Adam's Canadian 13B Rotary 818-SR Build

[Canadian818]

Good news is that the car starts and shuts down with the key, I've done it about 10 times today. Another good thing is I know my idle problems are related to my DBW system. My inability to fix my DBW problems however is due to the fact that my alternator isn't charging the battery. In fact voltage dropped at the battery when I started it. Pretty sure I have the alternator wired up wrong.

[Frank818]

Phew, that was close again, man you really know what to do to stress me out. lollll

What would the DBW have to do with the alternator? What's your voltage when the car runs? Between 13 and 15?
If the alt isn't charging the batt, where is its electricity going to?

It's normal to have the volt drop at crank, it will drop under 11v, cuz it pulls out a lot of juice from the batt.

[Canadian818]

Here's my list of problems.

1.Alternator isn't charging, no voltage increase when engine is running.

The AEM tech said that when the DBW setup is done, it needs to be done at whatever voltage the battery sees when running. He suggested to just have a charger hooked to the battery to simulate this, which allowed me to keep it at 13.6v

2. Pedal and TB work perfectly in the setup wizard, however at the end when I "commit" the changes it doesn't work. At all times, engine running or not, throttle position is at 20.9%. Thus why it wants to idle at 3000rpm

3. Map reading is stuck at 90kpa. New AEM 3.5bar sensor, properly setup and wired afaik.

4. Coolant leaking past my rad cap at the engine. This will be better sorted once I can maintain a reasonable idle.


AEM was supposed to get back to me, but no word yet. I suspect my grounding could be causing an issue, so I'm going to work on seperating a lot of them. I have two bolts on the engine that everything is grounded to. After some reading I think I made a mistake here. Coils are grounded to their respective rotor housings, and everything else is grounded to the same bolts, injectors, sensors, tb, and "sensor grounds" from the ecu. I'm going to separate everything and see if perhaps something is causing "noise".

[Toysrus]

I believe acc power is on in any position but off. Easy test put a multimeter on it. The ignition switch is generally a mechanical relay so each position turns on certain things. Only possible exception to this would be if a ecu had control of or a internal relay that it could open to cut power.

[Canadian818]

I believe acc power is on in any position but off. Easy test put a multimeter on it. The ignition switch is generally a mechanical relay so each position turns on certain things. Only possible exception to this would be if a ecu had control of or a internal relay that it could open to cut power.

I was thinking about this and realized that if I switched it to the ACC position, that would just become the new ON position. The car would keep running until I turned it to off. It's no longer an issue now that I've resolved the "run on" problem. Hopefully I can figure out the throttle and map problems tomorrow without having to replace parts. Once it's idling normal I can figure out the alternator and coolant issues.

[Toysrus]

http://www.teamhealeytexas.com/Techn...Wiring%201.htm

Disconnect the ignition switched power on the alternator check that it has no power key off , and power key on. The battery wire should be hot all the time.
You have it hooked up as a auto sensing 1 wire which should be ok but hooking it up as a 3 wire is simple and would bypass the alternators sensing circuit making it work sooner. Atleast this should eliminate the alternator as the problem and it's east to change or change back as all you need is a small jumper to #2 and move ignition power to #1.
When your are testing the power use the case of the alternator as your ground, if no power, ground your alternator case.

[Toysrus]

Describe the coolant issue a bit more. Is the cap leaking or is coolant being forced into the overflow tank? It could be a compression leak from the combustion or exhaust into the coolant.

[longislandwrx]

are you using stock rx7 alternator? do you have the L wire hooked up the the gauge cluster and the S to positive?

[Canadian818]

Describe the coolant issue a bit more. Is the cap leaking or is coolant being forced into the overflow tank? It could be a compression leak from the combustion or exhaust into the coolant.

It's not pushing into the the overflow. It could very well be an o-ring leak, however I did a pressure test on the cooling system with the exhaust manifold off and didn't hear anything. And I used the air lift kit to fill it and it held a good vacuum. I ran it once with the cap off, when running the coolant was sucked down the neck, and when I shut it off it shot out over the top. It'll be easier to diagnose once I can idle normally.

are you using stock rx7 alternator? do you have the L wire hooked up the the gauge cluster and the S to positive?

Stock FD alternator, L is left unhooked, s is to positive.

[longislandwrx]

with no reference off the cluster, I don't think it will charge the system. maybe an EE can chime in.

[Canadian818]

with no reference off the cluster, I don't think it will charge the system. maybe an EE can chime in.

I was reading a little on that last night. Seems to be a lot of confusion about "L". Some say it's irrelevant, some say it's needed. From what I understand the light gets power elsewhere and is grounded through "L", so I will try putting power to it and see what happens. I also read that "modern" alternators need some resistance on the signal wire like a light or the internal regulator will burn out.

