AFR, VE, Timing chart review before dyno tune
 

Hey guys,

I'm having a dyno tune done this week and wanted to post my map and charts for review before the tune. I've been using Tunerstudio VEAL to tune what I can with driving the car, but want a full range tune.

Specs:
2004 MSM
BEGI S4 manifold and downpipe
Garrett 2560R with ported wastegate
Artech 3" exhaust
FIC 775cc
MS3x ECU
BEGI #2 FMIC
3.63 gears
AEM UEGO WB 02
Hallman Pro RX MBC set to 11 PSI (would like to go to 12-13 after tune)
GM AIT
Turbosmart BOV


Thanks in advance.

I've also included my MSQ incase someone has a few extra minutes of review time.

Sounds like a nice setup you've got there, should be a fun ride.

Assuming this is on 93 octane, I feel like you could safely add more timing and save yourself some time on the dyno. Then again, I'm use to looking at 9:1 CR timing maps so you may be closer than what I'm thinking. I'd be interested to see some opinions from other NB2 guys.

You know, that turbo would really like 14-15 psi...

Unfortunately there is only 91 Octane here in Maine (at least around where I am). I know the timing is conservative, but in all honesty I understand the AFR’s and the VE fuel tables, but timing just gets me lost. How do I know what is good, what is bad, too much, not enough… I feel dumb when it comes to that.

I know the turbo would LOVE more PSI, but I’m not so sure about the stock MSM engine. 

Add timing until it either knocks or stops gaining power, whichever comes first.

Does these knock setting look correct? I'm having a hard time trusting the knock settings because it goes between %50-%80 at idle and I get the knock warning and it retards timing by 10 degrees. (This goes away as soon as throttle is given)

When tuning the timing and watching for knock, what perectage level is a good reference point to start backing off timing? %30, %50, %70 ???

That part, i don't know. And on top of that, i just ditched an ECU with knock control on my MSM in favor of one that doesn't have it.

I have it turned off at the moment because it isn't working correctly (in my thoughts).

What did you have?

Waht do you have now?

Had a Hydra. Now have a Haltech.

you can add more timing at the last 1000rpm. like 4* more
tuning with knock sensor is a bad idea. use det cans instead.

Am I missing something here? If it is retarding timing in idle, just move the minimum detection up to 1200 rpm. It's right there in the picture you posted.

Heres an updated spark map. I also moved the range up to 1500 RPM for timing retarding during knock.

I could never get away with that much timing int he 180kpa row. I'm interested to see what your tuned table looks like.

After reviewing that I agree thats a bit much. I'll update after the tune. I wish I'd have had more comments on the files before tomorrows tune.

Well?!

Sub'ed
I may cross reference to see if my settings for my MSM are good.
I have forged internals though.

Ok, I was able to get on the dyno for about 2 hrs. I made a headphone det the night before and they worked great and I'm glad I had them. With the spark map above I had a bit of det around 3500 RPM and found a fuel hole in the same area. I was able to run VEAL in "hard" mode and run the car through the gears at different boost levels to smooth my map a bit. I then reduced VEAL to "normal" and did a few more runs. I then was able to run the car in certain areas of RPM/boost that I was concerned with and tune them manually until I was happy with the tune and comfortable with the spark. I still am running a bit rich at WOT (10.5 afr) but will work that out little bit at a time. The end result was 239HP with 218 TQ. Not quite as much as I was hoping for (250HP) but I'd rather be safe then sorry.

Please note that this was done solo by me, the shop I used for the dyno doesn't touch Megasquirt and is a VW/Audi specialized shop. I'm sure I could get the tune a little better but I don't know of any shops anywhere near me that would do that. This was the only dyno shop I know of within a 70 miles radius.


I'll post charts when I can.

It will be interesting to see how much you backed down on that more aggressive spark map you use. My bet is you will be within 1-1.5* of the basemap I provided.

