The International Tin Whisker Symposium

I'm surprised that a whisker like that could carry enough current to short out the contacts without melting. Neat stuff.
 
Tin whiskers and solder balls - it's the end of the world as we know it.

--

When the powers-that-be decided to remove lead from solder, tin whiskers became a big problem.
 

Attachments

  • TinWhiskers460x276.jpg
    TinWhiskers460x276.jpg
    30.3 KB · Views: 35
Last edited:
Just to be fair.

Unintended acceleration was invented by James Watt, not Audi or Toyota.


Guess what happens when the belt breaks.
 
Last edited:
One of the reasons I don't trust electronics is that I am an electrical engineer, and I know that stuff can go wrong!
 
I'm surprised that a whisker like that could carry enough current to short out the contacts without melting. Neat stuff.
Not all connections require a lot of current; 10 megohm of resistance in the wrong spot can be just as upsetting as a dead short. Tin whiskers are bad news in the space world, and the rise of lead-based solder is a big headache (since it has more tin). Vendors don't want the hassles involved, and you're really got to ride herd on them.

Ron Wanttaja
 
I don't know why they use this technology in the first place. Why can't they use a freaking cable? There are only two failures associated with the cable, either it sticks and the throttle stays constant, or it breaks and the throttle spring closes the throttle automatically. Neither of those involve the throttle going WFO all of a sudden.
 
I don't know why they use this technology in the first place. Why can't they use a freaking cable? There are only two failures associated with the cable, either it sticks and the throttle stays constant, or it breaks and the throttle spring closes the throttle automatically. Neither of those involve the throttle going WFO all of a sudden.

Except for the times when it does go WFO.

Broken motor mounts, other things that I don't recall off the top of my head.

And the reason for the technology is to keep up with Government regulations for fuel economy and emissions. One can better optimize transmission shift schedules and stuff for fuel economy if you can better manipulate the actual torque put out by the engine. And, having control over how and when the throttle moves, you can better control the fuel / air mixing and avoid misfires.
 
Not all connections require a lot of current; 10 megohm of resistance in the wrong spot can be just as upsetting as a dead short. Tin whiskers are bad news in the space world, and the rise of lead-based solder is a big headache (since it has more tin). Vendors don't want the hassles involved, and you're really got to ride herd on them.

Ron Wanttaja

Even worse, there are proposed EPA regulations that would require changes to the flux-cleaning solutions and processes. In the space world, uncleaned flux can also cause huge issues - the stuff EPA is talking about would put major hurt on some space-equipment manufacturers. Too onerous or expensive, and the manufacturing is likely to go offshore where the regulations are less.
 
It enables a flux capacitor that's used in a modified DeLorian for time travel.





















Seriously, though, it's a substance or chemical (traditionally rosin or acid) that cleans the metallic joint being soldered to remove and prevent oxidation, and enable smooth flow & adhesion of the solder across the metal being joined.
 
It enables a flux capacitor that's used in a modified DeLorian for time travel.

Seriously, though, it's a substance or chemical (traditionally rosin or acid) that cleans the metallic joint being soldered to remove and prevent oxidation, and enable smooth flow & adhesion of the solder across the metal being joined.

Bill, thanks for your reply; folks, that was in reply to a post I deleted about "what is flux and what function does it provide". Love the BttF reference!
 
Even worse, there are proposed EPA regulations that would require changes to the flux-cleaning solutions and processes. In the space world, uncleaned flux can also cause huge issues - the stuff EPA is talking about would put major hurt on some space-equipment manufacturers. Too onerous or expensive, and the manufacturing is likely to go offshore where the regulations are less.

Look at who benefits from this and follow the money.
 
One of the reasons I don't trust electronics is that I am an electrical engineer, and I know that stuff can go wrong!

I'm an EMC engineer and Edsel Murphy is my patron saint.

