The knock sensor bolts into a bracket in the center of the V which in turn bolts to the block.
Only high compression blocks are machined for the bracket.
Early Mitsu knock circuits are analog, later are digital.
The sensors are different for each. I have not looked up exactly when the sensors changed.
Frank, it may be possible the later sensor would be compatible with KS.

im placing a engine of that type in my 1996 montero ls. im making a step by step manual well instructional on my own experiences.

ive got that so thats great its easier, also finally found my block castings, BLOCK CAST #6G74M, DRIVER SIDE REAR, HEAD CAST #G7S4FF, however the diamante intake said it was a 3.0, weird mitsubishi mislabeled their engine with a 3.0 intake, but low and behold its a 3.5, goodtimes. got a better engine for only 250.00 if anyone knows any wiring tricks lemme know. thx

ES, did you ever figure out your overheating issues fully? I'm considering this swap since I have the motor just taking up space in the garage, all looked pretty straight forward to me.

Sorry for the hiatus, but it's been insane this year. Bought a house, expanded the office, evacuated from said house for forest fire, caught in flood on I-70, and now dealing with flood damage on a house I've only had for five months. <img src="/forums/images/graemlins/frown.gif" alt="" />

Anyway, I'm still plunking along with the swap. I've learned some stuff and fixed some stuff and am STILL fixing some stuff ...

The overheating problem - yea, that's still a problem. The new radiator helped. Fixing two HUGE vacuum leaks helped. A few other minor things "helped". The problem remains - the 3.0L engine ECU has too much timing for the 3.5L engine. As I think I posted before - it's like tuning a 350cid Chevy SB vs a 350cid Pontiac. Sigh. The timing issue if not resolved will get worse the lower in altitude you get and the hotter ambient air temps are.

Another thing that came up this spring was the gearing. With the 3.0L I was running 5.38s. That gearing worked "ok" with the 3.5L, I guess. It was nice on the highway in the mountains at 7-8K+ feet, but the lower in altitude you went, the more annoying the gearing was. In a somewhat successful attempt to "force" the ECU to pull timing and reduce the overheating I dropped gears to 4.90s. That worked kinda - nothing great but it helped a tad.

It helped A LOT with the 'annoying gearing' part, though. The problem with the 5.38s is that from a stop I blow through the torque curve before you had a chance to get the vehicle barely moving. At 4K feet in Moab it was even worse. It was like driving a tractor or having a granny-gear first gear. PITA. Didn't do anything for my gas millage, either.

I would strongly suggest to anyone doing this swap at sea level going with 4.63s if you're streeting it and NO DEEPER than 4.90s for mixed use. There's enough torque in the 3.5L that with big tires and a little injudicious use of the gas pedal its possible to pop something in the drive train. In my case probably a u-joint. Keep in mind that messing with the overall ratio can do unpleasant things to the ECU which will affect the ability to pass emissions, so ....

I put the ECU/ECM swap on hold. Its mostly wired but I had to stop work on that to deal with a bunch of other things - including a minor redesign to my cross-over steering to fix a long standing problem that left me stranded. Amazingly enough, in May it DID actually pass emissions. Barely. I'm not entirely sure they didn't do some hokey magic to accomplish this. They tested it twice and fiddled with stuff on their equipment that looked somewhat suspicious. Given the way the thing smells I find it hard to believe it validly passed. But, who knows ....

I decided to go the AEM F/IC route as the 'simplest' and quickest possible fix. I ordered that and its showed up last week. Now I have an Innovate LC-1, the AEM F/IC 8 and the Boomslang patch harness so I should be good-to-go.

I used ScanXL and built a bunch of graphs to map timing, MAP, STFT and LTFT, temps, etc. to be able to figure out how much timing and fuel has to come out and where for the F/IC. I have the maps for my big Sport with the swap, but I still need to get the baseline mapped from my stock '97. Its currently on loan to my friend while I'm having her '98 rebuilt. Can't start tuning at least until I get the base maps. Sigh.

