Jump to content
HybridZ
mobythevan

LS1 MegaSquirt Install

Recommended Posts

UPDATED (3_4_16)

V3 pcb cam input JS2 label wrong, changed to JS8, thank you jartech007

UPDATED (12_7_11)
May be a problem with trigger return cranking and blowing coil/FP fuse, added some info about setting hold ignition to 2 to help prevent this.
UPDATED (3_24_10)
Added 24 vs 58 tooth crank sensor info (58 tooth will not work)
UPDATED (9_18_09)
Added better description for v2.2 crank sensor mods
UPDATED (3_16_09)
(fixed spark/coil wiring to be intuitive, fixed trigger return cranking for quicker starts)

MSQ files are attached at the bottom of this post for HiRes10e. That is what I run on all my cars right now. One msq is for gas, the other for e85. My car has 150lb injectors so you will need to change virtually all of the settings to make this files work. Warmups, accell, ASE, cranking PW, required fuel, etc. But at least it will give people a starting point. I also never got to tune this setup so the AFR tables are all 11.0.

The harness color codes in this document reflect my 99 LS1 harness, I have noticed the colors vary on some other harnesses.

running MS1 on V2.2 or v3 board using MSnS_extra029y4 or HiRes10e code (I have run with both versions and both v2.2 and v3 boards with no problems)

Section One, TRIGGER inputs to MS
1. The crank sensor input goes to the TACH input pin, pin 24 on the DB37. The megasquirt must trigger off the falling edge of the crank sensors signal because it is evenly spaced 24 tooth signal with no missing teeth. The rising edge of the signal is not evenly spaced. NOTE: These instructions are designed for LS1/LS6 type engines with a 24-tooth wheel, which can be identified by its black harness connector. This will not work with the LS2/LS7 and its 58-tooth wheel, which can be easily recognized by its gray harness connector (the crank sensor itself is gray on 58 tooth engines, thanks for the information "jartech007").

Crank Sensor Wires:
Signal - blue/wht
Power - green
Ground - black or yellow black in some harnesses

v2.2:
Remove the opto isolator from the board. Remove R11, D5, D8, and R10. Leave C11 installed for noise filtering. Bring the crank sensor input directly to the processor pin with a 1k series resistor just like the cam signal. This is pin 14 on the ECU (IRQ1). Since pin 14 is active low it automatically uses the correct falling edge of the crank sensor input. A convenient way to bring the crank sensor input(tach pin 24) directly to processor pin 14 is to install the 1k resistor from the non-banded side of D5 to proc pin 14. You may need some wire to reach.

v3:
Remove all jumpers that go from tachselect or tsel to vr or opto, depending on how your board was built it may or may not have jumpers installed. Run a wire and series 1k resistor between tachselect and tsel. Add the .01uf capacitor between tsel and XG2(gnd).

2. The cam sensor goes to second trigger input which requires a 1K resistor to pin 11 of the processor. It also requires a .01uf cap for noise filtering. On both of my setups I used the ignition pin 36 on the DB37 for the second trigger cam input. This is S5 on the terminal block of the relay board if you are using it.

Cam Sensor Wires:
Signal - brown/wht
Power - red
Ground - pink/black

v2.2:
Add wire with 1k series resistor from pin 36 of DB37 to pin 11 of the processor. Add .01uf capacitor from pin 11 to gnd.

v3:
Remove any jumpers that may be installed on your board to IGN. Run a wire from IGN to JS2 (pin 11).  Run a wire from IGN to JS8 (pin 11) with a series 1k resistor in between. Add a .01uf capacitor from pin 11 of the processor to pin 4 of U7 (gnd).  jartech007 pointed out an error here in labeling pin 11.  The schematic for v3 pcb shows pin 11 is JS8.  

Both crank and cam sensors are powered with 5 volts. It has already been confirmed by other people that both sensors return a 5 volt square wave logic signal. I did find new info that in the factory setup both sensors are powered by 12 volts, but 5 volts works without the need for other circuitry to interface with the processor.