Not all connections require a lot of current; 10 megohm of resistance in the wrong spot can be just as upsetting as a dead short. Tin whiskers are bad news in the space world, and the rise of lead-based solder is a big headache (since it has more tin). Vendors don't want the hassles involved, and you're really got to ride herd on them.

Ron Wanttaja

It's the use of lead-free solder that's the problem.

I don't know why they use this technology in the first place. Why can't they use a freaking cable? There are only two failures associated with the cable, either it sticks and the throttle stays constant, or it breaks and the throttle spring closes the throttle automatically. Neither of those involve the throttle going WFO all of a sudden.

The one WOT problem I had in a car was due to the throttle return spring breaking. A quick turn of the key to OFF solved that one.
 
I'm an EMC engineer and Edsel Murphy is my patron saint.



It's the use of lead-free solder that's the problem.

Not entirely. There are various ways to minimize or eliminate the problem.... it's just that it costs more money.
 
A quick turn of the key to OFF solved that one.

Does anyone know if the brakes were able to override the accelerations described? Or were the events typically too sudden or unexpected for the brakes to be timely enough?
 
I think the claim is that the brakes will always be able to overcome the throttle.

I can't remember for sure, but I think there was one actual case where an aftermarket floor mat 'probably' jammed the accelerator on the floor. The rest were considered unknown causes or assumed to be driver error.

--

My wife's Camry fell into the recall. They reprogrammed the computer a bit. They modified the s/w so that under acceleration, if the brakes were pressed, the throttle would cut back to idle. We went on a trip after the mods, and I decided to test it out. We were on the highway, cruising along, so I stepped on the gas, then tapped the brake. Yeah - it worked. I did make sure there was nobody behind me before I tried it, and it was a good thing since we slowed down real fast.
 
I think the claim is that the brakes will always be able to overcome the throttle.

If you step on the brake while the vehicle is stopped. There is no problem holding it against the throttle.

If you are going down the road at, say, 50 mph and go WOT then try to stop it's a whole different story (at least when I've tried it).
 
If you step on the brake while the vehicle is stopped. There is no problem holding it against the throttle.

If you are going down the road at, say, 50 mph and go WOT then try to stop it's a whole different story (at least when I've tried it).

I wasn't able to find any documentation or certification requirements, but I did come across this article from Car and Driver. They tested 3 vehicles at full throttle and then standing on the brakes:

http://www.caranddriver.com/features/how-to-deal-with-unintended-acceleration
 
I wasn't able to find any documentation or certification requirements, but I did come across this article from Car and Driver. They tested 3 vehicles at full throttle and then standing on the brakes:

http://www.caranddriver.com/features/how-to-deal-with-unintended-acceleration

Yea, if you just jam the pedal down hard once and stop, it will stop. It may not feel like it's going to stop, but it will. But, if you try to just slow down, control the speed, try to pull off in a "safe spot", or push the pedal a couple times (or "pump" it) becaust it doesn't seem like it's working right, then you run into lack of vacuum / brake fade real fast.

Kind of like when the engine quits at 600 feet - either you do the right thing, or you don't.

Of course, turning the key off (which can be entertaining with a keyless vehicle) or putting the transmission in neutral solves a lot of problems.
 
Last edited:
Another thing about those reports of unintended acceleration on Camrys - wasn't there a study that broke down the reports based on the age of the drivers? I thought I remembered that, and most reports were from drivers >60 yrs old - maybe that's due to the demographics of Camry drivers, I dunno. I'll see if I can find that.
 
