Mazdarian

iTrader: (1)

Hi all,

I've been busy trying to get my 'build' running, but sorta hit a roadblock. The engine catches within a few cranks, fires right up, and maintains a steady idle. It also responds to throttle input really well, so at least i've got some settings right and the engine seems healthy. It doesn't - however - brap at all, and doesn't sound crisp, like something's out of whack.

There's tons of specs I could rattle off here, and i've got ample maps and logs to share, so name whatever info you'd like and i'll happily provide.
I'm looking for someone to take a look at what i've got and tell me if something seems off, and maybe start a conversation about ways forward.

basic specs;

- 13B 6-port
- Half bridge
- 50mm ITB
- 4 x 550cc injectors
- 4 x IGN1A AEM coils
- MS3Pro EVO ecu

stuff i've checked/fixed;

- lost sync errors: checked, with highspeed logs, fixed, rock-solid now
- Ignition firing order: wiring got jumbled somewhere, fixed now
- Injector firing order: set as per manual, though I don't understand the what/why yet.

I'm still unsure about lot of things, so having someone to talk to about this would be invaluable..

Hope anyone chips in

elturbonitroso

did you sync the CAS and the ecu and what a/f you have at idle

Mazdarian

iTrader: (1)

@elturbonitroso if by synced you mean stab it correctly then i think yes, as per FSM. @ idle (currently ~2000/2200rpm) i'm getting ~12:1 AFR.

elturbonitroso

ok but do you lock the timing on the ecu and check that its sync properly with a timing light.

Mazdarian

iTrader: (1)

Ah like that, yeah timing was set and checked (:

peejay

Check your ignition timing, it might be really retarded.

If your port to throttle blade distance is really short, it will idle smooth as glass no matter what, so don't focus on that. I can get my full bridge to idle dead smooth at 900rpm using a Holley manifold, not even a "proper" ITB manifold. The rough idle comes from vacuum in the manifold pulling exhaust gases up into the chamber on the intake stroke, smaller post-throttle intake volume means vacuum dies quickly.

50mm is HUGE! I have 42mm throttle plates and it only just barely starts pulling a vacuum at WOT at the top end. Even the stupid high power Aussie IPRA engines only run 40mm ITBs, but they also have some seriously scienced-out intake manifolds.

If you mean that you only have two throttle plates and not four like ITBs.... on a half bridge, especially a six port half bridge, the two different port opening and closing times are going to fight each other in the intake manifold, and it's going to kinda suck, you need to isolate the primary and secondary ports from each other in order to make that work, either with a modified S4 N/A intake manifold or a proper 4 throttle ITB setup.

Mazdarian

iTrader: (1)

Hi Peejay,

Thanks for the input. I'll check my timing again but it's steady at 22 degrees in the logs. I do have an Atkins pulley that has marks at 10 ATDC, 0 degrees, 10 BTDC and 20 BTDC, I added a 5 ATDC mark myself and stabbed the CAS to that mark as per FSM. Can't imagine it being off that much mechanically..

Engine's pulling enough vacuum at idle, but i get what you mean about it being a small volume and thus dissapearing quickly. I originally planned this setup out with the guys over at EFIHardware to have a tuned length of roughly 17"/45cm to get me in the desired spot on the rev-range, idk if that's considered stupid-short but it's what we ended up with *shrugs*

Imma say the ITB bore size is up for debate, but i'll grant you they're not the smallest around

About the port overlap and timing on a 6port; yeah, I manually closed the throttles causing the rpm's to drop closer to 1k and sorta heard the onset of a clean non-brappy idle so i'll take your word for it!

peejay

It is critical to check timing with a timing light if you have not done so, you could be quite far off and not know it.

Mazdarian

iTrader: (1)

I know, and I did initially, thing twice actually, but will check again and update thanks for all the feedback so far!

driftxsequence

iTrader: (8)

Can you post a video of it running?

Mazdarian

iTrader: (1)

Yeah ofcourse, might take a while though as i'm not near the car atm

Malic

You did change it back off of fixed timing after you set the timing right? The timing should not be locked to 22 when it is moving around the MAP table

Mazdarian

iTrader: (1)

@Malic Yes, timing was locked at 5° ATDC for timing light check then unlocked after

I've got a video of it idling in the meanwhile, I'll try and post it tonight. If you look at the AFR gauge you can see it bounce around a LOT, in kind of the same rythm as the engine's spluttering.

I went to check the plugs and the front ones were visibly wet. This got me thinking maybe the front rotor had a major lack of compression. I got a rotary compression tester delivered and did a check; fine for a rebuild.
The front rotor actualy has a bit more compression than the rear one, so that kinda ruled out the bad compression leading to wet plugs / misfires idea.

So it's an ignition issue then. Here's some checks I did;

- Checked wiring of the coils
- Checked voltage to the coils
- Checked grounding of the coils
- Checked coil operation via TunerStudio
- Checked wiring of the CAS
- Checked stabbing of the CAS

No faults were found, so I'm left scratching my head again..

Malic

Just two statements from this, I thought that the timing marks are for 10 degrees, so wondering where the 5 degrees is coming from

Your described behavior sounds like other posts where the leading plugs were swapped. Would idle, but dies and sputters when trying to throttle up. Your first post said "- Ignition firing order: wiring got jumbled somewhere, fixed now" I would triple and quadruple check this.

