EVSE (Electric Vehicle Supply Equipment) for Home Use

Forums › Transport (cars, bikes and boats) › EVSE (Electric Vehicle Supply Equipment) for Home Use

1 | ... | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26

6754 posts

Uber Geek

Lifetime subscriber

#3128600 16-Sep-2023 16:00

Linuxluver:

I traded in the MG ZS EV as it was almost 3 years old. We now have a GWM Ora Cat extended range. It works great with the TWC, too.

I sat in an Ora Cat a while ago and thought it would be a good (and affordable) city runabout. Any chance you could give it a write up in the test drive thread? Of particular value from an experienced EV owner.

“We’ve arranged a society based on science and technology, in which nobody understands anything about science technology. Carl Sagan 1996

Linuxluver

5828 posts

Uber Geek

Trusted

Subscriber

#3128686 16-Sep-2023 18:34

Obraik: Oh yeah, I mean the new feature Tesla has been rolling out in the US where if your solar is generating any extra, it will send that to the car. I suspect that might require a Tesla vehicle as it adjusts the amps to make sure that it doesn’t use more than the excess solar. As far as I’m aware, there’s no way to dynamically change the amps on the Wall Connector, yet.

You can adjust the TWC charge rate using the web interface in the TWC. The TWC is also (ideally) on the home wifi…..so it should be possible to adjust the charge rate remotely but the TWC may not have the smarts to it by itself…. If that’s what you meant?

_____________________________________________________________________

I've been on Geekzone over 16 years..... Time flies....

Jase2985

13457 posts

Uber Geek

ID Verified

Lifetime subscriber

#3218011 14-Apr-2024 10:23

Im about to get an EVSE installed at home. I'm likely to go with a Fronius Wattpilot, as my solar system is going to have an 8kW Fronius Inverter and they interact nicely together to allow charging straight from the solar.

All the documentation says the Wattpilot has a type B RCD in the unit, so this should mean I shouldn't require one at the circuit board, and it would just need a standard Type A RCBO at the circuit board? is this correct? I already have the 32A RCBO on the circuit board and a 32A rated cable to where the EVSE will be installed.
It also just needs an isolation switch near the EVSE?

Linuxluver

5828 posts

Uber Geek

Trusted

Subscriber

#3218191 14-Apr-2024 14:12

I'm not an electrician.

I found that my 32amp Rolec Wallpod was on a 32amp Circuit breaker in the power board. That was fine for a couple of years, but then that RCD (not the RCD B in the Wallpod) started tripping after 30-40 mins. They seem to become less tolerant with age.

I had the 32amp circuit breaker and related wiring replaced a month ago with 40amp rated circuit breaker and wiring. No issues since.

I mention it because you said you had the same breaker and wiring as me. The Wallpods are different.

_____________________________________________________________________

I've been on Geekzone over 16 years..... Time flies....

dklong

167 posts

Master Geek

#3219599 17-Apr-2024 15:54

Jase2985:

Im about to get an EVSE installed at home. I'm likely to go with a Fronius Wattpilot, as my solar system is going to have an 8kW Fronius Inverter and they interact nicely together to allow charging straight from the solar.

Just an FYI that the EVNex E2 also allows supports 'Solar Only' charging and is about $1000 less installed than the Fronius Wattpilot. :-) A

lso made in NZ (Christchurch). Mine is installed with a Froniius 8kW inverter as well. They do put a separate CT clip in from the existing Fronius Smart Meter but it does the job quite happily.

Cheers

Jase2985

13457 posts

Uber Geek

ID Verified

Lifetime subscriber

#3219611 17-Apr-2024 17:10

dklong:

Jase2985:

Im about to get an EVSE installed at home. I'm likely to go with a Fronius Wattpilot, as my solar system is going to have an 8kW Fronius Inverter and they interact nicely together to allow charging straight from the solar.

Just an FYI that the EVNex E2 also allows supports 'Solar Only' charging and is about $1000 less installed than the Fronius Wattpilot. :-) A

lso made in NZ (Christchurch). Mine is installed with a Froniius 8kW inverter as well. They do put a separate CT clip in from the existing Fronius Smart Meter but it does the job quite happily.

