So obviously there are a lot of really good turbo LS tuners our there and a lot of really bad ones. I've been thinking more and more about doing my own turbo tune on the street and then just doing a couple hours on the dyno as most local guys in the Vancouver area want $1000 for a "forced induction" tune but have never seen a microsquirt, never the less a turbo LS; they're mostly honda guys. My last tune was done by a really reputable local guy, but he never even adjusted my AFR table from what I had! I didn't even know at that point in my life the importance of properly matched AFR/VE tables.

My basic thought process for basing my tune in would essentially be to setup my AFR table to acceptable values and pull all my timing in boost more or less. I'd also make sure to pull timing when IAT's get higher (and also display warnings on the shadow dash app which should be possible, haven't looked too far into it yet). I'd also do my best to incorporate some overboost protection in tuner studio. Then it seems like it's a matter of adding timing and watching power levels, listening to knock, maybe kill a bit of timing at peak torque. Is that the basic process?

So what do those really good LS tuners do that the bad ones don't? Is it simply the fact that a lot of guys just floor your car on the dyno and watch AFR's then send it home? Is it all in the drivability tuning? Or is there way more I'm missing here? I'm referencing tuning a standalone too for what its worth, not a GM ecu.

 

Start, and drive ability is where a good tuner shines. WOT fueling isn't too tough, but start, drive ability, ign curve, and cyl pressure management takes plenty of time to get right, and is also where a good tuners experience comes in.

 

AMEN x1,000 with the above! I tell people all the time and I don't htink they truly believe me, but I honestly think I spend 90% of my time tuning getting the perfect startup and idle. If you can get that part dialed in pretty close, then let Autotune get the driving VE pretty close. Then go in and clean that up manually. SEt your timing table so it runs pretty good WOT and drives great. Then go to dyno for WOT tuning.

Last car I dyno'd I picked up 4hp. FOUR HP on the dyno vs my street tuning.

Now, don't knock guys that are "Honda tuners". Tuning is more a knowledge base than a vehicle base. Once you know how to tune properly, you can tune anything with any system. Yes, there is a learning curve with every system, but Tunerstudio aint one of them. If you have a local Honda tuner that is lights out good with Hondas and is willing to tune your car, you're probably going to be OK. I personally feel I can tune anything. I do tune a lot of Hondas since they are so common and the owners aren't generally interested in tuning themselves.

The part i'd caution you on is still startup and idle. 99% of dyno shops don't do that. Even ones that will say they tune idle only tune it for warm idle in their shop. They don't tune for cold, warm, hot, lights on, in gear, AC on, etc. etc. etc. That's the hard part but most rewarding.

 

^all good points

Additionally, the tune that makes the most power may not be the "best" tune. Especially with spark timing.

Random number example:

510 hp @ 36* vs. 500 hp @ 31*

I will take the 500 @ 31 everyday of the week. The slight increase in power is not worth the much greater increase in cylinder pressure. Particularly in a street/endurance application. How much of an increase in pressure per degree is overwhelmingly dependent on the specific application.

Look up pressure vs. crank angle vs. spark advance diagrams......also look up MBT spark


Ive also seen "good tuners" (sarcasm) not increase spark timing at lower MAP for fear of damaging the engine. Additional timing is NEEDED at lower MAP due to the slower flame travel caused by the decrease in charge density and leaner A/F ratio.....how much is again application specific....but baseline rules of thumb are easy to find.

Cold start can be the most frustrating.....because you only get about one shot at it a day.

 

The knowledge to manage cyl pressure at a given rpm is also very important on most of these sbe builds. Just because it's makes good torque, and the a/f, and timing are "safe" doesn't mean it's a good tune for the build. I'll also add that the issue with ign tuning without a Dyno or track time is that you're relying solely on plugs, which is tough, and very time consuming.

 

Twice a day, AT BEST. Even then a morning cold start can act differently than an evening cold start. Getting all these little things dialed in is the hard part of tuning. It just takes a lot of time.

The car i'm doing now (79 Coupe 302/78mm/MS2) I've already emptied the fuel cell (10 gal) once and the car hasn't left the lift yet. All just startup and idle tuning. I'm just now as of yesterday at a point where i'm confident in it's idle enough to get out and drive it without fear of it randomly stalling in front of a dump truck.

 

Thank you all for your excellent replies. I understand the basics of cylinder pressure and totally understand how spark advance is changing your centre of pressure. From a practical tuning perspective however how could one ever measure cylinder pressure or even determine if it's high or low? Does that come straight from the dyno whp number?

 

Through torque output.

 

So here is my two bits, (weather you want it or not) its long so read it if you want and feel free to put as much stock in it as you like.

