Electrical

This is normal for most DSMs. The alternator output drops slightly when the engine is idling, and the headlights will dim slightly.

Similarily, turning on the brake lights or turn signal puts an additional electrical load on the alternator. As a result, the alternator output voltage drops slightly, resulting in slightly dimmer lights. This holds true even for Canadians-spec DSMs equipped with the larger 90A alternator, and especially at idle. It is not a sign of malfunction.

Most DSMs have this problem.  Here are the fixes, courtesy of Tom Stangl (the VFAQ man), from a post in the Digest of Sept 25, 1998:

"The fixes are, in increasing order of difficulty

1 - Clean and lube the rubber channels in the window frame everywhere you can get to them without tearing the door apart.  Do this by getting any rubber/vinyl cleaner, putting it on a rag, and wiping the channels until the rag comes out clean.  This may take a LOT of cleaning. Then lube with NuVinyl, anti-static ArmorAll, or even dielectric grease (YES, this grease works well and does not gum up if put on VERY lightly and then rubbed off).  

2 - Open up the door, and clean/lube the bottom section of the rubber channels you couldn't get to in #1.  This will take raising and lowering the window to get all the areas.

3 - Check the bolts that hold the window to the window guide bar to make sure they are not loose.

4 - Loosen the bolts that hold the guide bar to the door, and move it forward or back to get the      window to go up perfectly straight (not a lot of adjustment here, but it doesn't take much).

5 - Loosen the window guides (the metal brackets covered in bristly material) a little so they don't      push on the window so hard."

Michael Reisin reported good results with using silicone grease of the type normally used on stock plug boots. He recommends greasing the window glass run channels in the interior of the door very well with this grease for a semi-permanent fix to the sticky window problem.

For those who are truly sick of the problem and don't mind using a little judicious force to set things right, Jeff Earl's solution involves bending the window track (only a little, so chill out) and removing a fastener that appears to interfere with the window operation.

Tom Stangl has a FAQ for this particular problem up now. Also try using the FAQ Locator to find it, or other FAQs on this problem.

Problems with the power lock switches, lock mechanisms and actuators seem to be more common on 2G cars than 1G cars. Some of the problems have been caused by loose window switches, while others have been blamed on contamination of the lock mechanisms with debris, or simply worn out actuators.

Also, one DSMer reported a factory build problem on certain limited '99 model cars. Apparantly, the factory applied too much sealant to the inside of the door. The extra sealant (or 'gunk') can run and leak into the door locks, causing them to progressively freeze up or otherwise misbehave.

Unfortunately, the only solutions are to repair or replace the affected parts. Loose window switches can usually be fixed by repairing the screw mounting points that hold them in the door. Actuators generally must be replaced, as do door lock mechanisms.

If the doors unlock themselves, it's often because the alarm "thinks" there is a key in the ignition. Improper installation of a turbo timer can cause this. This is a feature designed to prevent you locking your keys in your car.

If the alarm does not go off when the door is opened, the door pinswitches require cleaning or replacement. The switches corrode over time. Sometimes cleaning with baking soda is all that is required.

The stock batteries in DSMs are not reputed to last very long. Many people experience failures within the first year.

Most people recommend replacing the stock battery with another brand, such as an Optima, Diehard or other aftermarket make.

It should be noted that poor contact (corrosion) on the battery terminals can cause the battery to behave as if it were dead, even though it may be ok. Also, persistently dead batteries may be the fault of the alternator or voltage regulator, not the battery. Poor idle and other problems can be caused by a defective battery.

Another little-known and highly annoying fact is that auto batteries, with few exceptions, are not designed to be run dead and then recharged, making them very different from most other rechargeable batteries. Recharging the battery too quickly will result in large amounts of internal heat, causing battery damage. Typical recharging methods (auto battery chargers, or running the car for a while) may damage even "bulletproof" batteries. Charge dead batteries gently to avoid this problem.

Those caught in the dead-battery trap will want this refresher course in how to properly jumpstart a car. A few notes on these techniques: the negative is connected away from the dead battery (onto the frame) to minimize the chance of creating a spark that might ignite hydrogen gas leaking from the dead battery. This is what can cause a battery to explode, not the 'parallel' nature of the batteries, as described in DDD#7. Fortunately, few modern automotive batteries are prone to leaking flammable gases, but better safe than sorry.

