M35 Throttle Restriction Removal, Maybe

[Commsman]

I've been thinking about the M35 throttle restriction for the last few weeks & reckoned it would be worth some effort to see if I could bypass it in some way. I mean, how hard could it be?

Last night I was convinced I had cracked it with a drive in a car that felt transformed . But as of this afternoon I'm not so sure . All sorts of gremlins have crept in to cause me a big headache so I've decided that rather than announce how I've found the Holy Grail, I'm going to turn it over to the community to see if we can solve it with a collaborative effort. That is if anyone is interested....

I need to do a lot more testing & gather more data to call this either worthy of pursuing or a total waste of time (don't stop investigating the G-Zone ECU flash quite yet)

Please don't flame me. I am but a simple man having a 'bit of a go' to try to further the cause.

[Commsman]

Preliminary investigation.

It started with me probing the most likely wire on the throttle body (red/white stripe/silver dashes) & found a voltage varying from low to high with more pedal depression. At warm idle it was about 0.65V and it increased by ~1V for each 20% throttle opening, as compared with my Greddy Informeter. The max I saw during testing was 4.04V which equated to 80% opening.

So if you assumed that all I needed to do was intercept this signal & increase the voltage to open the throttle more (like I initially hoped), then you'd be wrong the same as me!

I also probed a second wire (black / silver dashes) that had a voltage that varied from high to low. It was about 4.6V at idle going down to 1V at 80% throttle. Essentially these to signals are the inverse of each other.

These two wires are the sensor wires that tell the ECU what's happening with the throttle. The Infiniti G35 has the same setup (although the voltages are slightly different) & this is the description from the manual.

The throttle position sensor has the two sensors. These sensors are a kind of potentiometer which transform the throttle valve position into output voltage, and emit the voltage signal to the ECM. In addition, these sensors detect the opening and closing speed of the throttle valve and feed the voltage signals to the ECM. The ECM judges the current opening angle of the throttle valve from these signals and the ECM controls the throttle control motor to make the throttle valve opening angle properly in response to driving condition.

I thought about modifying the signal from the ECU that actually opens the throttle, but the ECU would know straight away something was wrong via the feedback from these two sensor wires. So instead I decided to try to modifying the signal going to the ECU from the throttle position sensors.

[Commsman]

Physical Setup

It's not a very elegant approach, but I used two voltage modifier kits from Jaycar. I already had one kit & the hand controller made up since Christmas, so all I needed was another. I wasn't keen to spend heaps of coin on what was essentially a whim.

These were installed in series with the two wires coming from the sensors (wires cut at ECU) so that the signal passed through them on the way to the ECU. I went for a few drives, with my son Jacob watching the hand controller & writing down the load points which corresponded with the 20%, 30% .... 70% & 80% throttle opening displayed on the Greddy unit. This was done for both sensors. When I graphed the results they looked exactly like the picture in the manual, funnily enough.

By subtracting voltage from the red wire (rising voltage sensor) and adding voltage to the black wire (falling voltage sensor) I hoped the ECU would think that the throttle wasn't open as far as it should be and open it more.

[Commsman]

Testing

We started by adding/subtracting 20 units (not sure exactly how this corresponds to voltage) ramping gradually up from 1 – 20 from the load point representing 60% throttle opening & went for a drive. We also monitored the voltage at the throttle body before realising this won’t tell us much anymore. What I should have done (& will do soon) is measure the voltage on the wire that is used by the ECU to actually open the throttle. This will provide some critical information I’m sure.

Anyway, at these settings there was no discernible difference. I did notice that when driving up a very steep hill it was impossible to get above 50% opening anyway. There are obviously other factors at play besides how far you put your foot down. All further testing was done on flatter roads after that.

Next we tried adding/subtracting 60 units (maybe that equalled 0.6V which would put us close to maximum opening (total 4.6V instead of 4.0V) I thought. We also started ramping up the changes from the load point which corresponded to the 50% mark of the unmodified system. That should be easier to test. I stopped the changes at the load point where it maxed out in the setup phase.

So how did this drive. Well, it went pretty hard but there was a couple of stutters and the figures in the hand controller where jumping around a lot. A head scratch later & maybe it was because the car had “dropped off” the edge of my ramped up/down settings.

We then added figures up to 65 units of change & then continued the 65 figures all the way to the end of the range. So what we were trying was changing the voltages slowly (both up and down) from after 50% opening till about the old 80% opening and then holding that maximum change beyond that.

What did it drive like? Absolute rubbish! The stuttering was really bad, even at low speed and moderate pedal depression. Obviously the car didn’t like that all.

Still too high? Let’s leave the starting position at 50% (it drives as standard below that) and only ramp up/down to 55 units, continuing that right through to the end of the range. Still rubbish. Bugger.

Beaten for now, we go home to forget about it for a few hours. Later last night I decided to give it another go. I dropped the max change to +/- 45 units, continuing the 45 settings all the way to the last load point on each interceptor. What did it drive like? Unbelievable! It was a whole new level of performance for my car.

[Commsman]

Sussing it out – The good, the bad, the ugly.

Now it occurred to me that some of what was experienced could be attributed to the colder night air (12 deg C) & I’d also noted that the injector pulse width was mysteriously lower than expected for going so hard. Was the car running lean & that was producing the extra power? It makes sense when you think about it. The ECU thinks the throttle is only open (say) 60% when it is actually open (say) 80% and so only fuels the engine for that amount. Now I’ve never experienced firsthand how much gain can be seen by leaning off the AFR, but surely I had done more than inadvertently changed the mixtures? Mmm, my stuttering problem from before was probably fuel cut because of changes being too high. Lean out perhaps? I’m not sure.

All this info was churning through my head overnight, but I decided that I had a win nonetheless.