I think the difference between a good tune and one that does not grenade is pretty strait forward. We all know what a good one is, we have all driven good running, stock applications. We all know how a early 90's honda runs. Its transparent, it starts hot and cold, drives smoothly in and out of traffic regardless of gear, speed, or accessory load. It idles hot and cold without missing a beat, passes emissions and returns decent economy. I think that's pretty much publicly accepted and agreed upon as being "good"

But I also think there is a lot of sides to this question, especially once you start talking a custom engine. A lot of factors play into it. Expecting a guy with little to no familiarity with your random combination of a parts to hit the nail on the head within few hours, when the above Honda example had 78468374638756387 engineering hours and $50968705783840987 dollars invested into the calibration is just silly. No other word for it.

Talking terms of megasquirt or microsquirt as that is what you are going to use on your Camaro, and assuming the mechanical, electrical, and ancillary systems are all correct (which from following your build it certainly looks like they are going to be) what I would call a "good" tune would go through a lot of steps.

First and most importantly is making sure the engine constants and settings are all correct, that the ignition input and output setting are right and working properly and the ecu knows where the moving bits inside the engine are. This is also the time to decide how you are going to tune this thing. Are you going A/N, speed density, Maf? Are you going to incorporate afr? (I recommend it) Real time baro? How big of a table do you need? What other features and extras are you going to use? Knowing this, make sure you have good injector data, while its not as critical on a MS as on say a factory ecu, it still will help with the tuning experience and help limit the amount of "correction" that has to be fucked into the VE table to make target afr match actual afr. Especially at lower rpm and loads.

Get the rest of the sensors, tps map, baro (if used) ect, act wideband, and so forth, all working and properly calibrated. With out good inputs you can pound the keyboard till your blue in the face and it will still not run right.

Set your afr table, ve table, and timing table up with good break points for your given engine combo both in rpm and load/map,
Once all the foundation stuff is laid, pop the the engine off, and get it to idle well enough to at least stay running.
I do not spend much time on the idle yet. As almost everything is based off the VE table, it needs to be correct first. Otherwise its just tail chasing.

After you get it running long enough to make sure all the oil and water stay in the engine, and the rest of the car is sound enough, you can either start roughing in a ve table on the street or dyno it. I recommend doing it on a dyno with a load cell and good accurate real time torque output. Street tuning can work just fine. It just takes a lot longer.

In the case of the dyno, set the gate up to run just spring pressure, turn off all your transient fueling (accel) and make sure you are out of any warm up or afterstart enrichments. You are also going to want to enter in a timing table that is lower then where you know you are going to end up. Set your steady state speed to get you into the first column over idle in an accessible gear, say 1500 rpm and 3rd gear. The dyno will hold you in that rpm column, adjust the ve table until your measured afr matches your target afr for the lowest cell you can get to, then apply a bit more throttle and move up to the next cell and repeat. Do this until you are at wot. After that column is done and all the cells match the target afr, you can copy and paste the entire column into the next 2 rpm columns, maybe 2000 and 2500 and then add 15% or so to them.

Bump the steady state dyno speed to the next rpm column, in this case 2k and do the same process here. Then once done copy and paste it to the next two columns. and add 15% or so to them as before. This will make sure that you do not end up in a situation where a lower or higher number in the next cell over skews your tuning and should be adding enough fuel to be too rich. You want to be pulling fuel if at all possible.

At this point you can either carry on until the cooling system cries uncle and needs cooling off, or you can bump back down a column and if the dyno has good accurate real time torque output, repeat the steady state load in each rpm column that has been already tuned for afr while adjusting the timing up until you get best torque.

Work your way across the ve table and timing tables until you are happy with how the car is running. Afterward its a good idea to revisit the ve table as changes in timing will have an effect on measured afr.

Once you've done as much of the ve table as you can, and you are still running gate pressure, you can start making a wot pulls while data logging.Once again, using the logged data adjust those cells until your measured afr equals your target, copy and paste into the next two rows above and add %15 or so. I recommend pulling at least a couple plugs between pulls and looking for tell tales. Bump the boost pressure up to the next row and repeat until you have made the power you are looking for, or reach a limitation of some sort. Sometimes where you stop is an arbitrary boost number, or power number or rpm, maybe injector duty cycle or clutch, its not necessarily the "max"

After all that's done you can turn on all the accel enrichment stuff and adjust any hesitation or hiccup out of the car during transient throttle. Its also when I will usually "grade" out the ve tables edges where I did not get to during tuning. This makes the hole table transition smoothly from the cells you tuned to the ones tyou did not get to. Doing this will help prevent any unexprected drivability symptoms that crop up if for some reason the engine ends up in some part of the table you couldn't reach on the dyno. I usually find this to be on decal but sometimes due to any number of reasons you might make an extra pound of boost and you do not want the car to do anything weird.

Next I like to disconnect the alternator and let the voltage fall while running steady state so I can check my battery compensation values for the injectors. You want the measured afr to continue to match the target even while the battery is going dead. This is not a critical step, but its easy to do and shows attention to detail

Now go ahead and spend some time on idle. With the area above idle on the ve table sorted out, and assuming you have decent injector data, you will have a good idea what values the engine will want to run at. Especially if you incorporate afr target.