For lots of info on batteries in general, read Alex's Electronic Resource Library

The Last Word: Some DSMs just seem to eat batteries. The proliferation of stock and aftermarket electrical accessories such as headlights, fog lights, in-car entertainment and navigation systems can strain a 12V system. That's why some car makers have changed to the 42V system. Storing a car for the winter with the battery connected can also draw down the cells owing to current draws from accessories in standby modes.

Battery-eating monster cars may also eat a corresponding number of alternators, making things extra hard on the battery. Investing a hardy deep-cycle battery like an Optima yellow or blue top might prolong your battery life - or, possibly, just give you enough to limp home the next time your alternator fails. Remember, the less times you discharge your battery, the better - even one discharge to 9V or less can be quite hard on it.

This is normal, and is a symptom of the alternator voltage dropping while under load - the same reason why headlights tend to dim at idle. Tom Stangl did some testing with a fuel pressure gauge and found that his fuel pressure remained 'rock solid' despite the changing noise of the fuel pump, indicating the pump is still pushing more fuel than the fuel pressure regulator needs.

Those who find this problem annoying or worrisome can fix it (or at least mitigate it somewhat) by doing the fuel pump wiring upgrade described here.

This is largely a 2G problem, where the fuel gauge sender unit has been knocked out of position. This often happens when the dealer does the gas tank recall on the car.

2G owners should look into getting the sender unit fixed.

If all your lights are out, all at once, it is likely that the fuse for either the rear brake lights or rear marker lights has burned out. This fuse is linked to the dash lights to alert the driver that something is wrong. Check the fuses.

Most DSMs will have at least one switch light burned out by now. Probably more.

This is often because your battery is sliding around and contacting the top of the hood momentarily. Check your battery and make sure it is secure.

This is usually an indication that the speed sensor [[What is a speed sensor?]] behind the speedometer has failed. Without an accurate speed reference the cruise control refuses to set.

Another possible problem is wear on the clutch pedal, preventing the clutch pedal from coming up all the way. If this occurs the pedal will not press the switch that tells the ECU that the clutch is engaged. Consequently, the ECU "thinks" the clutch pedal is down and it won't allow the cruise control to be set.

An easy test is to pull up the clutch pedal with your foot and try the cruise control again. If it works, you have clutch linkage problems. For more information on this problem, read the answer to this question.

If the cruise control will set, a split vacuum hose caused cruise control to continually drop the vehicle speed. Eventually the hose popped off completely, and the cruise would not maintain any speed at all. Several DSMers have run into this problem

If the cruise control sets and holds speed, but 'hunts' (oscillates) around the correct speed, don't worry too much. Even stock DSMs have a certain point where the cruise control will oscillate slightly. This is due to the nature of the turbocharged engine - when the cruise control speeds the car up, the turbo kicks in and the car accelerates faster than the cruise control expected it to. The cruise is forced to decelerate to compensate, but the car will also slow down faster than expected as the turbo pressure dies out. The net result is a slight, constant surging on and off boost.

2G cars have this sound. It is normal.

1Gs have this sound too - for the rear wiper.

Yes.  It should turn off after a few seconds.

If your "Check Engine" light does not turn on, the bulb is probably burned out. Replace it.

No. Get it serviced.

This indicator is lit every time the ECU detects a problem (any kind) with the car.  Often the cause is a sensor failure.

The ECU outputs a diagnostic code whenever the "Check Engine" light is lit. For 1G owners, you can:

• use a Chrysler scan tool to read the code. Not an option for most people, but still.....
• use a multimeter to read the code, as described in the shop manual, or as in this post by an early club member.
• use an LED to read the code, as described in this post by another club member, and this post from December 1998. The December post also details many of the output codes.
• use a buzzer to read the code, as described at the Diagnostic Engine Codes and Procedure page.  This page also has a list of many of the codes.

2G turbo owners need an OBD tool to read the ECU codes. Non-turbo owners can cycle the ignition key on-off-on-off-on, and the "Check Engine" light will blink out the codes. For NT owners, the codes are the same as a Neon. For the meaning of the codes, check the Digest archives, look in a shop manual, or check out this link from the 2G non-turbo pages.

