General > General Technical Chat

The US electrical system

<< < (21/29) > >>

richard.cs:

--- Quote from: Monkeh on June 29, 2020, 02:42:20 pm ---Now, I could be wrong, but as far as I'm aware most electric cars do not connect the charging CPC to the chassis. That coupled with plastic handles and extremely high quality paint finishes leaves me entirely unconcerned about taking a belt off a car. Oh, that and like most people I walk around on a big sheet of rubber, not my bare feet.

--- End quote ---

Unfortunately most electric cars do have that CPC to chassis connection. I suspect only because it makes EMC design easier, or reduces costs over double/reinforced insulation, or even just because of an "earth everything" mentality. I agree there is no sound technical reason why it is required.

Personally I think plastic handles, paint and shoes make the risk small compared to other things (outside taps are the classic worry, often being associated with wet skin and bare feet). I don't worry at all when mine is plugged into a known TNC-S supply.


--- Quote from: Monkeh on June 29, 2020, 02:42:20 pm ---I do, however, think we should never have used TN-C-S as an excuse not to properly control potential at the installation. It's a perfectly good system for ensuring a nice low impedance fault path for currents in normal operation without using lots of extra copper (or lead, or alu), but a poor excuse for not having good earthing at the property (and the American method of whacking a rod in quickly doesn't cut it).

--- End quote ---

Fundamentally it is a poor system for that reason, and if we were starting from scratch today this is not where we'd end up. Personally I'd be inclined to go for something with an impedance-earthed neutral and then we could distribute a low-current CPC at negligible cost. This would be somewhat similar to UK HV distribution. Such a system allows RCD-like fault protection with much greater reliability than common 30 mA RCDs due to having a few tens of amps of fault current to play with, and gives negligible rise of earth potential during a fault.

From where we are today though I think that will never happen, we'd have to return to TNS or more likely give up on supplier earthing and TT everything - and that has problems with unreliable RCDs.

tom66:
I think the reason for the CPC being earthed to battery negative is the sheer convenience of the communication interface of the pilot signal and cable detection. And for vehicles with DC-CCS, then they also need a communication bus with the charging station.  So you can keep all of this logic on the non-insulated side which reduces costs and complexity.

Of course the high voltage battery is always isolated from the rest of the chassis and this system is continuously monitored,  isolation drifting out of the specification will flag a service warning or disable drive.

Monkeh:

--- Quote from: richard.cs on June 29, 2020, 02:59:38 pm ---From where we are today though I think that will never happen, we'd have to return to TNS or more likely give up on supplier earthing and TT everything - and that has problems with unreliable RCDs.

--- End quote ---

The reliability (or lack thereof) of RCDs is why I'm not a fan of TT systems. Although I suspect a lot of that has to do with the price point of them usually being used for secondary protection. Quality of protective devices is a problem which sorely needs addressing (more so than 'ooh, AFDDs sound good', and 'oh no, plastic boxes are bad in fires').


--- Quote from: tom66 on June 29, 2020, 03:06:33 pm ---reduces costs and complexity.

--- End quote ---

Increases cost and complexity to the end user for the sake of the profit margin of the vehicle manufacturer, you mean. They know full well what earthing systems are in place, just like the manufacturers of bathroom fans know full well they'll be installed on 6A circuits.

tom66:
Well the costs only come to users for countries where PME is not standard.  I don't know what the situation is in the rest of Europe but I imagine this would have been a consideration.

It's worth noting that in the Type2 standard the CP/PP signals are referenced to PE, and a secondary pain happens should there be a short that leads to the ~10mA DC current from these saturating a downstream RCD's current transformer, preventing it from operating correctly.

What this really is showing is how outdated our electrical infrastructure is. Frankly given all of this I'm shocked (no pun intended) that "granny chargers" (i.e. portable EVSE) are still permitted and sold with cars, despite the well known fire hazards.

Monkeh:

--- Quote from: tom66 on June 29, 2020, 03:12:11 pm ---It's worth noting that in the Type2 standard the CP/PP signals are referenced to PE, and a secondary pain happens should there be a short that leads to the ~10mA DC current from these saturating a downstream RCD's current transformer, preventing it from operating correctly.

--- End quote ---

Ah, yes, because we still use type AC RCDs in blissful ignorance of the loads around us changing. Well, those who even have RCDs, because spending money on your electrical installation is daft, "It's worked fine for all these years".

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod