Hi guys I have a 2009 Triton with the 6G74 3.5 litre engine.

I want to swap the crank to a 6G75 Mitsubishi 380 one and also swap heads, given the '75 heads flow much better than the '74 ones

Can anyone tell me will the '74 flywheel bolt up ?

Will the cam timing belt be a different pitch or length?

Are the manifold bolt configurations the same ?

I plan on buying a 6G75 motor complete and will possibly use more parts.

I do want to retain the Triton ignition, Fly by wire throttle and (unfortunately) oxygen sensors.......

Any advice or thoughts would be appreciated

I want to swap the crank to a 6G75 Mitsubishi 380 one and also swap heads, given the '75 heads flow much better than the '74 ones.

I do want to retain the Triton ignition, Fly by wire throttle and (unfortunately) oxygen sensors.......

Swapping the crankshaft would also require new pistons too, surely? (remembering that the 6G75 pistons have a smaller crown to gudgeon pin centre height than the 6G74 to account for the extra stroke).

Just swapping the heads might get you a bit, at the cost of a fair loss of compression (which could be addressed by custom pistons). 380 (#15) or Ralliart (#7) cams into the 6G74 heads would probably be about as effective as swapping 6G75 heads on a stock 6G74.

Seems like a complete 6G75 swap would be more cost effective if you can make it work.

For a Triton, as with the Magna AWD, some camshaft advance with the #7/#15 cams should improve low end torque usefully.

By the way, what are the quoted power and torque figures for the 6G74 in the Triton? (older Tritons & Challengers typically had detuned engines compared to the Pajero with the same motor, such as the 6G72).

I'm also intrigued by what capability there might be for ECU modification on these vehicles.

Matey take a look at the 380 engine, it runs coil on plug ignition and drive by wire as standard.

Swapping the crankshaft would also require new pistons too, surely? (remembering that the 6G75 pistons have a smaller crown to gudgeon pin centre height than the 6G74 to account for the extra stroke).

Just swapping the heads might get you a bit, at the cost of a fair loss of compression (which could be addressed by custom pistons). 380 (#15) or Ralliart (#7) cams into the 6G74 heads would probably be about as effective as swapping 6G75 heads on a stock 6G74.

Seems like a complete 6G75 swap would be more cost effective if you can make it work.

For a Triton, as with the Magna AWD, some camshaft advance with the #7/#15 cams should improve low end torque usefully.

By the way, what are the quoted power and torque figures for the 6G74 in the Triton? (older Tritons & Challengers typically had detuned engines compared to the Pajero with the same motor, such as the 6G72).

I'm also intrigued by what capability there might be for ECU modification on these vehicles.

Yep definitely new pistons for all the above reasons and also planning to bore as far as practical.
The 6G75 heads flow way better than 6G74, up to 22%
The bolt on stuff is all different on the block and so I need to retain the original block......

Triton 6G74 (3.5 litre) is 135 Kw , very detuned for a two tonne vehicle....... I am aiming at somewhere around 200 Kw, focusing on low end torque.

I will be tuning via software which re flashes the ECU.

Matey take a look at the 380 engine, it runs coil on plug ignition and drive by wire as standard.

Awesome that takes care of that problem :D

Yep definitely new pistons for all the above reasons and also planning to bore as far as practical.
The 6G75 heads flow way better than 6G74, up to 22%
The bolt on stuff is all different on the block and so I need to retain the original block......

Triton 6G74 (3.5 litre) is 135 Kw , very detuned for a two tonne vehicle....... I am aiming at somewhere around 200 Kw, focusing on low end torque.

I will be tuning via software which re flashes the ECU.

Cams + adjustable cam gears, exhaust and a tune would get you a long way towards your objective for modest outlay given that level of detuning....

Are you able to be more specific about the reflashing setup?

Cams + adjustable cam gears, exhaust and a tune would get you a long way towards your objective for modest outlay given that level of detuning....

Are you able to be more specific about the reflashing setup?