Cheers

My issue is my meter is in the house about 25m from where the inverter and charger are, with no easy way to get a cable there. so unless it was wireless and could get that distance it makes it hard to have a system that uses solar. Hence, the wattpilot is going to be the easiest as i alreay have to have it there. With everything the Wattpilot is about 2.5k which isnt bad.

Zappi was looking promising as they offered a wireless solution, but their customer service has been atrocious.

Also go an answer from the supplier on the wattpilot, it does have a suitable type b RCD in the unit so existing type a RCD on the switchboard is fine and already have wiring in the walls to the charger location.

SaltyNZ

8218 posts

Uber Geek

Trusted

2degrees

Lifetime subscriber

#3219692 18-Apr-2024 08:10

I'm currently on my second Wallbox PulsarPlus, with the first failing less than a year after installation. The guy who came out to replace it (no questions asked at Wallbox) told me there is a known problem with the current sensor failing. The symptom is that the charger will just randomly stop after a few minutes.

It started doing it again in December and they asked to do a 'remote restore' whatever that is, but afterwards it was OK again (and by OK I mean it wasn't randomly stopping) till about 6 weeks ago. I didn't get around to doing anything about it since the Model 3 was away being uncrashed-into and the MG could just use the Tesla charger (with the 16A plug) but when we were trying to use the Wallbox I could see what the guy meant about the current sensor. Sometimes the app would show 0kW even though it was happily supplying the set current, and you could watch it bounce around from 0 to the highest I saw of 18kW (all the while supplying exactly the amount it was set to) so presumably at some point it must decide it's supplying 1 meellion kW and that's when it panics and stops.

Anyway ... I tried to log a ticket last week: radio silence. Have tried again this morning.

To cut a long story short, stay the hell away from Wallboxes.

iPad Pro 11" + iPhone 15 Pro Max + 2degrees 4tw!

These comments are my own and do not represent the opinions of 2degrees.

Equinox IT

Cloud spending continues to surge globally, but most organisations haven’t made the changes necessary to maximise the value and cost-efficiency benefits of their cloud investments. Download the whitepaper From Overspend to Advantage now.

cthombor

81 posts

Master Geek

Lifetime subscriber

#3220332 19-Apr-2024 14:57

SaltyNZ:

... when we were trying to use the Wallbox I could see what the guy meant about the current sensor. Sometimes the app would show 0kW even though it was happily supplying the set current, and you could watch it bounce around from 0 to the highest I saw of 18kW (all the while supplying exactly the amount it was set to) so ...

Hmmm.... I don't have a Wallbox, but I am curious about the accuracy of the current sensors in the EVSEs commonly used in NZ, as this will become important when (if!) our lines companies, our electricity retailers, and their governmental regulator get their act together on "remote-controlling" the slow-charging sessions. It's a much harder problem than just dealing to the long-deferred maintenance on the pilot-wire and ripple-control systems that are still operative on hot water cylinders, and then putting the EVSEs on the same circuit as your electric HWC. So many "kittens" (and more than a few "tigers") which must be "herded" somehow by a weakly-funded and not highly-empowered regulator! So many technical issues still to be sorted for an increasingly-diverse installed base of EVSEs and "home energy controllers" in NZ! From that perspective, the inaccuracy of an uncalibrated current sensor, its ongoing drift after each calibration, and its sensitivity to the "dirtiness" of the AC power it is sensing, are "minor details"! But even so... they're important.

My guess is that you are seeing a series of "instantaneous" current measurements -- which are varying widely due to "dirty power". The folks who maintain firmware for Wallbox have, I imagine, been issuing revs since 2022 or so, in an attempt to time-average these readouts sufficiently to tell the app's user what they probably really want to know: how fast is my car charging? (My dip in the blogosphere just now tells me that, in 2022, there was a well-documented defect in the Wallbox app whereby it would report wildly-fluctuating charging currents.) Revving the firmware for an EVSE must be a very delicate task... if the power is really dirty (i.e. with high harmonic distortion, or with wildly-varying phase angles and thus currents in a 3-phase supply) then I for one would really want to see some sign of this issue on the app's display! But then again, there are so many ways power can be "dirty" that it'd be a hiding to nowhere to do anything more than write firmware that'd keep the homeowner safe, and would cause the EVSE to shut down occasionally -- forcing the owner to reboot it or (in what is hopefully the worst usual case) to get a replacement or repair under warranty.