The mechanical swap itself is pretty straight forward. Just make sure you have all the little pieces, 'cause some aren't available from the dealership anymore and yards typically won't sell them individually.

AND, be prepared $$$-wise for the cost of doing the electronics piece.

Edward

I managed to get something done last week while waiting for contractors to tell me how many trillions of dollars its going to cost to take care of the flood damage. Sigh.

Here is the preliminary wiring pin-out and diagrams for installing an AEM F/IC 8 piggy-back unit and Innovate LC-1 wideband O2 sensor on top the existing 1997 5-speed ECM. I'm still waiting for a couple relays and momentary switches w/ LEDs but hopefully I'll have time this week to start wiring the F/IC and LC-1 into the Boomslang pass-through harness.

Montero Sport 1997 ECM & 1999 PCM w/ AEM F/IC 8 and Innovate LC-1 Wiring Diagrams

As always, let me know if any of you find any glaring error or typos.

The diagram includes wiring information for ALL applicable wires on the F/IC 8 and LC-1. It doesn't mean that all of them need to be used.

I selected the F/IC 8 because it has a much larger memory buffer for internal logging. I also selected the F/IC 8 because its required to interface with our frequency based MAF. I almost certainly will never need MAF control since I only need the F/IC to pull down the timing and zero out the fuel tables. However, I want to be able to log over a long - week or two - period of time and the F/IC 6 memory buffer isn't large enough.

The Innovate LC-1 is JUST the wideband (WB) O2 sensor - no gauge. That's fine for me since I do all my logging off my diagnostic software on my laptop. The diagram does, however, have the LC-1 tied into the F/IC so WB O2 readings will be logged by the F/IC. This setup will work for me because the LC-1 outputs both in analog and serial. The analog WB signal goes to the F/IC, and I can still plug the serial connection into my separate diagnostic software. You'd need to do something a little different if you wanted a gauge as well as the WB going to the F/IC.

The only connections I technically need to the F/IC will be the fuel injectors, TPS, cam and crank. The narrow band O2 connection isn't needed if all you're trying to do is decrease fuel supply - as fuel is decreased trims will move to zero 'on their own' as adjusted by the ECM.

The timing is pretty straightforward. I mapped my big Sport with the new 3.5L so I know where the timing is but I still want to map my stock '97 with the factory 3.0L so I have a better idea of what 'stock' looks like. Once that's done here in a couple days it's just a matter of doing some simple math, tapping into a couple decades of tuning Pontiacs and small block Chevys and modifying the appropriate cells in the timing map for the F/IC. This will be the time consuming part as unlike the fuel where its an across-the-board cut, the timing doesn't need to be modified everywhere by the exact same amount.

The point of having the LC-1 WB is to keep me from doing something stupid accidentally (hopefully) and blowing my new engine. If at some point I get this working as I expect it to I'll maybe consider tweaking things a little to better accommodate the modifications I've done and the weight of the vehicle. THAT may require the MAF and narrow band O2 hook-ups.

Anyway, since I was going through everything anyway I figured it'd be a good idea to document all the connections, used or not, since this information is pretty hard to come by and a lot of what I found on the 'Net isn't applicable to Montero Sports or is just flat incorrect.

KIM that I have not yet wired this up and run it yet. Thanks to all the Series 2 DSM guys since that's about as close a ECM match as us '90s Montero Sport guys will get. I presume since it works for them it'll work for us. My diagrams don't quite match up in some places with the couple guys that run these on their supercharged Sports - that's because I actually went through the service manuals for months and understand the factory wiring better, I think. Stuff is where it is for a reason. i.e. grounding is absolutely CRITICAL for proper operation of both the F/IC and especially the LC-1. Guess I'll find out.

Edward
www.4x4extremesports.com

Have you taken into account the functioning of the det sensor, or does your engine have one? My 3.5 dohc had one in the middle of the vee. Will the timing gizmo pull timing if det is detected like the stock ecu?