From the USA it is a small thing that has downloadable software that I assume plugs into the vehicle somewhere. It then reflashes the ECU with whatever tune is programmed into the software. I won't be playing with this myself and will be relying on a switched on technician at the dyno shop

Reviving this thread instead of starting another.

Well the 6G75 crank, pistons, heads and inlet manifolds are in and running :happy:

Had to rebuild my block as no way the 380 block would work, all bolt holes are different.......

So I have run the engine in and changed the oil to a full synthetic.

I have some fueling issues, when stopping at traffic lights etc the AFR's run down below 10:1 and often the engine will stall, too rich......

The other issue is fuel trims, was -12% to-6% so reset the injector scaling from 221 cc/min to 261 cc/min assuming the 380 injectors flow 305 cc/min....

Fuel trims are still low but a lot better. The ECU now brings the AFR to 14.7 a lot quicker and the stalling has all but gone.

Nut now I am seeing knock counts of up to 15 at WOT and up to 11 at cruise.

I am logging with Evoscan and not 100% sure I am logging knocksum.

Am trying to learn the dissassembly program to find the info myself but this is taking a long time.

Hopefully someone here has had a similar experience and can help !

Well the 6G75 crank, pistons, heads and inlet manifolds are in and running :happy:
Glad to hear that part of it has all gone well!


I have some fueling issues, when stopping at traffic lights etc the AFR's run down below 10:1 and often the engine will stall, too rich......

The other issue is fuel trims, was -12% to-6% so reset the injector scaling from 221 cc/min to 261 cc/min assuming the 380 injectors flow 305 cc/min....

Fuel trims are still low but a lot better. The ECU now brings the AFR to 14.7 a lot quicker and the stalling has all but gone.
Don't get too hung up on the absolute values - they're only as good as a lot of other calibration in the ECU. You want the fuel trims within +-5% from what I've read (and you need to allow for 1-2% variation); if it needs some more adjustment of the scaling to get that right then do it. You should also check out the injector latency table too (see the Evo Tuning Guide if you haven't already read it). I'm guessing that the enrichment and enleanment tables may also need a bit of tweeking. Also is the fuel pump up to the job? Likewise the fuel pressure regulator? (given the Triton 6G74 was so derated compared to the Magna/Pajero 6G74s...)


Nut now I am seeing knock counts of up to 15 at WOT and up to 11 at cruise.

I am logging with Evoscan and not 100% sure I am logging knocksum.
Have you been in touch with the guy who did your ECU definition? He should be well placed to advise you on the above... Presume you're logging with the ODB2 protocol rather than MUTII?


Am trying to learn the dissassembly program to find the info myself but this is taking a long time.
Hard row to hoe that, if its not your usual business. :headbange :badpc: I wish you success!

Lol Thanks Andy, I am taking it slowly and learning a little at a time, I am about three years from where I want to be though.....

Yep waiting to hear back regarding the addresses, I also emailed a couple of others and offered payment to discover the correct addresses.

I have read the guide and refer to it all the time, I was under the impression that I need to get the trims very close to zero before I started manipulating the latency table.
My fuel pressure is regulated at the pump and according to the manual has the same pressure as the 380. Seems to be ok as running rich through the entire rev range, might get a pressure test anyway.

I am logging with both the Evox can Mut3 and the OBD11, the rest won't log. With these I can log most things but not at the same time. Evox can Mut3 lets me log the fuel trims, speed, spark advance, MAP, knocksum ( I think it may be logging the knock base though) and depending on what boxes are ticked sometimes MAF. Seems a bit hit and miss.
The OBD11 lets me log coolant temp, air temp, ambient temp, Baro, speed, rpm and some others that I can't remember right now.

Anyway the 6G75 upgrade has been a challenge from the outset but well worth it !
Triton goes well now..... towing our camper around the countryside will be a breeze and it is lots of fun at the traffic lights:happy:

what did the compression work out to be?

I haven't tested but should be 10:1, definitely more than the 9.5:1 it was before:happy:

maybe u should check this.....