This morning, basking in the awesomeness of my success, I tidied up the wiring and mounted the interceptors either side on the ECU (double sided tape & Velcro) & did a few other things in preparation for upcoming mods.

A short test drive this afternoon revealed the same problem as before. Temperature was now 10 deg warmer than last night. What to do? Lower the max voltage change & test again or reset all figures to zero. I decided I needed my wideband meter installed (on order) before playing anymore so I reset everything to zero before leaving to drive to Gold Coast airport (about 45 mins away).

Now another problem had surfaced that wasn’t evident yesterday. Most noticeable at a steady cruise on the highway, I was getting slight variations in throttle openings, evident by slight “hesitations” as I drove along. Could this be because the interceptors were taking too long to route the signal through their electronics? I checked & it is <10ms from input to output so unlikely. Maybe more likely is because this morning I wired up ignition power from a source near the ECU that wasn’t as heavy gauge as the Acc power I used during previous testing. I must change this & test again.

So when nearly at Tweed Heads, the car starts to drop the throttle altogether for a few seconds before giving it back. Crap! Then it throws up a MIL & I’m down to idle power, ie no accelerator at all. I pulled into an exit & removed the fuse to the interceptors so the inbuilt relay would pass the signal straight through. All back to standard & no more problems on the way home. The MIL even went out when I started the car .

Where to from here? I’ll keep on reporting on what I find & if anyone is interested enough to offer suggestions I’ll do my best to try them out.

For now my head hurts. Cheers, Leon.

[KiwiRS4T]

Has anybody replaced the throttle body with an aftermarket one (or a mechanical Nissan one) and fitted a new accelerator pedal complete with cable?

[scotty nm35]

I went down a similar path in my head Leon, I wasnt too keen to see 300kw suddenly at a red light though if it all went pear shaped. Driving the electronic throttle manually isnt a good idea I decided.

The stuttering you experienced was the engine running lean for sure, get the wideband in if you decide to do further testing.

If the ecu sees any variation between the tps signals it will obviously go into limp mode which you experienced. Not fun on a busy road as you can imagine.

The cable throttle idea worked well, the engine went into limp mode, thinking it was doing 1200 revs. Cihan and I tuned the emanage around this taking fuel out at less throttle and increasing it as the throttle opened. Surprisingly it ran very well, using the pedal resistor as tps. The only draw back was the cams were in full retard and the gearbox had no throttle kickdown. (As I drive in manual anyway it wasnt an issue) I was hoping the Vmanage would take care of the cams which may still happen. At least if the cable snaps it defaults to idle. I always make sure any throttle mods failing have a similar outcome...

I still think a remap is the best option as all the factory parts and safety remain in place.

[Commsman]

Thanks for your reply Scotty. I knew you would have at least thought of this before since I knew you'd tried with a cable throttle.

So far every time there was an issue I lost throttle, ie no power, never the other way around. But it is a valid point & I was always careful with my testing done on, erm, the local test track in controlled conditions.

I too think (now) a remap would be the best option if you desperately wanted the restriction removed, but I may do a bit more testing after my wideband arrives, just to satisfy myself - one way or another- that it can/can't be done this way. Even if I do get it working properly, I may still remove it because at the end of the day there's probably easier/safer ways to make those last few kW.

It was a bit of a challenge & a sort of fun way to spend a weekend. I'll let others know what eventuates when I get time to have another play.

[zei20l]

Nice work Leon!

I asked a mate to look into this also, but after everything thats gone on, and your testing, its clearly so much bigger than we all thought!

[whiteaxis]

Is the throttle restriction in all gears and at all speeds or is it just first gear and under roughly 30km/h as it feels?

Perhaps it would be possible to trick the ecu into thinking it was in a higher gear or already going faster whilst still in first?

Does the throttle body ever open 100%

[Sunkist]

throttle doesnt open to 100%

i was playing with the car last night and it allows me a throttle of 78% before 30kmh. if your stalling it up it wont budge past 34%.

i think its more so that the Cam's are in retard below 30kmh. as im aware Scott proved this with the feedback he got from the V-manage.

[zei20l]

i was playing with the car last night and it allows me a throttle of 78% before 30kmh. if your stalling it up it wont budge past 34%.

if you wait long enough it will open further. just gotta get the car thinking its going faster than 30km/h. i got mine into 2nd gear at a standstil on saturday

[Sunkist]

nah it wont, i held it for a good 15-20 seconds. wouldnt budge. the boost wanted to go up but it kind of felt like it "wasnt allowed" lol yeh you and you Nismo ECU's

[zei20l]

your car is soft, i could also do it on the stock ECU.

might be the RX smaller rolling diameter wheels that help with mine.

also your turbo is poo

[Sunkist]

hows my turbo poo?

you could spin up all fours with the stock ECU? lol are you dreaming of my sister again?

[zei20l]

i could chirp the front wheels once the rears were spinning. with the nismo it chirps more.

dont bring your sister into this again, she told me she wants to stay out of it

[Sunkist]

mine never does it.

mine looses traction once its moving. if im travelling about 60km an hour and plant it, the wheels will just start burning up. but never when im stalling it. wish i could though..

or was it your Mrs that told you to stay away from my Sister??

[ironpaw]

I don't know how you guys get your wheels spinning in a straight line....

re throttle restriction...I covered about 500km over the weekend...and I have to admit that even with the restriction in place, the cold weather has almost nullified the sluggish start of the line...car's been awesome to drive!

[zei20l]

i dont have as much power as you to spin up all 4 while im moving. not yet anyway!

[ironpaw]

i dont have as much power as you to spin up all 4 while im moving. not yet anyway!

I don't get it though...I've dropped it down to 2nd gear when i'm at say 60 and planted my foot. Can't get the tyres to chirp. Car just takes off like a rocket...