First up this thread is just to get my head straight about a few things.. hopefully from those with direct experience rather than hearsay and conjecture.

I'm lead to believe that there is some difference in the actual valve timing events in the rear bank compared to the front bank in the 6G74/6G75 - is this correct?

If so is this something that is inherent in the camshafts themselves, like the lobe separation angle is just different front-to-rear. Or, is this difference a matter of installing adjustable cam gears and bringing everything back in line?

Next question;
Advancing or retarding the valve timing does what? Stick to our SOHC/6G motors if possible.

This isn't something I'm about to do, or want to do. This is just me getting my head straight on a few things from the ground up.

All input, as always, is much appreciated... but, like I said can I get more input from those with legitimate experience please.

I have seen two different schools of thought espoused: a) there is something like 3 degrees timing difference between the banks, and b) there is little real difference as designed but differences can appear to exist due to problems making good measurements.

I actually sought advice from 2 separate knowledgeable sources, both of whom are familiar with cam design, and got 1 of each of answer. In setting up the Ralliart cams I currently have I went with b) because that answer came from someone ex Mitsubishi...

I've also seen reference to the 6G75 heads having different valve train geometry to the 6G74 heads which may result in different measurements between the two.

One of the issues I've pondered is the behaviour of the timing belt as it stretches over its life, and related to that is how the timing measurements are made depending on the state of the belt tensioner. To say nothing of production tolerances of all the various components involved in this geometry...

Edit: within limits, advancing the timing boosts output at lower revs at the expense of output at high revs; retarding the timing does the reverse. From what I've pieced together, there probably isn't much gain to be had at the top end by retarding the timing more than it is from the factory, but particularly with the AWD the torque gains at lower revs arising from advancing cam timing is useful. I'm running the #7 cams with 8 degrees of advance, and previously ran #6s with 4 degrees of advance (both measurements in crankshaft terms, on the cam gear vernier the values are halved). In an FWD auto I'd probably use only 6 degrees of advance with #7 or #15 cams, but still use 4 degrees with #6 cams. Owners of manual cars may prefer not to advance cam timing at all...

Wow. Excellent response. Thanks.

8 degrees sure seems like a lot of advance though. Said the guy with zero experience

Wait up.

Another two questions;
Are the valve timing events listed in the workshop manual the 0.5" lift values, or, zero lift values?

And what are the "at the valve" lift figures. Or, for we have a definitive rocker arm ratio + cam lobe lift values?

Wait up.

Another two questions;
Are the valve timing events listed in the workshop manual the 0.5" lift values, or, zero lift values?

And what are the "at the valve" lift figures. Or, for we have a definitive rocker arm ratio + cam lobe lift values?

I meant 0.05" lift.... not HALF AN INCH OF LIFT. Sheesh.

Here are some 0.050" figures from a 380 6G75 #15 cams.

Inlet duration 187 degrees @0.050" with 9.70mm valve lift.

Exhaust duration 190 degrees @0.050" with 9.25mm valve lift.

Rear bank 108 degrees lobe seperation. Front bank 109 degrees lobe seperation. both cams just over 3 degrees retard.

Ralliarts # 7's were made in Australia and have around 6 degrees retard. The #15 are made in Japan thats why I think there is some some variants between the 2. The numbers do vary slightly between the 74/75 heads and also engine to engine depending on water pump and idler postion, thats why I would recommend a set of adjustable sprockets if yo want to play around.

Hope this helps, Cheers Dave

8 degrees sure seems like a lot of advance though. Said the guy with zero experience

8 degrees at the crankshaft is only 4 degrees on the camshaft... The 48 tooth cam gear used in the 6G74/6G75 engines means each tooth represents 15 degrees of timing change at the crankshaft, so 8 degrees is slightly more than half a tooth's adjustment. Fidanza adjustable cam gears have vernier scales of +/-12 degrees (ie +/-24 degrees at the crankshaft).

Timing belt stretch retards the cam timing - I was quite surprised at the extra "sparkle" in the engine response when the belt was changed in my car at 90000km/6 years (mechanic reported the tensioner was right at the limit of its travel with the old belt), and that was with stock timing on the original #6 cams.

Timing belts don't stretch, do they?

Here are some 0.050" figures from a 380 6G75 #15 cams.

Inlet duration 187 degrees @0.050" with 9.70mm valve lift.

Exhaust duration 190 degrees @0.050" with 9.25mm valve lift.

