Seriously Contemplating a Haltech

[eHoward]

We could go back and forth about this all year Will.

I have zero doubt you could tune an engine by yourself. And if it happened to break(and that's a real possiblity) it wouldn't even be that big a setback for you.

The question is, could a mere mortal do it AND get it right the first time, and my experience has told me no. That's why I would hire a professional to tune a turbo motor if it was mine.


btw, i don't even know what LTFT's are and I like to think I'm knowledgable.

Thought so (F4's)

You don't guess what it needs and where it needs it... that's what LTFT's tell you... where and how much fuel is needed.

Timing's a little different.

[The Dark Side of Will]

Long Term Fuel Trim
The premise is that although one measurement of AFR via narrow band O2 sensor is not necessarily accurate, the statistical average of thousands is accurate.

The integrator (STFT--Short Term Fuel Trim) subtracts stoich voltage from O2 sensor signal, then integrates the fluctuating result. It trims the fuel rich or lean to keep the O2 signal integral at zero. This is how the fuel delivery is dithered between rich and lean.

The LTFT performs that same operation on the STFT. If the STFT starts to build up a non-zero integral over a period of time, the LTFT adjusts to zero out the integral of STFT.

So if your fuel map isn't dialed in, drive around for a while datalogging and look at your long term fuel trims when you get home. If the LTFT is adding 8% fuel to a certain part of the map... bump the VE values in that part of the map up by 8%, smooth the surrounding area and go for another drive. When you start to get to high manifold pressures, bump the TPS % for PE up so that you're not running in PE and keep going. Eventually you get everything but WOT tuned... and then you go to the dyno for an hour, maybe two.

This is OEM techology from twenty years ago. VERY few aftermarket setups even today have this algorithm.

GM originally devised the LTFT arrangement to compensate for changes in temperature, humidity, air density and gasoline specific gravity from day to day and season to season. It happens to make a very good built in tuning aid when the computer is remapped, though.

Howard, your experience has mostly been with aftermarket computers and your comments reflect that. I'm not knocking what you have to say. You just don't seem to be familiar with the DIY community's tuning methods on factory computers. If Shaun did the research on www.diy-efi.org and to some extent the computer section of TGO, then he'd also ask why he needed a buttload of professional dyno time...

[whipped]

I've had some experience with both aftermarket and oem computers. I can tell you right off the bat that most aftermarket computers are shit. Not dilluted shit either, pure, rancid, green shit with crayon bits. The kind you see in a toddler's diaper.

Example:
Holley commander 950, perhaps the most widely used aftermarket computers, has a little problem that holley won't tell you about. Alternator noise can erase your memory. Not just the maps and all that, but the actual memmory it uses to compute timing and shit. I've seen it happen in my buddy's honda (don't ask). You have to send the computer back to holley to get programmed.

On top of that, the factory computers are by FAR more advanced than any aftermarket ecm. I could go into details, but I'm late.

[Shaun41178(2)]

so ther eis one problem with the holley that you mention and you say thats shit? So what other major problems are there?

[The Dark Side of Will]

On top of that, the factory computers are by FAR more advanced than any aftermarket ecm. I could go into details, but I'm late.

Yeah... aftermarket computers don't have the fuel trim algorithms. I think that the latest gen Accel DFI might have a similar autotune feature, but I'm pretty sure that requires a wide band and can't work with a narrow band like the OEM algorithms.

[The Dark Side of Will]

so ther eis one problem with the holley that you mention and you say thats shit? So what other major problems are there?

Well dude... having the computer's memory completely wiped and having to send it back to Holley to have the firmware reinstalled is a pretty big problem, IMNSHO.
I'd like to know how often this occurs and how bad alternator noise has to be in order to cause this.
Dirty power can wreak absolute havoc with ANY electronic device... so it's not like this is special to the Holley... it just may have lower tolerances for it than other systems.

Also, as we've just discussed, aftermarket computers don't have fuel trims that allow you to tune the thing yourself. Bummer...

whipped: What was up with your friend's alternator?

[Shaun41178(2)]

I dont' see how the stock stuff can work with a narrowband either. I mean how do you program something with a narrowband and know you are getting a true and accurate reading? You can't

I am sure gm figures out all hte tuning by using a wideband, burns the computer with those algorithms mass prodcues it based on their tests and then throws in a narrowband o2 to help keep it near stoich at idle and part throttle.

hell maybe I am wrong.

I dont' know I alwasy here of all these people that supposedly use these programs and such on their own cars from third persons but I have yet to meet one person who went out and bought all their own stuff and emulators and everythign and sat down and did all this stuff.

I mean it can't be as simple as you make it out to sound otehrwise more people would be doing it instead of jumping to a stand alone.