That said... it's certainly *possible* that some production batches of some models of Wallbox EVSEs had current sensors that often don't go the distance. You may have been unlucky enough to get one of these?

And... I wouldn't pay much attention to the current level that's displayed on a Wallbox app, except when diagnosing a software fault in the app, or except as a sign that the firmware in your Wallbox might not have been updated. What does your EV have to say about its measurements of its charging current? (Well ... most probably it will say nothing to you, because you don't have the necessary gear to "persuade" your EV to reveal this information".) It seems your sparky didn't have a clamp meter on hand, otherwise they might have measured the actual current flow into the Wallbox. The cheap-and-nasty clamp meters are about $20, and you can spend thousands of dollars on a power meter that'll diagnose the sort of power-quality issues that'll cause both a cheap-and-nasty meter and the current-sensor in an EVSE to report ... well not exactly nonsense, but some readouts that can be quite difficult to interpret with any validity.

Disclaimer: I have written tens of thousands of lines of software for a variety of instruments and devices; but I have no direct experience with any Wallbox EVSE or indeed with any EVSE. I have no professional qualifications as a power engineer, nor as an electrician. So -- here I'm just doing the usual thing online, of giving you my opinion ;-) "Take what you like and leave the rest!" But... if you ever do collect any documentary evidence (such as a series of screenshots) of wildly-varying current flows through an EVSE as reported by one current-measurement system, whereas another current-measurement system shows a steady flow of current... that would be an important contribution to the EVSE blogosphere, IMHO, and I hope you would take the time & trouble required to do this if you ever see it again! See e.g. https://electronics.stackexchange.com/questions/698644/why-do-different-multimeters-give-me-very-different-current-measurements

mkissin

388 posts

Ultimate Geek

ID Verified

#3220354 19-Apr-2024 15:40

I'm an electrical engineer who designs wireless EVSE's for a living. Metering current isn't hard (unless you need it to be metrology-grade for charging people, and then it becomes immediately incredibly difficult). Do people still manage to stuff it up? Of course, but it isn't actually difficult.

Calibration to any real degree isn't important because your lines company isn't going to be controlling the charging current of a single EVSE; they'll be batched into groups and statistics takes care of a lot. Various standards and local grid codes require current to be known to better than 1% accuracy if you wish to be used to support grid function, so you'll either comply with that or not.

cthombor

81 posts

Master Geek

Lifetime subscriber

#3220415 19-Apr-2024 19:40

mkissin:

I'm an electrical engineer who designs wireless EVSE's for a living. Metering current isn't hard (unless you need it to be metrology-grade for charging people, and then it becomes immediately incredibly difficult). Do people still manage to stuff it up? Of course, but it isn't actually difficult.

Calibration to any real degree isn't important because your lines company isn't going to be controlling the charging current of a single EVSE; they'll be batched into groups and statistics takes care of a lot. Various standards and local grid codes require current to be known to better than 1% accuracy if you wish to be used to support grid function, so you'll either comply with that or not.

Thanks for your comments!

Indeed, for billing purposes, it'll be the household power meter that measures the kWh. And the accuracy of the newfangled smart meters is pretty well controlled by governmental regulators -- with only the occasional scandal.

And... for purposes of load-shaping, as you say, it'll be the controllability of the aggregate load by the lines company that's important, not the particulars of how much power is drawn from each connection (except I guess on long LV lines, where it'll matter somewhat "where" on the line that power is being drawn or injected).