Rear bank 108 degrees lobe seperation. Front bank 109 degrees lobe seperation. both cams just over 3 degrees retard.

Ralliarts # 7's were made in Australia and have around 6 degrees retard. The #15 are made in Japan thats why I think there is some some variants between the 2. The numbers do vary slightly between the 74/75 heads and also engine to engine depending on water pump and idler postion, thats why I would recommend a set of adjustable sprockets if yo want to play around.

Hope this helps, Cheers Dave

More information - thanks!

Not wanting to play around though, this is more theoretical... which suits my budget and skill level.

This is me inputting data into some simulation software that's all, then tinkering from there. More detailed data input = more accurate simulation. Well, as accurate as a computer can get.

Doing things like this just keep my mind busy while my budget recovers between mods.

8 degrees at the crankshaft is only 4 degrees on the camshaft... The 48 tooth cam gear used in the 6G74/6G75 engines means each tooth represents 15 degrees of timing change at the crankshaft, so 8 degrees is slightly more than half a tooth's adjustment. Fidanza adjustable cam gears have vernier scales of +/-12 degrees (ie +/-24 degrees at the crankshaft).

Timing belt stretch retards the cam timing - I was quite surprised at the extra "sparkle" in the engine response when the belt was changed in my car at 90000km/6 years (mechanic reported the tensioner was right at the limit of its travel with the old belt), and that was with stock timing on the original #6 cams.

I did see the cam gears, and I have considered using them , but would only do it if there was a tangible result. And even then, the installation of the gears might be outside of my skill level just yet.

Timing belts don't stretch, do they?

They have been using them (timing belts) in motorbikes for a while. I don't ever recall coming across a bike that had failed due to a stretched/broken timing belt.... but I sure remember seeing a few bikes that had snapped cam chains.

While I have to agree with the evidence - that they don't stretch - my mind just wants to believe they do.

just yesterday i changed a timing belt on an accent and i noticed that the new belt advanced the cam about 2 degrees, so it was either awesome Hyundai/Dayco engineering or that they set up the cam timing with a little bit of advance to compensate for stretch.

More info Skapper? Tell me what you want and I see if I can help you out. What program are you using? interested to see what it spits out.

Cheers Dave

Dave - I'm keen! See attached imgur gallery of some screenshots, mostly showing the areas of input I'm unsure of or that are missing information. GALLERY HERE (http://imgur.com/a/P8fyT#0)

The Software is DynoSim Pro. It seems pretty good, but I'm not in any position to vouch for its accuracy until I have all the data to input. There are a few grey areas with the software; the intake manifold type, the combustion chamber design and the ignition timing. I've done what I can where I can using the information I've scoured the forum. Just getting stuck at the camshafts.

Changing the intake manifold or exhaust type can do some funky/amazing things according to this software :/ If I can get things to a base point (stock exec 3.5) and then work from there that would be good for hours of pointless but fun computer time. You know... add six turbos and see what happens kind of fun.

That looks pretty good, a stock 3.5 TJ on puts out 200 hp/5000rpm max torque/4000rpm @ the Flywheel. Thats with a slightly smaller camshaft #6, They have about 8.8mm intake lift and 8.5mm exhaust lift, with a few degrees less duration. Stock 3.5 head flows around 200 cfm @ 28" and a Ralliart has around 220 cfm @ 28". A Ralliart is good for around 235 hp/5500rpm max torque/4000rpm @ the flywheel with 9.4:1 compression.

Hope this helps, Cheers Dave

Okay, Dad Mode Friday night so after I endured a series of imaginary tea parties I was allowed to play with this some more.

The software has some whizz bang tool that runs a gazillion different cam timing combinations and presents the top ten results. Seems legit, so I ran it. The results are purely timing based - no lift change - but I believe it also alters the ramp angles or "lifter acceleration rate".

Results as a dyno;
http://i.imgur.com/ALUyNV7.jpg

Cam specs I used in my "standard" engine;
http://i.imgur.com/KxE8BBd.jpg

Camp specs calculated for best HP;
http://i.imgur.com/PFMTkIn.jpg

I'm still getting the feeling I'm missing something very important here....

Couple of runs; Standard Magna vs' RPW Profiles (http://imgur.com/a/IXjC8#0)

Hay mate. The stage 3 cams have a really nice gain in power from 4rpm and torque after 3500rpm. That seems like a typical loss off power in the earlier rpm witch u do expect. But it manual u don't seem to feel the torque loss as much and especially in a straight line when u can launch in a manual u don't feel the power loss unless your driving around town. Are these
Cam results with standard 3.5l heads and cam timing? any tuning done?

