DIY Electric Car Forums banner

Staying alive - electrical safety.

10K views 45 replies 14 participants last post by  DanGT86 
#1 ·
Though there are tons of topics about safety in a crash, safety for first responders, fuses, etc... there are actually strangely few threads about safety while building an EV.

So what things does one need to do (or not do) in order to not to kill themselves?

I was reminded of the power hidden in our batteries when my bus bar turned 90 degrees while tightening a bold and a shower of sparks and molten copper shot everywhere. I went back to my standard "cover everythign with electical tape" idea after that... I'd clearly become way too comfortable.

Things like:
Can you touch a positive as long as you are also not touching another battery terminal? Do I ever need to worry about touching ONE point in the high voltage loop? What can and can't you touch if you'd like to stay alive? Are the positive and negative terminals just as 'deadly' when it comes to human touch? Etc///
Most 'centralized' BMS wires, as far as I can tell, carry pack voltage -- how do you keep the dozens of wires that all need to be on terminals 2mm from each other from touching? If they did, what's the real risk?
We shoudl 'wrap' our wrenches and tools in electical tape -- but does that really stop 160V+ or is it just for our own 'feel good'? Is standard "automotive" wire acceptable for low amperage, pack-voltage connections?
Etc etc etc...

It just seems this 'basic' safety info is sparse on this forum... might be nice to toss it all in one place.
 
See less See more
#2 ·
The last really big pack I worked on. That is 3p51s of 200Ah cells.


All the spanners and other tools were insulated.
Plywood boards were placed over the pack so that only the area we worked on was exposed.
The pack was tested with many fuse links to split the pack into manageable sections.
As the cells were linked in groups of three parallel cells to get 600Ah we needed to make sure the bus bars were installed correctly. This was always done with two people, one fitting and one checking, to make sure the bars went across the right terminals.

When it was installed in the car the same process was used and each row of cells ended with a fuse link which was only put in afterwards. That meant that during installation there was only the voltage of one row.
Before connection of any cells two of us separately checked all the cells were correctly orientated relative to each other and only when we were both satisfied did we get spanners. Then it was two other people to make sure each connection was still right.
 
#3 ·
I heatshrinked all my tools after my first such incident (just this month, after nearly 2 years of working/driving).

Yes, you can touch a single point in the pack, not two. Don't touch one unless you need to, and when connecting things be sure to touch metal to metal first.

Try to size busbars or other connectors so they can't swing and hit another...btw, how did yours turn? busbars are usually connected to two points.

Put insulation on anything where a short could occur, such as a metal strap latch that rests between batteries.

Use mechanical disconnects, and use them whenever you're working around the pack.

Keep the pack covered, in a box or at least by a tarp or similar to keep things clean and safe.
 
#5 ·
Good ideas, all. Any others?

Try to size busbars or other connectors so they can't swing and hit another...btw, how did yours turn? busbars are usually connected to two points.
I put the (first) busbar spanning to the next cell , started screwing in the bolt on cell and the bus bar must have caught as I was turning. It turned with the screw 90 degrees before I realized "hey, that's not a good idea" and it started a little spark show. Given that's just the potential from ONE cell, and I've got 48 it really drove home how much deadly power we are messing around with here.

I've since gone through about 3 rolls of electrical tape. ;)


Yes, you can touch a single point in the pack, not two. Don't touch one unless you need to, and when connecting things be sure to touch metal to metal first.
At a high enough voltage just touching one terminal can shock you - if not electrocute (that is, kill) you - because it can force enough of a current flow through the atmosphere, your clothes, shoes, etc., for you to feel it. In general, we don't work with voltages in an EV high enough to do more than barely tingle if you were to touch one terminal (and polarity, by the way, is irrelevant).

That said, you still don't want to touch *any* terminal which is at traction pack potential just in case there is already a path to "ground" (ie - the vehicle chassis) somewhere else.
So that brings up the question that kind of was the impetus for this question. We aren't supposed to touch the terminals or anything at traction pack voltages (for me, 150V)... yet we do, in fact, need to assemle these things. How are you connecting the last few batteries if you aren't touching them? Ya know?
 
#4 · (Edited)
...
Can you touch a positive as long as you are also not touching another battery terminal? Do I ever need to worry about touching ONE point in the high voltage loop?
At a high enough voltage just touching one terminal can shock you - if not electrocute (that is, kill) you - because it can force enough of a current flow through the atmosphere, your clothes, shoes, etc., for you to feel it. In general, we don't work with voltages in an EV high enough to do more than barely tingle if you were to touch one terminal (and polarity, by the way, is irrelevant).