[DrunkenSailor]

I’m not sure being an EE will help, but I’ll give it a go. Without going into too much electrical detail, here is how I understand the alternator should be hooked up. The “L” connection is not irrelevant, it is tied to a transistor in the voltage regulator. This transistor acts like a switch, turning on and off the charging voltage. (As an aside, the warning light is grounded through the transistor when the switch closes, turning the light on) The circuit to “L” normally runs: battery, through the ignition switch, warning light, a diode, and then to “L”.

To make a moderately long story short, when you check the voltages on the alternator, “B” to ground should read the 14.1V-14.7V at idle; “S” to ground should be the same at idle, and “L” to ground should be at 12.9V-13.5V at idle.

Just putting power on “L” from the battery may be too much voltage, I don’t know how sensitive the transistor in the regulator is, 14.1V from the Battery shouldn’t fry it, but it might.

[Canadian818]

How would you suggest I get a clean 12v source? Any of the wires coming from the alternator or battery will be higher. I assume the draw would be minimal, so I could tap into one of the 12v lines coming out of the ecu for the injectors or coils?

[Toysrus]

http://www.rx7.org/public/altxref.html


http://www.ausrotary.com/viewtopic.php?t=33676

Unfortunately these 2 sources don't agree, but it seems to me you can run a ignition on wire to L and a separate B+ fused wire to S. If you have another B+ wire coming from your fuse block that should work for S. Just come off any ignition on source for L.

[Toysrus]

How would you suggest I get a clean 12v source? Any of the wires coming from the alternator or battery will be higher. I assume the draw would be minimal, so I could tap into one of the 12v lines coming out of the ecu for the injectors or coils?

Your 12v ecu source wires are regulated to 12v won't work. Just use a separate battery wire

[Canadian818]

Ok I'll try that once I figure out this DBW bs. What a headache it's been! AEM got back to me, but not with a fix. They suggested I manually tweak and setup everything. They also said that my pedal numbers are way off, however from what I can figure they're spot on. My "target" position is bang on to what the pedal is doing, but the TB stays locked at 20%. I would think that I wired up the TB wrong except it cycles perfectly during the setup wizard. However if I unplug the pedal and turn on the ecu, the TB does a full sweep cycle, but with the pedal plugged in it just clicks.

[DrunkenSailor]

If I knew the current draw of "L" I would be able to tell you how to get a constant 13v to it, without that variable I can't be precise. A wire from ACC on the ignition switch may do it, you could check the voltage on the ACC pin to ground at idle, and see if it falls into the correct range. I am hesitant to suggest any of the workarounds from the RX7 sites after looking at the technical manual wiring myself.

[Toysrus]

Seems like it's the pedal calibration not the tb. Rerun calibration and make sure you selected correctly

[Toysrus]

I believe the regulator is looking for variable voltage to determine when to charge so a steady state voltage would be counter productive

[Frank818]

What a headache it's been!

Damn you bet!

[Frank818]

For the TB, if you move the pedal, does the % change or really stays locked at 20% all the time no many at what physical position the pedal is?

[DrunkenSailor]

I believe the regulator is looking for variable voltage to determine when to charge so a steady state voltage would be counter productive

That is true, I should have been more clear, a constant 13.0v at idle.

Canadian, one other thing to look at is the voltage on the ACC pin, with the switch turned to that position with the engine not running. If you see apx. 1v that is good. Then with the the engine at idle, if the ACC pin to ground reads in the 12.9v-13.5v range, you could hook up "L" to ACC. This should satisfy the transistor in the voltage regulator, allowing it to charge the battery when the voltage drops.

[Canadian818]

For the TB, if you move the pedal, does the % change or really stays locked at 20% all the time no many at what physical position the pedal is?

I did make a little head way with it last night. 20% is the "resting" position for the TB. If I manually move the blade it changes, so the tps is bang on. The problem is the calibration wizard didn't really work the way it should've. I can adjust the P.I.D values and bias %, but without a reasonable starting point I'm flying blind. I played with it a bit, but I'm so far off that when I did get it to move with the pedal, it wanted to go to positions it couldn't and made some terrible sounds. This was what the engineer sent my contact at AEM about my logs;

"Based on the logs from that RX7, I think both the values and the breakpoints in the DBW_Bias table are wrong,
It looks like the DBW_Bias table is set to use E46 BMW throttle values... the E46 throttles have an odd actuator, so those settings are unlikely to work well with any other DBW throttle."

So I've requested they give me some better numbers to work with as a starting point. One of the main reasons I chose the infinity ecu was because of its DBW support, and the reason I chose the LS3 TB was because it was one they supported.

That is true, I should have been more clear, a constant 13.0v at idle.

