Making home EV ready

Open for any sort of non-technical discussion regarding EVs
T1 Terry
Senior Member
Posts: 978
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Making home EV ready

Post by T1 Terry » Wed, 16 Oct 2019, 13:29

I think with the particular inverter I'm looking at, PIP5048MG, the solar is boosted first, then enters what ever creates 240vac and 48vdc out of that. When the solar is not present, the inverter uses the 48v nom. battery voltage, so I'm guessing it boosts the 48vdc to what ever the feed in DC voltage is for the AC inverter part so the same unit does the job rather than having to double up on the inverter side. I doubt the 48v battery sees a big ripple across it so why would the solar or vehicle battery feeding in through the solar?

T1 Terry
Green but want to learn

User avatar
coulomb
Site Admin
Posts: 3762
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: Making home EV ready

Post by coulomb » Wed, 16 Oct 2019, 22:14

T1 Terry wrote:
Wed, 16 Oct 2019, 13:29
... the particular inverter I'm looking at, PIP5048MG... I doubt the 48v battery sees a big ripple across it so why would the solar or vehicle battery feeding in through the solar?
The isolating transformer is in the 48 V to/from 400 V DC-DC bidirectional converter. So the battery is isolated from the mains. In the models with a 145 V max Solar Charge Controller, the panel voltage is buck converted down to battery voltage, so the PV inputs of those models are isolated from the mains.

However, the PIP-5048MG and similar models with the 450 V max Solar Charge Controller boost the panel voltage to 400-500 V, which is the input to the inverter proper (DC-AC converter). So these would NOT be isolated from the mains, unless the Solar Charge Controller has another isolating transformer (and I doubt that very much).

Personally, I think that makes the idea of connecting these economical inverters directly to the car's 400 V battery a non-starter. You'd have to find a non-transformerless model (i.e. one with a transformer), and such inverters are rare these days. Sadly.

Edit: here is a presumed block diagram (which I drew for another forum) for the PIP-5048MG:

Image

Only the lower DC-DC converter in this diagram has the isolation.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

antiscab
Senior Member
Posts: 2479
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Re: Making home EV ready

Post by antiscab » Thu, 17 Oct 2019, 08:14

I'm not sure I understand the issue. The car battery is double insulated just the same as the panels on the roof are.

The cars own motor controller output swings from +400v to -400v, so the insulation would be designed to handle that kind of switching.

Coincidentally the both grid tie Inverters I use are transformer based, though that's mainly because they're not efficient enough for permanent use in a solar setup
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells

User avatar
coulomb
Site Admin
Posts: 3762
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: Making home EV ready

Post by coulomb » Thu, 17 Oct 2019, 10:10

antiscab wrote:
Thu, 17 Oct 2019, 08:14
The cars own motor controller output swings from +400v to -400v, so the insulation would be designed to handle that kind of switching.
But the motor is floating, whereas the mains has a low impedance connection to earth.

I worry about the modern cells with their metal cases (if they have any) being non-isolated from the terminals, and the very thin plastic that separates them from everything else.

You're probably right, it's probably no problem, but it just doesn't feel right to me.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

T1 Terry
Senior Member
Posts: 978
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Making home EV ready

Post by T1 Terry » Thu, 17 Oct 2019, 11:12

This set up isn't intended to feed back into the mains or even be mains connected, it is for an off grid stand alone system or SPS as the Clean Energy Council calls it. The output side of the inverter (240vac) must be isolated from the DC solar input, surely, otherwise there would be both 450vdc and 240vac on the same conductors and the solar isolators are only designed for max 500vdc so I'm not sure if they have an AC rating at all. The solid state relays are certainly only marked as suitable for DC and not AC so I'm not sure about the electronics within the inverter itself would handle AC on the DC side.

T1 Terry
Green but want to learn

User avatar
coulomb
Site Admin
Posts: 3762
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: Making home EV ready

Post by coulomb » Thu, 17 Oct 2019, 12:12

T1 Terry wrote:
Thu, 17 Oct 2019, 11:12
The output side of the inverter (240vac) must be isolated from the DC solar input, surely, otherwise there would be both 450vdc and 240vac on the same conductors and the solar isolators are only designed for max 500vdc so I'm not sure if they have an AC rating at all.
The AC component is only from each PV terminal (positive or negative) to earth, not between terminals. So the isolator doesn't have to switch the AC at all, just not break down or leak to earth or a user touching it.

If you believe that all inverters are isolated between PV in and AC out, then what is the transformer / transformerless designation for? It's not just about the weight and cost.