The '97-'05 Montero Sports never came with knock sensors. Kinda lame, IMHO. So, no, there's nothing to pull timing within the ECM/PCM other than via the normal programed in timing/load maps. In addition, there's no way to hook up a knock sensor(s) unless you go with the Megasquirt since the F/IC doesn't have any circuitry for that. I wish they did, but knock sensing is a science in and of itself and most turbo or supercharged engines come with their own knock sensors already so I can understand their rational.

The general operation of the ECM/PCM is to increase timing (up to a point) as load decreases. Timing runs from 0 to 40 degrees. My reasoning behind changing gearing from the 5.38s to the 4.90s was an attempt to increase load thereby getting the ECM to decrease timing. Which, as I said earlier, kinda worked somewhat.

I see this in action going from 5th gear down to 3rd gear on the highway on protracted inclines. In 5th it's MOSTLY ok, but on dropping to 4th (decreasing engine load) the timing goes up and it really starts heating up. 3rd is impossible. I had to stop three times on Eisenhower Tunnel to let it cool down and ambient was in the low 50s. <img src="/forums/images/graemlins/frown.gif" alt="" /> <img src="/forums/images/graemlins/frown.gif" alt="" /> It's fine in town because of the stop and go but going over the passes is terrible. <img src="/forums/images/graemlins/frown.gif" alt="" /> That's 15 miles of 4-12% grade at 9000+ ft.

Even that isn't the worst part. Off road its REALLY bad. Even in freezing temps at 12K ft. with the fans on high and with the new radiator it's a big problem. It MIGHT be less of a problem with the 1.92s and stock diff gearing (kinda like my change to 4.90s), but with my gearing I can do a 50 degree incline without even touching the gas pedal so the load is SO low that the timing pegs at 40 degrees - which is the maximum on the timing table - and WAY WAY WAY too much for a 3.5L.

In CO I can work around it off road, but in UT its a real problem. Lots of trails don't have safe 'pull-offs' and you just can't stop on some inclines for the hell of it to let the engine cool down. And the hotter ambient is, the worse the problem. Coming out of Potato Bottom or up from Potash really sucks in the summer. <img src="/forums/images/graemlins/frown.gif" alt="" /> I've had to get creative with the fans and air conditioning fan to keep the temps somewhat in control.

Towing BTW is out of the question with things the way they are. <img src="/forums/images/graemlins/frown.gif" alt="" />

Edward

So, they released a new version of DashCommand yesterday for the Droid. It's still kinda crappy, but they seem to have fixed a few things - like the busted fuel trim issue I've been having. <img src="/forums/images/graemlins/rolleyes.gif" alt="" /> Anyway, I plugged it in for the hell of it to test it out. It was interesting to see how badly the timing is out of whack under light throttle. <img src="/forums/images/graemlins/frown.gif" alt="" />

Brief explanation of WHY this is a problem ...

Engines need timing advance. You can just set the timing to zero so the spark is triggered at top dead center (TDC) of the stroke because of several physics issues. Primarily its because fuel doesn't burn instantaneously. So the mixture has to be ignited slightly BEFORE the piston reaches TDC to get the maximum amount of 'push' on the down stroke.

It's a trade off. Since the combustion process starts BEFORE the piston reaches TDC the engine is 'fighting' against the force of the expanding combustion process as the piston moves upwards to TDC. But, the mixture must be ignited BEFORE it reaches TDC; just not TOO MUCH before.

What's happening in this case is the combustion process is too far along while the piston is still on the upward stroke to TDC. This is causing the engine to fight against the combustion process. In other words, it's like a big hand pushing down on the piston while the crankshaft is still trying to push the piston up. This generates an enormous amount of heat and stress - more than the engine and cooling system are designed to get rid of.

Ignition timing that is advanced too far wastes energy in the combustion cycle trying to push the piston the wrong way.

Edward

Long time, no post. So, here is where its at ...