Yup, I figured that would be the trade off.

I'm just tinkering with stuff... stuff I really dont know enough about. But, yes, the "standard 3.5" torque curve is supposed to represent an untuned/modded Magna.

I'd still be silly enough to drive an auto with a Stage 3 cam :P IF I had the money to do that sort of mod.

cams make the 3.5 come to life (with other mods of course), ive been using Tighe Cams for a while now with a bit of timing tune for 98 ron the cams are timed 4&6 degree advanced 6 being the rear bank, not by choice its how it fell into place, below 2.5k rpm feels like im driving a stock 3.5 then it all comes to life after that

cams make the 3.5 come to life (with other mods of course), ive been using Tighe Cams for a while now with a bit of timing tune for 98 ron the cams are timed 4&6 degree advanced 6 being the rear bank, not by choice its how it fell into place, below 2.5k rpm feels like im driving a stock 3.5 then it all comes to life after that

I'd love to have the time, money and skill to do a performance build on either a 3.5 or 3.8. One day maybe. Right now I'm just dreamin' :P

I let the software run through 12 million different cam timing variables (took three days) and it came up with a cam that (supposedly) produces more power than the RPW Stage 3 HKS cam. ALBUM HERE (http://imgur.com/a/SIMzD)

Are any of these damn cam specs close to reality/functional?

I'd love to have the time, money and skill to do a performance build on either a 3.5 or 3.8. One day maybe. Right now I'm just dreamin' :P

So would i :) i do like the idea of retiring the 3.5 and droping in the 3.8 with all the bolt bits my cams etc, still needs to be dyno run and tuned with the cams... but i only drive the car once a month

Can't comment much on the power figures as I'm running a 3.8 not 3.5, but the torque and power curves themselves look similar in shape and crossover points to what I experienced in my car.

I went from standard #15 cams to a set of graham bells prototypes.

Below around 4000rpm there was a bit of a loss in power but once you hit 3500-4000rpm the power starts rising very quickly, peaks at 6000 then starts to drop off slowly.

Can't comment much on the power figures as I'm running a 3.8 not 3.5, but the torque and power curves themselves look similar in shape and crossover points to what I experienced in my car.

I went from standard #15 cams to a set of graham bells prototypes.

Below around 4000rpm there was a bit of a loss in power but once you hit 3500-4000rpm the power starts rising very quickly, peaks at 6000 then starts to drop off slowly.

Nice. What was that like to drive in traffic? Auto or manual?

The engine mods - mechanically - were just the camshafts?

Nice. What was that like to drive in traffic? Auto or manual?

The engine mods - mechanically - were just the camshafts?

It is manual.
TBH it is pretty good in traffic. Can cruise in a school zone (40km/h) in 3rd gear and then accelerate out of it without jerking etc.

It isn't gutless under 4000 or anything, its just noticeable for two reasons.
1 The stock 3.8 has great low down torque which you really notice after the swap from 3.5. After cams it dropped a bit again.
2 once you get over 4000 the power comes on very quick so it highlights the difference. Stock was progressive build of power, cams meant a bit more of a slow build up then it just pulls.

The only change in driving really is that you would drop down a gear more often tan with stock cams. For example when you really want to power out of a roundabout, or overtake.
It's all what you would expect when you go with a bigger cam.

At this stage yes just cams (and fairly tame tune to suit). Had exhaust with HMs already done.

Very happy. Fuel economy only marginally affected.

I'd love to have the time, money and skill to do a performance build on either a 3.5 or 3.8. One day maybe. Right now I'm just dreamin' :P

I let the software run through 12 million different cam timing variables (took three days) and it came up with a cam that (supposedly) produces more power than the RPW Stage 3 HKS cam. ALBUM HERE (http://imgur.com/a/SIMzD)

Are any of these damn cam specs close to reality/functional?

Any chance you could run the program looking for maximum torque between 2000-4000rpm ?
I have converted my Triton from 3.5 to 3.8 and being a 4wd I am looking for more torque low in the rev range. It has quite a bit more than the 3.5 already...

I have the adjustable cam gears and stock 380 camshafts.

I'd be curious too if Skapper were able to find "optimum" figures for the factory cams from the software.