Lots of 2G turbo owners mistakenly try to cycle the ignition key to get the 'Check Engine' light to blink out the codes. It doesn't work on Mitsubishi engines - only Chrysler engines. 2G NT cars have 420A Chrysler engines very similar to the Neon cars, while 2G T cars have 4G63 Mitsubishi engines similar to 1G cars. For this reason the codes are also different between 2G turbo and non-turbo cars.

Check this section of the Club DSM Homepage [Technical Difficulties/ECU Problems/Error Code Readers] for more information.

It is usually not a good idea to reset the ECU without first figuring out what is wrong.  See "How can I find out what the "Check Engine" light means?", above.  In many cases, resetting the ECU will increase the learn time, not decrease it.

For instructions on resetting the ECU and what this can accomplish, read this issue of The Diagnostic Port, and this section of the ECU Theory Series, both by Technomotive

This is usually a 2G problem. For a good explanation of why it happens, read the comments at the bottom of this post, and Scott Evans' description of ODBII functions. So far the only "fixes" are to replace the missing oxygen sensor, or to try and create and electronic replica "sensor" that makes the ECU believe the original sensor is still in place. Since the OBDII software in the second generation ECU also does sensor diagnositics, creating a fake sensor can be a touch tricky. Information on how this might be accomplished can be found here courtesy of Blake Heisler.

Here is an excerpt from message #7 of the December 22, 1998 Digest, where Todd Day (the 'talon mgr') summarizes this problem:

" What to do if you get a code 44? Well, it seems that no one on this list (myself included) ever got this code thrown for a legitimate reason, like the coil being blown or the drive transistors being dead. I would start with the connections that go from the ECU to the drive transistors, the connections between the drive transistors and the coils, and finally the "tachometer" feedback link from the transistor unit to the ECU. Lastly, check the ground on the transitors as well as the power lead to the coils.

Oldtimers on this list might remember that this very problem happened to me the morning of the Virginia City Hillclimb a few years back. I tried messing around with a lot of stuff, including wiggling all the coil connections. The problem magically went away and didn't come back until a few days later. I wiggled the connections again and it didn't come back until the next roadtrip I took. Again, after that, it happened on a long roadtrip. I've not since seen it in over four years. I have done nothing special to solve the problem other than wiggling the connections. Guess the last time was the charm."

Fortunately, Darrick Yezak has come up with a more specific answer to the code 44 gremlin. On his 1990 AWD, the wiring harness leading to the power transistor was short enough to actually end up pulling the wires out of the connector. After he extended the wires to eliminate the tension, his code 44 problem went away. He said he had similar success with two different cars. Aaron Litt found a similar problem - the bottom two connections on his '90 pack connector became corroded and caused a code 44 problem. Those plagued with the code 44 will want to check their harness and connector ASAP.

Code 44 CAN come up legitimately, but you will know it, because you'll suddenly be driving a 3000 lb go-kart. The power transistor is easier to change, so try it first.

If you're having problems with a Greddy Turbo Timer, you can check the manual from the Greddy website.

This is a 'feature' of the DSM power steering system, which includes a speed-sensitivity feature that is intended to decrease operating effort at lower RPMs.  This also means poorer response at high RPM, including the low speed/high RPM combinations required forautox.

DSS sells a modified power steering pump that fixes this problem - other vendors may as well.

Victor Del Col has worked out a fix for the high RPM cutout problem on 2G cars. There is a similar fix for 1G cars; a few details are available at this Calgary Area DSM FAQ page, and there is a VFAQ on how to modify the pump here, courtesy of Matt Price. [You would have found the latter VFAQ (hint, hint) had you checked the FAQ Locator.]

According to Robert Arrowood, Tyler Hodgson, and Micheal Astor, 1G Talons and Lasers have the fog light wiring installed, even if they did not come with the actual lights. They also give a list of components that must be added to enable the harness to work properly. Since another user reported the same was true for his 1994 Talon DL, this may be true for all 1G Talons.

Unfortunately, Micheal reports that Eclipse cars may not have the harness. This is only the impression of an anonymous Mitsubishi technician, however, and may or may not be true.

2G owners might also be out of luck as both Patric Sansoucy and David Parker both had to run new wiring when installing fog lights on their cars.

The Last Word: Brian Chapman states definitively that the 1990-1994 Eclipse cars do have the fog light wiring harness installed, even when the fog lights themselves were not installed. [Thanks, Brian!]

Some people have. Note that this mod is usually illegal - do it at your own risk.

Read the FAQ.

