kevosiroc

OK tuning pros out there. Any advice on my guidelines for tuning the ECM on a dyno? I am tuning using TunerCats RT tuner software and will have a SnapOn scan tool.

The engine will be in a 1990 Vette. Using AXCN bin to start with.

The einge set-up is (infront of a T56 manual):

ZZ4 short block
Edelbrock 8085 heads - ported
Edelbrock lower intake - ported
SLP siamese runners - ported heavily
52 mm throtle body
Crane Cam 2032 cam w/1.6 rollers
No EGR
No smog
10.33:1 compression

First Order of Business (VE tables):

set PE enable to TPS greater than 100% TPS.

set load and RPM to desired increments and tune to 14.7 A/F ratio starting in the lower table first. Smooth out the VE curve and recheck.

Next tune the upper table the same way as the lower.

Second order of business (PE):

set PE vs TPS% back to stock.

Tune the PE% change to A/F vs RPM and the PE% change to A/F vs coolant temp the best I can aiming for 12.5 A/F.


Third order of Business (Spark Advance):

Advance spark at many different lower load vs RPM points until knock retard occurs and back of a few degrees.

Repeat lower point strategy for extended tables.

My last checks:

Go back and check all A/F at many different places to verify the VE curve and Spark advace curve.

Make changes as necessary.





My major question is, what would you guys do to set the multiple accel enrich tables? I am a bit unclear as to what methods to use.

Grumpy

I wouldn't set the PE TPS% to 99.6% for any reason.
Especially when first roughing in a tune.

You can do alot of the 14.7/VE/idle timing/cruise timing, before going to a dyno. Doing that stuff on a dyno isn't going to thermally load the engine like just driving as you would normally.

Get the light load stuff all squared away before even thinking of dyno time, unless you just like spending money. And for the dyno time to set the low load stuff you could buy you own WB, and really know where you are tune wise.

WOT, fuel and timing is absolutely the last things to tune.

And tuning to the limits of detonation ain't the way to go unless you like short engine life, ie racing.

TRAXION

I totally agree with Grumpy. Get all the part throttle cruise stuff worked out before going to the dyno. I would only use the dyno for WOT tuning. That's it. You would need to spend hours and hours of dyno time doing the part throttle tuning stuff ... and that wouldn't even be enough.

I also wouldn't set PE Enable that high until I was sure that my upper VE curve was darn close.

I would drive around town collecting data via datamaster. I would then run VEMaster against it to bring the VE tables in-line. Then I would look at the overal fueling curve in GRAPH form and ... errr ... interpolate those values which appear that they SHOULD change but you basically never got the car into that range while driving around and recording data. I would do this approach a minimum of 3 times.

Tim

proryder2

kinda off the subject but can you use a mallory unilite distibutor with a msd 6a ignition system and still tune the spark advance tables in the computer? or am i wrong?

kevosiroc

Look, with an engine dyno (NOT chassis dyno), I can exactly adjust load and engine speed to exactly what I want. That means that I can hit each point on the VE table and optimize it to 14.7. That will nail down the VE tables 95%.

After further researching spark timing, and talking to my buddy who is an engine tuner for International Trucks (formerly Navistar - International Harvester) I have found the way to really tune the spark tables. The process involves getting a special spark plug that can record cylinger pressure. With this, you then can graph the P-V diagram and calculate the indicated HP. Then you find the max indicated HP and set it to there. Really it is not all that simple, but I do not have time for the whole explanation.

Thermal loading is not really an issue to the engine tune unless you are worried about prototype part thermal fatigue, which anyone in our shoes is not. The thermal conditions of the pistons, heads, coolant and intake air will be damn near correct (close enough not to worry about it). How do you think that automotive manufacturers tune an engine?

Seeing that my Vette is in pieces, no tuning on/in vehicle is possible. I am stuck with Engine Dyno Tuning. The vette will be racing this spring and I need a good tune on the engine that I can perfect over the race season. I am getting a great deal on price and have it for 2 days if need be.

Engine dynos are much more useful than just PE tuning, and I was hoping that someone that has tuned these engines a lot such as Triaxon or Grumpy or someone, could review my plan of attack and make any suggestions. I do not mean any disrespect to anyone and am certainly not trying to flame anyone, but I AM going to the Engine Dyno with my engine to tune it and was not looking for criticism for using an engine dyno.

I am really looking for some constructive criticism on my method for tuning. How would you start tuning an engine on the engine dyno?