I will test when the inlets are off for any reason, which might not be far away if I can't get the logging sorted. I will make a knock listening device....

It definitely has more compression than before, like having an engine brake when the fuel cuts off decelerating.

I will test when the inlets are off for any reason, which might not be far away if I can't get the logging sorted. I will make a knock listening device....
Because you are logging knock I'm presuming the original engine had a knock sensor. Given the engine changes, I wonder if the ECU's knock sensing filters need tweaking as they would be optimised for the response of the block with the original bore and compression.

I remember seeing an article on autospeed about making a knock listening setup from a cheap "personal amplifier" (if I remember correctly the device was sold by DSE and was known as a Whisper???? - no longer available of course, but very similar devices are readily available on fleabay...). I had thought about trying to adapt this to directly tap the ECU's knock sensor...(if I can get my hands on a knock block).

Yep this one....http://www.autospeed.com/cms/article.html?&A=0353

I have a spare block and sensor if you want it ?

I think this will be the only way I really know for sure whether knock is present. I am not even sure that I was logging knocksum, I tend to think I may have been logging the knock base looking at the timing behaviour. 11 counts of knock and the ECU was not pulling any timing. 16 counts and the ECU pulled 2 degrees, as it should for 6 counts ?

Does the knock base get reset when the ECU is flashed ? I also disconnect the battery for 30 seconds to clear the ram after flashing, would that reset the knock base ?

Yep this one....http://www.autospeed.com/cms/article.html?&A=0353
That's the one I was thinking of.


I have a spare block and sensor if you want it ?
PM sent.


I think this will be the only way I really know for sure whether knock is present. I am not even sure that I was logging knocksum, I tend to think I may have been logging the knock base looking at the timing behaviour. 11 counts of knock and the ECU was not pulling any timing. 16 counts and the ECU pulled 2 degrees, as it should for 6 counts ?

Does the knock base get reset when the ECU is flashed ? I also disconnect the battery for 30 seconds to clear the ram after flashing, would that reset the knock base ?
Sadly you're in territory that I haven't investigated yet. I do have a couple of Diamante ROMs which have knock sensor support and I know that various people would like to have this in Magnas, so maybe one day I'll get to dig into it.

I do remember reading (on the EvoM forums I think) that the retard applied in the presence of knock was about 1 degree per 3 counts, so 2 degrees of retard for 6 counts of knock "delta" does fit. I'm wondering whether the relatively constant "knock base" of about 11 means that you've got effectively a nearly constant retardation of about 4 degrees? Can you correlate your logged ignition advance with the values in the main high octane ignition map (perhaps in conjunction with the EGR advance map)?

Having some sort of direct listening tool does help identify whether you've got real or phantom knock. When it comes down to it, I gather human hearing is still generally considered to be more effective at distinguishing knock from engine noise than any electronic gadgets yet available :(

What fuel are you running at the moment? In the absence of the knock ears, are you able to try a different (preferably higher octane) fuel and see whether that changes the logged behaviour?

I know with the older ECUs clearing the RAM required a power disconnection but I think the newer ECUs do clear the RAM after a reflash. Disconnecting the battery does make sure though ;)
Whether the "knock base" is reset would depend on whether it is calculated or preset in the ROM. You might be able to check this by disconnecting the battery, then starting logging before you start the engine but with the ignition key ON. Once the ignition key is ON, even if the engine isn't started, the ECU is running and can be logged (though most of the values won't be updated until the engine is started) - at least this is true with the SH2 ECUs.

OH:shock: I assumed they all had knock sensors....

No wonder they go so hard, not worried about detonation lol.


I do remember reading (on the EvoM forums I think) that the retard applied in the presence of knock was about 1 degree per 3 counts, so 2 degrees of retard for 6 counts of knock "delta" does fit. I'm wondering whether the relatively constant "knock base" of about 11 means that you've got effectively a nearly constant retardation of about 4 degrees? Can you correlate your logged ignition advance with the values in the main high octane ignition map (perhaps in conjunction with the EGR advance map)?