Many people have done this mod. However, there have been occasional reports from 2G owners that the 100W bulbs drew enough power from the car to damage the headlight wiring harness.

To combat this problem a few people have rewired the headlights with larger gauge wire. Curt Shambeau did so on his 1996 Spyder convertible and reported an increase in light output of 250 foot-candles on both stock and aftermarket bulbs. For the electrically inclined, this is obviously a good modification. For the non-electrically inclined, some companies now market plug-n-play wiring harnesses to upgrade your wiring. Often marketed as "SUV" or "heavy duty" lighting harnesses, they usually come complete and ready to put in.

Lots of people have. While they obviously work, opinions on their utility are still a matter of some debate. A few owners claim they are vastly superior, while a few claim they are vastly inferior. Most people fall into a moderate camp, claiming little or no difference (good or bad).

For some more information on blue bulbs, Superwhite bulbs and HID lighting systems, please consult the bulbs page of Daniel Stern 

A few people have installed HID systems in their DSM. These systems are often installed in place of the factory fog light system, rather than as a headlight replacement system. There is at least one person that is selling HID conversion headlight systems for 1G DSMs, however.

Current information seems to indicate that some of the HID kits are not worth the cost. Some owners have complained that the HID kit was "worse than stock". Research carefully before buying.

Since many Eclipse cars came with power antennas, changing a manual-antenna Talon or Laser to a power antenna is fairly easy. According to Sam Reed, you need:

• the power antenna unit;
• the mounting bracket;
• a chrome nut (for the outside of the car);
• a wiring harness adapter; and
• a black plastic 'plug', or cover, for your spoiler (if any).

Clear corners are available for both 1G and 2G DSMs. Or, at least, they are at the time of this writing.

According to Brad McIntyre, stock 90-91 DSM corner lenses are two part assemblies that can be converted to clear corners. All that is required is to soak the lenses in boiling water to soften up the glue. The lens halves can then be separated, and the amber lens removed. He also suggests painting the lamp housing chrome, and cleaning the lenses really well before putting it all back together.

Unfortunately, this mod does not work for 92-94 DSMs, or 2Gs. You can often find the clear corners you want on eBay, though.

Those installing clear corners will want to polish their headlights. The headlights will not appear faded or yellowed, until you put the brand-new clear corners next to them. To fix up this problem, wet-sand the headlights with 1000-2000 grit sandpaper, available at any Canadian Tire or other store that carries basic autobody repair materials.

The Last Word: With aftermarket replacement headlights now available at low cost on eBay, polishing the headlights is probably more trouble than it's worth. Just replace them.

Some people have. It's not high on the list of things to do, since the ignition system is DSMs is reportedly very strong. In general, if you don't have a problem with it, there is no need to replace it.

Those interested in the theory behind ignition systems should read Dennis Grant's Ignition Theory Series.

Information on alarm and radio wiring for DSMs can be found at the Club DSM Southeast subregional web page. More information is atDSM Alarm Wiring Codes page, written by Jamey Vester, and from this post from the archives.

You would already know the answer to the 'install' question (hint, hint) if you had checked the FAQ Locator. The tachometer display will be wrong until you reset the AFC for a 2 cylinder engine. DSMs only have 2 coils for the four cylinders. People who try to install shift lights have the same difficulty.

Those experiencing other problems should look up the on-line documentation on the A'PEXi USA website and the wiring diagramsprovided by the Super-AFC-DSM users group.

AFC owners should check out the Super-AFC-DSM mailing list on Yahoo! Groups. You need a Yahoo! ID since membership is now restricted. To join, click here.

Lots of DSM owners have installed the PMS. In fact, there are so many people using the PMS that the it has its own mailing list ategroups. There also exists two page on PMS tuning: here and here.

The manufacturer, EFI Systems, apparantly provides good support for the PMS. Unlike the AFC, PMS owners rarely plead for installation instructions.

It is in the nature of the PMS that it takes time to fully understand how to tune with it. Most PMS owners are willing to share their experience to save others from the same mistakes.

The Last Word: The PMS is a fairly primitive tuning system with only three RPM points. Do yourself a favour and get a DSMlink (available for 1G and 2G) or at least an AFC.

The ITC is generally only useful on 2G DSMs from late 1996 to 1999. All prior model years have adjustable base timing.

