Ready to start collecting parts for GM ECM in toyota 4runner

I know that 1project has done a gm ecm swap on a toyota, so it must be possible. We (my son and I) want to try that on his 4runner now that the new engine is up and running. What GM ECM should we use, and what other parts do we need. It is an 86 22re so it looks kind of like a tpi motor that was cut in half. Will the toyota MAF interface with the GM ECM? Will the GM ECM control the spark timing thru the distrubutor properly?
Lots of questions, but we are ready to start collecting parts and trying to make this work. The 86 toyota ECM is really limited, so this will be a fun project and might make a good how to article.

There was a guy who did this about 100 years ago? Don't know if he's still around? :innocent2:

http://home.metrocast.net/~shannen/y...ECM/gmecm.html

I really doubt the toyota MAF could be used, have you looked at that thing? Want to make the Toy run better, do the TPS adjustments and then peel of the cover of the MAF, loosen the screw holding down the spring and twist it till AFR comes in line! Works great! :laugh:

Yeah, we did the tps adjustment a long time ago, it helped a lot, however you can't datalog a toyota. We also will custom tune it for the bigger cam and higher compression with a mild port on the intake side of the heads. If we have to change MAF's, that is ok as long as we can hook everything up. I would really like for it to look stock as much as possible.

If I'm not mistaken that uses a trap door style MAF, which will not interface with the Delco ECM. As other's have done with this swap, ditch the MAF and go SD.

Which ECM to use depends on what your ultimate goals are, and what you can find.

The '7730/'7749/'7727 style ECMs are popular for ease of tuning and being able to use them in various applications, and they work easily with MPFI set-ups.

To control timing you need to use a GM Ignition Control Module. IIRC, the 22RE dizzy will trigger the GM ICM without issue, just pay attention to the polarity of the trigger from the dizzy to the ICM. This will give the RPM signal to the ECM and allow timing control as well.

In addition to Six_Shooter's advice, I can offer this:

The 22RE / 22RTE Airflow sensor is unique even in Toyota land. It uses a 9V signal that just doesn't interface nicely with anything. Dump it.

Toyota's CTS curve is the inverse of GM's. Easiest solution is to use the GM sensor. Look at my truck page (link previously posted) as I detail installing this sensor in a later style intake.

Toyota's idle air adjustment device is a mechanical valve that's thermostatically controlled. It's independent of the ecm. If you choose to eliminate it and use an ecm controlled IAC I've detailed an easy solution using a Jeep 4.0L IAC and housing. You will also need to install a thermostat bypass but that's not a challenge.

The '86 intake manifold, idle air valve, and TB was simplified and improved around 89 or 90. You might want to switch.

The Toyota TPS is usable if you choose a P4 ecm with "auto zero" code. The upper and lower voltage provided by the sensor "just fits" the factory settings for min and max. In fact, sometimes my truck would set a TPS low voltage code during cranking when system voltage was low on very cold mornings. The values to trigger the code are adjustable.

The cold start injector can be completely disabled and removed. I used the injector hole for the IAT sensor. I also used the port in the fuel rail as a handy place to install a fuel pressure test port.

Toyota wiring is absoultely nightmarish. I don't have time right now but later I'll post some of the issues you'll need to work with. Things like dual coil fuel pump relays and redundant power and ground circuits really add the the fun.

I can verify the distributor will trigger the GM EST, just did one on carb/TBI Toy engine conversion.

For apperance the MAF could be gutted and left there with wiring to no where. Inspections must have lightened up there? They used to be very hard on visual and didn't even care to look at emmissions until that was done.

I know there no data or at least any usable and no way to retune. When my son and I built his 22RE same as your doing and I wanted to do a GM ECM conversion which we know can be dialed in! But with higher compression, .030 over bore, port work and cam we did have this thing run great all the time with TPS adjustments done as per internet writeups with feeler gauges and volt meter. That was the overall fix. Then loosening up the spring on trap door MAF fixed the other fueling issues. It really ran good overall everywhere. His Toyotas off road freinds would come over for a tweak and results were the same every time. IIRC timing was bumped a little but the real timing issue was the dual vacuum diaphram adance cans on distributor... seemed like everyone had a bad thermastatic vacuum valve that never opened or never shut?

