. After rereading it a few times I noticed the author used the term "Earth" instead of "ground". I believe the assumption here is that when I said "ground" I meant connected to a ground rod. In the electrical world, the term ground can mean many different things. There are many types of ground. Earth ground, chassis ground, high resistance ground, digital ground, analog ground etc. This may seem like an extreme example, but I'm sure many electrical circuits on the international space station have references to "ground". In my post whenever I meant connected to the earth I would say ground rod. When I said ground I meant neutral bonded to the ground of the generator. Maybe I should have used a term like chassis ground.

You do not have to ground the car to charge it from a generator. You only need to fool it into thinking it is grounded. The "basic bonding requirements" referred to here is connecting neutral to the frame. Is that safe? Well no it isn't and that's why if you hadn't fooled the car into thinking it is grounded it wouldn't allow you to charge. Are you going to get killed? Probably with liklihood about the same as being struck by a meteorite. But people have been struck by meteorites. It's up to you to decide whether you want to go to the trouble of properly grounding the generator or not.
First, no, the "basic bonding requirements" I was referring to was that all non-current carrying metal parts need to be bonded to the generator frame. Since this is the second time the words I used to describe this were misunderstood I will accept blame.

Second, OSHA seems to disagree with the opinion that this is unsafe. OSHA specifically states you can use a portable generator without connecting it to a ground rod driven into the earth if you meet certain requirements (see OSHA link above). Also, one of the requirements listed by Generac to use a portable generator without a ground rod is that you have a "...bonded neutral portable generator..." Meaning neutral is bonded to the generator ground or chassis ground in case that isn't clear.

I cannot find a way to comment on the meteorite analogy without sounding mean. I've tried.

No generator requires a ground rod.
Not sure what is meant by this. NEC requires a ground rod for a generator in many applications. I guess technically it is correct. A GENERATOR does not require it. In case there is confusion, I meant check the manual to see if a ground rod is required for your application.

It has been proven that the first bridge has diodes. It has been proven that it has transistors. Both right here on the internet. There are many reasons why the input bridge could use active devices but the most likely one in this context "power factor" correction. That's a misnomer as in the case of the Tesla the true power factor is 1.0 but it has come to refer to correction of distortion of the current waveform. To do this, of course, the system must be able to fire correctly when the waveform is distorted as that is what it is designed to do.
I have not seen anywhere on the internet that has proven the first bridge is diodes. Please reference your sources. I believe it was the third bridge that has been proven to be diodes. But the bridge that utility power first touches uses transistors in every example or investigation I've seen.

The reason the Tesla power factor is 1.0 (or at least very near 1.0) is BECAUSE the first bridge is an active converter. If it was a passive converter it would lower the power factor, along with create harmonics.

The active converter is designed to not be the cause of poor power quality. It is not specifically designed to accept poor quality already on the power system. Now I admit these are assumptions on my part. I do not work at Tesla or have access to their designs. But the circuits do resemble circuits I am familiar with that function as described.

When I expressed dismay in No. 4 at having this topic come up again so soon it was because I knew someone would come out of the woodwork with the pure sine inverter nonsense again. And Voila! You do not need a sine wave inverter generator! Were I to buy a portable generator I would probably get one as they are advantageous in other ways. But the power they produce is better than your utility's.
Another assumption here. I said "a true (pure) sine wave generator". I never used the word "inverter".

When I said "a true (pure) sine wave generator" I meant any generator that was designed to create a sinewave voltage waveform. If the generator, under proper operating conditions, produces a voltage waveform shaped like a sinewave, it is a true sinewave generator. Some generators, mainly cheap inverter duty generators are not designed to create a sine wave. They are designed to make a square wave or a square wave that dwells at zero volts for a few milliseconds before continuing to the opposite polarity. Those are what I highly recommend not using.

4)There are generators and generators i.e. some very good ones and some real POS. It seems a good metric is THD. If THD is under 3% [Edit] 5-6% then the generator is pretty good. Note that purely mechanical generators can produce that so that the common advice that you must use a pure sinewave generator is bad advice but pure sine does convey other advantages (and cost).
Here's the funny thing, we were arguing the same point the whole time from different perspectives. At least I'm assuming the recommendation is to use a pretty good generator after it was defined as a generator that specifies less than 5-6% THD. I don't have hard numbers, but asking around with other engineers at work, they expect a square wave generator to have at best 30% THD. (in case it isn't clear, %THD is like golf, the lower the number the better).

If you have constructive criticism on anything I've said please post. If you are going to attack my post for reasons of ego or fear it threatens your established superiority - Please don't. I do not have any more time for such nonsense.