There are many videos on the relocation of the battery to the trunk. See Jafromobiles GSX battery Relocation Video. It is important to choose the right gauge of wire to use.

To choose an adequate wire gauge, determine the amp draw (amperage) that the wire circuit will carry.  Then measure the distance that the wire will travel (length) including the length of the return to ground (the ground wire running to the chassis or back to a ground block or battery.  Using these two numbers, Amps and length, locate the nearest gauge value in chart below.  For 6 volt automotive systems typically a wire gauge 2 sizes larger than what is shown should be used.

SRC: http://www.offroaders.com/tech/12-volt-wire-gauge-amps.htm

Other items you will need are a marine battery box (or fab your own), fusible links and distribution blocks), venting tube if not a sealed battery.

Since many proponents claim this reduces the understeer on DSMs, there has been quite a bit of discussion over the years on this subject.

This works - kind of. Read this post by Todd Day (of Technomotive) for the results. Technomotive recommends that only 1990 ECUs be used in 1990 cars, and that only 1991-1994 ECUs be used in the later cars.

However, you can make the switch, if you want to. Mike Hamilton has a FAQ on installing a 1991-1994 ECU in a 1990 DSM. This may be an option for people who have difficulty finding a 1990 ECU - they are becoming increasingly rare.

Those few that have tried it have not reported any significant side effects from interchanging ABS and non-ABS ECUs and cars. However, several of them had already disabled the ABS systems.

The engine ECU does control the ABS on the DSM cars. It is not known for certain if a non-ABS ECU has the ABS circuitry installed at all. As in the case with knock sensor circuitry, different ECU models (in that case, turbo and non-turbo) might be the same ECU without certain parts.

In theory, an ABS ECU should work on a non-ABS car just fine. However, there is a strong possibility that the ABS ECU might "see" the lack of ABS equipment and cause the ABS light to turn on, because the ECU tests the ABS system on each startup.

Also in theory, a non-ABS ECU should work on an ABS car. In this case the ABS system might not work since the ECU may not have the circuitry necessary to operate it. If it did work, warning lights indicating an ABS malfunction might not work as intended.

As always, YMMV. There are few examples of DSMers trying this swap.

This is possible, and the Canadian part is a direct bolt-on. Some dealerships may get confused over what you are asking for, however. You may be able to get around this by ordering a Galant VR-4 alternator instead - it is also 90A and will bolt right on.

There are some indications in the archives that the 90 amp alternator is the same as the 75 amp U.S. stock alternator, except it is rated for 90A. (Read about this here.) Wes Grueninger asserts that the two alternators are, in fact, different, with the 90A having an extra output tap attached to the wye terminal in the diode rectifier pack. In any case, the 90A alternator has a rated minimum output of 50A, with a 90A typical output rating. Even Canadian DSMers complain that the alternator on the car is undersized.

Read this post by Gary Selph for a description of possible alternator replacement options, including the Mitsubishi part number for the Canadian alternator. Also read Dean's Daily Discussions #8 for a little encouragement on how simple the replacement process should be.

For auto owners, Kyle Zingg posted that the alternator for the auto cars has a different part number, at least for 1Gs. The part number for the 1G auto alternator is MD153843. Manual owners have the same alternator on both the turbo and 2.0L non-turbo engines. It has also been reported that 1G and 2G alternators are the same aside from the actual output plug, which can be swapped.

It has been reported that Mitsu alternators rotate in the opposite direction of most alternators, making aftermarket units problematic. Read about this here. For this reason it is possible that aftermarket parts may not ventilate correctly, leading to premature failure.

With the creation of at least two Canadian sub-regional chapters of Club DSM, those who find themselves in a bind may be able to work a deal with a Canadian DSMer to provide a Canadian alternator. Owing to logistical difficulties, currency exchange and the possibility of taxes and/or duty on the imported part, this should be considered as a last-resort option.

As a companion to this question, you will likely want to read the Battery FAQ, and the dead battery question in this file.

You would already know this (hint, hint) if you had looked at the FAQ Locator. Alexander Shikhmuradov, Lowell Foo, and John Christou (among others) have done this, and helpfully provided instructions to Eric Porter, who made this Mini How-To page of how to do the wiring. The full VFAQ is here, courtesy of Dallace Marable.