Canadian, one other thing to look at is the voltage on the ACC pin, with the switch turned to that position with the engine not running. If you see apx. 1v that is good. Then with the the engine at idle, if the ACC pin to ground reads in the 12.9v-13.5v range, you could hook up "L" to ACC. This should satisfy the transistor in the voltage regulator, allowing it to charge the battery when the voltage drops.

I will try that today. I seen some places online that "s" can go directly to "b" instead of to the battery. Any thoughts on that?

[DrunkenSailor]

Both B and S are direct to Battery (+) connections. Connecting "S" to "B", then "B" to the Battery is ok. Per the schematic.

[Frank818]

"Based on the logs from that RX7,...

Well your engineer is wrong right here. lolll

But yeah the DBW is not always simple to configure, I know with my ECU it's not that great either, but I have a cable so I'm lucky.
Hopefully AEM can help more.

[Canadian818]

I'm making some progress today! I have my settings close enough that it "tries" to hold the positions I request of it, there just seems to be some struggle. It's as if the open and close motors are fighting each other, and after 30sec the motors turn off and it rests at 20%. While setting up a DBW from scratch is out of the realm of AEM's tech support, and admittedly over their head, one of their engineers is a rotary guy and has offered to help.

Fingers crossed that I can have it idling normally today and put everyone's helpful alternator suggestions to work!

[Lumpyguy]

awesome to here bud, hope to see a video of it running good

[Canadian818]

[Canadian818]

AEM took good care of me, above and beyond! They exchanged quite a few emails, I sent a log, they sent me changes to make, I sent another log, etc. The DBW is setup bang on now. I haven't tuned the idle yet, and I still need to check timing, but I'm feeling optimistic. Oil pressure is now good, around 85-90psi on cold startup, settling to 50-60psi.

The map sensor problem I thought I had was just me being ignorant to the metric system. 90kpa is just atmospheric at this altitude.

I wired the alternator like we all discussed, "S" to "B" and "L" to ACC. Works good, 14v on at idle.

Coolant leak was also solved, and buttoned up.

I jacked up the wheels to try putting it in gear, seems to work ok, but I despise that damn shifter! Does everyone else actually hear their clutch engage? I was surprised to actually hear the wooshing noise of the clutch engagement and disengagement.

[turbomacncheese]

Oh man, I can't wait to hear that thing wind up!!

[Toysrus]

Sounds mean. Actually made me think about getting a rotary lol

[Mitch Wright]

Congrats, been enjoying watching your build. What shifter are you using?

[Canadian818]

Thanks guys! The idle note is quieter and deeper than expected.

Congrats, been enjoying watching your build. What shifter are you using?

It's the one that came with the kit. But I think an mr2 one is in my future.

[Mitch Wright]

BIG difference with the MR2 and shorter cables.

[Frank818]

Awesome! You fixed more issues and faster than I did on my VR6!

[Canadian818]

Awesome! You fixed more issues and faster than I did on my VR6!

Still lots to fix.

So I was attempting to tune my idle today, and came across some trouble. I'm not sure what normal numbers are, but after 5 mins or so my oil temp was around 60c and water temp was around 90c. Just after shut off (key off did nothing again, had to pull the coil fuse), I noticed the front rad cap was pissing out coolant. My front cap is a 20psi with overflow capped and rear is 16psi with overflow tank. Thoughts? I can't think why the lower pressure cap would bleed off pressure, air pocket maybe?

[Mitch Wright]

Air pocket is my bet.

[Frank818]

That's hot. Aside from air pocket which is well possible, would it warm up that much if only the coolant in the block/head would be circulating around instead of the thermoswitch opening to let coolant flow to the rad? Nah, even that it wouldn't warm that much that fast. But no coolant (or air), yes.

[Toysrus]

Is there a coolant bleed point on the motor? Is the rear cap too tight to function. I once pushed down too hard and my cap ended up resting on the stops. 1 cracked plastic radiator end cap later.... or a bad/ stuck 16psi cap.


bleed the system. If you have a pressure tester put that on in place of the cap. Could be a bad thermostat causing it to not flow and overheat or could be gasket or coolant seal allowing high pressure combustion leak into coolant. http://www.nopistons.com/2nd-generat...ressure-68270/

Also check that your radiator isn't clogged. Or maybe the extended coolant system builds pressure at the front first? Stuck thermostat would cause pressure and heat to go out the motor inlet... pinched or kinked coolant hose...

[Canadian818]

I filled the system using vacuum with the Airlift kit. New rad caps front and rear. The cap seems to function fine, it's spit some coolant into the overflow as it warmed up. T-stat and pump are also new.

A real bummer is I went to confirm timing today and realized that my underdrive pulley doesn't have timing marks on it. So now I need to remove it and transfer some marks. It's going to be a pain to remove. I'm actually looking forward to body work, lol