I ran the new Boomslang/AEM F/IC 8 harness installed with the loop-back plugs installed for a week to make sure that there weren't any apparent miss-wiring problems. No issues there, so apparently I got something right. <img src="/forums/images/graemlins/smile.gif" alt="" />

I will get back in touch with Boomslang after I'm done and get them my documentation so they can wire up a complete Boomslang/AEM F/IC 8 harness for anyone that doesn't want to go through the headache and expense of doing their own at least for the '97 and '99. If I can find a set of '98 service manuals, I'll add the wiring for '98.

There are two problem with installing a piggy-back. The first is making sure that the unit is grounded perfectly and in the same ground loops as the ECM/PCM. This is REALLY, REALLY important!!!!!! There's lots of documentation regarding this and not just for the F/IC so I won't beat the whys and wherefores into the dirt. My wiring diagram explicitly addresses this.

The second problem that no one ever points out is that when building harnesses to plug into or wiring in a harness to the stock ECM/PCM wiring is resistance. I'm sure with a little digging it wouldn't be hard to figure out that this is a big problem and likely at the root of a lot of barely working or flaky installations where the correct wiring has been checked dozens of times and there's still bad behavior.

Lots of sensors on a vehicle change resistance as conditions change. The ECT is the perfect example. So, any change in resistance between the ECM/PCM and the sensor changes the actual data ariving at the ECM/PCM. The manufacturers factors in added resistance between the ECM/PCM and sensor in the ECM/PCM - for the factory installation, some degradation, variation in sensor output, etc.

I've already run across this. I'm getting a CEL (P0203) for injector 3 on ignition on with the F/IC plugged in, but not with the harness installed on loop-back. Several things are tested on 'ignition on' before start for 'presence' - resistance through the circuit to confirm that they're 'there', 'good' and reading withing a pre-defined range. Injectors and transmission solenoid system valves are two things that Mitsu clearly documents.

In this case it appears that I have an injector that's right on the hairy edge and the extra resistance of the F/IC is JUST enough to push it over.

My suggestions to those wanting to do this (or anything like this): Make sure you're real good with a soldering iron and are capable of making 30 odd perfect joints. Have and know how to use a good multi-meter 'cause you're going to need it to check and confirm every one of your joints - if they're measuring over 1.3/1.4 ohms you probably screwed up and it/they need to be redone. Get a good, lighted magnifying glass 'cause this is intricate, tiny work. <img src="/forums/images/graemlins/smile.gif" alt="" /> And, most importantly - be patient and TAKE YOUR TIME. Believe me, this is NOT something you want to have to tear out 10 times before you get it right. <img src="/forums/images/graemlins/frown.gif" alt="" />


AEM F/IC 8 Harness for Montero Sport



The green wiring and ECM connectors make up the Boomslang straight-through harness. I know guys have 'sucessfully' done this but there's no way I'd try hacking the stock wiring. I can't imaging trying to do all the soldering required while wedged under the dash. <img src="/forums/images/graemlins/frown.gif" alt="" />

You never use all the wires that come on the AEM F/IC harness, so I snipped off the 'never gonna be used' wires at the F/IC plugs. Those were the wires for all the alternate things like cam and crank sensors, etc. The small group of unattached wires are the ones that I MAY use in the future so I want to keep those around. They'll get bundled inside the sheath so I can pull one or more out in the future if necessary.

I did maybe make one 'mistake'. I sniped off bank 2 O2 sensor wire. I wired in bank 1 O2 to the harness in case I ever wanted to change the fuel delivery to other than what the ECM does on it's own. In hindsight, I probably should have wired in bank 2 just in case. Some ECMs use the post-CAT O2 in combination with the pre-CAT O2 sensor when calculating fuel delivery. I don't THINK anything pre-'05 does that on Montero Sports or Monteros - but I've already run across at least one thing with the post-CAT O2 that isn't supposed to do something it obviously does. <img src="/forums/images/graemlins/rolleyes.gif" alt="" />


Completed AEM F/IC Harness for Montero Sport



This is the completed harness plugged into a spare ECM for picture taking purposes. The flat Weatherpack connects to the coil of white and black wires. The white and black are the actuators for logging on the AEM and the calibration/indicator for the Innovate LC-1 Wideband O2. The flat Weatherpack is also wired for an intake air temperature sensor (IAT). So, two for the WB O2, two for the AEM logging and two for the IAT. The round gray connector goes to the Innovate LC-1.