Software modelling aside, I've seen reliable advice from people who know a bit about cams and 6G7x engine performance that the 380 cams work very well advanced 6°. This is based on a target camshaft timing of 3° advanced, taking into account 3° of measured retard on these cams.

With your Triton, provided that you don't mind trading a bit of top-end power for low down torque, an extra degree or two of advance might be helpful based on the information I have - but too much advance is as much a problem as too little.

With Ralliart cams, which are the same profile as the 380 cams but with more ground-in retard, I'm running 8° advance. Low down torque is noticeably improved compared to stock.

NOTE: all timing values above are relative to the crankshaft; divide by 2 to get the timing adjustment on the camshaft.

You will find references to using more advance on the rear bank, however I haven't done this as I received advice from someone ex-MMAL that the difference is more measurement artifact than real with factory cams. However if you go to the trouble of properly timing up the cams to get to the 3° target (plus any extra you might want) you will probably end up with different settings on the two banks as a result of manufacturing variation.

Ah, sorry guys, I ran this today. My nerd box is probably sitting there with the results. I'll post what I have, and anything else that might be of value. Unfortunately the software doesn't run a series of optimizations that represent the effects of adjusting cam timing via adjustable gears. It just burns numbers based on a range. To find the best torque between 2000 and 4000 I just tell it a range within which to try as many variables as possible until it finds the best result. I can however adjust the timing at the input stage willy nilly. Today I input the data to create the stock 380 engine. I'll list all of that input data as well.

Awesome stuff !

Yes I am aware of the difference in timing between banks. My Triton has the engine in a North/South position so I will be timing each camshaft the same.

Looks like 6 deg (crankshaft) will be a good starting point. I can't find the little circle on my keyboard for the degrees.

I swapped the inlet manifolds recently from the 380 one to the Triton one thinking that the smaller diameter tubes of the Triton would increase velocity and torque. Well I was surprised at the difference, waaaay more torque !

https://imageshack.com/i/ex9hwclrj

Hope that link ^ works.
Anyway I am not concerned about high rpm power, it is all about getting 2.5 Tonne moving !

Salt - link works, more information is good.

Anyway I let the software try to develop a cam profile for maximum torque between 2000 and 4000rpm. This software I dont think is super accurate, and really relies on having accurate data for input.

Some PDF's to start with - I'm hoping somebody can confirm and correct anything they see wrong in all of this. I've input any data I could find with google, but I'm sure there's more, and, its possible some of this input data could be very wrong.
Stock 380 6G75 Engine (https://dl.dropboxusercontent.com/u/1479025/380%206G75%20-%20standard.pdf)
Stock 380 6G75 Engine with cam timing retarded by 8 degrees using cam gears (https://dl.dropboxusercontent.com/u/1479025/380%206G75%20-%20standard%20-%20Cam%20Gears%20-%208deg%20red.pdf)

Camshaft Profile used in "Stock 380" (Base Camshaft Profile)
http://i.imgur.com/k6vWf06.png

Camshaft Profile calculated from Software - Best Torque between 2k and 4k
http://i.imgur.com/VhQgJfo.png

Software Setup #1 - Setting the cam timing ranges within which to search. I'm going to set these a bit finer and with the maximum range, so lookout for those results.
http://i.imgur.com/BwP4IWK.png


Software Setup #2 - Setting the requirements of the new cam profile.
http://i.imgur.com/aPjPCNV.png

Resulting Torque/HP Graphs Comparing Stock 380, "Best Torque" Result Cam Profile and the stock engine with cam timing retarded 8 degrees
http://i.imgur.com/xHtDuTY.jpg

Any help tightening up my input data would be greatly appreciated. Any input also appreciated.

Looks complicated...

What about cam timing 6 degrees advanced ?

Looks complicated... What about cam timing 6 degrees advanced ?

I know right? The understand the bulk of this, but it's annoying that I don't have more information.

https://www.dropbox.com/s/sqqliphelpf7w0h/380%206G75%20-%20standard%20-%20Cam%20Gears%20-%206deg%20adv.pdf?dl=0

That's awesome mate, The dyno page certainly shows torque ramping up quicker to 3000 rpm.
Starts to fall behind the standard profile after 3000 though.

I will experiment a bit with the cam timing and find the best compromise.

Thanks for spending the time to help Skapper :thumbsup:

I have just installed the adjustable cam gears on my 6G75 (6G74 block)

I found some interesting results, these cams are the stock 6G75 camshaft supplied in the 2007 AU Mitsubishi 380 sedan.

