Speedo Correction option

[linuxgod]

From a different thread discussing suspension, I saw the Superior Engineering kits come with a CANbus speedo correction adapter. It looks like this is produced by Marks 4WD out of Australia and sells for 313 AUD (about $220), plus shipping.

Speedo Corrector - Toyota LandCruiser VDJ70 Series | Marks 4WD

Our Marks4wd SPEEDO CORRECTOR allows you to adjust the TOYOTA LANDCRUISER VDJ70 series speedo to compensate for tyre size and/or differential ratio changes.

www.marks4wd.com

I dug a bit and found these installation instructions. Looks like the hardest part is disassembling the dash to get to the connector:

https://cdn.shopify.com/s/files/1/0309/4370/7180/files/mfk1050-1.pdf?v=1603926303

FYI it looks like there's a German company which also manufactures a similar calibration kit for the 200, though I don't see any installation instructions on it. It's entirely possible the Marks kit is this plus a wiring harness and instructions.

CAN-bus Speedometer calibration

E1-certified CAN bus control unit for calibrating the speedometer and the odometer in the combination meter. With our EC type approval, different tyre rolling circumferences or modifications to the differential can be road-approved in many countries worldwide.

www.wenning.de

With my truck on 34s my speedometer reads about 5% low, so this is tempting to try. It's easy to adjust so if you go up/down in tire sizes you can flip a few switches to adjust the correction amount.

 

[Markuson]

From a different thread discussing suspension, I saw the Superior Engineering kits come with a CANbus speedo correction adapter. It looks like this is produced by Marks 4WD out of Australia and sells for 313 AUD (about $220), plus shipping.

Speedo Corrector - Toyota LandCruiser VDJ70 Series | Marks 4WD

Our Marks4wd SPEEDO CORRECTOR allows you to adjust the TOYOTA LANDCRUISER VDJ70 series speedo to compensate for tyre size and/or differential ratio changes.

www.marks4wd.com

I dug a bit and found these installation instructions. Looks like the hardest part is disassembling the dash to get to the connector:

https://cdn.shopify.com/s/files/1/0309/4370/7180/files/mfk1050-1.pdf?v=1603926303

FYI it looks like there's a German company which also manufactures a similar calibration kit for the 200, though I don't see any installation instructions on it. It's entirely possible the Marks kit is this plus a wiring harness and instructions.

CAN-bus Speedometer calibration

E1-certified CAN bus control unit for calibrating the speedometer and the odometer in the combination meter. With our EC type approval, different tyre rolling circumferences or modifications to the differential can be road-approved in many countries worldwide.

www.wenning.de

With my truck on 34s my speedometer reads about 5% low, so this is tempting to try. It's easy to adjust so if you go up/down in tire sizes you can flip a few switches to adjust the correction amount.

Sallah to Indy: “You go first...”

 

[Moby]

So close... but you have to cut the stock harness. Why oh why. I hate cutting stock wiring.

I realize it is a pain to source the correct terminals to make this plug and play but it would be so much nicer to pull the stock wires from the stock connector and plug them into the calibration unit. And then run new wires back to the stock connector. I eventually found terminals for the stock plugs in my FJ-62 so that I could tap or repair stock wiring without having to hack things up.

Unfortunately the German version doesn't seem to have online instructions and says only suitable for install by a shop...seriously?

 

[empty80]

I’ve been looking into the logistics of creating an inline harness from something similar. I can find a 28-pin female, but I can’t find the 28-pin male. In essence you are a only intercepting power, ground and two CANBUS wires. It would look like 24 wires straight through from male to female and then “plug” the Yeller Box, Mark’s box or Hypertech in the appropriate pins.

If you look at the Hypertech system for the Tundras they use a printed circuit board that allows them to use two female connectors. Essentially, they’ve created their own male connectors. Tundra is 32 or 40 pin, so that’s right out. The principle would be the exact same.

 

[grinchy]

Has the q been answered yet if this corrects items that aren’t Speedo and mid based. On the lx there are suspension adjustments based on speed, would love to know if these are also corrected.