[eHoward]

Will, the AEM offers autotune with a wideband if you have one. I hear autronic does sophisticated self-tuning also. I have seen the AEM do it and it's impressive.

I think you are oversimplifiying things. What you are talking about doing is far from a FULL tune. Furthermore, WOT is pretty much the biggest concern for someone like shaun.

And you don't mention monitoring exhaust gas temps? What about wide 02 numbers? Spark plugs need to be checked after each run. And each of thier relationships to fuel and spark maps.

About shitty aftermarket units, I've been there. Electromotive Tec 2 = Garbage. Haltech experience was A-OK.

Hey, I will admit my use of OEM computers has been fine. We used the Power Comanader in 2003 which is a piggy back computer for the formula car and it worked great. I also used a hacked 85GT ECM (i know you know this) with miller woods code scaled for bigger injectors and it didn't kill anything.

I'm sorry, but I've seen too many engines break. That's what it comes down to. To me, this isn't about the hardware, but who is tuning it. And I know of exactly zero dyno operators/tuners that specialize in tunercat.

[The Dark Side of Will]

Will, the AEM offers autotune with a wideband if you have one. I hear autronic does sophisticated self-tuning also. I have seen the AEM do it and it's impressive.

I think you are oversimplifiying things. What you are talking about doing is far from a FULL tune. Furthermore, WOT is pretty much the biggest concern for someone like shaun.

And you don't mention monitoring exhaust gas temps? What about wide 02 numbers? Spark plugs need to be checked after each run. And each of thier relationships to fuel and spark maps.

About shitty aftermarket units, I've been there. Electromotive Tec 2 = Garbage. Haltech experience was A-OK.

Hey, I will admit my use of OEM computers has been fine. We used the Power Comanader in 2003 which is a piggy back computer for the formula car and it worked great. I also used a hacked 85GT ECM (i know you know this) with miller woods code scaled for bigger injectors and it didn't kill anything.

I'm sorry, but I've seen too many engines break. That's what it comes down to. To me, this isn't about the hardware, but who is tuning it. And I know of exactly zero dyno operators/tuners that specialize in tunercat.

I'm sure it is impressive to see a computer tune itself... That would take the human out of the loop and save a whole lot of time and mental interpolation.
What I'm talking about is the OEM computer giving you valid tuning info based on the feedback of a narrow band sensor. It won't tune itself... mostly because it can't write to program memory... you still have to make the changes to the map manually, but to get the same information from a narrow band as the aftermarket systems do from a wide band is pretty sophisticated.

I know I'm not talking about a full tune. I'm just talking about getting the VE maps right. There are timing maps, idle parameters, AE parameters, PE parameters....
You've got to mess with ALL of it when doing a full tune... It's just that you've got somewhere to start when working from a factory calibration. That's why EGT monitoring isn't an absolute must.
Yes, plugs ought to be read when fine-tuning the engine.

WOT is nice, but for a car to drive, part throttle is WAY more important. For instance, with the 275 chip a 300 HP Northstar has a nasty idle surge. I spent a lot more time hating that idle surge than I did with my foot on the floor...

I think that the Haltech is crap. Part of that may just be lack of familiarity with it. A local guy in P'cola screwed around with an FD3 with Haltech and took weeks fixing a problem that would have been all too obvious with an OEM computer. There was an intermittent wiring fault in his CTS circuit... the kind of thing that would have popped a trouble code and triggered a compensation routine on an OEM computer... only way to tell on the Haltech was to notice that the coolant temp was way off... Which was difficult because the diagnostic program couldnt' communicate with the ECM most of the time... and all this while the car was undriveable.

As whipped said... the OEM computers are immensely more sophisticated than aftermarket units...

No, most dyno operators don't know how to use specific DIY software... that's 'cause they never see it because the DIY guys are doing just that... doing it themselves...

[The Dark Side of Will]

I dont' see how the stock stuff can work with a narrowband either. I mean how do you program something with a narrowband and know you are getting a true and accurate reading? You can't

I am sure gm figures out all hte tuning by using a wideband, burns the computer with those algorithms mass prodcues it based on their tests and then throws in a narrowband o2 to help keep it near stoich at idle and part throttle.

hell maybe I am wrong.

I dont' know I alwasy here of all these people that supposedly use these programs and such on their own cars from third persons but I have yet to meet one person who went out and bought all their own stuff and emulators and everythign and sat down and did all this stuff.

I mean it can't be as simple as you make it out to sound otehrwise more people would be doing it instead of jumping to a stand alone.

The OEM systems don't base fuel trims on one reading... they take thousands and put statistics on their side. Also... there's one point at which a narrow band O2 is accurate: Stoich. And really, that's all you need.