3. Megatune settings. Under "wheel decoder settings" , set "wheel decoder base teeth" to 24. Turn on 2nd trigger. Set 2nd trigger edge to "rising and falling". Set to "no missing" teeth. I also have added a timing mark and indicator tab on the LS1 engine block and harmonic damper so that I can determine the relationship between megasquirt and the crankshaft. There should be no need to do this on installs, I just did it on the first one to make sure everything worked correctly. The LS1 damper is not keyed so if you pull and re-install it your timing mark will move anyway.

Trigger Position settings(updated 3_16_09):
Trig Pos A 21, Return A 24
Trig Pos B 3, Return B 6
Trig Pos C 9, Return C 12
Trig Pos D 15, Return D 18

This yields a trigger at 60 BTDC and a return trigger at 15 BTDC. Under Spark settings the trigger angle is set to 60. Use trigger return mode with hold =0* and cranking advance angle = 15. *NOTE(12_7_11): I have read about problems with trigger return cranking and hold set to zero. I believe hold should be set to 2 to prevent problems at slow cranking or beginning of cranking with blown coil fuses and backfires. If you still have problems use time based cranking to avoid these issues.

Section Two, SPARK
1. Under "basic settings" then under "codebase outputs and functions" , run wheel decoder code type by selecting generic wheel. Set LED17 to Spark A, LED18 to Spark C, LED19 to Spark B and output3 to Spark D. Spark E must be set to shiftlight, Spark F must be set to knock input. I have X2 set to fan control and X4 set to boost control. FIDLE is set to idle control. To setup output3 for spark D control, build another 2222 transistor circuit like the LED circuits in the proto area on v3 board or add a proto circuit to the v2.2 board. Spark D is output 3 (pin 15 on processor) which has a resistor in circuit already that needs to be removed, remove R14 for v2.2, remove R1 for v3. You can omit the LED and 330ohm resistor from the spark D circuit. Put a 1K resistor from pin 15 of the processor to pin 2 of transistor, another 1k resistor from VCC to pin 3 of the transistor and tie pin 1 of the transistor to GND. Take your wire for spark control from pin 3 of the transistor just like the spark A,B,C LED mods.

Picture for LED spark mods, do this mod to all three LED circuits for spark A,B,C. The LEDs on v2.2 are D17,D18,D19. The LEDs on v3 are D14,D15,D16.
LED_spark_mod.gif

On my v2.2 and v3 setups I ran the four spark connections to the DB37 on pins 25, 27, 29, 31. On v2.2 these are X11,X12,X13,X14. On v3 these are IAC1A, IAC1B, IAC2A, IAC2B. On the relay board if you use one, these connection become S1, S2, S3, S4 on the terminal block.

2. The LS1 firing order is 1-8-7-2-6-5-4-3, so I set the spark outputs as follows for wasted spark:

(Updated 3_16_09 for intuitive spark and coil wiring)
Spark A(S1/DB37 pin 25) connects to coils 1 and 6, blue/wht and purple
Spark B(S2/DB37 pin 27) connects to coils 8 and 5, green and purple/wht
Spark C(S3/DB37 pin 29) connects to coils 7 and 4, green/wht and red
Spark D(S4/DB37 pin 31) connects to coils 2 and 3, light blue and red/wht

The coil ground wires black and black/wht are grounded to the chassis with a lug. The coil reference low wires brown and brown/wht are grounded at the DB37 on any available ground pin. Coil power is pink and should be powered from the fuel pump relay.

3. Megatune settings. This setup has the processor firing the transistor, then the transistor fires the LS1 coil ignitor. With this setup I believe the "spark output inverted" box should be set to NO in megatune. I used the dwell settings from Z-ya--> Cranking 5ms, Running 4ms, Minimum 1ms.

NOTE on coil power wiring: my LS1 harness has the injector and coil power tied together in the harness. If you connect these wires to the inj power terminal you will get backfires when you turn the key on because the coils power up before the processor is booted. I separated the coil power wires at the big connector on the coil bracket and then powered the coils from the fuel pump relay power as recommended in MSnS_Extra manual.

Section Three, Wiring INJECTORS, TPS, CLT, IAT, etc
1. Connect the injectors so that 2 on each fuel rail fire together for the least amount of pressure drop in one rail. Run alternating injection so you don't hit the rail all at once.