The reasons for doing this are threefold:

• the 2G MAS, being shared with the Stealth, flows much better than the 1G MAS.
• the 2G MAS does not have difficulties with very high flow rates, where the 1G MAS will eventually start 'missing' counts.
• The 2G MAS, apparantly, has lots of 'whoosh'.

Doing this swap generally requires some type of fuel management computer, as the 2G MAS is not a drop-in replacement for the 1G MAS. However, Keith McDonnell reported that the 2G MAS will operate almost perfectly with the stock 1G computer when larger 550 cc injectors are also fitted. The bigger injectors add more fuel to offset the additional 'uncounted' air flowing through the larger MAS. Keith was experimenting on a Galant VR-4, , which could possibly behave slightly differently than other DSMs, but his results were confirmed second-hand by Dallace Marable. For more details, read his Keith's post here.

1G and 2G alternators have been reported to be the same aside from the actual output plug, which can apparantly be swapped. The information is unconfirmed so this experiment is best done with junkyard parts first.

According to some DSMers, you can use a one-wire oxygen sensor on your car instead of the four-wire OEM version. However, there may be some restrictions on their use.

Stock DSM oxygen sensors include a heating element that allows them to heat up to operating temperature faster, especially in cold weather. It is almost certain that a large number of owners have oxygen sensors with broken heaters. They don't notice the lack because the heater is not essential for O2 sensor operation. At worst, the oxygen sensor will take some additional time to heat up to operating temperature, and gas mileage might drop a little bit. So two of the four wires on the O2 sensor may certainly be considered optional, especially for those living in warmer climates.

The one wire on the non-DSM sensor is the oxygen sensor signal. Since there is no ground wire, the sensor must use the mounting point as ground. There is a small possiblity that this point might not be a good ground on some cars. Cars with upgraded downpipes might be suspect, as there is a grounding strap on the OEM downpipe that is frequently removed during the upgrade. This may affect the ground reference of the oxygen sensor to some extent.

Even in the worst-case scenario, this is highly unlikely to affect the operation of a stock or near-stock DSM. The precise reading of the oxygen sensor is not important, and is not used by the engine computer, so the ECU will not 'see' any shift in ground potential on the single-wire O2 sensor.

Owners of upgraded cars who use the O2 sensor for tuning purposes might have to be a little more careful. In many cases, owners rely on their oxygen sensors providing a consistent (if not accurate) reading. A shift of 0.100V might be enough to make their tuning more difficult. Thus, individuals who switch to a 1-wire sensor may have to spend some time re-learning their tuning methods to compensate for any differences in the new setup.

It must be noted, however, that tuning by OEM oxygen sensor is quite possibly the worst method of tuning a DSM. Owners with upgraded cars will hopefully have better and more reliable methods than relying on their O2 sensor.

Several 1G owners have found that the electrolytic capacitors in their ECUs leak after roughly seven years. This problem is not limited to DSMs, engine computers or even the automobile world - many types of equipment suffer from similar problems, as electrolytic capacitors are commonly used in electronic equipment.

Excessive leakage can cause very troublesome damage to the printed circuit board (PCB) inside the ECU.  This is very difficult to fix unless you have access to professional soldering equipment and tools.  For those that do, they will find the repair difficulty on a par with any typical conformally-coated through-hole PCB repair.

For more details see the "How to Replace the Capacitors in a 1G ECU" page, from our friends at Technomotive.

Andreas Santoso offered to provide the required replacement capacitors for $1 plus a SASE. He later changed this to $2. See the archivesfor details. This offer is old and is likely out of date, but others may make the parts available.

As a special note for 1999 owners, the 1999 ECU may have had some type of build problem involving the throttle position sensor and the ECU. The single owner that reported the problem apparantly had it fixed under warranty. As a bonus, the replacement ECU was EPROM-based. No other owners have reported any problems, so this was likely to be an isolated case.

The Last Word: Still need caps? Try here.

Several manufacturers of electrolytic capacitors have confirmed that 105 degree capacitors will last roughly twice as long in any given application as 85 degree rated capacitors. This holds true even if the capacitors are not operated anywhere near their rated temperatures. (Note that capacitors are invariably rated in degrees Celcius, not Farenheit.)

The difference in longevity is directly related to the temperature ratings of the device because 105 degree capacitors must be of much better construction in order to survive at that high a temperature. The highly robust design requirements naturally lead to a longer-lived part. 105 degree capacitors can also survive higher ripple currents, a prime suspect in the premature failures of the factory DSM ECU capacitors.