 

[empty80]

Has the q been answered yet if this corrects items that aren’t Speedo and mid based. On the lx there are suspension adjustments based on speed, would love to know if these are also corrected.

Not sure how applicable this is for the LX. I can’t get into the LX files, but if it looks similar then there’s a possibility.

If I’m interpreting this correctly, then the altered speed signal from the Combination Meter Assembly would be an input for the skid control ECU.

 

[grinchy]

I wonder if skid control is muxing the 4 speed sensor signals and sending 'the truth' out on SP1, so that is the single point of correction required. <edit 2009 LX looks the same to me.>

 

[Markuson]

Oye!!!
These last several posts right here are why we still don’t have readily available speedo corrections... heh...


Imagine if wired telephone service had been like this—
-If each home dweller had to climb telephone poles and hack their way in amongst power and other lines...

...We’d still be using snail-mail, telegraph offices and homing pigeons.

Even electrical engineers hate hacking up harnesses and wondering what it might impact.

 

[empty80]

I wonder if skid control is muxing the 4 speed sensor signals and sending 'the truth' out on SP1, so that is the single point of correction required. <edit 2009 LX looks the same to me.>

That’s actually a pretty solid interpretation. Is anyone smart on NPN transistors and skid control ECUs?

 

[grinchy]

I dug around a bit to see how difficult it is to take a given square wave and create a fixed 1.1 multiple of it (10 in 11 out). Looks like analog can transform the shape and magnitude but not the frequency.
probably a trivial activity for a dsp, but there are so many potential partsto choose from. I do have some analog devices Dsps left over from a stereo project, I’ll have to look to see if the can count square waves and output square waves. If so they’ll work with the simple multiplier calculation. They do have 50 ms processing delay, I wonder if that is relevant.

 

[Markuson]

I dug around a bit to see how difficult it is to take a given square wave and create a fixed 1.1 multiple of it (10 in 11 out). Looks like analog can transform the shape and magnitude but not the frequency.
probably a trivial activity for a dsp, but there are so many potential partsto choose from. I do have some analog devices Dsps left over from a stereo project, I’ll have to look to see if the can count square waves and output square waves. If so they’ll work with the simple multiplier calculation. They do have 50 ms processing delay, I wonder if that is relevant.

I hereby volunteer @grinchy to build...test and mass-produce these...then send us all a bill.

 

[ttrmike]

I’ve been looking into the logistics of creating an inline harness from something similar. I can find a 28-pin female, but I can’t find the 28-pin male. In essence you are a only intercepting power, ground and two CANBUS wires. It would look like 24 wires straight through from male to female and then “plug” the Yeller Box, Mark’s box or Hypertech in the appropriate pins.

If you look at the Hypertech system for the Tundras they use a printed circuit board that allows them to use two female connectors. Essentially, they’ve created their own male connectors. Tundra is 32 or 40 pin, so that’s right out. The principle would be the exact same.

The Tundra Hypertech option uses a pretty generic picture. When I purchase mine, it didn't include the additional circuit board. It uses an adapter harness with male/female ends and the Canbus corrector spliced in the middle. Easy enough to install without cutting though. Just de-pin and re-pin.

I wonder if skid control is muxing the 4 speed sensor signals and sending 'the truth' out on SP1, so that is the single point of correction required. <edit 2009 LX looks the same to me.>

I'm not 100% on the electronic wizardry within the Canbus corrector, however I'm not sure they'll modify the SP1 signal in the manner we need. The off the shelf solutions tap into the CAN High/Low (pin 25 & 26 on the E7 connector) signals rather than a speed specific signal wire.

If I read MPH from the ECU is still displays the uncorrected speed sensor value, while the gauge cluster displays the corrected value.

Maybe the solution is intercepting the Canbus signal prior to the ECU instead?

 

[TN LX570]

.

Maybe the solution is intercepting the Canbus signal prior to the ECU instead?