The problem with computer control of everything is when the computer goes haywire (or whiskers - which is a real and known problem in electronics, not just cars) is that a whacky computer will refuse to obey driver commands... What is needed, and what will eventually be in all cars, is a mechanically operated disconnect at the drivers hand that will physically interrupt computer commands to the fuel injector system - and leave the engine idling through a separate controller for steering and brakes... It is only going to become more critical when electrical steering goes completely fly by wire through the CPU...
Well, just to put it in perspective, think that you are driving a mini AirBus - now that ought to make you feel all warm and fuzzy...

denny-o
 
The problem with computer control of everything is when the computer goes haywire (or whiskers - which is a real and known problem in electronics, not just cars) is that a whacky computer will refuse to obey driver commands... What is needed, and what will eventually be in all cars, is a mechanically operated disconnect at the drivers hand that will physically interrupt computer commands to the fuel injector system - and leave the engine idling through a separate controller for steering and brakes... It is only going to become more critical when electrical steering goes completely fly by wire through the CPU...
Well, just to put it in perspective, think that you are driving a mini AirBus - now that ought to make you feel all warm and fuzzy...

denny-o

I work with embedded systems and industrial controls - it's really amazing what happens to sensors and digital I/O when current leaks in places it's not supposed to. I've seen some behavior that is totally unexplainable unless you finally realize there's a connection someplace where there isn't supposed to be a connection. So yeah - a tin whisker creating a path that isn't supposed to be there can really cause problems.

Here's some info on the reported accidents and age-related charts:
http://www.theatlantic.com/business/archive/2010/03/how-real-are-the-defects-in-toyotas-cars/37448/
 
Another thing about those reports of unintended acceleration on Camrys - wasn't there a study that broke down the reports based on the age of the drivers? I thought I remembered that, and most reports were from drivers >60 yrs old - maybe that's due to the demographics of Camry drivers, I dunno. I'll see if I can find that.

In the words of Frick and Frack, the NPR Car Guys: "Nothing says 'grandfather's car' like a gold Camry"..... :rofl:
 
In the words of Frick and Frack, the NPR Car Guys: "Nothing says 'grandfather's car' like a gold Camry"..... :rofl:

It's Click and Clack, and our Camry is silver I'll have you know!!

(insert image of old man waving fist, wearing shorts, sandals, black socks, with waistband pulled up to armpits.)

The two high school boys (twins) across the street bought their grandfather's Cadillac. I think it's pretty funny to see them cruising in an old Caddy to and from school.
 
And let's not forget putting the transmission in Neutral, turning off the ignition, etc.

However, if you're entering a curve on a cliff, and the accelerator goes WOT, you may not be able to react in time.
 
The one WOT problem I had in a car was due to the throttle return spring breaking. A quick turn of the key to OFF solved that one.

Didn't think about broken motor mounts scenario.

As for the throttle return spring breaking, I would think the cable and the friction within it would mean that the throttle stays open at the last setting you put it at with your foot. A broken throttle spring won't snap the throttle wide open all of a sudden, it will just keep the throttle from closing when you relieve pressure on the pedal.

^ This scenario is still dangerous, i've had the throttle stick on a motocross bike more than once. However I think it would be less of an emergency than the throttle all of a sudden firewalling itself.


turning key off works, shifting to neutral works - most vehicles are still pseudo-mechanically controlled in these areas. For those that aren't in most vehicles just stepping on the brake pedal with all your might will overcome the power produced by the motor.
 
Last edited:
I don't know why they use this technology in the first place. Why can't they use a freaking cable? There are only two failures associated with the cable, either it sticks and the throttle stays constant, or it breaks and the throttle spring closes the throttle automatically. Neither of those involve the throttle going WFO all of a sudden.

First of all, there are advantages to TBW (throttle by wire) that can translate into better performance, emission control, and fuel economy. And I think you can rest assured that few if any auto manufacturers employ advanced technology simply because they can or because they want the car to appeal to techno geeks.

Second, I've seen nothing (including the referenced report on "tin whiskers") that shows conclusive evidence that TBW has a greater propensity for a stuck WOT condition or other "unintended acceleration" events as these have indeed occurred with mechanical linkages.

Third (and this is a biggie), not only is it a fact that a stuck throttle by itself shouldn't cause a loss of control since normally functioning brakes always have the power to overcome engine torque if properly applied. But the "biggie" aspect goes a bit further in that on many investigated UI episodes it was possible to determine that the brakes hadn't been applied at all even though the driver strongly insisted that they were pressing the brake pedal as hard as possible. What that clearly indicates is that in many if not most instances the driver was pushing with all his might on the accelerator pedal while believing they were stomping on the brakes.