Edit: you should be able to convince yourself of this by using a multimeter on AC volts, with one terminal on earth, and the other on a PV input terminal of your PIP-5048MK (either terminal will do). I think you'll find over a hundreds volts solid. An isolated one would show a few to perhaps 20 volts and varying.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

T1 Terry
Senior Member
Posts: 978
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Making home EV ready

Post by T1 Terry » Thu, 17 Oct 2019, 12:57

Ah, interesting, I haven't even bought the MG inverters yet but I do have a couple of MS inverters with the lower solar open circuit voltage, will these have the same effect of AC on one leg of the solar to earth?

T1 Terry
Green but want to learn

User avatar
coulomb
Site Admin
Posts: 3762
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: Making home EV ready

Post by coulomb » Thu, 17 Oct 2019, 15:15

T1 Terry wrote:
Thu, 17 Oct 2019, 12:57
I do have a couple of MS inverters with the lower solar open circuit voltage, will these have the same effect of AC on one leg of the solar to earth?
No. These connect to the battery directly, and the battery is isolated from the AC output.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

T1 Terry
Senior Member
Posts: 978
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Making home EV ready

Post by T1 Terry » Fri, 18 Oct 2019, 09:03

coulomb wrote:
Thu, 17 Oct 2019, 12:12
T1 Terry wrote:
Thu, 17 Oct 2019, 11:12
The output side of the inverter (240vac) must be isolated from the DC solar input, surely, otherwise there would be both 450vdc and 240vac on the same conductors and the solar isolators are only designed for max 500vdc so I'm not sure if they have an AC rating at all.
The AC component is only from each PV terminal (positive or negative) to earth, not between terminals. So the isolator doesn't have to switch the AC at all, just not break down or leak to earth or a user touching it.

If you believe that all inverters are isolated between PV in and AC out, then what is the transformer / transformerless designation for? It's not just about the weight and cost.

Edit: you should be able to convince yourself of this by using a multimeter on AC volts, with one terminal on earth, and the other on a PV input terminal of your PIP-5048MK (either terminal will do). I think you'll find over a hundreds volts solid. An isolated one would show a few to perhaps 20 volts and varying.
coulomb wrote:
Thu, 17 Oct 2019, 15:15
T1 Terry wrote:
Thu, 17 Oct 2019, 12:57
I do have a couple of MS inverters with the lower solar open circuit voltage, will these have the same effect of AC on one leg of the solar to earth?
No. These connect to the battery directly, and the battery is isolated from the AC output.
Hmmm..... maybe a time for serious rethink on the MG inverter for the solar to EV battery link, although I doubt anyone would be touching the solar wires with up to 450vdc across them anyway. There is a Morningstar 600vdc max solar input that the engineers there say can be connected across a battery or solar to act as a battery to battery DC to DC charger, so maybe a better option, although I haven't checked the price yet.

T1 Terry
Green but want to learn

T1 Terry
Senior Member
Posts: 978
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Making home EV ready

Post by T1 Terry » Fri, 18 Oct 2019, 10:52

Checked out the price for the 600vdc 60 amp Morningstar MPPT controller .... ouch, around the $2,000 mark, would something like this block the AC on the solar/battery lines https://au.rs-online.com/web/p/rectifie ... s/8776418/

T1 Terry
Green but want to learn

User avatar
coulomb
Site Admin
Posts: 3762
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: Making home EV ready

Post by coulomb » Fri, 18 Oct 2019, 17:54

T1 Terry wrote:
Fri, 18 Oct 2019, 10:52
would something like this block the AC on the solar/battery lines ...
No, that AC has to be there by the design, you can't safely shunt it to earth or anything.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

T1 Terry
Senior Member
Posts: 978
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Making home EV ready

Post by T1 Terry » Sat, 19 Oct 2019, 10:44

I don't want to shunt it to earth, just block it from being present on the solar/battery lines. Just DC to one side of the diode and mixed AC and DC inside the inverter covers where it would involve removing an access panel to be exposed to the AC element on the DC lines.

T1 Terry
Green but want to learn

T1 Terry
Senior Member
Posts: 978
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Making home EV ready

Post by T1 Terry » Tue, 22 Oct 2019, 11:21

More thinking on this topic. Why would AC be seen on the DC side? It must pass through a DC to DC converter before it is fed into the inverter itself to make the AC wave form, so wouldn't that block any AC waveform. I can understand DC ripple being seen on the DC side, but if the wave form doesn't cross the zero volts line it isn't classed as AC is it? I guess an oscilloscope would be required to definitely determine if the wave crossed the zero voltage line or not, maybe the ripple is enough to cause an AC effect that the multimeter interprets as AC.
The next thought is the operation of EV itself. The DC battery is powering an inverter to drive the 3 phase motor with an AC wave form, so would the battery see an AC waveform anyway rather than the DC ripple?