I do believe this is the correct path. The gauge cluster is read only. So the Canbus signal going in is not sending any data back out. If you spliced in the hypertech earlier in the bus you could send the modified signal to other systems. IRC the VSS feeds to the Main ECU and then the signal is passed on via Canbus.

On a similar note, I tried splicing my hypertech box in directly before the AHC ecu, when I was trying to override the H mph restriction. This was unsuccessful. The AHC ecu is not really reading the VSS directly off the canbus. Rather it seems it gets another signal similar to a telling it an allowed stage (example: Stage 1,2,3 etc)

 

[grinchy]

I’m not optimistic we can change the value once it’s on the canbus. It will be in a message with a header etc. that is probably more investment than any of us want to make.
We can absolutely change the signal between the sensor and skid ecu. Maybe intercept and scale after the skid ecu and before it’s gets on the canbus, but I think the low hanging fruit is at the easy to access sensor wires. They already have voltage, and the signal is just a plain old square wave. It’s just a tiny bit of electrical engineer we need . . . To pick the dsp and program a few hundred of them. The dsps are $2 or $3 each or less.

 

[grinchy]

Here is the huckleberry we seek.

Phase-locked loop - Wikipedia

en.m.wikipedia.org

 

[grinchy]

An nice overview. Off the shelf parts more than I though anywhere from $2 to $100. Not educated enough to see which would be most suitable, but I suspect the very lost basic could handle the frequency of the wheel sensor pulses.

http://cmosedu.com/videos/s17/ecg721/APLL.pdf

 

[linuxgod]

Sallah to Indy: “You go first...”

I bookmarked their site for later purchase. At the moment my speedo reads about 5% low, so not a huge deal. In fact it's helpful because my wife always drives slower than me, often a bit under the speed limit, so it encourages her to drive slightly faster Also the odometer reads a bit in my favor. Also it's $300 and involves me taking the dash apart.

If I go to 35s I'd be inclined to make this part of my purchase. At that point I'd be off 6-7% where the mental effort to correct it in real time starts to become a hassle. I might do it sooner as we're looking at a second vehicle later this year as my daughter starts to drive, at which point I won't be monkeying with our only vehicle

 

[linuxgod]

So close... but you have to cut the stock harness. Why oh why. I hate cutting stock wiring.

I realize it is a pain to source the correct terminals to make this plug and play but it would be so much nicer to pull the stock wires from the stock connector and plug them into the calibration unit. And then run new wires back to the stock connector. I eventually found terminals for the stock plugs in my FJ-62 so that I could tap or repair stock wiring without having to hack things up.

Unfortunately the German version doesn't seem to have online instructions and says only suitable for install by a shop...seriously?

It's cost, I'm sure. And the fact that apparently the 2008-2015 harness is different than the 2016+, which means sourcing and stocking two different harness versions.

I'm not against cutting and splicing a couple wires. Yeah I'd rather the solution be 100% plug-and-play, but so long as I've used good quality connectors, crimpers, etc I've never had an issue. FWIW I'd likely splice in connectors so that reversing it would allow me to disconnect the wires and plug the harness back together cleanly, rather than needing to re-cut and re-solder, but to each their own.

 

[TeCKis300]

Curiously watching this thread. With 35s, the offset is more than I like. I find myself driving faster, but I'll likely adjust in due time. Still interested in a possible fix, and I don't mind cutting and splicing as I trust my wiring work. Some of these PnP harnesses may invite some of their own vulnerabilities.

Part of the reason is that LX has other systems reliant on CANBUS speed, namely the AHC ECU. It should be lowering the suspension for freeway speeds at ~62mph, but it's currently happening closer to 70mph. As I rely on this behavior when towing, for the side effect of dynamically increasing weight distribution hitch tension at higher speeds for more stability. I've had to adjust my hitch head for more static tension.

 

[sdnative]

I have used one of these CANbus analyzers. Once the relevant CAN message and data is identified, it may be possible to intercept the speed message and re-transmit a different speed message. Would need to be done correctly though as wrong speed data could cause unintended and dangerous behavior.