the thread where Savington asks a lot of really, really dumb questions.
 

I am putting an AEM 30-1710 EMS into my '94 chassis with a 2002 motor on stock sensors. I have a couple of early CASs laying around if I need them as well.

Let's start with the dumbest of the dumb: MAP sensor input, hooks up to 4F, labeled "Spare MAP", right? Or is that for a spare map? :hustler:

IAT I think I have figured out, I wired it straight to the ECU connector when I did my rewire in the summer.

VVT and the stock wheel is what I'm most confused about. Where do I feed the signal for the stock trigger wheel into the ECU? Through the OEM CAS connector? I've seen Jason's thread on VVT settings. Can I just use an OEM exhaust-side CAS with the VVT stuff and somehow lock the adjuster in place?

I'm using TurboTim's calibration, rescaled for 750cc injectors, so I think I'm in the ballpark for stuff like idle, etc.

Trying to keep it as simple as possible to start out with - I want to try to lock the VVT at zero advance/retard, and if I have to add a couple of "unnecessary" sensors to make things work that's fine with me.

Goal is to get the car running with zero bells and whistles. Get the new motor broken in, rough the fuel in, win the Miata Challenge championship on Sunday, and then pull the head back out for headwork and ECU bells/whistles.

Jason C, Julien, if either of you have some time to hold a baby spoon for me, I'd be eternally grateful. I wish there were more info on basic stuff here.

Yes, 4F is your MAP sensor input. Use the sensor wizard to set the config table for whatever MAP sensor you are using. You don't use the MAF input, pin 2O.

I am no help with setting up the CAS sensors for anything other than the 92 style CAS. You will have to change the tooth counts for the fuel injectors and coils. Are your COPs wired sequentially? I assume not cause of the megasquirt. Using your old style CAS should get your car running, but my map as-is has coils 1-4 fired sequentially (not in that order obviously). If you're batch fire, use the standard config/tooth counts in the default maps, or I can get them for you. My fuel injectors also fire sequentially but your '94 should be wired fine.

My injectors are 440cc @ 54psi static. These COPs don't like battery voltage below 12volts. My red top optima is below that when cranking even though the battery tender has been on it forever. It will crank then backfire bigtime. Jumpstarting it works. Anyway...best of luck. I'll help whenever I can.

I presume you have the 01+ cam (rear of intake cam) and crank sensors.

Yes you can send the signals through the chassis' wiring which is originally meant for the 2 NA motor CAS signals (the CAS has 2 signals). The one with more teeth is the CKP signal.

These will go to the pins for CKP and CMP. CKP is the crank signal, and CMP is the cam signal. Check that this is where it routes to in the AEM schematic/manual.

Load the cam and crank setup variables that I showed in the post I made. I presume you're running COPs. I show a setup for that too.

I would at the least, run the VVT, to run fully advanced between 1500 RPM and 4500 RPM or whatever y8s found was optimal, and retarded outside that. You can use any of the "low side" switch outputs of the AEM and program it to do be an RPM window switch.

Use the stock harness to feed the crank position and the cam sync signals. And I guess for the VVT engine as a start the best is to use Jason's setup from here:
https://www.miataturbo.net/forum/t27028/

The variables are found in the cam/crank setup page.

The coil setup is in the coil phasing page.

The injector timing setup is in the injector phasing page.

Be sure to check timing before revving the motor and going into the ECU setup -> adjust timing page.

Oh and start now and use "boost comp" so that your fuel pulse page appears flat wrt MAP.

Yes, I think - my valve cover did not come with the rear of intake cam sensor, but the one off the front of my '99 valve cover looks mighty familiar. I can't find the part in the comp diagrams, so I'm not 100% sure if it's the same or not.

OK, cool. I didn't know the CAS produced two signals.

Logic check:
-Hook the crank sensor signal wire into the OEM harness' Crank input (2E, White wire)
-Hook the cam sensor signal wire (from the valve cover sensor) into the OEM harness' Cam input (2G, Yellow/Blue wire)

I've got my coils wired for wasted spark. I can rewire to sequential without a big hassle, but the less I change the better (for now). If I do wire them sequential, I'll just use Coil 3 and Coil 4, 3A and 3B on the AEM's 3rd harness?

ok. My goal with the VVT is to keep my fuel maps as intact as possible - I'm going to attempt to graft the MS fuel map over using datalogs, although I fully expect this to fail in every way possible. When that fails, I have Tim's TT fuel map which should be somewhat close.

As far as boost comp, I want to fiddle with that, but I have plans to try and get some fresh parts and +1 valves into this head, so the less time invested I have into the fuel map the better.