Just for kicks someday look at factory service manual for 1985 toyota. It came with carb or EFI. There is 1000 more switches and vacuum lines on the carb! I think I still have it in zipped pdf if you need one...

Quote:

---

Originally Posted by 1project2many

The 22RE / 22RTE Airflow sensor is unique even in Toyota land. It uses a 9V signal that just doesn't interface nicely with anything. Dump it.

---

I am thinking maybe using a MAF out of a Ford Tauras as they are easily available and cheap at the JY. We are changing the air intake to make room for a 2nd battery, so that wouldn't be too difficult to adapt in.

Quote:

---

Originally Posted by 1project2many

Toyota's idle air adjustment device is a mechanical valve that's thermostatically controlled. It's independent of the ecm. If you choose to eliminate it and use an ecm controlled IAC I've detailed an easy solution using a Jeep 4.0L IAC and housing. You will also need to install a thermostat bypass but that's not a challenge.

---

That ought to be the easiest part, my son does heavy line at the local Dodge Chrysler Jeep dealership and can just get the parts off a core motor.

Quote:

---

Originally Posted by 1project2many

The '86 intake manifold, idle air valve, and TB was simplified and improved around 89 or 90. You might want to switch

---

I think we already have one, I will look at that.

Quote:

---

Originally Posted by 1project2many

The Toyota TPS is usable if you choose a P4 ecm with "auto zero" code. The upper and lower voltage provided by the sensor "just fits" the factory settings for min and max. In fact, sometimes my truck would set a TPS low voltage code during cranking when system voltage was low on very cold mornings. The values to trigger the code are adjustable.

---

Which P4 ecm would you suggest?

Quote:

---

Originally Posted by 1project2many

Toyota's CTS curve is the inverse of GM's. Easiest solution is to use the GM sensor. Look at my truck page (link previously posted) as I detail installing this sensor in a later style intake.

---

That makes sense. I wonder also if a little piece of code could be written to correct the Toyota CTS. I will probably install the GM sensor.

Quote:

---

Originally Posted by 1project2many

The cold start injector can be completely disabled and removed. I used the injector hole for the IAT sensor. I also used the port in the fuel rail as a handy place to install a fuel pressure test port.

---

This sounds really cool, how did you seal things to keep the fuel from leaking, or letting in extra air?

Quote:

---

Originally Posted by 1project2many

Toyota wiring is absoultely nightmarish. I don't have time right now but later I'll post some of the issues you'll need to work with. Things like dual coil fuel pump relays and redundant power and ground circuits really add the the fun.

---

Somehow that doesn't surprise me. That may end up being our biggest challenge.

Quote:

---

Originally Posted by Six_Shooter

To control timing you need to use a GM Ignition Control Module. IIRC, the 22RE dizzy will trigger the GM ICM without issue, just pay attention to the polarity of the trigger from the dizzy to the ICM. This will give the RPM signal to the ECM and allow timing control as well.

---

That's a relief, something that won't be horribly complicated.

And now for some more questions I thought about.

1) what memcal should I try to find for this project, and how about a knock sensor. Also, what BCC should we start with?

2) VSS-- I thought about using a VSS that was cable driven. The only ones I find are 4000 pulse, and I think we need 2000 pulse.

This sure sounds like it will be loads of fun, as well as a challenge.

Do you need a MAF sensor? I don't know of any factory four cylinder plus MAF combination from GM. I'd probably use a 1227165 and a V8 MAF if I had to go that route. The 165 is not a P4 and you will not get fast ALDL data for tuning. You will also need to determine which knock sensor and external ESC module are appropriate for the 22RE. Choose your parts by matching cylinder bore as closely as possible.

There may be a definition file for a 3800 V6 with MAF but I haven't tuned any and can't tell you if the code and cal will adapt easily to a four cylinder.

If using a P4 the 7730 or 7727 are both good choices. I used a memcal and knock sensor from a 2.3L quad 4 since the bore size is the same. I can generate a list of BCC's, part nos, and ecm nos if needed. Choosing which mask id to use might take a little consideration but ultimately there are several that will work. These are all speed density. Is it possible to hide a MAP wire and sensor under the manifold and use a hollowed out MAF as Mark suggested?

It may not be a challenge to alter the voltage <-> temperature calibration in the ecm without a pile of work. Look around the web and try to find the megasquirt 22re conversion page. I believe there's a temp -> voltage chart for Toyota sensor there.