That said, you still don't want to touch *any* terminal which is at traction pack potential just in case there is already a path to "ground" (ie - the vehicle chassis) somewhere else. The two most pervasive cause of such current leaks are dust from the brushes and condensation/spilled electrolyte on top of the batteries.

Use insulated tools and, where possible, try to follow the old electrician's rule of keeping one hand in your pocket when working on even supposedly dead traction pack circuits.

...Most 'centralized' BMS wires, as far as I can tell, carry pack voltage -- how do you keep the dozens of wires that all need to be on terminals 2mm from each other from touching? If they did, what's the real risk?
Assuming the wires have an insulation rating appropriate for the total pack voltage, they can all touch each other.

A centralized BMS that only monitors cell voltage should use - in my *opinion* - teflon insulated wire, which can be purchased for a reasonable price as surplus. If the BMS also actively balances the cells with a shunt then you should add a fuse to the cell-terminal end (said fuse needs to be capable of interrupting the full pack voltage, btw, which can be a tall order for a low-amperage fuse combined with a high pack voltage.

We shoudl 'wrap' our wrenches and tools in electical tape -- but does that really stop 160V+ or is it just for our own 'feel good'? Is standard "automotive" wire acceptable for low amperage, pack-voltage connections?
Any tape that says it is for electrical applications must actually pass some dielectric strength and flammability tests so, yes, it will stop 160V+ (a typical rating for vinyl electrical tape is 600V per single layer).

Standard "automotive" wire is generally *not* acceptable for carrying pack voltage. Anything rated for "AC wiring" use is going to have much tougher insulation, such as THHN, THWN, MTW, etc. Alternatively, the teflon insulated wire above is my number one choice.
 
#9 ·
...Are they really required, or are you probably safe at 150V using the 1-hand trick?
Depends. Seb shocks himself all the time working on the battery pack in the Porsche 911 because those new Helwig brushes seem to be throwing off a huge amount of carbon dust. He neither wears gloves nor puts one hand in his pocket and suffers the consequences. Maybe one day he'll depolarize his atrioventricular node and that will be the end of tinkering on that infernal Headway pack for Seb. :D
 
#11 ·
If you aren't going to wear gloves, at least cultivate the habit of checking continuity between the pack and chassis ground each time to ensure it is an indefinitely high resistance before doing any work on the pack. Check that the point is a good ground by checking continuity between it and another ground point.

Also plan out step by step exactly what you are going to do and how you are going to do it, thinking about possibilities for mishaps, then execute those steps carefully. Many times it is after things become routine that we become complacent and make mistakes due to lack of thought.
 
#15 ·
Along that line, I use simple vinyl gloves on my hands with common cloth gloves over them. Plenty of di-electric strength and easy to work in.

For more serious things, I have an issue pair of 480 volt certified {but out of date} linemans gloves. Cloth inner, rubber middle and leather over gloves, all kept in the mesh bag.

My only real shocks on an EV have come from supposedly dead circuits.....I assume they were powered by those caps in the controller. They were traction cables to chassis ground with the pack disconnect off.

Just a "tickle" really.

Miz
 
#18 ·
brace yourselves there is a potentially stupid question coming.

I had thought that people where relatively well insulated, and i had assumed that made us relativity safe at low voltages, and that 3.7v even at high currents couldn't hurt us.

if this isn't the case that's cool, if this is the case what voltage is it safe to work with. Could you break the pack up into a safe voltage, then use simpler Anderson connectors to conect the harmless low voltage packs to create the high voltage pack? this way you aren't fiddling with bus bars and a spanner on a 300V pack.
 
#19 ·
brace yourselves there is a potentially stupid question coming.

I had thought that people where relatively well insulated, and i had assumed that made us relativity safe at low voltages, and that 3.7v even at high currents couldn't hurt us.
That is generally correct. The danger is that you contact different parts of the pack with larger voltage difference, or your pack is not isolated from the chassis and you touch a terminal on the pack while in contact with the chassis. With the potential for thousands of amps to flow, you don't want to explore what amount of voltage is too high though, and skin conductivity is increased if your hands are sweaty.:D I suppose you could use large Anderson connectors rated for your magnitude of discharge current, but they are another interface so another power loss.
 
#21 ·
I believe that the fire department is more concerned about electric vehicles than the state DOT or DMV here in Arizona.

I have had several conversations with firemen at Hot Rod functions and they all tell the same story. Inconsistent construction among "home builts"

Factory cars are somewhat engineered. They are more numerous and can be trained for easier.