It must also be noted that each different capacitor series from each different manufacturer usually has a different rated lifespan. Those who have the means should attempt to procure high-temperature, high-reliability capacitors when repairing their ECU.

For those who cannot procure 105 degree rated capacitors, 85 degree capacitors will work. They will simply wear out faster. Those who can practice cap replacement as part of a routine maintenance schedule will certainly be able to utilize 85 degree caps will virtually zero risk of ECU damage.

Read the FAQ page at Technomotive.

Unfortunately, electrical schematics for the ECU are not available from any known source. No one has seen fit to release such documentation from within Mitsubishi, and the ECU is too complex to be easily traced out by hobbyists. If someone does have such diagrams, they are keeping them a secret.

One small consolation may be that the ECU pinout is available in every shop manual.

You can check http://www.lilevo.com/mirage/

Some of the parts can be found at mail-order retail shops such as Digikey, Newark Electronics and Allied Electronics. Please note that these shops cannot help you identify parts inside the ECU. If you require such assistance, try a local electronics parts store.

You can't get code for the ECU, unless you can download it yourself. There are several people and businesses who have downloaded and disassembled the ECU code for the various different DSM cars, but they have not made the source code public.

While this might seem like an unfair thing to do, it is important to realize that some of these individuals spent years figuring out the DSM ECU, which is based on a proprietary microcontroller not generally available to the public. They had to figure out not only how to access the program code, but how to decipher it into instructions, and how the instructions operated the engine - all without any kind of documentation or support from anyone. This amount of effort is not to be taken lightly.

Also, many people have either built this specialized knowledge into a business, or was operating on the DSM ECU as part of a business venture. Private individuals are understandably reluctant to share their hard-earned work for nothing, and employees have legal and ethical responsibilities to their employers to not divulge trade secrets. Either way, they have little incentive to give away the information.

It is important to realize that this is very much like any business operates - on the strength of specialized and/or restricted information that they have gained through hard work. There is nothing stopping anyone from following in their footsteps, if they are willing to invest the necessary time (and money) . Most people would rather do other things, of course, which provides a foundation for commerce.

The Last Word: DSM code may be available from various websites or vendors. At least, the people I know didn't have a lot of problem getting it.

These are all listed on the Club DSM Error Code Readers page.

The 1G ECU codes can be found at Brad Bauer's Diagnostic Engine Codes & Procedures page, and also in this post from the archives. You can also read your shop manual.

The Last Word: OK, so Brad's site is gone. What do you want me to do about it?

The 2G ECU codes can be found at the 2G NT page, and also at the Avenger/Sebring Owners Group homepage. You can also read your shop manual.

Here's what Todd Day of Technomotive has to say on the matter:

"As far as I can tell, there are only three ways that you can see a CHECK ENGINE light, but then not be able to clock out the code.

1) The ECU for some reason loses power or otherwise crashes and does a hard reset while driving. The CHECK ENGINE light comes on for five seconds because the ECU always turns it on for five seconds as a bulb test when power is first applied.

2) The ECU flags a legitimate error. After you turn off your engine, a Mitsu shop tech sneaks into your car with a MUTII scan tool and clears the codes while you are not looking. (1991-94 only)

3) The ECU flags a legitimate error. You turn the key to ACC or OFF, wait for the relay to go *CLICK* after seven seconds. The fuse that runs your radio memory and footlights and ECU BACKUP MEMORY is burnt out and your ECU loses its mind. You turn the key to ON to clock out the codes, but everything is gone.

If you have a case of #3, try turning the key to ACC to kill the engine, but immediately bring it back to ON. This will keep the main power supply to the ECU going, and you should be able to clock the codes."

The ECU only checks the oxygen sensor under specific circumstances. If it doesn't get to check the sensor, it can't tell if the sensor is dead.

According to Todd Day of Technomotive:

" Your car must undergo a "cruise" session above 45 MPH for 30 minutes or two "cruise" sessions for 20 minutes (depending on [model] year) before the ECU will flag a dead O2 sensor. At least for the 1Gs - this probably got a lot tighter for 2Gs. This is why a lot of people won't ever see a code get thrown for O2 in their daily commute driving."

The best way to determine if an O2 sensor is dead is to monitor it with an air/fuel (A/F) gauge.