Speed sensor can be 2K or 4k pulse. But use a sensor that generates an a/c signal rather than the "reed switch" type.

The intake air temp sensor I used has 3/8 pipe thread so once I tapped the manifold it was self sealing. The fuel pressure test port can be found on many 2.8 and 3.1L GM vehicles as well as the TPI cars. I believe I had to drill out the existing hole then run a tap through it. Sealing was probably with an O ring or nylon washer that was GM OE equipment.

I will never miss the days of diagnosing leaking vacuum lines in early and mid eighties carbureted asian vehicles. One of the very few upsides to living in the rust belt is that most of those cars are gone now.

I just looked, and it appears that the 7165 is a P4 processor, capable of 8k baud, and using a memcal. I could use the memcal from the 2.3L Quad 4 if I can find one, and adjust the code for a 4 cylinder (maybe, I will look at it and see if it is even possible) Otherwise, I suppose we will do the MAP conversion.

How do I tell if the VSS is a reed switch type?

7165 may be P4 but it was something of a hybrid. It uses an external ESC module and IIRC the 8192 comms sometimes doesn't happen. The Q4 code might not work as the 7165 may not have the hardware necessary to support all the inputs or outputs that code uses. Safest answer is to use code intended to work with that ecm. There is a four cylinder TBI application that uses the 7165 but it's speed density.

The aftermarket often applies the term "reed switch" to any speed sensor that produces a 0 - X Volt square wave signal. If the sensor voltage doesn't exceed 5V nominal value and produces a 2k pulse per mile it can be used. That's the only configuration that works correctly with the "optical" speed signal input. If you get a speed sensor which produces an a/c output of 2k, 4k, or even a few other multiples of 1k it can be connected to the "analog" input and the software can be adjusted accordingly.

1project2many,

I probably worded that wrong last night. What I thought I would do is use the Memcal from the quad4 with a $32B prom from a camaro 5.0 TPI. The 32B code has code for 8 - 6 - 4 - and 2 cylinders in the calibration. Set to 4 cylinder, it might be close, since the cylinder volume is close. and external ESC would be no problem either. I am not sure how I can tell what type of VSS the aftermarket VSS's I found are, nothing in the description. Do you mean the reed type output a squarewave and the A/c output is a sine wave? I will look at the 32B code to see if the pulse count can be adjusted. Also, on another note, apparently some newer OEM O2 sensors are wideband, and my son has some used ones laying around at work, so if they are indeed a wideband O2 sensor, we are snagging a couple to use for tuning.


>>>edit<<<< just looked it up, the 7165 used a 2 pulse DC square wave VSS. now I just have to figure out how to adapt one inline to a toyota speedo cable.

There's no need to use the quad4 memcal if you're using an external ESC module. We can modify any memcal to work.

Quote:

---

Do you mean the reed type output a squarewave

---

The aftermarket often applies the term "reed switch" to any speed sensor that produces a 0 - X Volt square wave signal.

Quote:

---

and the A/c output is a sine wave?

---

Yes, A/C is a sine wave. The other type of speed sensor is constructed similar to a distributor pickup coil and shaft and is known as an "analog" sensor.

There are multiple speed sensor inputs in most of the ecm's including the 7165. One expects a square wave 2k pulse per mile on the "optical" input. One expects an A/C signal which can be more than 2k pulses per mile.

Wideband O2 sensors require a driver circuit to operate correctly. The sensor driver will maintain a specific level of heat across the sensor by varying current through the sensor. The amount of current relates to the AFR. The stock gm ecm does not provide this functionality.

If the Toyota cable uses "standard" speedometer threads then you can find an inline VSS which sandwiches between cable and speedometer.

Many people that have '7165 swap to a '7730 to improve their tuning experiences, if that means anything to you.

Common issues are ALDL not connecting, regardless of resistor use (10K) or PC re-starts, software restarts, etc. Also it's commonly reported that comms will randomly drop out while driving along, by people using the '7165.

In a conversion, this ECM would be very near the bottom of my list to use.

Why do you want to use a MAF sensor?