LS = low side?

Is the closed-loop fueling a big whore to set up, or is it actually decent like I hope it would be on a $2k+ ECU? :)

Thanks a ton, Jason.

For fuel start using the "boost comp" method as Jason already suggested. It makes your life a million times easier, because you don't have to deal with a full 3D map. Very easy to tune the fueling this way.

Setup boost comp now, it will make troubleshooting easier.
Translate your old MS maps to boost comp.

Boost comp if setup correctly will multiply the fuel injector on time by MAP.
Warning DO NOT do the "hybrid" setup discussed in the AEM forums wherein the boost comp curve is not as I describe below. That will just confuse you.

Setup the boost comp table like this:

Assuming you have a manifold referenced fuel pressure regulator.

Fuel - advanced - fuel trims - boost fuel - table

For every value of MAP, enter above it, the value MAP-101

For example, if you have an entry for 150 kPa, above it you enter 49
For an entry of 50 kPa, above it you enter -51.
This is important. The reason for this funky equation is that the numbers you enter are interpreted as %fuel to add. Thus at 1 bar of boost (202 kPa), you add 100% fuel as compared to 101 kPa. At 50.5 kPa you add -50%. (subtract 50%)

This will flatten your fuel numbers wrt MAP. The resulting fuel map will only curve as a function of VE (you will see it peak vs RPM, at 5500 RPM or so), and if you want to enrich it in boost, or if you want to deviate from 14.7:1.

Then you take your MS fuel table and correct it to work with a boost comp setup.

What you do is for every fuel pulse width, multiply it by 101/MAP.

For example, if you have 2.00 mS for some RPM, at MAP=50, enter into the AEM fuel map a value which yields 4.04 mS.

Now in the AEM fuel map, you have 3 ways of displaying the numbers. Width, duty cycle, and "raw". Right click to change.

Raw is a number between 0 and 255. This number is multiplied by the variable "microsec/bit" which is found in fuel-advanced-trims-options. The result is the actual on-time. You will want to choose a value for microsec/bit such that the maximum raw value exceeds 100, preferably above 200. This gives you better resolution.

As you can see, the variable microsec/bit is like a gross fuel multiplier which you can change when you change injectors.

Yes LS = low side.

Another tip. After installing the software for the first time, do Configure-Units and set AFR to gasoline.

easiest way to change my map for injectors is to change said microsec/bit. EDIT: Sorry I see JasonC said that in his last line.

There's also a feature that rescales the entire map such that your max ms value entered (assuming you entered ms values when viewing ms values, not bits when entering bits) gets set to 255 (or so) and the microsec/bit gets set accordingly, so your ms values are close to what you entered. It automatically sets the resolution as good as possible. My max bit is 212 so you can see I haven't done this in a while. You boost past 18psi so you'll have to rescale your breakpoints.
EDIT: right click the fuel map, select "rescale fuel map". I just did it, it rescaled my fuel map so the max bit is now 245 @ 65 microsec/bit instead of 212 @75.

I do not know if this feature would work with the boost comp method of setting up your map without thinking about it which isn't going to happen at the moment. I don't use boost comp, after seeing it demonstrated at AEM and trying it on my car, I prefer the standard map layout where a higher load row has a higher ms (bit) value in it, and keeping my boost fuel correct table zeroooooo. While it could give you significantly higher resolution (needed?), I don't see how it makes tuning simplier, other than a single change in the boost fuel correction table will scale an entire load row. I don't know when i've ever had to do that, and if I did, I would just highlight the entire load row in the fuel map and hit the "+" key. If you need to adjust the fueling in one specific cell you still need to do that in the fuel map, not in the boost fuel correction table.

It will work with boostcomp.
All the rescale function does as you point out, is change usec/bit and the "raw" values.

It makes tuning easier bec for a given VE and a given target AFR, the value in the fuel table doesn't change with MAP.
So if you have the same AFR target say at 50 kPa, and at 80 kPa, the curve across RPM will be the same ... and when you examine the fuel map in 3D view, you can easily spot bumps and dips. Your fuel map only changes by maybe 2:1. Without boost fuel comp, your fuel map goes from 0.5 mS, to 15 mS, a change of 30:1, and so you can't see a 1 mS dip which is a 50% dip in a 2ms region, when your scale goes to 15 mS. With a boost comp map, the dip is pretty much 50% of full scale, easily visible.

Makes sense. My head isn't quite into the tuning, but I will definitely be reading that tomorrow. Thanks a lot, guys.