INJ1 = 1,5,4,8 - blk/wht, black, blue/white, blue/blk
INJ2 = 3,7,2,6 - pink/blk, red/blk, yellow/blk, green/blk

The injector power wires are pink.

TPS Wires:
Signal - blue
Vref - grey
Ground - black

Coolant Wires:
Signal - yellow
Ground - black

Intake Air Temp Wires:
Signal - purple or orange on some harnesses
Ground - tan

MS ECU grounds:
Bring at least a couple ground wires from the DB37 back up the harness to some chassis lugs(I brought six ground wires up and tied to larger gauge wire then eventually two lugs). Grounding between the battery, engine block, and chassis should always be a top priority for EFI installs. With proper grounding it is no longer an issue where to ground the ECU, sensors, etc.

Section Four, General NOTES
1. Don't remove the LS1 IAC and use a block off plate. I tried that before thinking it through and it left the entire hole open so idle was too high. Also it makes a deafening whistle as air goes through the weird shaped passage. I just left the IAC installed but not connected to anything.

2. Watch settings under "more settings" then under "shift light/fan/outputs 3+4" , the output 4(LED 18) settings are confusing, this could cause a conflict between spark C and fan settings.


Section 5, MISC circuits
1. A circuit for controlling a relay for your cooling fan. Picture for Taurus Fan control using X2 output
x2_fan_control.jpg

2. A circuit to control a stock tachometer. A 1K resistor seems to work fine for my rx7. The 029y4 code provides a tach output signal that can be set to 4 cylinder mode, like the LS1 PCM does. My dash tach reads about 300 rpm high at any spot using this circuit. I will determine the correct resistor to calibrate this shortly. Picture for the circuit to drive the a Tach
LowVTacho.gif

MSQ FILES HERE, (Updated 3_16_09)

updated_msq.zip

Edited by mobythevan
cam sensor for v3 pcb worng pin label JS2

Share this post


Link to post
Share on other sites

interesting-i am still playing with old ms 2.2 but i would like to go cop ignition on my l28et without all the edis stuff.i have 6 cop units from 300hp merc outboard that have onboard transistors-looks kind of like chevy lsx cop units.these units fire at 6000+ rpm on a 2 stroke.will be upgrading some day.will have to read on cid #1 input or code wheel for ms 2 to sync it for dis ignition.

Share this post


Link to post
Share on other sites
I'll be very interested in reading a good write up about this moby, going to be sure to include back links from your own research in other LSx based forums?

 

Once I get it running here I will do some type of writeup, at least the basics for wiring. I was hoping dieselgeek would do a writeup first so I could plug'nplay. He did most of the work for the 4G63 so it made my install a breeze. If my flexplate makes it in this week I will be installing the engine this weekend and then wiring up the MS box. Maybe start testing next week. The worst part is cutting the factory PCM connectors off my perfectly good LS1 harness. Just seems like a waste, but better than building a new wiring harness I guess.

Share this post


Link to post
Share on other sites
I'm in the same boat. I'm buying a l92 416 longblock and would like to run Megasquirt. Any word on the router board for MS2? Is that still a ways out?

 

Justin

 

The router board project has pretty much turned into a stand-alone ECU on its own; it ended up with so much processing power that it can handle all the calculations without an MS-II. We're expecting it out next year.

 

On the other hand, MS2/Extra code that will handle LSx wasted spark ignition is in the testing stages right now, ready for the more adventurous users to download and try out.

Share this post


Link to post
Share on other sites
On the other hand, MS2/Extra code that will handle LSx wasted spark ignition is in the testing stages right now, ready for the more adventurous users to download and try out.

 

Unfortunately I will be using the old MS1 extra code and on top of that I will be using the old v2.2 board. That is what I have in hand.

Share this post


Link to post
Share on other sites
Unfortunately I will be using the old MS1 extra code and on top of that I will be using the old v2.2 board. That is what I have in hand.

 

works for me, i like experimenting w/ the 2.2 stuff anyway. it may not have all the bells and whistles of the newer units, but it's cheaper to do the initial running with and the initial screw ups that always come w/ a new build.

Share this post


Link to post
Share on other sites
Unfortunately I will be using the old MS1 extra code and on top of that I will be using the old v2.2 board. That is what I have in hand.

