Megasquirt Ignition Timing

[Red_Bullet]

The topic of ignition tables hasn't been covered in the Meagasquirt thread so far, please correct me if I'm wrong.
Consequently I thought I'd start one right here right now.
Car: NA8 Rotrex C30-74 Supercharger, Compression 9:1
I basically got my table from the big forum from illuminaries like Braineac.
I haven't had the car on a dyno so feel this is an area where gains might be available.
Anyway, for what it's worth here is my ignition table, feel free to comment, discuss etc:

[StanTheMan]

ooops not sure I really qualify here just yet.

[Red_Bullet]

What? No takers? Anybody running anything radically different or similar to this table?

[StanTheMan]

For what it's worth.

I don't have a knock sensor.

I'm also N/A

I run a slightly more aggressive spark table on 98
A significantly more aggressive table on E85

I've hunted around for concrete walls for the engine noise to bounce of so I can listen for detonation.my hearing is not so good.
As far as I can tell there is none.

I feel your table is a decent one to start off from. It is just a tad more aggressive than the usual base map. Which runs at about 37 deg max. Running it richer or leaner may have an effect on detonation.

When it comes to Megasquirt & cam choice. I don't know. The peeps that do know make it their livelihood and the others pay for it tend to not want to share that info.
I'm in a place where I now know a little. But unsure.

I found a few guys like miataturd quite knowledgable on the big forum and also happy to help out.

[ManiacLachy]

Spark isn't something I really understand and so it's an area I mostly leave to a tuner. The exceptions being overrun and idle, where loads are quite low and experimentation is fairly safe.

Everything I've read says to aim for close to 10 degrees in your idle cells, though my DIYPNP map had it around 22 where yours is. I've brought mine down a bit now, and use Idle Spark Advance to bump it up/down a few degrees as needed to keep the idle steady. In the overrun I've been told you can reduce it quite low if you want to induce a few pops and burbles on the overrun.

For the rest of it, I'd keep it on a conservative base map (DIY/Braineack) and let a pro tune it on a dyno.

Why are your X & Y axies oddly numbered? By that I mean not rounded to 5 or 0. It doesn't really matter if the Z axis is tuned correctly, I just mentally think in round numbers ie 60kpa/2600rpm instead of 58.7kpa/2593rpm.

[Red_Bullet]

My AFR's and VE are pretty much dialled in over a couple of years now on this Ignition table. I consider it fairly safe. Not interested in inducing detonation to gain a few more KW (already broken one 5 speed gearbox with the (unknown) power I already have). BTW it's running on 98. Not sure why my X axis has non rounded columns, from memory my first MSQ came from glsender so I suspect gslender probably set them up like that. I should open the original MSQ and have a look. Just typed the intervals into excel and calculated the deltas between columns, they are linear above 2593 revs.
RPM Delta
7200
6613 587
6027 586
5440 587
4853 587
4267 586
3680 587
3107 573
2593 514
2200 393
1867 333
1567 300
1200 367
993 207
847 146
700 147


I recall I think I have pulled a few degrees from gslenders original table after reading and re-reading miata turbo discussions. I figured better safe than sorry. I have played with the timing in idle, it seems to like 22 degrees, idles smooth, stable AFR. I have added a couple of degrees at times to loaded areas but my seat of the pants dyno noticed no real gains so I've reverted back to what you see above.

Nitro Dann would be the man to sort it further but it's along drive from Brissy! Imagine turning up there with a damn supercharger! Plus I'm running too much camber for a long road trip.

[Red_Bullet]

Here where I began 2 years ago, courtesy of gslender who provided me with a base map. I've since gone more conservative over time. This also confirms where the unusual increments of RPM came from. NOTE:Use at your own risk, this is for information only and is not to be regarded as any recommendation whatsoever!

[StanTheMan]

Knowig this was written by Gslender there must be good reason & method for it. Didnt he rewrite the code for megasquirt?
On one hand the internet tells us for perfomance cars we should have full spark advance by 3000 rpm. But here we have a gradual advance. I dont think ive ever reached the parts wehere hes got 42 deg.....

Sent from my SM-G935F using Tapatalk

[ManiacLachy]

You would only reach those 42s by snapping the throttle shut at redline. I don't know how much advance is safe down there in vacuum though

Red_Bullet, I don't think Dann would have any reservations about working on a SC setup, and the drive down isn't too bad. However if your camber is that far dialed that it'll kill your tyres then I understand. If you're happy with the tune and the power you currently make then rock on, no need to get it professionally done just for a few KWs. Maybe one day take it to an open dyno day to get a read out?

Any one know of any good reading resources on Spark Advance theory? I feel I should educate myself.

[StanTheMan]

http://www.innovatemotorsports.com/resources/myths.php

[Red_Bullet]

Thanks for the link to the Innovate info. What is clear is that the flame front travels fastest at AFR 12.5-13. It gets complicated with the interaction of timing and AFR at any given point. That's why I just figured to use something cloned/evolved off the Miata turbo forum, there is too much going on for newbs like me.

[Red_Bullet]

For anybody new or for anybody who hasn't come across this yet.
https://www.miataturbo.net/megasquirt-18/good-afr-target-table-61587/page3/

[Cus]

Knowig this was written by Gslender there must be good reason & method for it. Didnt he rewrite the code for megasquirt?

He didn't re-write it, but he did do a bunch of good work, like A/C Idle-up - his work ended up in the mainline releases, so we're all using a little bit of G's handiwork! (side question, where did G go?)

As for the "good reason" - the Megasquirt (all versions) do Bilinear Interpolation from 4 cells to determine what the actual figures it's using are.

So, if you've got a spark table like the example below, it will draw a line from the closest cells to the actual engine state, and work out how much timing is required.

Code: Select all

50kPa        20         30
40kPa        10         20
        1000RPM    2000RPM


at 45kPa / 1000RPM You'd have 15deg of timing
at 45kPa / 1500RPM You'd have 20deg of timing
at 45kPa / 2000RPM You'd have 25deg of timing

If you want to fix/change the bins, click on the "up/right" arrow in the bottom/left corner of the spark/ve tables, then enter the new bin numbers, then press "Interpolate" and it'll work out what the values should be for the new bins.

Side note; I'm going to be having a go at braineack's AFR table, mine goes stoich at 70kPa, and rich at 80kPa, which means just about any time I'm cruising up hill, i'm soich or richer, which could be why I'm completely unable to get back to my stock ECU fuel usage stats.

Is there some kind of consensus when you should go from lean to rich in terms of engine load? I figure anything under 50kPa is going to be safe, but it quickly starts getting into napkin-math and knowing-things territory, which means I'm out of the conversation!

My current spark table is basemap + mt.net + gslender telling me to advance it + bartmanftw telling me to retard it. If it's hot, and the car is heat-soaked I was getting some pinging at light throttle at low RPMs. I pulled some time out of the affected area, but it hasn't been hot enough since. I've no (noticable) pinging @ WOT (100kPa) but also my table is topping out at 30deg

On one hand the internet tells us for perfomance cars we should have full spark advance by 3000 rpm. But here we have a gradual advance. I dont think ive ever reached the parts wehere hes got 42 deg.....

I've always assumed they meant at Wide-Open Throttle, which means only the 100kPa row for N/A users, and whatever rows are maximum boost for a given RPM. (but then the density of the A/F mixture changes the required spark timing as well once you add boost, so timing will be lower)


Also, here's a good (or, at least, long) thread from MT on spark advance: https://www.miataturbo.net/megasquirt-18/lets-discuss-spark-advance-69662/