I'm not sure you understand what the MEMCAL does, or contains, due to the way you want to mix and match parts. The PROM is part of the MEMCAL, so trying to use a $32 PROM with a Quad4 MEMCAL (Weren't most of those $58?) is counter productive, since you want to tune it anyway. I'm not sure the LHM or RFD circuitry will work with the '7165 ECM correctly either. It's better to use a MEMCAL that would have been used in the ECM you are using, to keep variables to a minimum.

If you do indeed have an OEM WBO2 sensor, they are useless without a controller. In OEM applications that use WBO2 sensors, the controller is part of the PCM itself, in retrofit applications, such as we as tuners do, an external controller is needed. At this point it's just better to buy an entire package, controller and sensor, since you know that the sensor will be new, and that the controller matched to it. I am fond of the 14point7.com products.

I adapted an optical VSS to my Datsun speedo. I did have to change to a later Datsun speedo, to have the correct hardware to function with the optical sensor that is looking for a pulsed reflection. I don't believe the pictures are online anymore, but I'll try and get them back up again to show one way of adding a VSS. Basically, I copied what many mid and late '80s GM vehicles had, especially the S-series, that had an optical VSS in the speedo itself and reflected off two arms that are part of the speedo mechanicals to provide a VSS signal to the ECM. I did have to set the BIN file to recognize the type of sensor though.

If there is a later version of the transmission, that has an electronic VSS, you might be able to slip that VSS into your trans to get a VSS signal, though this usually also means that the speedo itself needs to be changed to an electronic unit. If I knew then what I know now, I probably would have used a later VSS from an RB trans that will fit in my trans to get a VR type VSS signal, and then changed my speedo to match.

Well, I guess I need to rethink this, as everyone giving me help indicates that speed density is a better way to go. I have speed density on my TBI truck and have looked at that code a lot. I guess I can convince my son that SD is the way to go if we want to proceed with the conversion. We will just need to make certain that it looks like a stock setup if somebody looks under the hood. Stupid, as it will definitely be a clean running vehicle. I am aware that the OEM wideband O2's will need an external controller, and IIRC there is an inexpensive DIY controller available that we could use. I'm just cheap and broke, so anything I can get free I am willing to try. If it doesn't work, nothing lost. The memcal choice was because of Knock sensors, but because the 165 used external knock I would have used a regular memcal. I think if I go to a speed density system, I will have a lot less problems. I can pick up something like a 730 or 727 for $50 or so, and then worry about the other parts. The other reason I looked at a 165 was because I could get one for free. You guys have pretty much convinced me that I should drop that concept and go with a 730 or 727. I guess I need to start looking at what mask to use next. Thanks to all of you for your patience. Believe me I appreciate it. This is an experiment, and probably not something I will repeat.

From a performance standpoint I believe MAF is fine. That engine isn't going to suffer from the choice and other than potentially having to adjust a bunch of variables initially, the MAF tuning process is likely to be quicker. But you're still restricted to using a GM MAF sensor which will need to look like a stock Yota sensor. I've worked with both vehicles enough to know the Delco sensor does not look like its Denso counterpart so you've got some work to do. The SD system can be hidden more easily by using a late model MAP sensor and mounting it somewhere under the plenum out of sight. It's ironic to think you could swap in 1990 TPI complete and it would be ok, but using the controller for the same system could get you into trouble.

I'm not sure what a "regular" memcal is. The 7165 memcal has no knock filter on board so you are able to use almost anything imo. Check bore sizes. Maybe a 3.8 V6 or a 2.5 matches the bore. Or maybe there's a version of the Q4 that uses a KS of the correct resistance for the external box.

In short, if you have access to the 7165 and related parts for short money then use it. But if you have to buy the parts then I'd recommend the SD system.

I've done much of my tuning without a WB sensor. I believe that anyone who's successfully tuned a carby engine can tune an EFI system to a reasonable level. I'm just as responsible as anyone else for making the world think the only way to tune is with a WB. "Back in the day" we'd almost yell at people for thinking they know what AFR is being delivered during open loop or WOT conditions. "The only way to tell is with a WB." Well, that got turned into "the only way to tune is with a WB" which is just not true. Your NB sensor switches at stoichiometric AFR so you tune for that first. Do the steady state and low speed tuning first using closed loop. The go after the open loop stuff. Force OL and adjust the desired AFR tables or the % change to AFR tables. You can still use the NB readings to tell you if it's rich or lean. You just won't know the actual AFR. Tune for lean best power by starting rich and reducing fuel a little at a time. Believe me, I've logged many hours tuning with no WB on board.