Home builts are generally all different. Other than all having batteries and electric motors, they are different from each other in component placement, battery mounting and safety features.

The fire Department teaches their personnel to cut the main pack cables if any smoke exists. They use standard safety gear,cable cutters and rubber gloves.

They are taught to not spray on anything if possible, use CO2 as second choice, foam as third choice then to simply stand back and wait for it to stop reacting.

Their biggest complaint is no standard disconnect placement. The NHRA requires one on the rear/center of the cars external body work. Clearly marked and RED. I would add it to be up higher than bumper height and out of the "crash" zones for street car use.

Or maybe an extra contactor mounted mid-pack. Have it connected to an inertial switch like all cars have when using high pressure fuel injection. When a large enough crash happens, it cuts the pack circuit, but is instantly reset-able if it is safe to do so {it seems to work on ICE vehicles}.

Miz
 
#22 ·
Their biggest complaint is no standard disconnect placement. The NHRA requires one on the rear/center of the cars external body work. Clearly marked and RED. I would add it to be up higher than bumper height and out of the "crash" zones for street car use.
My concern with an externally visible disconnect is "vandalism" if that is what it could be called. It would be readily accessible to just reach out and kill it as someone walks by. How do you place such a device where it can be used in an emergency but passers-by won't know it is there?

Or maybe an extra contactor mounted mid-pack. Have it connected to an inertial switch like all cars have when using high pressure fuel injection. When a large enough crash happens, it cuts the pack circuit, but is instantly reset-able if it is safe to do so {it seems to work on ICE vehicles}.
I remember seeing on a Tesla Roadster that there was a loop of small wire, under the hood IIRC, that had a label that said something to the effect of "cut wire for pack disconnect." I assume it was for such a contactor.

I guess for a mid pack contactor to be a reliable disconnect it would have to be energized by the pack, maybe with its own DC-DC, so that when the inertia switch tripped it would open up. The problem is, how to initiate the initial closing of the contactor? Are there contactors which have a manual enable option but which need power to remain energized? Maybe use a shorting bar across the main posts to get started.

As I think about this more one could have several places through out the vehicle where the coil wire could go with loops which could be cut in an emergency, much like on the Tesla Roadster, in addition to the inertia switch. For manual disconnect just pull out the fuse operating the contactor.
 
#25 ·
Mine looks like this. It's pretty much a a worldwide standard that a round red button stops stuff. Mine's mounted on the left of the instrument panel, easily to the driver or from outside (once door is opened).

I like what was mentioned above though, about cutting wires with isolated tools, etc. People seem to freak out about it alot, but I'm pretty sure most 1st responders aren't cutting cars open with their bare hands.

 
#26 ·
Ah, yes, I've seen a similar button. However, in order for the first responders to be absolutely sure, they need a button on the outside.

The first responders are actually being trained on how to cut The Big Wires, and when to just let the car burn. I was a volunteer fireman for 6 years, and we got training in just that sorta stuff. Lots of fun. :)

I guess I'll go mouse around and find out what the NTSB and others are looking for us to have.

Cheers,
Peter
 
#30 ·
After doing some research, I have found some interesting things.....

Most OEM disconnect switches or controls are orange, not red.

Most OEM built cars have the disconnect switches or controls in the trunk areas.

Many have external markings along the HV current cable paths.

Most all have a combined system to shut off low and high voltage as well as gasoline flow on a large impact.

http://evsafetytraining.org/Resources/Auto-Manufacturer-Resources/Tesla-Motors.aspx

This link leads to almost all electric or hybrid car safety systems in PDF form.

Miz
 
#31 ·
So this thread seems to have taking a turn into the typical 'safety' discussions we have here -- first responder, fuses, cutoffs, etc.

Does that mean we've exhausted all discussion on how to BUILD your pack without killing yourself? Just put on some nitrile gloves and hope? That doesn't sound right...
 
#33 ·
So this thread seems to have taking a turn into the typical 'safety' discussions we have here -- first responder, fuses, cutoffs, etc.

Does that mean we've exhausted all discussion on how to BUILD your pack without killing yourself? Just put on some nitrile gloves and hope? That doesn't sound right...
There were quite a few other suggestions beyond the use of gloves. I suggest you re-read.
 