Test install AEM F/IC Harness for Montero Sport



Everything except the Innovate LC-1 plugged in for my first test run last Saturday. I had the vacuum hose for the on-board AEM MAP sensor running out the window and under the hood - that has since been fixed. <img src="/forums/images/graemlins/smile.gif" alt="" /> Totally 'getto' - you have to love it! <img src="/forums/images/graemlins/lol.gif" alt="" /> Good thing the AEM isn't that big - even so it's going to be a PITA fitting everything back together and mounting it. <img src="/forums/images/graemlins/frown.gif" alt="" /> If I get real enthusiastic next weekend, I may mount the two lighted switches (AEM logging, and Wideband) in the dash. Not looking forward to drilling two more holes. <img src="/forums/images/graemlins/frown.gif" alt="" />


I already had my base map built so I loaded that in the F/IC, ran the vacuum hose, cleared the ECM and drove it to work last Friday. So far, so good. I started tweaking the fuel and timing last weekend.

Fuel is at -7.8% so far. LT trims are coming in +/- 2/2.5% on the outside. That's all good so far. 'course long term fuel trims are the easy part. <img src="/forums/images/graemlins/smile.gif" alt="" />

I'm really working on the timing now. As I expected, that's turning out to be much more difficult. It was reasonably easy to get a map that works decently on the street, but getting on that works on the street AND on the trail may not be possible.

The big problem is the amount of gearing in my big Sport. Its extremely difficult to work out a static map that will accommodate high-medium load (low vacuum) and very low load (high vacuum). For instance, on the street its in the 4-15 psia range while driving, but on the trail its in the 1-6 psia range. Since the Sport doesn't have a MAP per say (if at all), its coming up with an Engine Load data point somehow.

When on the street Engine Load is high, so the ECM pushes the timing down. However, when in Lo, Lo-Lo, or God forbid Lo-Lo-Lo, Engine Load is almost non-existent and vacuum is in the .5-3/4 psia range so it cranks the timing WAY up.

So, I need to pull a lot more on the trail, versus on the street. I have a lot more work to do on it is what I know. <img src="/forums/images/graemlins/smile.gif" alt="" /> If the weather stays good this weekend I'm going to take my laptop with ScanXL and the custom OBD maps I've built for this up on the trail and see what I'm getting.

I suspect that this isn't going to work EXACTLY like I hoped and that I may end up with two maps. One for the street and one for the trail. While I hadn't planned on this eventuality, the F/IC 8 has the capability to have two completely different maps loaded and two switch between the two at the push of a dash button. That means that I'll have to give up my ability to do logging on the F/IC independent of a laptop, but its not that big a deal for me anyway.

So far, though, everything is working as AEM advertises. Its completely transparent to all the factory electronics. Everything in the ECM clears as it should - no interference with any of the OBD drive tests. Guess the reall test will be in two and a half months when I go in for emissions testing. <img src="/forums/images/graemlins/scared.gif" alt="" />

Below is a link to the wiring doc on www.4x4extremesports.com. I've confirmed everything I have wired up to the vehicle so far (so, non-aftermarket stuff like the WBO2 yet), except the connection to the factory (forward) O2 sensor. At some point I'll have to try playing with that before I know if the resistor is needed or not. According to one of the members on the Montero Sport forum with the supercharger install its not supposed to be needed on Sports (and I presume Monteros), but ....

'97/'99 Montero Sport -> AEM F/IC 8 Harness Pin-Out Doc

I'll post a base map for the F/IC 8 for DL on my site when I'm done.

If anyone has any ideas on working through the timing map, feel free to chime in. I can post screen shots of timing maps from ScanXL if its wanted.

Edward