T1 Terry
Green but want to learn

User avatar
coulomb
Site Admin
Posts: 3762
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: Making home EV ready

Post by coulomb » Tue, 22 Oct 2019, 19:04

T1 Terry wrote:
Tue, 22 Oct 2019, 11:21
Why would AC be seen on the DC side?
It won't be pure AC. It happens because the inverter makes the AC output by connecting the live and neutral outputs to bus positive or bus negative, via the IGBTs, for microseconds at a time, with varying pulse width to create the sine wave. The opposite polarity half cycle of 50 Hz output is achieved by using the other pair of IGBTs, connecting live-via-inductor and neutral to the negative or positive side of the DC bus, the opposite of the way they connected last half-cycle. All this is filtered at the AC output end, usually by an LC filter, but the DC bus doesn't see that filtering. Remember that neutral has to be connected to earth at the output of the inverter, so that would include the metal chassis of the car.

During a half-cycle of AC output, you are connecting one side of the EV traction battery to earth/chassis for some microseconds, and it's floating for the rest.
During the other half-cycle of AC output, the other side of the EV traction battery connects to earth for some microseconds, and is floating for the rest. I hope I got that right. So what each side of the battery sees with respect to earth is a mess of lethal, high frequency (~30 kHz) pulses whose width varies over 10 milliseconds.

I just realised another problem. When an EV goes to "ready", it does that ~6 step dance with the contactors. Part of what it's doing is checking that there isn't too much leakage to chassis from either side of the traction battery to chassis. If there is, it won't go to ready, and will instead throw a fault code. [ Edit: and the inverter will almost certainly cause that test to fail if it's connected. ]
It must pass through a DC to DC converter before it is fed into the inverter itself to make the AC wave form, so wouldn't that block any AC waveform.
? I thought we were discussing connecting the traction battery directly to the inverter bus. But if it's through the PV input terminals of a model that takes up to 450 V, then it's through a boost converter. Boost converters aren't usually isolated, and the ones in PIP-5048MGs almost certainly are not.
I can understand DC ripple being seen on the DC side, but if the wave form doesn't cross the zero volts line it isn't classed as AC is it?
It doesn't really matter if it's "real" AC or not; the issue is that it's high voltage and referenced to earth with a low impedance. Also, I forgot that it's also a high frequency, high dV/dt signal, which will cause all sorts of problems.

And again, the problem isn't from terminal (of bus or PV input) to other terminal, it's from both terminals to earth. I'll grant you that the boost converter in the solar charge controller would smooth a fair bit from terminal to terminal, but hardly any from terminal to earth. [ Edit: in other words, it's a sort of "common mode" noise, except very high amplitude "noise". ]
The next thought is the operation of EV itself. The DC battery is powering an inverter to drive the 3 phase motor with an AC wave form, so would the battery see an AC waveform anyway rather than the DC ripple?
Again, the EV motor is isolated from earth, the AC output of the inverter can't be. So that's why the motor inverter won't cause the same problem.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

T1 Terry
Senior Member
Posts: 978
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Making home EV ready

Post by T1 Terry » Wed, 23 Oct 2019, 11:14

So, if the inverter remained isolated and remained a floating system, line 1 and line 2 rather than neutral and positive, then the AC would not be seen on the vehicle negative side so it wouldn't trip the leakage to the chassis problem. If an RVD rather than an RCD was used on the 240vac output the supply would still be safety protected yet remain floating so only the battery cables would see DC or AC ripple .... could/should this be filtered? I know the Winston LYP cells don't like to see AC on them but aren't too bad with DC ripple and actually recharge faster with it as it acts like a pulse charge

T1 Terry
Green but want to learn

User avatar
coulomb
Site Admin
Posts: 3762
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: Making home EV ready

Post by coulomb » Wed, 23 Oct 2019, 13:24

T1 Terry wrote:
Wed, 23 Oct 2019, 11:14
So, if the inverter remained isolated and remained a floating system, line 1 and line 2 rather than neutral and positive, then the AC would not be seen on the vehicle negative side so it wouldn't trip the leakage to the chassis problem.
Yes, except I think that the high dV/dt combined with unavoidable capacitance to chassis would cause the earth leakage test to fail anyway.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

T1 Terry
Senior Member
Posts: 978
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Making home EV ready

Post by T1 Terry » Wed, 23 Oct 2019, 14:11

Hmmm...... could be a rather expensive "suck it and see" type of thing. If the inverter was not actually in the vehicle but rather plugged into an Anderson plug connected to the battery with a switching relay that isolated either the Anderson plug via a pr of contactors or the Ev's contactors if the Anderson plug was live, would that be enough isolation to avoid the EV earth leakage problem? Possible the EV's cell sensing harness and electronic components could still be affected .... all getting to hard ... pity, it was a good idea if it could be made to work without too much drama.

T1 Terry
Green but want to learn

Post Reply