11sORbust

First get the a/f ratio lined out THEN move on to spark. AE will need to be done while in car,generally. You might also need to cut back on the spark once the engine has the load of chassis. You'll need to have the engine in the car to tune the part throttle so just try to rough that out, don't try to tune your ve tables while on the dyno. The engine will have different fuel requirements once in the car.

I know this is general stuff. So ask specific questions to get the desired response.

11sORbust

who makes that pug?

kevosiroc

Kistler makes the spark plug. They are not cheap, but have the timing curve dead on can mean added engine life and power output.

kevosiroc

Let me make a correction... the spark plug measures pressure, to record it you need instrumentation that the DIY'er will not have.


About load. Load is load. No matter if from a dyno brake or a chassis. When the computer sees a certain value from the MAP sensor, that is the load. The map sensor does not care if the car weighs 10,000 pounds, or 5 pounds. Now since load is adjustable on the engine dyno, I can get the engine to run at any load and RPM I want. That means I can get to the areas of the VE table that are really hard to on the street.

GM tuned the TPI motors, and all of thier motors for that matter, on an engine dyno way before they even got close to the chassis with it. I am just trying to find a plan of attack to bastardize what GM would have done to tune the engine on thier dynos.

11sORbust

you have tunercat RT so I would just simulate the load and start tuning. I wouldn't go crazy and try to hit load and rpms that never/hardly get used on the street. Even though you can simulate load I still think you'll need some street time to get the tune nailed......(as you may know)

Good luck.

Doctor J

That sounds like an interesting project. I'll bet it keeps you busy for the full 2 days.
The first part of the testing sounds fine.

The goal of the AE/PE testing is generally to achieve the best change in crankshaft
angular velocity per unit of time, while imposing a constant load on the engine
(representing constant vehicle mass), IMHO.

I put a few modest thoughts on the subject here: http://www.corvetteforum.net/c4/doctorj/cartest.htm

I'd familiarize myself with the instrumentation available at the shop, and set up my
spreadsheets (as much as possible) before testing. That will maximize time available
for actual data collection, once you are bolted down & ready to run.

Have fun.

Grumpy

A dyno room (and even a chassis dyno [typically]) is not like being in an engine compartment, so yes the thermal loading is different.
You'll probably wind up with too much timing, and probably not enough fuel.

I've previously posted about dyno *errors/conciderations, etc* a search might find that posting for you.

Dave_Jones

When I read your initial post, I also assumed you were referring to a chassis dyno. Yes, I know the title was "Engine Dyno Tuning" but commonly when people discuss "tuning their engine on a dyno", they're often thinking of a chassis dyno, just because those are way more common. So please excuse any misunderstandings!

I would absolutely love to have unrestricted access to an engine dyno -- I'm drooling, just thinking of the possibilities <g>

Now, as to the original question, of testing methodology, I'm not smart enough to suggest the most efficient sequence, BUT it seems to me, that if you can walk away from the dyno session knowing the optimum SA and INJ PW for load and RPM, that you would basically have what you need, since you could then reverse engineer the rest of the tune (and most of that work could be done on an ECM bench, with no wear to the engine) so as to have the net result of the various corrections end up at that known fuel & spark target. Especially if you're paying for dyno time by the hour, it'd be a lot less expensive that way. Make sense?

kevosiroc

Well, the small difference I will experience in thermal loading will be easily corrected after I get the tuning 90% done unsing the dyno.

In no way do I think the engine dyno is the start and end of tuning, I just think it is the right place to start. It will get the tuning very close. It will also save me lots of time and chip burning.

My time constraints are the driving force for the engine dyno. I need to be able to drive the car and race it ASAP. The engine dyno will get me close enough on tuning that I will be safe. Time will make it perfect, but the engine dyno tuning will keep my engine from burning up.

Now, does anyone have any suggestions that I may do while at the engine dyno (based on my outline)????

Dave_Jones

Well, as far as your original outline, I would think it better to work SA more in conjunction with the base VE, before moving on to AE and PE.

kevosiroc

Yeah, I was kind of planning SA after the VE tables.

Ed Maher

Thing is, you're not talking about a radical combo, those heads are basically a wash with the stock AL vette heads and the comp 2032 is IIRC a smallish grind. Such a combo will no doubt run and drive just about fine on a completely stock chip.