There are several versions of WB sensor. I know of at least two NTK types and two Bosch types. The controller must match the sensor. I may be able to help ID the sensor once you get it.

Okay, I think we are going to go SD, and just hide it. The 7730 ECM looks good, we can do that. I assume that AUJP is the code to use, but can't find anything about manual trannys and AUJP, is it not a good idea to use AUJP with a manual? I just scanned the AUJP disassembly, but can't find any settings for manual trans. (only a quick scan however) I got offered an 8051 out of a 95 firechicken, but will pass as I don't want to deal with optispark and all the other stuff it does.

22RE = 91.9mm bore.

quad4 = 92mm bore.

i'd say that's a pretty nice match for the knock filter.

I would say so. So stroke doesn't matter, just bore diameter? Also, do you have to modify the memcal for the frequency of the knock sensor you use?

as long as you use an appropriate 3900 ohm sensor, you should be fine.

Okay guys, BIG question...... I was thinking about all the reading I have done in the past, and seemed to remember a $0D maf setup. So I looked on this website at Marks $0d thread and found $0d MAF and MPFI . It would make sense to me to use that as I have 100's of hours staring at $0d code. Am I missing something here? We can use a GM MAF, as we are modifying the AIR intake anyways. I doubt most smog techs would know the difference anyways, MAF sensor.....check.

And there's your sign! :happy:

But back to what memcal for 4 cylinder that will be compatable with those mods? All that stuff was done for V8 and memcal that came with them.

Just to clarify I saved and organized other peoples work in that thread and had no part of the cool development.

grab a Q4 MEMCAL and try it?

are there actual mods to the MEMCAL? i only briefly looked at the subject and it looked like there were internal mods to the PCM, but that was about it.

I remember an internal mod to the drivers for the injectors, and a jumper on the memcal for MPFI, is that what you are talking about? I actually did that for a friend for a 383 TPI project he did, but it was MAP. IIRC the memcal selects the number of cylinders correct? There is a parameter in $0d at L4142 for number of cylinders, but not sure if that is all that needs changing.
<< edit>> I looked at the code a little more, and it appears that 8 cyl is hard coded in several places and may need some modification to make it work. I haven't looked at a 6 cyl hac, but I guess I should to make sure.

There are three reasons to select the correct memcal. First is to ensure you have the correct fuel delivery mode enabled. Second is ensuring you can match the knock filter circuitry with the sensor and engine. Third is to have redundant fuel / limp home mode matching the engine configuration so the vehicle can be driven if the primary control logic fails. If we postulate that 99% of mildly to wildly modified systems have done nothing to match RFD to configuration, then the job gets easier. The knock sensor circuitry is a soldered in board which can be removed with less effort than installing a ZIF socket for the main eprom. Fuel mode select is a simple resistor network which has been decoded and measured across several memcals providing a consistent pattern. http://www.gearhead-efi.com/Fuel-Inj...ull=1#post3374 Interesting to note is that one of the Quad 4 memcals used "single point TBI" as the fuel delivery mode rather than "four cylinder pfi" mode. Armed with this information, memcals become nothing more than physical devices to store and connect various hardware to the ecm, and a source for parts for future modification.

Knock frequency has a very direct relationship to the speed of sound. Sound travels slower in warm, less dense air and takes longer to reach cylinder walls with a larger bore diameter. Discounting cylinder temperature you can use the following formula to calculate a theoretical frequency for knock based on bore size: 900/(3.14*cylinder radius)

Or head to this handy online calculator: http://www.phormula.co.uk/KnockCalculator.aspx
Mark just posted some info confirming that GM knock sensors are primarily tuned to 5kHz, 6kHz, or 7kHz. http://www.gearhead-efi.com/Fuel-Inj...ll=1#post18135 As we learn more about knock and sound filtering I'm sure we'll be able to make better choices using the additional information in that chart but for now the 2.3 sensor and filter seem to work well enough with my engine that I'd recommend them.