#32 ·
Hi Lowcrawler

I have the bits for my pack but I haven't yet assembled it

My pack is 88 Headway 16Ah cells 44S2P (little pack for short range)
It is split into two halves with an Anderson connector
The full pack voltage goes to an old forklift connector thing

Assembly
Cells in support structure
square copper connectors bolted onto the bottom, the rest of the bottom structure is screwed on - pretty safe so far

square copper connectors bolted onto the top - definitely need safety glasses for this step - don't think I can electrocute myself but a carelessness could result in a spark
Rest of the top structure is screwed on

The battery box is now compete and safe until I plug in the Anderson connector

I can make the car safe by removing the Anderson connector and doubly safe by undoing the fork lift connector
 
#35 ·
I understand the frustration, but seriously - most of the precautions are what should be common sense (yeah, yeah, I know it *isn't* common. :)

I'll try to summarize:

Work rules:
1) Don't wear jewelry.
2) Wear anti-conductive gloves.
3) Use tools that have anti-conductive coatings.
4) Cover areas you aren't working on with non-conductive covers.
5) Use one hand at a time if possible.

Battery and Connector rules:
1) Always have a fuse inline with your pack.
1a) More fuses are fine as long as they don't have too much resistance.
2) Always have a safety disconnect.
3) Invest in an intertial breaker to shut off your contactor.
4) Make sure your batteries are someplace that doesn't get wet.
5) Make sure your batteries are someplace that people can't easily get into.

I'm sure there are more, but those seem to be pretty good ones to me. :)

Cheers,
Peter
 
#36 ·
There have been so many common sense but good tips on pack building, it could really fill a book.

For my own part, After I receive my cells, I plan to equalize them, then assemble them into 5 cell units. These 5 cell units fit into the space of 1-6 volt golf cart battery. I must make 6 of them, then 2- 4 cell units, giving me 38 cells on a 120 Volt pack.

My plan is to use 1/8" aluminum plates at each end with 3- 3/8" plastic shipping bands around them. This stuff is strong. I use it every day and it still amazing at the strength.

At this point I plan to put the cell electrical straps on the 5 cell units, then install into car before installing the inter connecting straps. This gives me only about 18 volts to work around during the bulk of the hand work.

After all units are installed in the battery boxes, I can inter connect them, still leaving the contactor, main disconnect and inter-pack disconnect open {off}.

If I have any reservations, I can simply lay an old blanket over the cells in the battery box and work through a hole, like surgeons do when they operate {except the majority dont use an old blanket}

If you want to complicate things, lets talk about adding capacitors to the pack and charging them....Batteries are pretty straight forward if you remember that they are a little more than an old flash light.

I see no real reason to get crazy if routine precautions are observed. Then again, years ago, I used to work as a powder monkey and you develop decent safety habits....static grounds...non sparking tools...Etc.

Miz
 
#37 ·
Tess said:
Standard "automotive" wire is generally *not* acceptable for carrying pack voltage. Anything rated for "AC wiring" use is going to have much tougher insulation, such as THHN, THWN, MTW, etc. Alternatively, the teflon insulated wire above is my number one choice.
This page recommends auto wire and specifically says not to use thhn.... thoughts?


http://www.austinev.org/evinfo/build/eva-safetyfirst.html
 
#39 ·
This page recommends auto wire and specifically says not to use thhn.... thoughts?


http://www.austinev.org/evinfo/build/eva-safetyfirst.html
There are lots of good tips on that site, but the author of that page did not specify *why* the insulation on THHN, etc. might eventually crack; my suspicion is he (she?) is concerned that extreme vibration or repeated flexing of the wire will crack the insulation.*

I suppose that is a valid concern, but my counterargument would be that you can take steps to minimize the amount of flexing, abrasion, etc., that your wiring must endure, but you can't do anything to minimize the voltage it will have to carry. :D


* - more specifically, it is the nylon outer jacket on THHN wire that tends to get damaged from mechanical abuse. Indeed, the nylon overcoat is routinely damaged when pulling the wire through conduits (aiding such pulling is the purpose of the overcoat) but as long as the main insulating layer of PVC underneath is unharmed the installation is code-acceptable.
 
#38 · (Edited)
Is the any concerns about dangerous arc faults at DIY EV power levels?

My research reading shows 120vac is the lower end of where you cease to have to concern yourself with arc faults or massive explosion type stuff... but i'm having no luck finding comforting info on DC stuff. (not because I'm finding bad stuff -- it's just the main computational method of arc fault protections for AC involve things DC doesn't have.

(Better to be over safe thab under - never assume anything)
 
#40 ·
Thanks for the quick answer. Any thoughts on Arc Faults or other electrical explosions at all?