However I suspect that there'll be severe difficulties with the EVSE's controllability if its local controller ever sees wild swings in the EVSE's apparent power consumption. If the +ve swings (apparently) ever exceed 16A, then I'd want to be very careful about the number of sequential out-of-band readings which are accepted (for purposes of the usual time-averaging of a sensor's outputs) before invoking an emergency shut-down procedure.

If the -ve swings (apparently) push the EVSE's AC output below 6A, then I believe that's out of spec for compliance with the EV's AC charging standard. I am not at all sure about how any EVSE manufacturer could adequately assure that there are no possibilities for causing any model of EV to go into an error state with (say) a charging session that's drooping below 6A. Exiting that error state might be very inconvenient for the user, as well as potentially very expensive. At the very least -- the EV driver is not unlikely to find that their vehicle hasn't been charged sufficiently for their morning commute.

Well ... I'm blithering... as per usual... I'm sure you're all over these issues when designing a wireless EVSE, unless perchance it's DC-coupled.

One of the lovely things about DC-charging an EV is that you can keep a session alive on negligible power -- there's no 6A*100VAC = 600W lower limit on the charging rate. I'm still waiting for some EVSE manufacturer to produce a low-power (maybe 3kW max) DC charger which (in my dreams) also has an onboard MPPT -- so can take a DC feed directly from a small rooftop PV array. That would allow my wimpy 1.6kWp rooftop array to solar-trickle-charge my EV even during winter; whereas I'm now able to 100% solar-charge only through a 3:2 stepdown tranny... albeit rather inefficiently due to the DC-AC-DC conversion losses. The plot below is of a session at a (nominal) 8A rate and a (nominal) 230VAC supply ... and I see pretty big swings (+/- 500W) in the VA product (green) as measured by the Leaf at the terminals of its traction batteries, but quite a stable reading of the EVSE's supply to the vehicle (blue) -- which apparently is measured with a granularity of about 100W.

All to say that -- as I'm sure you know! -- if your wattage-measurement system is not measuring voltage and current at the same instant and at the same point in the circuit, you can have some non-trivial measurement errors -- especially when your sampling frequency is "beating" with the noise (perhaps 10kHz in the Leaf's DC-DC converter) that's injected by an SCR clipping a 100A supply. Or even a 10A supply.

As you say -- it's dead-easy to measure current, unless you want an accurate measurement! The wild fluctuations in the green VA curve on the plot below suggeste to me that there's some inductive coupling to a hefty source of square-wave "noise" in my Nissan e-NV200 while it's AC charging. I do understand there are some regulatory limits on the EMF noise radiated by the EV -- or injected by the EV into the mains supply through the EVSE -- but the charging of a lithium-ion cell isn't gonna be disturbed by a few millivolts of 10kHz noise. It's "just" a measurement problem ;-)

I wouldn't expect any EV manufacturer to do a much better job than Nissan's (+/- 500W, by my estimate) at measuring the power reaching the terminals of the traction battery... and even with the aid of a dongle and an app like LeafSpy, I think EV drivers will *always* be somewhat in the dark about the actual charging rate of our EV's traction battery -- especially in an AC charge where there are non-trivial losses (hundreds of Watts) in the vehicle's AC-DC converter.

SomeoneSomewhere

1802 posts

Uber Geek

Lifetime subscriber

#3220594 20-Apr-2024 10:20

My understanding is that 6A is the minimum current that the EVSE can advertise. The car must stay at or below the advertised current, but if the car elected to draw, say, 2A, that's entirely acceptable. It just can't be externally enforced.

During initial ramp-up and final constant voltage charging, you would expect the current to sometimes be below 6A, especially when very small batteries are involved.

I don't believe undercurrent is given as a fault or Ramon for the EVSE to disconnect.

Running for extended periods at low charging currents is to be avoided, however. As you note, the quiescent losses become a large portion of the actual charging power, resulting in low efficiency. Part of that is the need to keep the 12V system fully operational. The charger for the 12V battery will be cycling on and off, and that could be leading to instability in your DC side data.