If i were in your shoes, i guess i would really concentrate on WOT and heavy load high RPM stuff. All the light load and low RPM stuff should be plenty close enough as is and could easily be completely dialed in less than an afternoon with VEphd. You could try checking a few points in the ligher load regions, but i'd think you'd quickly bury yourself in data if you tryed to get comprehensive, and then you'd still have to do final tweaking on the road anyway.

kevosiroc

Ed,

I forgot to mention the 1.6 roller rockers. The lift is .482 intake and .496 exhaust with the rockers. duration at .050 is 214/220. I think the cam is a good step up from stock.

Also, the heads have been ported quit a bit.They should be flowing pretty darn good now.

I do not htink the stock chip will run this thing well at all, do you?

kevosiroc

Ed,

I forgot to mention the 1.6 roller rockers. The lift is .482 intake and .496 exhaust with the rockers. duration at .050 is 214/220. I think the cam is a good step up from stock.

Also, the heads have been ported quit a bit.They should be flowing pretty darn good now.

I do not htink the stock chip will run this thing well at all, do you?

Ed Maher

Yes, i do think 214 / 220 is pretty tame for a 350 with what started as 170cc intake runner heads. The stock cam would run that thing acceptably enough that 5-6 years ago you would have done that and tuned it with a FPR and been done with it and been pretty happy. Ask Traxion, he ran the 218 / 224 comp and AFR 190s with the TPIS chip which turned out to have basically stock tables. Wasn't optimal, but he ran it on purpose because thats what people did. Then chip burning came along...

I mean, you could definitely dial the idle VE in, which would be by far the biggest issue, and on a small cam like that that will barely lope if at all. Probably just have to drop 10% or so there, which is still in eth range of the stock BLM adjustment. Cruise range and such will also be well within BLM adjustment range, in other words normal deviation from stock built into the computer. It won't flood the engine or anything weird like that, not even close. And it literally takes maybe 3 20-30 minute drives with VE phd to dial in cruise VE. And your WOT stuff will basically all be above what you'd care about for part throttle VE so those adjustments are almost independent as long as you think straight.

JPrevost

The cam alone will move your VE peak up enough that you'll want to change it in the chip.
I don't (hope) I'm not the only one that's used an engine dyno but from my experience it's best to be all ready for data aquisition and all tools ready. Get completely familiar with the tools and have a precise plan of attach.
I don't have much (1 engine) tuning time with a v8 and a water break dyno (which I believe you're using) but try and get EGT on the headers, at least one port should have a port to keep your engine in it's thermal limits.
Just this past weekend I through our little Honda F4i 600cc backup motor on a DC electric dyno and had it purring at 12000rpm... lots of fun. I use National Instruments Data aq and our engine runs off of a pe-ltd ecm that allows us to tune real-time. In other words you're setup is very similar. Other than the spark plug pressure transducer/sensor.
I honestly feel the pressure transducer is nothing special unless you are planning on spending days and weeks tuning an engine. The only time using one would be if you DON'T have access to dyno and want to derive torque... but you should be getting that information so what else do you need .
Here is how I plan on doing my motor in a few months;
Chunk my tuning into area's/quadrants. Use my current calibration with 2 degrees less timing as a baseline. Since the ecm does interpolation I'll be tuning cells around cells to keep from developing divots in the maps. I'll disable PE and go to points and tune in stoich. First VE then timing, then back to VE, then back to timing, next cell, etc. After getting the ruff in I'll enable PE and do some dynamic pulls watching the AFR and torque. One key idea is to use the least amount of timing required to get the same torque! Actually I shouldn't say that, it isn't the LEAST amount but it should be a medien between the SA where your entine makes no torque and when it pings. With a little Honda F4i motor it's amazing how much advance it can take and the window is smaller than some of you would ever be able to imagine. I'll go from making gobs of power to like 3/4 power in just 2 degrees too retarded, and the little motor will let you know it's unhappy, the big v8 should be less sensative so watch the torque closely when you tune your SA.
Once you're all done ruffing it in and feeling good, save your calibration as "stoich" and then go back and start trying to run leaner in cruise but look at the torque and then you might even want to disable PE AFR and just tune in open loop under the 100kpa areas for a richer AFR. This is what I plan on doing after I get stoich. I'll be leaning things out to an acceptable AFR of like 16:1 and watching the thermal limits while recording torque. I have other research I plan on doing but what I've just told you is what I plan on doing the first day it gets running.
How much time do you have on the engine dyno? If you have all day and you want to do some research you might look into changing your oil to a lighter weight synthetic and/or trying lower octane fuel, or higher octane fuel (just don't use leaded).
Most importantly (other than having fun) is to post back here with your dyno charts and calibration information .