Two more questions:

-Where do I plug the VVT solenoid into the ECU?

-Jason, what did the marks on your '01 cam gears look like for setting the timing? There is a couple of marks and a dot on the exhaust side, and a mark and a dot on the intake side.

Hmm, I marked them with some paint, but didn't take pix, and it's at my machinist now.

VVT solenoid, use any LS (pulldown) output that isn't being used.

I think I figured it out. It's tricky because the VVT cam only has those two marks. No clue what the dot is for, but I aimed the line at the black "e" mark on the plate behind the gears and the cams seem to be actuating at a logical point. Exhaust opens after TDC, then it closes right as the next TDC comes up, then the intake goes for half a stroke then closes up for the compression. Mi logic iz gude.

Jason, assuming I can just set that output in the software no matter which one I use?

Also, does anyone know what the size and thread pitch is for the VVT oil feed banjo bolt at the back of the valve cover?

Jason, I am not going to have time to deal with tuning around the VVT. Is there a way to default lock it to 0 degrees advance/retard?

So, how did this go?

Discretion is the better part of valor. I came down with a sinus infection while I was doing this work and didn't make my deadline, so I took a break for the holidays. I will probably end up pulling the head and having it built before I fire everything up, so it may be a couple of months before anything comes to fruition.

You may have already dealt with the issue, but you can use the MAF input (2O) as the input for your MAP. Althought the AEM does have the MAP input on plug 4, you can just go to the sensor setup and set the input to the MAF signal and it works. I found it much easier to use the MAF sensor wiring instead of running another signal wire.

I wish I thought of that. Damn you!

It's really simple to take a wire out of the terminal plug and move it to whatever terminal you want.

But is it easier than just telling the ECU it is a MAP and not a MAF?

Might be - my wiring harness has been stripped down and I'm not even sure 2O exists on my car.

e: i'm a moron.
[/idiot]

rofl, I had that same problem when I started using it :)

dont you have to hit enter?

:P

Yep. I was really confused for a couple of minutes though.

Oh snaps, it fired up. New MAP sensor fixed it. Swapped injectors as well so I set all that up again. Looks like I am seeing too many RPM, though - the engine sounds like it's lugging along at ~1000ish but RPM on the AEM says ~2000+.

If the ecu thought it was running twice the speed, wouldn't it be firing on the compression stroke too? I dunno.

I'm glad to hear the MAP sensor fixed that issue.

Make sure all of the sensors are set properly (cam sensor etc). For some reason I seem to remember having this problem myself, but can't remember how I fixed it...

yeah, all my cam/crank sensor settings reset themselves. Set it all back and now RPM is half of what it should be. Car is giving little hints of firing but its not quite there.

OK, got all the settings back to where they need to be as per Cuadra's 2001 VVT settings. It cranks and syncs but all I get are backfires. I have spark, I assume the backfires mean I have fuel, but I'm starting to wonder whether everything is firing in the right order. What datalog do you guys (Jason, Julien, et.al) need to help me diagnose this?

I'm so close...

Maybe you have the COP signal wiring wrong.
I presume you used the COP/VVT settings? (as opposed to the non COP settings)

When I first installed my COPs in sequential I had major backfiring. I played around until I got them to work. It wasn't until I reinstalled the OEM coils a month ago that I figured out what was wrong. The default AEM wiring/naming has "coil 1" being the wire that runs to the coil pack that fired cyls 2&3. "coil 2" fired 1 & 4.

This may not be your issue at all, and I run my COPs where coil 1 fires cyl 1, etc.

I was able to run the car on 1 or 2 cylinders while diagnosing the issue (activating one coil/injector at a time to see what cyls are backfiring). Definitely a scary time and sounds like rods are being bent. I blew my boost gauge hose off too.

The setup I posted requires that coil outputs 1,2,3,4 connect to COPs 1,2,3,4:

AEM trigger and dwell setup for '99~'05 with Toyota COPs - MX-5 Miata Forum

I saw your post at the AEM forums first so I answered there. Give me that list, then I'll tell you whether what you have in your ignition and injection positions is correct :) If you're just copying you might be off - i.e. Jason's wired his ECU different than mine (different outputs fire different cylinders/coils) so just copying numbers might not be a good idea without understanding what they mean first

Yes, I get SYNC ON and it stays that way while cranking.

Lots of fuel smell and backfires,eh? Are you hiding in my garage? ;)

The motor is a 2002, but I'm not using any of the 2002 wiring. Wiring is pretty much all 1994 stock, except for the cam and crank sensors which I built a harness to feed into the OEM CAS harness.