You might be able to fit a hall effect sensor around one of the battery leads?

cthombor

81 posts

Master Geek

Lifetime subscriber

#3220662 20-Apr-2024 14:38

SomeoneSomewhere: My understanding is that 6A is the minimum current that the EVSE can advertise. The car must stay at or below the advertised current, but if the car elected to draw, say, 2A, that's entirely acceptable. It just can't be externally enforced.

During initial ramp-up and final constant voltage charging, you would expect the current to sometimes be below 6A, especially when very small batteries are involved.

I don't believe undercurrent is given as a fault or Ramon for the EVSE to disconnect.

Running for extended periods at low charging currents is to be avoided, however. As you note, the quiescent losses become a large portion of the actual charging power, resulting in low efficiency. Part of that is the need to keep the 12V system fully operational. The charger for the 12V battery will be cycling on and off, and that could be leading to instability in your DC side data.

You might be able to fit a hall effect sensor around one of the battery leads?

Hmm, I'm now thinking the VA readouts on the traction battery of a BEV will often be quite noisy, due to the incompletely-filtered harmonics of the AC-DC converter, and also due to the cell-balance shunts (roughly 340 ohms?? i.e. a 10mA balancing current to/from each cellpair?) flipping at some rate (perhaps 10kHz). There's always *some* variability (typically in excess of 10mV, rarely above 100mV) in the cell voltages. 10mV/400V isn't a large swing, but I'd guess the net charging current into the pack will vary by somewhat more than 10mA. On a 2kW charge there's only 5A of current into the battery pack so 10mA/5V = 0.2%. Not quite negligible -- and this noise could be greatly magnified, or even become a bias that persists for seconds, if the sensor-sampling clock is beating with the cell-balance clock, or with the frequency of the chopper in the AC-DC converter. As you say -- I *could* rig up my own sensors on the 400V DC supply, but that's way past my pay grade ($0, I'm retired ;-), level of interest, to say nothing of my near-total lack of expertise in automotive measurement systems (or for that matter, in any of the analog circuitry of a measurement system. My expertise kicks in at controlling the sample-and-hold of an ADC; and if the noise filters in the analog circuitry aren't adequate for the limiting ("Nyquist") rate of what my ADC can handle, then I consult with an electrical engineer!

I do know enough to be worried about EVs that'll (at least indirectly and thus "dirtily") cause a charging session to terminate whenever the EVSE isn't supplying "enough" current. How much is "enough"? Well I think there are many factors... but one is the make & model of the EV. The Renault Zoe is notorious in this regard. One research group measured 0.6A H3 on its AC charging supply when a Zoe is actively charging, with H5 spiking to 2A when it is "idling". Ouch! (The article is paywalled at https://ieeexplore.ieee.org/abstract/document/9334798) Anecdotally, an AC charging session on a Zoe is marginally stable when its EVSE is supplying 10A, and becomes reliable at 16A or above.

I have zero interest in trying to measure, to better accuracy and precision than my BEV does already, the instantaneous charging power reaching its traction battery during an AC charging session! However when (if!) NZ's domestic mode-2 and mode-3 EVSEs are routinely being throttled by a lines company, the added inefficiencies of a throttled charge -- and the huge inconvenience of a charge session that's aborted and won't "automatically" restart (without the owner unplugging and replugging their vehicle to their EVSE) -- will I think become very important indeed.

Well... in my dreamworld future, we'll mostly be mode-4 DC charging at home! (Tesla v Edison, see https://www.youtube.com/watch?v=gJ1Mz7kGVf0 ;-)

Today I noticed on Alibaba -- for the first time -- a lowcost low-powered DC fast charger which claims to have some "solar charging" feature (which I presume means it's either DC-coupled to a PV array or, more likely, is "happy" with accepting a trickle (or even an intermittent) AC supply on its charging-supply terminals, so long as its controller is reliably powered by an amp or two of 230VAC). USD 500 FOB Shanghai, for a wall-mounted 9kW DC fastcharger. That's a very attractive price-point... but it surely won't have an SDOC, and it'd require a seriously-bent registered electrician, or a seriously-deranged do-it-yourselfer, to install it in a NZ garage. But within a few years... I can hope!