Ingineer
Well-known member
It is not overblown. Clearly you do not possess enough engineering skills to even be qualified enough to make such a statement! If you did, you'd know that all common "GFI's" (GFCI) are NOT sensitive to DC leakage, and will NOT trip if you were to touch the EV's body (which is not grounded, as you don't seem to think it's needed, and in fact CANNOT be grounded if you are using a non-isolated design!) this could then conduct a large and fatal DC current through a person to something that is grounded. There would be no trip. Try it if you have doubts!valerun said:First of all, the danger is hugely overblown. If the system does not have any electrically live components accessible to the user (which it doesn't in any of these cars), there is no possibility for a shock. Isolated or not. Furthermore, if protected by a proper GFI, non-isolated is nearly as safe as an isolated design - even if there are live components accessible (again, not in these cars). If that wasn't the case, we would still all have inductive charging...
Risking some engine damage is not the same as risking LETHAL ELECTROCUTION!valerun said:Secondly, if everyone thought like this, we wouldn't have automotive performance aftermarket industry at all. For example, increasing pressure by 2x in one's turbo engine has risks. Same for putting in a NOX system in your car. Some engines blow up. Yet hundreds of thousands of people do that. It's all about trade-offs. Learn about the potential risks, weigh them against the benefits, and make an informed decision. Not based on hearsay, ideally...
Paranoia starts because of people taking unreasonable risks.valerun said:EMI/RFI - I have a couple of problems with the related statements in your post:
1. To uninitiated (i.e. not EE types), saying 'radiation' and 'bad consequences' without explaining is fear mongering. We have a pretty paranoid society here as it is.
If you use a non-isolated DC supply on the LEAF, you will have to intentionally NOT ground the LEAF's body to an RF sink (Earth ground) to make it not trip the leakage detection in the LEAF. This will turn the LEAF's body into a passive radiating antenna, re-radiating all the RFI from your poorly designed non-isolated converter. Note that I am more concerned about RFI, not EMI as you are claiming figures for. Apparently you don't know the difference, thus further proving your lack of qualifications to be working in this area. RFI is Radio-Frequency interference, EMI is Electo-Magnetic Interference. Milli-Gauss is measuring EMI NOT RFI!valerun said:2. 'A ton of RFI' is not a very scientific term. Have you measured it? Have you compared it with the background levels? Across the spectrum? We have measured radiated EMI from our non-grounded units - in a 0-10kHz band, at full 12kW power, at 3 feet, we measure less than 2 milli-gauss. Background level in my last 3 houses was 1.5-3 milli-gauss depending on how many lights are on...
You also haven't directly connected the AC power lines to the car while you are driving, but when using a non-isolated converter, this is exactly what you are doing! All modern EV's use fully shielded equipment and transmission lines, everything is double-insulated and there always is a fault detection circuit that will cut power to the EV is a fault is detected. There is NO current path to ground or the car's body in normal operation, and during any connection to an external power source referenced to ground, there is always a bonded connection to the car's body to ground, as well as leakage detection that will cut power in the event of a fault. You will have to effectively disable all this protection in order to use your device and create a dangerous condition.valerun said:3. Lastly, when your car moves, it's not grounded. Yet there is up to 80kW of AC power transferred around you (in a Leaf, up to 400kW in a Tesla S). And you are sitting IN the car (as opposed to you being in your HOUSE when you car charges). At freeway speeds, that AC also has much higher frequency than line - and radiated power is roughly proportional to the square of the frequency.
No regulatory agency on the planet would ever approve your device, and there are good reasons for that! It would be trivial to prove your negligence in a civil suit, and you'd be liable for damages if anything bad happened. The FCC would easily issue you a cease and desist as soon as your converter starts blasting RF interference all over.
If you want me to prove the danger and give you quantifiable measurements, I'll be happy to. Send me one of your converters and I'll give it a full and proper analysis for free and publish the results here. My concern is that people understand the risks before attempting dangerous scenarios. You are dangerously downplaying safety, and you clearly are not qualified to make such assessments. Your initial offering of your kit EVSE wasn't even going to include GFCI protection until you were called out on it! A $10 Chinese-made hair dryer even includes this essential safety device. That you could consider skimping on such a critical thing to save a few bucks shows where your sensibilities lie.valerun said:So let's try to not generalize things into non-quantitative categorical statements. Especially if you decide to launch a direct assault on other people like you did.
-Phil