Throttle Translator - Getting the British to speak Japanese

I am in the process of installing a TVS1900 Supercharger from a 2011 Range Rover Sport on my 2001 Toyota Tundra, it has been a series of bad decisions supported by emotional or impulse purchases.

The project is now a year in and I am tying up loose ends and I am not happy with how the stock throttle body is looking/fitting on the new supercharger.

Stock 01 Tundra Throttle Body

11 RRSC Throttle Body

The stock 01 Tundra throttle body is giant in size, small in bore, and requires a cable to run out to it to actuate the accelerator pedal position sensor. I can fix the cable thing by moving the accelerator position sensor into the cab using a 05/06 Tundra throttle pedal, a Subaru Legacy one, or just make a mount for the 01 Tundra accelerator position sensor that normally lives on the throttle body to mount it to the pedal. The electrical side has proven to be an issue.

The Toyota ETCS-i system used on the early model Tundra, 100 series Land Cruiser, and some Supra's use redundant sensors for both TPS and APPS so you have 4 signals to the ecu, APPS1 and APPS2, TPS1 and TPS2. The ECU has a fault checking function that looks at how APPS1 (which is used for actual pedal position sensor) and APPS2 (which is used for fault checking) compare to each other and the ecu decides if the input is valid. The same goes for the TPS signals. The throttle motor is then actuated based on what the ecu thinks you actually want. It is not a 1:1 pedal to throttle position relationship. Toyota also didn't trust the system to function 100% at the time so they added an electromagnetic clutch to the system so if the throttle motor went all Rise of the Machines, they could just disconnect the motor from the throttle blade.

The Toyota manual specifies this which is helpful but not the whole story.

To make things interesting, the Toyota throttle percentage vs voltage and accelerator pedal position vs voltage are kind of oddball curves. The Toyota manual only specifies the max and min voltages of the throttle but when you log the curves with a microsquirt you get this.

01 Tundra TPS

05/06 Tunda TPS

11 RRSC

So I exported all those to excel and plotted them out. The Range Rover Sport throttle position sensor stock output won't make the Tundra ecu happy, even the later model Tundra one won't work. I plotted TPS1 vs TPS2 to figure out what the curves needed to look like so I could figure out the math. Spoilers in the images, but there is a corrected version (yay, I was successful, well partially anyways)

I also did the same thing for the APPS sensor.

Essentially, I wanted to create a new curve that perfectly matched the stock 01 Tundra part. Tuning the throttle position sensor input to output and the accelerator pedal position sensor input vs output is just a matter of some math and electronics.

I started with an Adafruit Feather RP2040 to do the math, an Adafruit ISO1540 I2C Isolator to keep the 3.3v feather from letting the magic smoke out when connected to a 5v sensor or when trying to provide 5v to the stock ecu, an Adafruit ADS1015 ADC to read the inputs from the TPS and APPS, and an Adafruit MCP4728 DAC to make the new signals to feed the stock ECU happy thoughts. The nice thing is that this whole package is low power enough to run on the stock TPS ref feed from the stock ecu. All this worked brilliantly once I had the tricky stock sensor curves figured out. There are some bonus parts in the fritzing sketch that are for other functions which is why there is an I2C address changer and another DAC on it.

I took this string of controllers and wires out to the truck and manually cycled the throttle body and it all worked, OBD reported happy and no CEL.

Here is where it gets tricky, and where I am stuck. The stock 01 Tundra throttle body, the 11 RRSC throttle body, and the 05/06 Tundra throttle body are all just dumb dc motors. They all open at about 2.2v constant voltage, and they all run on a PWM singal from the factory. The stock 01 Tundra motor measures at 1.8ohms and so does the 11 RRSC motor so I was thinking "it would be great if the motor would just drop in with the stock setup" but alas, twas not to be. When I tried to actuate the throttle the blade did this lovely number.

While it would have a sick idle chop that only the biggest cam could provide, it is less than ideal for drivability.

I knocked the supply side down to 5V using a mosfet triggered off the stock TB wiring and that made it kind of work but still not driveable.

And here is what the stock one does.

I noticed that the stock tb "locks" in place and you can't manually move it once it is on. That makes me think the ecu is feeding an open and closed signal to it at the same time to fix it in place. I am unsure if the RRSC one works like that or if there is something else internally in that circuit that is defeating it.

So I wired up a current divider using some 2ohm resistors and added a second 2ohm resistor to get the system current back to a manageable level and gave that a try and it didn't help much if at all.

So, where I am now is I have the TPS and APPS happy but I can't get the stock 01 Tundra ECU to move the throttle blade in a normal manner. I need to keep the stock ECU for the gauges/odometer/ and cruise control so running standalone is not an option on the throttle. I can try to bodge together throttle control on my feather using a motor control feather wing I have but I would need to take input from the stock ecu control circuit to trigger that or the toyota side will throw a fit if it calls for different throttle percentage than I give. 

Any ideas on how to make the stock circuit drive a different motor?

Sean

Is it possible to adapt the stock Toyota sensor and motor to the Land Rover throttle? This would have the fewest potential failure points.

In reply to MadScientistMatt :

TPS maybe, motor likely not. I have a secont yota tb on the way now to play with to test that theory though.

Sean

My second throttle showed up and I got to work dissecting it. My initial thinking was that I would just do a motor comparison and see where that got me as far as motor spec and control.

Stock Tundra throttle body

Tundra APPS, everyone assumes this is a cable driven throttle but that is just a sensor with a "limp home funciton" on it. If the sensor fails, you can get 20ish percent throttle if you go WOT with the pedal. We can mount the APPS sensor wherever, its big and annoying in the engine bay.

Jumping over to the drive side, the gearbox, motor, TPS, and motor clutch are in this gearbox.

Splitting the case gives you this, the motor and clutch are there in the bottom, and the dual TPS is up top.

On the other half is an integrated shaft, min/max throttle stop, and the gears as well as the tps fingers (nut and lock washer removed already, no proof)

Then I took the range rover throttle apart to see what if anything could be transferred. Start by drilling out the 6 rivets and popping the lid off.

Under the lid is a digital, hall effect/prox probe type sensor for the tps and the motor contacts.

Nylon/glass fiber reinforced gears, the motor, and throttle stops.

With the gear removed you can see the motor mounts and the gear a little better.

Motor removed, nothing special.

The gears mesh but the ratios are wrong, the stock tps on either setup would clash with the gears and the net result would still be not great. I went to sleep dejected.

Sean

I woke up itchy today and had to go to work. Then at lunch I broke out the portaband and cut a big chunk out of the Tundra throttle. I ran out of time at lunch but came home and did the sketchy manual mill (a router with a carbide bit and some 1-2-3 blocks as spacers) to flatten the butchery out. Then I put everything back together and came up with a plan.

Now I just need to design a bearing retainer to hold the second bearing in place on the face, and make a shaft coupling to hold the Tundra shaft to the new TB shaft and mount the gearbox to the new TB.

Sean