On the subject of constructive criticism, maybe it's a generational thing or maybe a geographical thing but I find the overuse of the words "Obviously", "Clearly" and the phrase "of course" somewhat condescending. The word "obviously" in particular I've always found interesting. If something is truly obvious it shouldn't need to be pointed out, effectively eliminating the need for the word. To me, when someone makes a statement with those words it often comes across as making the reader feel dumb for not knowing that. Or, sometimes I think someone might use it to validate a statement they feel is true, but do not or cannot confirm it is. So by saying their statement is obvious no one should be able to refute it. Obviously.
 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,173
Reaction score
2,283
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
You start with a diode bridge which turns the incoming mains waveform into a series of positive voltage pulses. Those are then chopped in a second active bridge which effectively modulates those rectified voltage pulses on to a 40 kHz or so carrier. The modulated signal is stepped up by a transformer and then synchronously demodulated in another active ring demodulator (bridge). The phase of the reinserted carrier is between 0 and pi/2 so that the average output voltage is always positive and that voltage passed to a small inductor thus producing a unidirectional current for the battery. The inductor smooths out the 80 kHz and higher harmonics from the demodulator. If the phase of the reinserted carrier be the same as the phase of the chopper carrier then the original rectified pulses are reproduced. If it's shifted by pi/4 then the output current pulses are scaled by cos(pi/4) = .707 etc. Thus by phase modulating the reinserted carrier one can regulate the output current. As there are no energy dissipative devices upstream of the demodulator the shape of the input current pulses must follow the current pulses going to the battery and they can be made to be any shape one wants. In this application we want them to be sinusoidal so a reference is constructed by phase locking a local oscillator to the input voltage waveform. The phase angle is controlled by a feedback loop which in addition to imposing sinusoidal shape sets the amplitude of the pulses to reflect the desired rms charging current. Obviously small inductors in the input lines smooth out the 40 kHz chopping.

Obviouly I don't know any of the details but it is very likely, based on what we do know about the Tesla supplies, that this is how they work. Thus there is no need for the input bridge to be active and that's why it probably isn't.
 
Last edited:

Bill906

Well-known member
Joined
Mar 21, 2020
Messages
93
Reaction score
123
Location
Wisconsin
Vehicles
Jeep
Country flag
You start with a diode bridge
If you start with a diode bridge you start with a passive converter. All evidence points to the first converter using transistors and being an active converter in the Tesla charging circuit.

If anyone would like to see an example of this evidence I recommend watching this clip:

At around 3:02 they go over the first converter.
 

Bill906

Well-known member
Joined
Mar 21, 2020
Messages
93
Reaction score
123
Location
Wisconsin
Vehicles
Jeep
Country flag
First gates are on the left most part of the board. The diodes are at the end.
 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,173
Reaction score
2,283
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
I wouldn't put too much cred on that video. It's probable that all the power chips are FETs as that would make the device 2-way capable for eventual V2G but probably more likely in the shorter term as the source of AC for the in bed power on the CT.

One very interesting aspect of this that hadn't occurred to me before is that the device in the video is clearly a 3ø device (each of the 3 rows is 3ø) but clearly Tesla would want to be able to use it in biphase vehicles as well. What's interesting is that the two modes require very different approaches to "power factor" correction. In 3ø mode the waveform isn't distorted (well it is but THD is small) but the current is shifted 30 ° relative to the voltage so there is true "power factor" - about that of an induction motor. In biphase mode there is no phase shift but the system draws current in pulses such that the THD is high and the rectifier shapes it. I don't know how this is done (and neither does anyone else who doesn't work at Tesla). It could be done in any of the three bridges or in a combination more than one.

This stuff is very interesting to me but I am sure not to most. The thing to take away is that yes, you can charge your car from a generator if you wish to defy Tesla and that there will not likely be any consequences and that the generator does not have to be a pure sinewave generator but that pure sine genny's are pretty nice.
 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,173
Reaction score
2,283
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
OK I think I've doped it out but again it is only reasonable assumption (and quite different from my previous "reasonable" assumption) based on my now more current knowledge of the state of the art. Where a vehicle battery is to be charged from a 3ø supply the modern designer uses a "Vienna rectifier" which is a modified diode bridge in which each phase connects to the common point between two series diodes through an inductor (which the video linked above mentions are present). The diodes outrer ends go, as usual, to the positive and negative ends of a capacitor but in this case the capacitor consists of two caps connected in series. The diodes inner (common) connection point can be connected to the capacitor center point through a FET. Thus each phase is connected to at least two but as many as 6 diodes and one FET. The FET is gated by a PWM waveform. By controlling that waveform properly input current can be made to follow input voltage thus realizing power factor (of both types) correction. Current waveform THD can evidently be controlled to about 2% (which is just about what my X yields).

Now how does this work in a biphase system? That's something for me to amuse myself with for the next couple of days.
 

Advertisement





 


Advertisement
Top