No. Or, at least, not directly. The operation of the BCS has virtually nothing to do with intake pressure. The only method of doing this is to disconnect the BCS and wire it to a customized boost controller. Theoretically, it could be done by by reprogramming the ECU, but the new ECU would also have to accept inputs from an additional intake pressure sensor. To date, nobody has created such an ECU modification.

The AFC works by changing the mass airflow sensor (MAS) signal going into the ECU. This signal is a frequency proportional to the amount of air passing through the sensor.

Picky tech-heads will realize that the MAS puts out three signals: air flow, air pressure, and air temperature. The ECU integrates all of these to calculate air mass. The AFC alters only the air flow signal.

Different levels of adjustment to the signal occur for the different RPM ranges. In-between ranges, the AFC uses linear interpolation to smoothly transition from one adjustment level to the next. Thus, if the AFC is set to +10 at 2000 RPM, and - 10 at 3000 RPM, the signal correction at 2500 RPM will be zero.

The Super AFC also has two different correction maps based on throttle position. The "Th" (throttle) points set inside the unit by the user determines the low-throttle and high-throttle boundaries. The AFC also interpolates between the low and high settings at part throttle. Thus, if the low throttle is set at 10% and has +10% correction, and the high throttle is set at 90% and has +20% correction, the correction at 50% throttle will be +15%.

Despite this flexibility the AFC suffers a few drawbacks. It does not "know" the engine load, only engine RPM. For this reason it is hard to tune each gear ideally, since the engine load at any RPM in 1st gear is different than that in 3rd or 5th gear.

AFC owners should check out the Super-AFC-DSM mailing list on Yahoo! Groups. You need a Yahoo! ID since membership is now restricted. To join, click here.

The S-AFC is set up to accept the lowest reading as 0% throttle. The TMO and Pocketlogger read the throttle setting from the ECU, which knows better. In other words, the S-AFC is wrong. However, this discrepancy does not affect the S-AFC operation in any way, so don't worry about it.

AFC owners should check out the Super-AFC-DSM mailing list on Yahoo! Groups. You need a Yahoo! ID since membership is now restricted. To join, click here.

Here is the "basic" tuning drill for the AFC, courtesy of Kyle Tarry:

"Ok, for all of you guys who are asking what to do since you just got a SAFC, here's the drill:

FIRST, BUY A LOGGER! Non matter what people say, you cannot maximize the capabilities of your setup without one. And, for the price of a good EGT gauge, a pocketlogger can be yours.

Now, on to the good stuff. Set the SAFC up, with the throttle poits at about 30% and 80%. Spread the RPM points across the rev range from 1000 to 7500 or 8000 rpm.

Start by getting the fuel trims in line. Adjust the 1000 rpm point with the car at idle until you get the low fuel trim in the 100%-115% range. Then, do they same for the mid and high fuel trims while moving in their respective zones. (probably about idle, then slow cruise, then a faster cruise). Once you have them all around 110%, then you can carry those numbers you used across the low throttle map, and also use them on the high throttle map.

Now, you should have the high throttle map set up with the same numbers you used on the higher RPM's of the low throttle map. Make a WOT run, and log it. Now, look at the log, and at every 1000 rpms simply decide if the engine is rich or lean. If it is lean (knock) then richen the AFC up a click or two, depending on the severity of the knock. If it is rich (no knock, good O2's) then lean it out a click. Keep doing this until you have all the rpm points set up to a few % right above where it starts to knock.

That's really about it!"

AFC owners should check out the Super-AFC-DSM mailing list on Yahoo! Groups. You need a Yahoo! ID since membership is now restricted. To join, click here.

You can also check out Road Race Engineering's FAQ page on how to initially tune your AFC, and TMO's guide to datalogger tuning.

Get home asap! Classic alternator dying symptoms.

Check the output voltage while running.

Now would be a good time for saturn alternator & relocation! Some have reported great success with 160 AMP Saturn Alternator. Not quite simple plug and play though. 

Reference: http://www.orldsm.com/showthread.php?13969-Saturn-160AMP-alternator-installed-with-GREAT-RESULTS!&s=8dcfd4229bbcd7a2d710821e649a7ae4

I can not find one good vfaq for this upgrade.

Those interested can also check out the install instructions from Black Cat Custom, a manufacturer of custom gauge faces for DSMs in both Canadian and American versions.