FWIW, that throttle pedal sensor in my car employs two resistive sensors that operate in a complementary fashion (one puts out more voltage as you press the pedal and one decreases it's output as the pedal moves down/forward) and the ECU which processes the pedal movement always checks that the two inputs make sense together and will switch to a "limp home" mode which significantly limits engine output if there's a discrepancy. In addition there's a similar sensor on the throttle itself which is compared with the commanded throttle position and the ECU is programmed to kill the engine by cutting off fuel delivery if the throttle is open when it shoudn't be. I cant' say with certainty that all such systems operate in a similar fashion but I think that's likely. And although I couldn't honestly state that there are no undetected failure modes which could lead to a WOT condition when the driver isn't pushing the gas pedal I'd be willing to argue that it's very feasible to design such a system so the chances of that happening are less than for the driver to be struck by a piece of a falling asteroid.

One design aspect I do feel was inadequate from a safety perspective is the push-on push-off ignition switch that prevented the driver from turning the engine off manually when the transmission was in drive.
 
I think the claim is that the brakes will always be able to overcome the throttle.

I can't remember for sure, but I think there was one actual case where an aftermarket floor mat 'probably' jammed the accelerator on the floor. The rest were considered unknown causes or assumed to be driver error.

--

My wife's Camry fell into the recall. They reprogrammed the computer a bit. They modified the s/w so that under acceleration, if the brakes were pressed, the throttle would cut back to idle. We went on a trip after the mods, and I decided to test it out. We were on the highway, cruising along, so I stepped on the gas, then tapped the brake. Yeah - it worked. I did make sure there was nobody behind me before I tried it, and it was a good thing since we slowed down real fast.
Kind of hard to "flash" the tailgater your brakes without causing a big problem. I'd have to learn a whole new style of drivin'.
 
Kind of hard to "flash" the tailgater your brakes without causing a big problem. I'd have to learn a whole new style of drivin'.

Nah - no problems with that at all.

What I was doing was flooring the accelerator, then tapping the brakes. Verified the new operation - brakes do take hold, and accelerator shuts off to idle, even with pedal on the floor.

Normal driving and flashing tailgaters shouldn't be any different.
 
Kind of hard to "flash" the tailgater your brakes without causing a big problem. I'd have to learn a whole new style of drivin'.
Not that I'd ever do this :rolleyes: but with a manual transmission dropping to the lowest gear that will allow you to speed up 10 mph, dumping the clutch with the engine at idle and then flooring the gas is far more effective at getting the attention of a tailgater. Instead of the warning flash of red brake lights that may just get them to close up tighter, the first thing they will see is the rear of your car leaping at them and by the time they lock up their brakes and brace for collision your car is zipping out of harms way. Of course this is best performed when there's traffic close behind the villan turned victim so they get distracted by a collision at their own rear or at least when you're about to depart the road you're both on.
 
Nah - no problems with that at all.

What I was doing was flooring the accelerator, then tapping the brakes. Verified the new operation - brakes do take hold, and accelerator shuts off to idle, even with pedal on the floor.

Normal driving and flashing tailgaters shouldn't be any different.


This is probably a good thing..

Although I can think of one instance where this would cause some issue. Starting on a hill. I have to start from a stop on a steep hill, I give the car just a tiny bit of throttle before releasing the brake, for a smooth start. Otherwise i'll roll backwards as I shift my foot from brake to accelerator.
 
This is probably a good thing..

Although I can think of one instance where this would cause some issue. Starting on a hill. I have to start from a stop on a steep hill, I give the car just a tiny bit of throttle before releasing the brake, for a smooth start. Otherwise i'll roll backwards as I shift my foot from brake to accelerator.