Now that the car runs well warm and does what it should, all the optional stuff like two step, boost builder, rev limits, shift lights, idle adder for A/C, ect can get sorted out and tested.
Hot start is next as the engine is already warm. Adjust idle valve position and inj PW until you get a clean restart every time. Don't be afraid to extrapolate what you think you need as the engine temp drops. A lot of the default "curves" in tuner studio are pretty close, they usually just need some global up and down changes to the entire curve to get pretty close. At this point pull the car off the dyno and go drive it while logging. I like to set my phone on voice record and record my "bitch list" as I drive. Just the little things I notice that should be fixed. After I'm happy with the test drive I park the car and let it cool off until the end of the day. Right before heading in or going home I do my first "sorta cold" start. Log the start and see how it does. Again make some adjustments if needed before letting it cool overnight. Next morning log your cold start and adjust accordingly. Normally its at this point I treat the car like its mine. Leave the laptop in the passenger seat and anytime I get a few free minutes, take it for a drive and keep knocking down the bitch list. Most of the time this really only takes a couple of days, but depending on how radical the combo is and what the goals are, it can take longer. If you are running real time baro, and live where altitude changes are going to happen, its a good idea to find a hill and set the baro correction table as you log afr while the altitude climbs.

There are a bunch of other things to consider depending on what the goals are, but I consider this to be a good start towards what I would call a "good" tune file.

 

Fastspec2, thank you so much! That's one of the most informative posts I've seen and your methodology seems sound!

I'm totally going to try and tackle a solid base map in my car. If I dont feel it's going well I'll throw in the towel. I actually based my car in NA and it was really easy (seemingly). My only concern with my new engine is I also have to break it in. I'm a not scared because I'm going to need to get the na tune done pretty quick so I can break in the engine properly without too much idle time. Any suggestions on that?

 

And that makes sense, not sure why I was thinking hp output and not torque. So could I generalize and say the ign advance curve should be more or less an inverse of the torque curve? Ie lose advance at peak torque and start to add advance after (for an na application). Then for a boosted application drop timing at peak torque and probably not add much after due to increasing cyl pressures from boost?

I actually ordered the high performance fuel Injection book to read let's hope it helps me out. Thanks again

 

I'm not sure if i understand a hundred percent what you are asking, but it sounds like you are trying to get it fired and driving asap. Am i right? If that is the case, just make sure your constants are correct to start. You can look at another running file from someone to double check those. I have a few if you want to pm me an e-mail. Once you know that stuff is correct, go ahead and start it. I find the LS stuff all wants to idle pretty easy and stay running unless it has a BIG camshaft. Once running baby-sit it as you walk the rpm around in neutral or whatever your break in procedure involves. Assuming you want to break it in under load, I.E. driving it, once its up to temp you can cheat and just turn on real time VE correction. Set the cell change to easy and go drive it around. You will watch the VE table bouncing around but it will clean up pretty quick and you'll be able to put some mild to moderate load on it before you know it. There are some filters for the real time VE that you will want to make sure you are clear of. Things like load, ect, rpm limits ect. Just make sure you do not get any little yellow flagged messages in the bottom-ish right side of the screen.
I am not a huge fan of VE analyzer (I know alot of guys are) especially when you are trying to learn/figure this stuff out, but for times like this it works pretty well.

 

I think it should be, bad tune, good tune, wrung out tune. Wrung out is a tune that is been tuned in the actual environment it is intended for using data , reading plugs, and paying attention to the characteristic how the motor responds.

 

What about closed loop turbo controllers vs. open loop.

I'm still kinda in the early learning curve.

 

Kinda depends on what you are trying to do.
Depending on the system it can take alot of time to get the P.I.D. controller tuned. For most street applications open loop will work fine. It also depends on how its plumbed and setup as to weather or not you have "boost bleed" before target pressure.

I think Tbird hit on something that i left out and that is tuning is never really done. You can always fuss with it. I have cars I've tuned years ago that I have not so much as hooked the laptop up to since, and i have cars i won't drive around the block without the logger running. Just depends.
Even OEM's are still tuning cars after you buy them. You take it in for an oil change and there is a ecu flash update for something. Might just be emissions might be something else. Point is even after 327845683725469287 engineering hours they are still making changes. So if you are paying someone to tune your car, keep that in mind.

 

Thanks again for the detailed responses!

Ya I've got a brand new engine and need to break in the rings. I'm going to use one of those tunes you gave me and make sure it's not running pig rich. After a minute if all gauges look good I taking to the street with the charge pipe off and do a quick and dirty break in. I've used the auto tune function before and it's good enough for this I think. Once the rings are seated that charge pipes going in and its game on!

 

Leave the charge pipes on. Just pull the gate spring. You don't want to overspeed.