Toyota uses low impedance injectors and attaches resistors in series (the aluminum box on the RH inner fender) so high impedance drivers can be used. I chose to remove the resistors and use the peak and hold driver ability of my 7749. If you use the 7730 you will likely want to leave the resistors in and keep the ecm operating in saturated mode (no injector sense jumper attached). Additionally, because the 7730 only uses one injector driver you should choose "single point tbi" or "eight cylinder pfi" mode as opposed to "four cylinder pfi" which alternates injector firing between two drivers. Choosing the 7427 means you can leave the fuel select mode on the V8 memcal intact (the same voltage is used for 4 cyl PFI or alternating bank TBI) and use peak and hold mode to drive the injectors. You shouldn't need to change sense resistors in the pcm or upgrade the drivers because the total current draw through (4) Toy injectors is similar to (2) tbi injectors.

I cannot add much about $0D with MAF operating a four cylinder engine. The calibrations I've studied which have a knumcyl variable for the number of cylinders will do a conversion to the time variable representing engine speed. After conversion it's equal to an eight cylinder value so this may be an option if not already included. I might have time to dig out my notes later today or tonight.

Found this in a 22RE with a home made turbo!

GM 7730 style ECU from a 92 Lumina Z34
Memcal (chip inside ECU) from a 91 Cutlass Quad 4

http://www.22rte-trucks.com/simplema...php?topic=1636

a 92 Z34 would be a 16149396.... good choice IMO!

a 1227727 would work underhood then, 1227730 if in-cabin.

I put my 7749 in the truck where the OE ecu was but it's still not attached to anything. I'm going to have to build a bracket and shields to protect it from passenger's feet. The underhood version would be nicer but I'll bet the inspektors in the Republik of Kalifornia would catch that modification pretty quickly.

Quote:

---

Originally Posted by EagleMark

Found this in a 22RE with a home made turbo!

GM 7730 style ECU from a 92 Lumina Z34
Memcal (chip inside ECU) from a 91 Cutlass Quad 4

http://www.22rte-trucks.com/simplema...php?topic=1636

---

That's Lucky's Land Rover.

That was a really neat project, I remember running across it in some other forum. I wonder if the 4.3 MAF from a 96 blazer will work for this project. I know that I may need to adjust the airflow table for it, but that can't be that bad, can it? Since we are completely redoing the air intake when we move it and switch to a cone type filter, it ought to look fairly stock. I need to find 1project2many's truck link so that I can look at his other mods. The IAC and coolant bypass especially. We are hoping to start collecting parts this month, I may start playing with code as soon as this weekend. I am excited, it would be cool to stealth convert this truck. California smog laws are very restrictive, but if it looks correct and passes the smiffer then we are good to go. We live in one of the least restrictive smog areas of California.

Post 2 of this thread...

Not quite sure what you mean Mark? I did look at that, but it was a turbo setup.

Quote:

---

Originally Posted by jim_in_dorris

I need to find 1project2many's truck link so that I can look at his other mods. The IAC and coolant bypass especially.

---

My first post to you (Post 2 of this thread) had a link to 1project2many Toyota EFI.
http://home.metrocast.net/~shannen/y...ECM/gmecm.html

Boy now I feel really Stoopid!!!!:mad1:

Well I sure didn't want to cause that. Blame it on me being blunt, or maybe it's the oldtimers syndrome? Been happening to me too! :laugh:

I just did a facepalm because once I really looked at the site I understood what I missed. I understand, I have part-timers myself. Tell the new kid where to find his ECM, then he can tell us what we need to know, but I bet you are correct, it will probably be $42 stuff. Good thing that's your specialty.... I am really busy right now, and it might be a while before I get back to $42 stuff, but I want to play with the single VE table concept and see what happens.

Part timers doesn't hurt as much as old timers! :thumbsup:

Your $42 hack stuff really helped me! Wish diassasembly was my specialty! I got a soft spot for kids, always have. I'm throwing him loads of bones to keep him going. We'll see how he does, I'm not about to hold him back!

Since I'm older than you, I can relate to oldtimers, especially in the morning trying to get all my kinks out.

My three back surgeries trumps you old kinks! Mornings suck...

</back to fuel injection!>

That is true sir, however, I have broken my neck and both feet as well as numerous other injuries, so those old kinks do indeed suck. Back to fuel injection, and have a good night, I have to be to work early tomorrow. so 5:30 will get here way to early.