Mayhap include a tune file so someone can look over it and see if anything stands out

Mazdarian

iTrader: (1)

@Malic thx for the reply.

To clarify;

- The 5° ATDC derives from the way the FSM describes setting base timing;

These same numbers are used in Aaroncake's MS guide to set base timing in the MS3;



The wiring checks out, i've confirmed this through Tunerstudio's ignition/injector-test app;


No luck on the video yet, will try again tonight AND add an msq and log.

Thanks so far,

Mazdarian

iTrader: (1)

Right.

Here's the vid;


.msq and log attached below.

Malic

Am seeing a few things that look off to me, just adding them in as I go though each part of the tune.

The fuel map, starting at 800 rpm, it cuts the fuel to half by the time you hit 1200 rpm, so after 800, it just ramps down to half fuel. TPS does not move during entire log, so it is at the 6.5 row the entire time, so engine is starving for fuel. Timing map is way to advanced.




Most of the Rx7 timing maps I see would be in the 8-10 degrees at idle, but yours is 22? There is only a 4 degree swing from idle to redline.

Would also recommend to make the RPM scale on all the tables the same, your ignition table goes down to 1000 but your fuel table goes to 750, and then 8500/9000 even though all are 16 columns?




Overun fuel cut is also on, do you have a separate system that us running the oil metering pump at all times when fuel is cut off?
You are tuning Alpha-N, but state settings show it is looking for MAP of 20 or below to activate overrun?
Log is showing 102 MAP, idles at 76, so is telling me something is not calibrated unless you live below sea level on startup

Back to, TPS during the entire log is showing 6.5, is it calibrated? This requires TPS to be giving a reading to work. In the log, as RPM raises, throttle position is not moving at all
As said before, as RPM goes up, fuel goes down, opposite of what you want, and your timing is likley just blowing a bunch of unburnt fuel out the exhaust and is why the wideband is showing as getting richer. Timing is way off.

When you said "stays at 22 degrees" for the log I thought you had locked your fixed advance when setting your timing as per one of my previous replies, the car is wayyyyyy too advanced and ignition table needs to be looked at

I would say start off with flattening your fuel to mid 50's across that blue 4x4 dip and fix the timing map to start with.

Mazdarian

iTrader: (1)

The 4x4 dip in the VE was a remnant of the last thing I tried, that was already reversed afterwards.

I've gone and rebinned the VE and Ignition tables to both show 750 > 9000rpm. I've also reversed back to an older timing map, with more conservative numbers down low. I'll try and get a log later today. Overrun shouldn't trigger either way while idle tuning, but i've switched it off nontheless.

About the MAP values; I do live below sealevel.. about 5 feet lol

thx for the input.

driftxsequence

iTrader: (8)

I think i did this before.

Your ignition settings are incorrect you either have opposites for the "going high vs going low" option, or the tooth picking rising/falling.

I'm like 97.3% sure thats whats going on here.

Mazdarian

iTrader: (1)

@driftxsequence Thanks for chipping in, and a good point at that. Still though;

Spark Output is set to 'Going high'. Setting this to 'Going low' on a set of IGN-1A's will fry them, as will 99% of other coils.

Ignition Imput Capture is set to 'Rising edge', as this proves to be the most crisp and noticeable signal from the CAS. There's no teething issues, i've already confirmed this before with the high speed logger / composite logger. If there were teething issues and the ECU was unsure of a signal being a tooth or noise there would be 'lost sync' counts in the logs, no?

I agree that these issues are common and often overlooked, but i sadly don't think it's the case here.

Thanks.

Mazdarian

iTrader: (1)

For posterity I did a spark check, partially to confirm my ignition wiring wasn't somehow crossed.

As shown, from left to right, with the MS3's outputs listed;

Spark A - Leading 1
Spark C - Trailing 1
Spark B - Leading 2
Spark D - Trailing 2

elturbonitroso

change your ignition capture to falling edge ,thats what i use all the time and never have a problem

Mazdarian

iTrader: (1)

@Malic So I gave it another shot today with a revised timing map. Ran pretty horrible at lower rpm's, with the engine physically shaking quite badly. I've attached a log and msq of said run, with a video to boot. Still feel like something's up with the firing order, but i've checked it in every way i know of to no avail..

Still tho, I recently came across someone on NoRotors forum stating the actual firing order and tunerstudio's ignition test app firing order differ, can anyone here attest to that?

Anyways, here's the vid;


cheers,

driftxsequence

iTrader: (8)

Messing with your timing map is only masking whatever issue is actually occuring.

Possible clogged/messed up injector?

Try adding 180* to your offset to see if youre firing out of sync?

Malic

Yeah, something is not right, can try switching to trailing edge instead of rising edge like elturbonitroso suggested (do not confuse this with going high and going low, one fry's your coils, the other looks at when the sensor reads or stops reading the cam signal)

Could also unplug your trailing coils, because it should be able to run on the leading alone, and rule that out as well, make sure that they are hooked up right.


Just for a sanity check, you are not reading off a timing wheel on the eccentric shaft by chance, and timing wheel was flipped 180 degrees on install? I see this happen on the Miata Turbo Megasquirt forums a lot and it throws off timing creating this kind of behavior there too.