Specifically:
>>
For example, the brake override feature does not operate if the brake pedal is depressed before the accelerator pedal. This logic allows for vehicles starting on a steep a hill to safely accelerate without rolling backwards, otherwise known as a hill start.
<<

Here's the Toyota description of how the brake override s/w works:
http://www.prnewswire.com/news-rele...complaints-in-remedied-vehicles-86474747.html

I think it had already been a feature built into the Prius at the time, then the updates were made to the Camry line as part of the recall.
 
Kind of hard to "flash" the tailgater your brakes without causing a big problem. I'd have to learn a whole new style of drivin'.

Just wait for an opening in the next lane and turn on your turn signal. They will change lanes.

(not always, but as often as not...)
 
The problem with computer control of everything is when the computer goes haywire (or whiskers - which is a real and known problem in electronics, not just cars) is that a whacky computer will refuse to obey driver commands... What is needed, and what will eventually be in all cars, is a mechanically operated disconnect at the drivers hand that will physically interrupt computer commands to the fuel injector system - and leave the engine idling through a separate controller for steering and brakes... It is only going to become more critical when electrical steering goes completely fly by wire through the CPU...
Well, just to put it in perspective, think that you are driving a mini AirBus - now that ought to make you feel all warm and fuzzy...

denny-o

That's silly. Stick it in neutral and let it rev up to the electronic rev limiter. There's no additional controls required. They're already there.

It won't break the engine in the time required to pull over from highway speeds. It'll hit the redline, drop, redline, drop, until the cows come home in a modern car. Redline is meaningless on a modern tach, the driver can't exceed it if they want to.

Then shut the thing off after you're stopped.

Try it. See if you can rev your engine past redline in any modern family car. If you don't believe me, try it in a rental. Haha.

As far as steering goes, the only reason to make steering fully fly by wire in cars is to save space or weight for something other than the steering mechanism. The disadvantages in making things too tight under the hood are massive complaints from owners and mechanics that simple jobs now take many more hours of shop time which negatively impacts the perceived value of the vehicle. So it typically doesn't get done. It's not practical.

Any controllability issues are done via "steering assist" add ons, or differential braking combined with throttle control, and rarely by changing to complete fly by wire. It would raise cost significantly for no distinct return on investment in any other than marginal ways.

Modern "traction control" usually modulates the brakes and throttle to affect direction changes in sideways acceleration.

My wife's Jetta wagon, for example, will modulate the rear brakes to straighten a skid while also ignoring a full throttle input if any tires lose traction, especially the front drive wheels. It'll basically whip the vehicle straight without a need for any connection to the steering at all.

You can try to oversteer it or under steer it while it's trying to correct the skid and it'll adjust via accelerometer data, faster than you can attempt to fool it. I've gotten it to skid more than 2' sideways with the rear end only once, and I was trying really hard in an empty parking lot. On snow, it can be coaxed into sliding out but it often limits the slide to about its own width in back unless you just crank it over in slick conditions. It's pretty smart code.

It also handles loss of traction on only one side on the front drive wheels very well. It has no way to only apply power to one side only like some systems but it knows how to apply braking to the appropriate side to utilize the differential as an energy transfer mechanism. It also modulates the throttle to lower the torque being applied to the spinning wheel overall. It'll climb an icy hill with the accelerator on the floor and the engine will spool up, back off, brakes forcing it to track straight at each spinout, as it goes from low throttle to spin out speed, and back. I believe it also up shifts and "lugs" the engine a bit in this condition, but don't hold me to that one.

So, not much need for a full fly by wire steering. It's being essentially done anyway buy other cheaper methods.
 
That's silly. Stick it in neutral and let it rev up to the electronic rev limiter. There's no additional controls required. They're already there.

True dat.

But then, people still stall, spin and die trying to make the so called "impossible turn".:dunno:

So, not much need for a full fly by wire steering. It's being essentially done anyway buy other cheaper methods.

It will be here before you know it.
 
Back
Top