Injector wiring is stock 1994:
Injector 1 is 2U
Injector 2 is 2V
Injector 3 is 2Y
Injector 4 is 2Z

Coils are wired wasted spark through the '94 wiring:

Coil 1/4 is wired through 1G
Coil 2/3 is wired through 1H

I am not adverse to re-wiring the coils in sequential - if I did that, I'd follow the AEM pinout sheets, so:
Coil 1 to 1H
Coil 2 to 1G
Coil 3 to 3A
Coil 4 to 3B

Interesting to note that the AEM pinout says that 1G is for coil 2, and 1H is for coil 1 (opposite of what I have), but the 1994 wiring diagram puts coils 1/4 on 1G and coils 2/3 on 1H. Typo in their pinout?


OEM crank sensor signal is wired through 2E
OEM cam sensor signal is wired through 2G

If you can bear to explain it, I would love to hear how all of these settings relate to the motor. This is sort of a new area of the car for me - I have never been a big wiring/circuits/electronics guy.


Schwing, thanks Julien. I don't have an auxiliary tach in the car so 600 is a guess, and I haven't heard the car run in 2 months. Better to be off myself than have the ECU be off, though. :drool:

Let's say long time ago I know of a garage which was similar to yours :laugh:


Not a typo. That's why I asked you about the info above, but looks like you're advancing fast by yourself too :p
Someone at AEM just decided that in the Miata wiring the EMS output named "coil 2" should be wired to fire the 1/4 coil and the output named "coil 1" should be wired to fire the 2/4 coil.
So, if using the '94 AEM startup calibration as a base, but you wire the harness differently, you have to change the numbers in the software calibration to match your new wiring ('cause you'll have the spark by 180 degrees otherwise)

I'll look at your numbers in the calibration and will get back to you

In that calibration you posted at AEM forums your ignition is off by 180 degrees for sure.

Go to: Ignition -> Advanced Ignition -> Ignition Phasing -> Options

and in the table:

Ign Tooth #1 (with your wiring this is your cylinder #2) should get 0.00
Ign Tooth #2 (with your wiring this is your cylinder #1) should get 1.00
Ign Tooth #6 (with your wiring this is your cylinder #3) should get 2.00
Ign Tooth #7 (with your wiring this is your cylinder #4) should get 3.00

Ign Tooth #3, #4, #5, #8, #9, #10 values don't matter because you're not using Coil 3, Coil 4 & Coil 5 of the AEM EMS for now

A difference of 1.0 is 180 degrees (4 tooth engine cycle in that Miata setup, so 4.00 difference = 720 degrees, 1.0 difference = 180 degrees).

So your current numbers are by 180 degrees off, which means you fire on TDC exhaust ... a nice full time anti-lag you have currently setup there ... :laugh: you fire the fuel right when you blow it in the tail pipe

Those numbers for the each Ign Tooth specify where in the engine cycle each output fires.
Ign Tooth 1 and Ign Tooth 6 are pin 1H (the name of which is "Coil 1")
Ign Tooth 2 and Ign Tooth 7 are pin 1G (the name of which is "Coil 2")

I haven't checked your fuel injection positions yet

Your fueling positions ( Fuel -> Advanced Fuel -> Injector Phasing -> Options ) are correct.

With the current numbers you're always firing the injectors at 79 degrees after TDC exhaust, which should be an ok startup setup (the exhaust valves are closed by then)

So, just fix your ignition numbers and start that engine :)

"Tooth #1" really means "TDC for IG output #1 is at:"
Then the number you enter to the right is the crank tooth #, for TDC. The TDC point is further fine tuned by the variable which you adjust in the timing setup screen.

The AEM on a 99-05 miata counts only 2 of the 4 teeth, so it virtually has a 4 tooth engine cycle. (It can't time off of uneven teeth, so it has to ignore 2 of the teeth on the crank) The gap between teeth is 180* on the crank. (Aside: I'd wanted to get someone to build me a 12-1 trigger wheel for the ATI, for more resolution)

So

"tooth #1" is where you fire IG output #1
"tooth #2" is where you fire IG output #2
"tooth #3" is where you fire IG output #3
"tooth #4" is where you fire IG output #4
"tooth #5" is where you fire IG output #5

"tooth #6" is the "virtual" ignition output, for waste spark, and its signal comes out on IG output #1.
"tooth #7" comes out on IG out #2
and so on

If you have waste spark where #1 and #4 are paired, and if you connect its coil to IG out #1, then you will use "tooth #1" and "tooth #6" to set their timing respectively. The reason for considering them separate despite sharing an output wire, is so you can do cylinder trim and distinguish between 1 and 4.

