“Why does a car like the Tesla still have old-style fuses?”
facebook.com
facebook.com
When you blow a circuit in a house, you have most likely massed up and plugged too many things into one outlet, or your device is drawing too much power. You learn from your mistake, unplug the device, and reset the circuit. If a car blows a circuit, then there is something seriously wrong with some component of the car or the wiring. You don't want the user to be able to reset the circuit, you want someone who knows what they are doing to figure our what went wrong.
I fix all of my own cars, and never once have I ever had a blow fuse "just because." There always was an underlining cause which needed to be addressed. Except once, when a 10A fuse was used in a circuit that required a 25A fuse, on a window motor.
So, basically, in a well functioning and well designed car a fuse will not fail just because. So why bother replacing a part that costs less then a cent with a part that costs several dollars. I hate this attitude, just because something CAN be complex, does NOT mean it has to be complex.
Take car modules for instance. It used to be that back in the day (the 90s) your headlights were operated by a mechanical relay. This relay was expected to fail at some point (though they very rarely actually did fail) and as a result this relay was installed in an easily accessible place. If it did fail, it would cost $10-15 to replace it plus 5 minutes of labor.
New cars nowadays have solid state switching modules to operate headlights, and everything else in your car. These components are not expected to fail, even though they often do. Don't believe me, talk to any mechanic. However, because they are not expected to fail, they are often installed deep inside the car. So, now, if your headlight module fails, it costs $400 in parts, and several hours of labor to fix the same problem. Progress?
Stop making things needlessly complicated. Blade fuses are a fine solution to a problem.
Yes. For the companies selling the parts and the repair services.
The resulting fire and shock hazard endanger not only the occupants but the general public and firefighters responding to calls.
I looked at a fixer upper house where the mains still had an ancient fuse box. Most of the high draw fuses had old pennies wedged in with a blown fuse. How the place was standing, I don't know!
http://www.eetimes.com/author.asp?section_id=4&doc_id=128389...
But, still, you need someone who understands what't being done.
Hell, when the HEADLIGHT BULB fails, which is something that IS expected to happen, it can be a pain in the ass to deal with on some cars. Some have sealed headlight units that have to be replaced when a bulb dies, costing hundreds of dollars.
Even ones where the bulb can be replaced sometimes make it ridiculously annoying. A low beam went out on my 2006 Honda CR-V. In theory it is easy to replace. Reach in, disconnect a cable, pull off a rubber seal, release a metal retaining clip, and the bulb can be pulled out. Reverse those steps with the new bulb, and you are done.
In reality, there is a bunch of stuff in the way, making it very hard to actually reach the damn thing. On one side, I was only able to do it by wedging my arm in, leaving it bruised and very sore for days. At least I could actually reach the low beams. While there, I tried to reach the high beams and could not get anywhere near them--I think when they go out I'm just going to give up on them.
Once you are in, it goes well, with the possible exception of the retaining clip. This is not unique to Honda, BTW. A lot of cars use the same kind of retaining clamp. Here's a drawing of how it works [1]. It is hinged at one end (the red circled part in that image), and it hooks under a piece of metal at the other end. It takes a fairly large force to make it hook there.
Here's the problem. You can't see this damn thing when you are trying to unlatch or latch it. The headlight housing is in the way. You can maybe see the end of it that is toward the bottom, but that is not helpful because they orient it so the hinge is on the bottom, and it is not the hinge end that people have trouble with.
Looking at YouTube videos of people explaining how to change headlight bulbs, and reading forums, it looks like it either latches easily, or it just won't latch and you are in for 30 minutes of flailing.
All the car maker would have to do is flip the orientation 180 degrees, so that the hinge is on top and the latching part on the bottom. Then you could see the latching part, and it least know what you are trying to do. The way they are now, you are going in blind.
I can understand that when they lay out components in the engine compartment, they have many constraints, so I can believe that all those things that made it hard for me to reach the bulb had to be there. I can believe that making an opening big enough for easy bulb changing might have made them have to make the car longer or something, that would have added a noticeable cost, and they could reasonably decide that it wasn't worth it.
The retaining clip orientation, though, is almost certainly completely arbitrary, and they just did not consider the effect on the ease of bulb changing (or deliberately picked the worst orientation to encourage people to visit a dealer instead of changing their own bulb).