I also wonder about fuses -- the Soliton Jr in my car can peak to 600A... yet I was recommended to get a 400A fuse. First, this makes no sense to me... but when you are in the theoretical stage and haven't seen dozens of videos of firey electrical explosions you just take it as a given that what other say is okay with very little need for 'supporting evidence' or understanding it yourself. From what I can guess, it appears it's simply a 'time delay' type thing... This "fuse opening time" section on this page seems to support that theory.

Given that's the case... and in an arc-fault or other plasma-inducing situation (think: car accident) wouldn't you want a fuse that popped IMMEDIATELY once over a level you know your car should NEVER pull?
For instance --- have your normal 400A A30QS as everyone recommends, (which if what I read is correct (please correct me if I'm wrong), can pull up to 1500A for a tenth of a second (times 150V is a heck of a lot of energy... no?) ) and then something like a fuse that blows IMMEDIATELY if seeing anything over 700A (for a 600A soliton jr-based system)? Does a fuse like that exist? Is it a worthwhile concept?

Given most of us have two main fuses anyway (a main and then a backup 'mid-pack' one... why not try to meet both these goals? Any how?)
 
#41 ·
Thanks for the quick answer. Any thoughts on Arc Faults or other electrical explosions at all?
This is too open-ended of a question; please be more specific.

I also wonder about fuses -- the Soliton Jr in my car can peak to 600A... yet I was recommended to get a 400A fuse. First, this makes no sense to me.......
I give a very brief explanation as to why I recommend a 400A fuse in the battery circuit on page 7 of our manual. To recap for those too lazy to read it: the fuse is there to protect the insulation on the wiring from melting due to prolonged overcurrent, it is not there to protect the controller or motor. For example, this datasheet shows a 300V rated 400A Class-T fuse from Mersen will pass 1000A for 10 seconds before blowing; a 600V version of that fuse will pass 1000A for *50* seconds before blowing.

NOW do you see why a 400A fuse on the battery side is fine for even a Soliton1 installation? If you are pulling 1000A of battery current for 10 seconds then I guarantee you just cooked your motor, and unless you have 4/0 cable, you probably cooked the high power wiring, too.

Given that's the case... and in an arc-fault or other plasma-inducing situation (think: car accident) wouldn't you want a fuse that popped IMMEDIATELY once over a level you know your car should NEVER pull?
There is no such fuse; only an electronic circuit can respond to an overcurrent that fast.
 
#42 ·
You need to consider the source of arc faults or explosions. An arc fault would mean a spark gap that is insufficient for the voltage you're using, so use stuff that's rated for your voltage. An explosion requires explosive materials, typically gasses, so don't have sparks around flooded batteries. Also don't tax your electronics.

For the fuse, most use a slow blow type fuse, that means it's designed to allow a variable amount of overcurrent depending on the time. In a full on short it would blow pretty quick.

There are quick blow fuses, but they would probably be expensive, and say you have one of those sparky events, do you really want to have to replace a very expensive fuse vs just feeling embarrassed?
 
#43 ·
You need to consider the source of arc faults or explosions. An arc fault would mean a spark gap that is insufficient for the voltage you're using, so use stuff that's rated for your voltage. An explosion requires explosive materials, typically gasses, so don't have sparks around flooded batteries.
It's my understanding the in an arc fault situation you actually end up vaporizing your 'wire' into an electrically conductive plasma at insanely high temps... it's not a hydrogen explosion, that's tame by comparison. (again, just going by the research I'm doing... looking at gaining insight into exactly why/how things end up safe or not)

A standard 480V three-phase faulting at 20kA is roughly equivalent to .8 pounds of TNT going off. I'm just looking to see if we are in the 120V/125kVA tranformer range (where it doesn't matter) or in the "TNT explosion" area...

This is too open-ended of a question; please be more specific.
Specifically, arc faults in industrial/commercial wiring are extremely dangerous. But in <120V installations with less than 125KVA transformers, it's really not an issue because there isn't enough energy to really create an arc fault or blast that really does damage enough that the explosion/blast causes much damage... thus leaving electrocution/shock as the only real concern.

... but that's AC... not DC. I was wondering if you knew of any guidelines about DC stuff in this regard. I haven't been able to find any yet.


I give a very brief explanation as to why I recommend a 400A fuse in the battery circuit on page 7 of our manual.
That was actually the first thing I ever read that gave me reasonable understanding of the issue back in the day. Drives home the idea of 'fuses protect the wires'. Well done.

There is no such fuse; only an electronic circuit can respond to an overcurrent that fast.
Bummer dude. Just to verify my understanding, however... if one did exist, it would be a reasonable idea, right?

Ziggy -- I think the A30QS is labeled 'quick blow' ... right? am I mis-reading something?
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top