Which actually is the "Ignition Sync" field in the screen I mentioned above :)

It just shifts all the Ign Tooth XX values by adding the value it contains.
It's just an extra variable (added there for convenience) so you can quickly shift all those values at the same time by a single number.

BTW will it accept a negative value?

Brand new motor, different ignition sensors, brand new ECU, new 1000cc injectors... and with Julien's settings, it lit right off. :) Need to sort out the idle and a radiator leak that I caused, but it lives. :) I owe you a beer, Julien.

:bigtu:

Btw, be careful not to overheat that engine.
With my '94 back then I found that the OEM Miata coolant sensor calibration in the AEM Wizard was wrong.

Here are the numbers I am using in the Coolant Sensor Cal table (the values are in degrees F) and this way the AEM via the stock '94 sensor reads exactly as the aftermarket coolant sensor/water temp gauge I have:

261 259 221 190 180 167 154 140 133 127 118 111 109 102 95 88 79 72 64 57 50 45 37 30 25 18 12 7 1 -4 -9 -15 -20


The radiator fan in my software calibration is set to turn on above 201 degrees F, and turn off below 196 degrees F.

I am running the stock '94 thermostat (which IIRC is set to maintain 190 degrees F)

In the AEM startup calibration you'll see that the fan is set to come on above 171 and turn off below 165 which just ain't right, cause the stock thermostat keep point is above 165. I guess it somehow works with their wizard calibration numbers, but be careful :)

Do you have a good aftermarket gauge and sensor besides the stock ones?

I do, yeah. I don't use a thermostat on this motor so I'll set the fans around ~170. Thanks.:bigtu:

Ok, so after you put my coolant sensor calibration numbers in, monitor whether the Coolant Temp parameter is showing the same number as your aftermarket gauge from cold to warm while your engine is warming up. Should be close with my numbers and it should be quite off with the AEM numbers.

Yes as a logic, but not literally

The period of the tooth counter in the Miata case is 4.0 (because you have 4 teeth), which means all the numbers describing a position in the engine cycle are positive numbers in the 0 .. 3.99(9) range.

4 is the same as 0, that's why the max number is 3.99(9)

And if you want a negative number -0.05 for example then it is the exact same thing as 3.95 ( 4.0 - 0.05 = 3.95 ), so you put 3.95 in the field
If in the "Set Ignition ..." screen you keep pressing the "-" button and monitor the "Ignition Sync" option, you'll see that it is what exactly will happen once you request smaller than 0 number.

Dunno if my explanation was clear, if not tell me I'll try again :)

Wow, that could explain why my setup is always running super hot. I have a second fan hooked up to a manual switch, and find I'm using it all the time. Will have to redo that table.

My car with my numbers from above - the fan works like stock. I calibrated the numbers by looking at the aftermarket gauge.

I guess some day I should just get a simple lab thermometer a container full of coolant, dip the Miata sensor in it, put it on the stove and record the readings in AEMPro vs the thermometer. That's what people usially do at the AEM forums for a new platform when if want to be precise :) Any volunteers to do it? ;)

Yes with a 4 tooth cycle it's modulo 4. :D

Do you use a restrictor? I'd run something to get enough pressure differential, so coolant goes through your CHRA.

I'm glad to hear the settings worked! Same thing happened to me being off 180deg. Got confused with coil 1 running 2&3 by default wasted spark, so when I switched to sequential, I kept the tooth count on coil 1 the same and assumed it was correct for cyl. 1.

My sequential ignition tooth counts match my sequential injection tooth counts.

Yeah, I use a gutted thermostat to restrict the flow slightly. I don't have a heater core or any way to bypass the thermostat when cold so the easiest way was just to gut it.

Stupid question time. Two wires on the VVT - purple and yellow. Which goes to INJ9 on the AEM, which goes to INJ10?

There should be just one wire to a LS output (INJ9). The other LS output (INJ10) is if you have an engine which has a second VVT cam (i.e. Honda K20).
I guess talk to Jason, he might've checked which wire in the '01 harness is what already.

Also maybe check in advance that INJ9 is already wired from the board the EMS connector - use a multimeter, while manually toggling INJ9 from the software (when you turn it on, double check that the corresponding parameter in the parameters window shows ON too). Don't forget to turn off the manual override once the test is done so it doesn't fight with the VVT functionality.

Does the VVT solenoid get 5 or 12 volts?

Does anyone have an online copy of the 2001+ wiring diagram?

12V sir.

Jim