[1] http://www.suzuki-forums.com/attachments/2g-2006-grand-vitar...
They're playing jigsaw with an engine compartment trying to get all the parts in in the most compact, but still workable, way possible.
It's sad when you think about it, because not only does it squeeze a lot of at-home maintenance out, which used to keep vehicles affordable, but now almost everyone discussing new vehicles discusses the maintenance costs.
Honestly, I'm not surprised the domestic car market essentially failed, because they're $40,000 printers. No one gives a crap about the up front price anymore, because they know they're getting screwed on the after-expenses so everyone's looking up how much their yearly maintenance expenses are going to be to make sure they don't end up paying more than the vehicle in a decade.
Should you need to do a high-beam, you may need to pull the battery on the (US) driver side, or the power steering reservoir on the other side. The PS bottle comes right off (no tools, IIRC), so it's only the one high beam on left side that's particularly tough.
it was only 20 years ago that a local company stopped making knife edge switches (the sort used as props in Frankenstein movies )
> These components are not expected to fail, even though they often do.
I think this is wrong. If you look at QA statistics, you will find that parts are getting _more_ reliable. The old relays were expected to fail becuase they _did_ fail. The new modules are not expected to fail, and this is indeed the case. They fail much less often, and of course this means that in the occasional cases where failure occurs, it will be difficult to replace or repair.
Overall reliability is increasing and to say otherwise would seem to be a glib misstatement or misunderstanding.
"So there’s a few things at play here. For context, I run the Product Security team at Tesla and I’m safety-trained on the HV systems - I’m also working hands-on with a small drive inverter on a hobby project right now.
First and foremost, our large drive unit pulls about 1000A at full load, and switching that with silicon is tough. We use a bank of custom IGBTs on each of the high/low sides of each of the 3 rotor phases in order to handle the power, and that’s with active fluid cooling. You can switch that much current with silicon but it ain’t cheap, and you’ll need either active cooling or a bunch of thermal mass if you want the thing to switch more than once. http://www.teslamotorsclub.com/attachment.php... is a decent pic, the object on the left is a single-phase switch, you can see 6x transistors laying flat at the front for one side of the phase (the other bank is behind).
Secondly, Model S is an AC induction motor so the current through the winding ramps up more-or-less linearly over time until the phase switches off (or changes direction). You’re at high power but you’re not switching the load at zero-crossing as you would in a resonant load such as a Tesla coil, instead you have to switch at an increasing current depending on how much power you want to the wheels. You now don’t just have to switch a lot of power, you have to switch it FAST so that the resistive losses in the FETs don’t blow out the power channel due to ohmic losses. Your switch is now not just big and bulky, it’s complicated (since you need an additional HV supply) and pretty sensitive to things like stray capacitances. On the previous pic the big black brick on top of the PCB is the capacitor that dumps into the IGBT gates to make them switch fast enough.
Finally, I believe there’s a regulatory issue. I think I’m right in saying that automotive standards around the world require that all electrical systems are fused, and considering that there’s multiple separate power rails it’s not inconceivable that an event could take place that leaves the HV drive rail powered on but kills the 12V accessory rail that powers a lot of the CAN systems. You could end up disabling your active fuse while the HV system is still energized, and considering the amperage our lithium packs can deliver (P85D draws up to 1.5kA) that’s not going to end well.
Woz: I would LOVE to put you under a Tesla NDA and then give you a _real_ tour of the vehicle - ping me at kpaget@teslamotors.com if you’re interested. I’m curious, do you still have one of my RFID cloners on your shelf somewhere?"
And link to the comment: https://www.facebook.com/stevewoz/posts/10153145090701282?co...
But I think something that people forget is that generally, fuses are there to protect wiring and fixtures (switches, connectors, etc) from fire. Preventing the device on the circuit that you, the user, cares about from completely melting down is just a nice bonus.
That's why building wiring codes generally spec fuses to cope with the capacity handled by wiring and switches, rather than the loads attached to it.
EDIT: In any case you NEVER want the car to silently and automatically try to "re-set" a tripped breaker, surely. You want this kind of fault to present itself noisily and obviously; it's a precursor to a potentially dangerous condition.