I think my map is mutch closer to the original map that was posted. I used the ms3 knock sensor. I had to pull quite a bit of timing out of my map.

The problem with dynotuning is that you dont really know competancy of the tuner.

Using a dyno to tune fuel is a huge waste of money (unless the car is not street legal).

I don't have the laptop I did the tuning with me at the moment. I'll post both maps for comparison. I used the dyno to be able to tune the spark and upper boost fuel levels. I figured that way a could hold RPM and boost is certain areas I knew we're off a bit to tune them safely and legally (the street isn't a place to tune those areas.

Corrected:

Here is the spark maps to compare. The top map was my starting map from Rev (good map). The bottom map is my tune map.

Note: the boost values are different

I had a map that looked like your final map a while ago. I was told that it looked like I took the spark map from a smallblock chevy. Well, I had created my spark table based on my Mopar 440 pushrod V8. Your map looks like a pushrod V8 default map that was then modified based on your detcans. I don't really know what should be run anymore. It seems many are now saying you can run a lot more timing in the non boost cells. I tuned mine based on my MS3 knock sensor. I pulled timing out of cells that in the past I would have considered absurd. However, when I reduced the timing, the knock spikes went away. I'm mentioning this because I'm surprised at how close my spark table is to your original spark table that was created by a very knowledgeable person.

You may want to run a hybrid of the two. Keep your newly tuned cells, but revert back to the Rev non boosted cells. Your 210kpa from 3100 to 4300 worries me a bit.



https://www.miataturbo.net/attachmen...ine=1370959169

I too think Rev gave me a very good map to start with. I'm having issues with the MS3 knock settings and can’t seem to get any help or advice. I can’t trust the knock sensor because the signal is all over the place (even with the base spark map). It's all over the place and I can't get any real readings because all it shows are spikes from beginning to end.

I did find that in the areas that I was having knock (from Det can) that after I reduced the timing I was very close to the original settings from Rev. I find that in the non-boosted areas of the map more timing was available and seems to run very smooth.

As for the 210kpa area, I'm not really getting to that high of boost to use it, I'm at 11psi at the moment, so I'm only around 180kpa.

djp: Would you mind sharing your knock sensor settings with me for comparison, I must have something set differently than yours? (I'm using the stock MSM sensor)

I'll take a snap shot of my knock sensor settings. I too am using the NB sensor. Post a screenshot of a knock sensor log. I miswired mine initially and it gave me nothing but garbage.

I would lower the timing in your higher kpa ares if you aren't getting there just in case of an overboost situation.

y8s or joe parez has a thread on the knock sensor. I modified my sensitivity settings from there.

I'll also repost my spark map. I don't know why it resized to something stupid small

Thank you. This is a great example of forum collaboration at its best. I'll take a datalog showing my knock sensor garble and post a shot of it.

Here is the chart and knock settings.

1) I don't think you want to look at knock in. There should be options to look at the knock on a per cylinder basis.

2) I hope you deactivated the knock sensor from pulling spark while you were on the dyno. It's set to pull 10 degrees in 1 to three degree increments. The knock thresholds that you are using will most likely need to be scaled higher as rpm increases.

Look for an MLV output like knock cyl1. I forget what the default knock sensor outputs are labeled, but there are four of them.

Start here: https://www.miataturbo.net/megasquir...-thread-67946/

Yes,
knock was disabled during tuning. I'll take a look at per cyl viewing. I know that when activated and hitting the %50 it pulls 10 degrees of timing. Thats where the problem lies, it seems that its hitting the %50 mark anywhere below 2000 RPM and jumps around at RPM's higher than 2000, so I'm having a hard time trusting it.

The good news is that you tuned with the det cans and feel confident that you are not experiencing knock. This will allow you to do a nice long datalog flogging the car to determine your knock window thresholds. Just make sure you set it to pull 0 timing before you start the log.

I actually don't run with mine active. i just use it for tuning.