Yep this is in Merlins tuning guide as well, where I am referring from, also Merlin touches on how the knock base works and how the ECU learns what is knock and what is not...it has three different multipliers to assess the knock and switches between the three at will depending on learned value which makes me wonder if it is reset when flashing/battery disconnect. It could take a couple of tanks before the ECU learns what is knock.......I am guessing atm.....
My timing is very close to the High octane spark advance map in all situations, I don't have the EGR spark map but the two EGR fuel maps instruct negligible change to the AFR(14.3:1 @ 3500), EGR is blanked.....
The logs show knock count above about ten is when my ECU starts to pull timing.
Unfortunately my laptop died yesterday and I lost all the logs but can remember most of it or I would post the log pics.

I am running 95 octane atm and wanted to tune on it. When I can afford it the LPG will be hooked back up, just need to pay for the injectors to be placed in the 380 manifold and tuned. The rest of the LPG system is in the car already from the previous engine. That is probably too big a jump in octane though, I think LPG is around 115 octane ?


Whether the "knock base" is reset would depend on whether it is calculated or preset in the ROM. You might be able to check this by disconnecting the battery, then starting logging before you start the engine but with the ignition key ON. Once the ignition key is ON, even if the engine isn't started, the ECU is running and can be logged (though most of the values won't be updated until the engine is started) - at least this is true with the SH2 ECUs.

The way I read about knock base in Merlins guide is that to some extent the ECU does learn. Which makes me wonder if it is reset each reflash.....

Mine won't log anything with ignition 'on' not even TPS, has to be running for some reason:roll:

OH:shock: I assumed they all had knock sensors....

No wonder they go so hard, not worried about detonation lol.

None of the 3rd gen Magnas or Veradas have knock sensors (though the EFI 2nd gens do) - they're just tuned for 91RON (even the Ralliarts).


Yep this is in Merlins tuning guide as well, where I am referring from, also Merlin touches on how the knock base works and how the ECU learns what is knock and what is not...it has three different multipliers to assess the knock and switches between the three at will depending on learned value which makes me wonder if it is reset when flashing/battery disconnect. It could take a couple of tanks before the ECU learns what is knock.......I am guessing atm.....
My timing is very close to the High octane spark advance map in all situations, I don't have the EGR spark map but the two EGR fuel maps instruct negligible change to the AFR(14.3:1 @ 3500), EGR is blanked.....
The logs show knock count above about ten is when my ECU starts to pull timing.

That at least is positive.


Unfortunately my laptop died yesterday and I lost all the logs but can remember most of it or I would post the log pics.

Bummer :(


I am running 95 octane atm and wanted to tune on it. When I can afford it the LPG will be hooked back up, just need to pay for the injectors to be placed in the 380 manifold and tuned. The rest of the LPG system is in the car already from the previous engine. That is probably too big a jump in octane though, I think LPG is around 115 octane ?

Try a couple of tanks of 98RON? If it is harder to provoke the timing retard with 98RON, that would be another indicator that the knock sensing is behaving predictably.

I would have expected the LPG to have its own piggyback ECU which can add extra advance to take advantage of the much higher octane rating?


The way I read about knock base in Merlins guide is that to some extent the ECU does learn. Which makes me wonder if it is reset each reflash.....

Mine won't log anything with ignition 'on' not even TPS, has to be running for some reason:roll:
Bummer! If you can log from as soon as the engine starts you might catch some stuff (such as fuel trims) before they get updated - perhaps trying this with a cold engine might work better, as the fuel trims won't get updated until the ECU goes into closed loop. If the fuel trims get reset after a reflash, as they are after a battery disconnection, then I would expect everything stored in RAM (which includes nearly all the loggable values other than IDs) is invalidated after a reflash and reset.

Hmm, yes good thinking on the logging form a cold start, when I get another laptop i will try it.

Yep the LPG has it's own ECU, am trying to wheedle the software from the supplier, they are hesitant so far.....