Any automatic re-set will have to factor in some cool-down time for the wiring in between attempts... what is the temperature of an overheating pair of conductors in the wiring loom when the breaker tripped? What rate of heat dissipation is there allowing them to cool down again? A wire which has experienced overheating will have a different (higher) resistance after the short-circuit event. Even if the wire isn't permenantly damaged, the temporary increase in temperature will still guarantee a momentarily higher resistance. Will the wire still have low enough resistance to trip the breaker again when the short is applied again?
After reading your comment it did make me wonder as I plugged our coffee machine this morning how the design substantial chunk of plastic and metal I was holding was influenced by the aftermath of WW2.
Now of course it was "overengineered" by the standards of the time, to compensate for our ludicrous, dangerous, and still-legally-required "ring main" system, which as you say was due to aftermath-of-WW2 copper shortages. But the plug design is the part we got right.
Evidence: Socket: http://i.imgur.com/zwRYC25.jpg Mask: http://i.imgur.com/b0RAZ9k.jpg
What makes you think ring circuits are dangerous? Incompetent installation?
We really, really don't. The most charitable of us manage to appreciate it as a quaint example of British discomfort, like those sinks with separate boiling/freezing water taps at opposite corners.
Woz is proposing that Tesla take well proven, very reliable, simple, inexpensive fuses that are very rarely even seen by customers and replace them with complex integrated circuits and software that will need a lot of testing, almost certainly be more expensive, and almost certainly be less reliable. For what? Very little benefit for a very small subset of customers.
KISW! Sometimes a small length of wire in a plastic holder is the best answer to a problem.
EDIT: I was thinking of the 12V accessory fuses. In light of the reply from Tesla, I can imagine where if the drivetrain fuse blew you might not want an average user replacing it ... and I imagine that for something like that, it probably already throws a code that gives you an idea of what's wrong.
Suddenly your "inexpensive" adc had become a $200 ecu that merely tells the driver wheat part of the car isn't functioning, right before they check the most common cause of the problem.
That is one of the downsides of ev's whist an electric shock from mains is unpleasant its unlikely to be fatal higher voltage and amperage shocks are very nasty.
My father the mechanic used to like to tell me about an exception to the rule as an object lesson about the trade off between risk and reliability. Back in the days of breaker ignitions the ignition circuit was almost never fused. A blown fuse in the ignition circuit could strand people out in the middle of nowhere. The extra risk was acceptable to eliminate the situation where the fuse blew when the ignition might of been able to continue to work in some sort of degraded mode. It was OK that that degraded mode might involve smoke and flame.
I don't believe in smartness for smartness' sake. For a fuse to fail you need 1) wrong fuse 2) physical damage to the fuse (not sure if there is a case in which this would cause it to not break the circuit), different set of laws of physics.
http://en.wikipedia.org/wiki/Resettable_fuse They make these in automotive format, or you can go to Digikey, get a PTC of the current you like, and solder it onto a blown fuse.
PTCs have the advantage of being straight replacement, and if you want, you can add a Hall sensor to see if current is flowing without creating paths to ground or other problems.
Anyone knows how to add this answer to Woz's thread? The "awesome power of social networks" isn't that awesome, since I can't answer on the thread linked...
Blade fuses are 100 for a dollar. PTCs are a handful for a dollar. Infineon ProFETs are a couple dollars per.
When's the last time you blew a fuse on your car? For most of us, it's never. For those of us who have, there was almost certainly an electrical fault that was the root cause.
I say "actually" because I presume it's possible to have a false failure now and again - there may be a nonzero number of times in which the fuse just pops, that fuses just wear out.
Fuses will, barring ludicrous overload (putting a 12V fuse in a 10kV DC circuit), fail open - a definite positive for a safety measure.
Using fuses makes sense in a lot of applications because they are (as noted above) safer, cheaper, and more reliable. It probably doesn't hurt that a fuse is easier to diagnose; control circuits can intermittently fail but an "intermittently blown" fuse would be an odd thing indeed.
Here's the real deal: a fuse costs cents, and a breaker costs dollars. Fuses rarely blow, so increasing cost two orders of magnitude for an extremely minor benefit doesn't make a lot of cents (har har har). Even Tesla cares about cost control - imagine if the whole car were two orders of magnitude more expensive. I expect that a feedback mechanism would probably actually be cheaper than breakers, since it could be a single, cheap chip rather than expensive mechanical devices.
