Originally Posted by docmirror

Well, yes I did. When I got to the part where the guy is injecting different resistance on the two circuits, and where Toyota describes a 'short' as a high resistance I kind of discounted the whole thing as two media talking heads without an EE degree anywhere in sight as a lot of hot air.

The fault has to be troubleshot to what's called 'root cause'. It's tedious, slow, difficult work that won't be fixed by some guy with a few feet of wire and a resistance box, or by some Toyota marketing hack. The solution will come when there is an exhaustive test of the whole system. That kind of stuff is not played out in the public. I know, cause I do this kind of work sometimes.

So, let's look at what they are working on.

We start with the actual pedal arm, that's connected to the pedal box, with two independent resistive elements. These elements are swept by two independent motion arms that change the resistance in the reference voltage circuit that is provided to the pedal box on two separate wires. They may or may not share a connector, but are still electrically independent.

Next, the change in voltage provided by the different resistance based on the position of the pedal is converted to a digital signal by two independent ADC(Analog to Digital Converter) chips. The output of this chip is a set of digital signals. These two signals enter the throttle ECU which does a set of comparisons repeatedly. One of the comparisons is to determine that the two signals are within a few bits of each other. There is a tolerance allowed, but it's fairly tight. The next comparison determines if the signal is within the valid range of the pedal position. For example, once converted to digital, the idle signal where no pedal movement is sensed may be H 0x00001000 and full throttle at the pedal may be H 0x000fffff. If there is a value of H 0x00000010 on one side, and within the range on the other side, the low side will be ignored and the other side will be used to determine throttle position. If I was designing this, after a few iterations, if the out of tolerance value was still bad, I'd set the throttle to idle, and set the check engine.

Now, what about the situation where both signals are out of tolerance, and they are both out of spec between each other. In this case, once again we have to reject the values, set the throttle to idle and set the check engine light.

What needs to happen for the unintended acc to occur is this: Both signals need to be 1) within spec between the two sense values. 2) Within allowed range of the ADC output(00001000 - 000fffff). 3) uncommanded by the pedal, where the pedal is at idle. 4) A value which would be considered full, or near full throttle(say, 00020000 - 000fffff).

The ECU for the throttle position also has a DAC(Dig to Analog Controller), and a driver for the electrical solenoids that move the throttle plate. I assume there are also two for redundancy again, and the same matching setup is in place to move the throttle.

Now, this is a simplified explanation of what goes on with the gas pedal and throttle in a drive-by-wire setup. There is more that happens to the signals, more smoothing, more checking, and more deterministic evaluation along the way. If any of it goes wrong, the default action is supposed to be to set the throttle to idle.

Who knows where the issue is? It could be pedal mechanical, pedal electronic, ECU, smoothing code, sensor code, throttle DAC, solenoid drivers, or mech throttle. And there's prolly more that I haven't covered exactly.

Toyota wants to get the word out that it's a pedal mech problem. It could be, but it could be a cascaded failure also. We'll see, but I doubt you'll ever get the root cause from Toyota, even if the senate asks, I still doubt you'll ever get the real story.

Originally Posted by docmirror

No, one of the DOT requirements is that the brakes are able to overcome the engine only when assisted by the vac servo, and only one time. On very old cars without vac assist this was not the standard before the DOT rules.

Once you've lost vac assist(or boost), you're screwed. There isn't enough force applied to the pedal to overcome the frictional heat dissipation of the pads. Typical vac assist is a multiplier of 3 in most cars, but it can be much higher of course for trucks and heavy vehicles.

So, if the throttle sticks open, you've got one pedal push only, and you have to really push very hard, very fast to get the cars brakes to overcome the torque of the engine. Once you let off the brake, or don't push hard enough, the loss of vac assist or from heat dissipation means your screwed. That's why the solution is to shift into N. My solution is to turn off the key and deal with a dead engine and no assist, but I'm weird.

Originally Posted by docmirror

The ECU is in charge of setting the position of the DAC values that are translated to the servo actuators for the throttle control arm. There are several ways it can screw up.

1) ECU sends wrong signal to DACs, signal is correctly interpreted by the DACs and it goes full, rather than idle.

2)ECU sends right signal to DACs, and it's misread, and goes full rather than idle.

3)ECU and DACs work, solenoid driver amp rails, and goes full.

4)ECU and DACs work, solenoid driver amps work, and mech stick in the throttle arm.

5) I could go on, and on, and on.....