So the results are as follows;

Intake lobe lift at camshaft is 6.3mm

Exhaust lobe lift at camshaft is 5.6mm

initially I set the adjustable gears to zero or stock and took some measurements with the dial indicator.

The Drivers side or no. 1 piston Inlet valve started moving at 6 degrees BTDC (crankshaft)
The Passenger side or no. 2 piston inlet valve started moving at 0 degrees or TDC

So I advanced both camshaft gears to have the Inlet valve starting to open at 15 degrees BTDC (crankshaft)

These measurements are all taken with the dial indicator sitting on top of the rocker arm directly above the valve

Intake opens 15 Deg BTDC
Intake closes 80 Deg ABDC

Intake centreline is 115 Deg

Exhaust opens 55 Deg BBDC
Exhaust closes 50 Deg ATDC

Exhaust centreline is 100 Deg

What is interesting is the duration of both intake and exhaust is a lot longer than any specs I have seen.

I double checked both banks and got slight variations but only by a couple of degrees.

Not really sure why, any thoughts ?

I have just installed the adjustable cam gears on my 6G75 (6G74 block) I found some interesting results, these cams are the stock 6G75 camshaft supplied in the 2007 AU Mitsubishi 380 sedan. So the results are as follows; Intake lobe lift at camshaft is 6.3mm Exhaust lobe lift at camshaft is 5.6mm initially I set the adjustable gears to zero or stock and took some measurements with the dial indicator. The Drivers side or no. 1 piston Inlet valve started moving at 6 degrees BTDC (crankshaft) The Passenger side or no. 2 piston inlet valve started moving at 0 degrees or TDC So I advanced both camshaft gears to have the Inlet valve starting to open at 15 degrees BTDC (crankshaft) These measurements are all taken with the dial indicator sitting on top of the rocker arm directly above the valve Intake opens 15 Deg BTDC Intake closes 80 Deg ABDC Intake centreline is 115 Deg Exhaust opens 55 Deg BBDC Exhaust closes 50 Deg ATDC Exhaust centreline is 100 Deg What is interesting is the duration of both intake and exhaust is a lot longer than any specs I have seen. I double checked both banks and got slight variations but only by a couple of degrees. Not really sure why, any thoughts ?

Whoa. This is great, thanks for sharing.

My thoughts? That I might get some adjustable can gears....

I will input this into the software though, just to see what happens.

Hey that's great your having a go at understanding what's going on Salt.

Your method of measuring is a little off that's all.

Here is how I do it. I'm not saying it's the only way but it's the way that seems to be the most consistent for me.

I measure off the valve retainer using an extension on the dial indicator making sure it's set up at the same angle as the valve stem. (Remembering it is really the valve timing we're looking for).
I also use a gutted valve lifter with a rocker arms with a adjusting screw and lock nut fitted so I can zero the lash on the base circle before starting measurements. This takes away the possibility of the hydraulic lifter bleeding down while measuring.
The next thing is manufacturers/cam suppliers/computer program usually specify an amount of lift before they start the measurement. Most European and Japanese manufactures use 1.00mm(0.040") lift before measurement and most American and after market manufacturers 0.050"(1.25mm). I also remove all the spark plugs to make it easier to rotate the engine and also confirm TDC whith a piston stop through the spark plug hole.

There is always going to be some variation you only have to look how vague the rocker gear is located. Graham Bells excellent advise with the slight variants between the banks is just make sure your inlet valve closing points are same, this gets the cylinder pressure closer to each other.

Hope this helps, cheers Dave

Thanks guys.

Yes I used a piston stop to find TDC on both banks.

I did not realise the stock figures would be at 1mm lift though, that throws my timing out by a good margin. Also the stock timing would be retarded by more as well.

I need to do it all again, when I have time, big job stripping manifolds, wiring etc to get to the valves.

Very good advice to keep the inlets closing at the same time, will be aiming for this for sure.

I'm glad to see this thread grow, loving the information and genuinely appreciate the efforts made to clarify and share the information further.

To chip in, I'll source the tools mentioned and run a few measurements on the standard 6G74 siting in my shed.

The dyno software I have has the option to set the valve events to either contact (zero lift) or at 1.25mm lift, as DaveTJ mentions. And I feel this is the key to cracking the code.

You see, I want to have the software produce an accurate representation of the 6G74's stock output. As accurate as possible anyway. From there, predications could be made with some confidence at least. Taking into account the softwares limitations of course.