Solid-state circuit breakers do exist, and would solve the vibration problem, but some quick Googling didn't turn up anything that looked at all inexpensive.
They are self-resetting, but make a fairly obvious noise when they reset.
Or rather, this is the internet at work, not social media. The only thing social media has done is reduce the time between question and answer. And that's impressive, but it's not the fundamental shift the internet itself was.
The high-power system uses fuses because, if things overload there, you want power disconnected reliably and you want it to stay off. Lithium batteries do not handle overloads well. Not at all.
And apparently some of them have 160 or 8192 baud serial on that connector, with really expensive cables, unless you're handy with an arduino or similar. Now that the truck is working, I'm not sure I need to do this... yet.
I think WOZ is suggesting an active current control that, while doable, is probably too risky (and you really don't want to blow the expensive parts)
If you have something drawing too much current in a plane, and so it is a fire risk, it still might be overall safer to continue to use it. For instance, if it is something involved in lowering the landing gear you might decide it is safer to risk a small fire in order to get the gear down than to do a gear up landing. Something resettable, like a circuit breaker, therefore makes a lot of sense in a plane.
With a car you are almost never in that kind of situation.
For an important item, I'll reset a breaker once in flight. Nothing electrical is that critical, and I carry portable AHRS, GPS, and handheld VHF radio.
Few things are worse than fire, so if I have any concern about the electricals, they can all stay off and I'll extend the gear manually with the crank system. Other planes have gear that free falls or has an N2 bottle to blow the gear down.
I don't think any civilian aircraft have had fatalities from a simple gear-up landing.
Really... have fun. I rather check the fuses and trace the wiring for that part using a meter.
Marc Rogers[0] Woz - Kristin[1] cant post to your thread but here is her answer:
So there’s a few things at play here. For context, I run the Product Security team at Tesla and I’m safety-trained on the HV systems - I’m also working hands-on with a small drive inverter on a hobby project right now.
First and foremost, our large drive unit pulls about 1000A at full load, and switching that with silicon is tough. We use a bank of custom IGBTs on each of the high/low sides of each of the 3 rotor phases in order to handle the power, and that’s with active fluid cooling. You can switch that much current with silicon but it ain’t cheap, and you’ll need either active cooling or a bunch of thermal mass if you want the thing to switch more than once. http://www.teslamotorsclub.com/attachment.php... is a decent pic, the object on the left is a single-phase switch, you can see 6x transistors laying flat at the front for one side of the phase (the other bank is behind).
Secondly, Model S is an AC induction motor so the current through the winding ramps up more-or-less linearly over time until the phase switches off (or changes direction). You’re at high power but you’re not switching the load at zero-crossing as you would in a resonant load such as a Tesla coil, instead you have to switch at an increasing current depending on how much power you want to the wheels. You now don’t just have to switch a lot of power, you have to switch it FAST so that the resistive losses in the FETs don’t blow out the power channel due to ohmic losses. Your switch is now not just big and bulky, it’s complicated (since you need an additional HV supply) and pretty sensitive to things like stray capacitances. On the previous pic the big black brick on top of the PCB is the capacitor that dumps into the IGBT gates to make them switch fast enough.
Finally, I believe there’s a regulatory issue. I think I’m right in saying that automotive standards around the world require that all electrical systems are fused, and considering that there’s multiple separate power rails it’s not inconceivable that an event could take place that leaves the HV drive rail powered on but kills the 12V accessory rail that powers a lot of the CAN systems. You could end up disabling your active fuse while the HV system is still energized, and considering the amperage our lithium packs can deliver (P85D draws up to 1.5kA) that’s not going to end well.
Woz: I would LOVE to put you under a Tesla NDA and then give you a _real_ tour of the vehicle - ping me at kpaget@teslamotors.com if you’re interested. I’m curious, do you still have one of my RFID cloners on your shelf somewhere? [2]
[0]https://www.facebook.com/marcrogers?fref=ufi
[1] https://www.facebook.com/kristin.paget
[2]https://www.facebook.com/stevewoz/posts/10153145090701282?co...