 

The 029y3 code should work for your setup - I just wanted to let Megasquirt-II users know they also have options without waiting for a router board.

Share this post


Link to post
Share on other sites

I got set back about a week because I bought the wrong flexplate for the engine. I will hopefully have the correct flexplate by this weekend and get the engine installed, then I can start wiring up MS. I don't have an IAC on my LS1 engine, it was missing when I got the engine. As usual I am going to do something a little strange and use the purge solenoid as my FIDLE valve. Since it is just a solenoid that allows air into the intake, the only thing is to see how much it raises rpms. It looks big enough to work as a FIDLE valve. I will not have it connected to the gas tank purge setup like factory car, so I'll just put a little filter on it.

Share this post


Link to post
Share on other sites

No problem, I know other people have all this info, but they don't write it up for some reason.

 

BUT wait until I get runtime on my setup before we call this good. Of course people that are up for an adventure should be able to take these settings and get there LS1 running.

 

For some reason I couldn't get trigger return cranking to work, even though the trigger return is at 15 BTDC and I set the cranking angle to 15 to match. Not sure about that yet, but I just put it back to time based cranking for now.

 

Thanks to James for putting in the rising and falling second trigger mod. Thanks to Mathew for helping me understand the LS1 trigger positions.

Thanks to Z-ya for advice on LS1 coil dwell

Thanks to the LS1 for teaching more than I ever wanted to know about wasted spark/wheel decoders/second trigger

 

Ha Ha, just had to have an oscar moment :mrgreen:

Share this post


Link to post
Share on other sites

Well, the fun is over. Don't wait on me to finish testing out the LS1 megasquirt setup. You can read the entire horror story and see pictures at the follwoing link. It had nothing to do with megasquirt. Basically I have to find a new engine or block.

 

http://www.v8rx7forum.com/v8-rx-7-build-threads/43173-mobythevans-ls1-fc-megasquirt-5.html#post413520

Share this post


Link to post
Share on other sites

OK, I got over my whining and bought an LM7(5.3) engine last week. Swapped all the LS1 parts over to it (intake/water pump/crankshaft pulley/oil pan/etc). Got the engine installed in my RX7 over the weekend and it is back up and running on Megasquirt. So I updated the instructions above with a couple new things I learned.

 

I let the engine idle for several minutes and it sounded really good. The MS setup seems to be very close. Next step is to tune with my wideband and see how it drives.

Share this post


Link to post
Share on other sites
BTW, thanks for the encouragement on the project. Maybe the first test drive tonight, if not definitely tomorrow. My wife is really supportive of me with the cars, she has her own hobby/business with horses so it goes both ways.

 

That's funny, I have the exact same relationship with my wife! No matter what I spend on cars it'll never come close to the horses!

 

Sorry to hear about your LS1 troubles, I've been looking into options now that I had troubles of my own. Maybe I'll pull the 5.3 out of my Tahoe and put it in the Z! Okay maybe not, but I need to find something.

Share this post


Link to post
Share on other sites

I am very close to having this running correctly. I have spent the last two weeks working out issues with the new 029y4 code and the LS1 sensors. The car has been running, but the trigger inputs are not decoded correctly so RPM is not steady and therefore it doesn't run very good.

 

I finally completed experiments last night that are very promising, so I am building a little extra circuitry and should be running 100%.

Share this post


Link to post
Share on other sites

Fixed 3 things and now I get steady rpm and the car is running pretty good. I am working on tuning now. I updated the settings in this thread to reflect what I found.

 

1. I had a bad opto transistor, it finally quit all together.

2. I had some settings wrong in megatune.

3. I had a bunch of higher frequency noise all over my MS setup. I added bypassing caps to the following wires on the relay board (.01uF and .22uF): battery 12volt, switched 12v, crank sensor input, cam sensor input, injector bank 1 control, injector bank 2 control.

 

Now I am going to install the knocksense module to monitor while tuning.

 

Also, I get a little bit of popping out the pipe at about 80kpa and 2000rpm. Could that happen if I'm too rich, it is definitely rich right now as I am just starting tuning.

Share this post


Link to post
Share on other sites

×
×
  • Create New...