GM obviously uses fully instrumented dyno cells to figure out their initial calibrations... but in that case they're operating in Howard's world: filling in a map full of zeroes. You wouldn't work that way because you have a factory program from which to start... this program is already mostly tuned and pretty much just needs VE and timing adjustments.

I am simplifying to some extent. However, the tuning operation is the same with OEM or aftermarket computers. The OEM just feeds you the information you need, rather than having to go to a dyno like you would with an aftermarket computer.

As far as coming from ground zero to tuning a GM computer? I haven't done that yet, but I'm walking my friend Steven through it. He's building a TT SBC Camaro (1990) with 7730 computer using $58 code (Syclone). he's going to be into the engine management for less than $500 but will have finer control of engine operation than he would with a Haltech... and he won't need a boost controller.

[whipped]

whipped: What was up with your friend's alternator?

Nothing. It was a brand new alternator. Holley just sorta shrugged, and said it sometimes happens. I guess it's more likely to happen if you run the power to the fusebox, and not through a relay directly to the battery. Or something. even though NONE OF THAT IS MENTIONED IN THE MANUAL.

Now riddle me this: can you call even one SINGLE INSTANCE when something like this has happend with a factory computer? I don't care if it was a friend of a friend of a cousin of a girlfriend's ex prison lover's car.

shaun: this wasn't the only problem we had with the holley either. Even after we figured out this problem, which was a huge problem, I don't care what holley says, we still had problems with the idle. ANy time there was a change in load, like flip on the headlights, or push in the clutch, and it would stall. We asw the datalogs, and it was closing the IAC when it should've been opening it. honestly, wtf holley?

my friend and I still don't know how to get that fixed. thank god it's not my car, I probably would've taken it off a cliff.

And just fyi, it's not like there was a bad throttle sensor or maf sensor or anything, everything was brand new on that engine. The datalogs showed everything operating correctly, except the holley. Maybe there's a box I need to check that says "open idle control when stall detected" :cussing:

Just for the record, I have nothing against haltech, they're probably better than holley.

[The Dark Side of Will]

The failure with the Holley involves the wiring to some extent, as well. Was it wired just like the factory computer?

Are you sure you had the IAC wired straight? Did you try inverting it? What was the spark advance doing when it would stall?

[Indy]

You may want to take a look at some of the advances that MegaSquirt has made in the past year. It's becoming a VERY viable system. (There are EGO autotune programs out there that will tweak the VE table(s))

It can now control a GM HEI system, or Ford EDIS, with compatibility with other systems in progress.

There's also support for three VE maps, two ignition maps, injector staging, bla bla bla...

Just look at the -Extra firmware. Keep in mind that's just for MegaSquirt 1. MS2 was just released for sale.

http://www.msefi.com

Not for everyone, but hey, it's there.

[whipped]

The failure with the Holley involves the wiring to some extent, as well. Was it wired just like the factory computer?

Are you sure you had the IAC wired straight? Did you try inverting it? What was the spark advance doing when it would stall?

It was wired just like the original computer. And yes, the iac was wired straight. The ecm wanted the iac closed: the datalogs showed the iac closing and it was stalling. If we flipped the wires around, I don't think it would've been able to control idle, it would've shot through the roof after startup.

I don't remember what the advance was doing, but we had a pretty flat curve, it should've been steady around 10-15 degrees at idle. don't remember.

For what it's worth, if he were to do it again, he'd go with megasquirt. Mostly the same features as the holley with 1/5 the cost, if you don't mind soldering of course. Right now though, I'm having a hardtime trusting any aftermarket computer.

oh: we did figure out how to stop it from stalling. we just blocked the throttle open enough so that the engine still runs when the iac is fully closed. It's not a great solution, because then the idle's high until it warms up, but it works.

[THE PUNISHER]

You should get a Haltech , it would be sokick ass sitting on your shelf of ""Things to do"" I can see the dust collect on it now.. along with the aluminum block

You are Teh WeakSauce.

[Nemesis]

You should get a Haltech , it would be sokick ass sitting on your shelf of ""Things to do"" I can see the dust collect on it now.. along with the aluminum block

You are Teh WeakSauce.

No shit. Just wrap that aluminum block and send it my way and bill me.

[Shaun41178(2)]

You should get a Haltech , it would be sokick ass sitting on your shelf of ""Things to do"" I can see the dust collect on it now.. along with the aluminum block

You are Teh WeakSauce.

:sad2: :whip:

[The Dark Side of Will]

The failure with the Holley involves the wiring to some extent, as well. Was it wired just like the factory computer?

Are you sure you had the IAC wired straight? Did you try inverting it? What was the spark advance doing when it would stall?

It was wired just like the original computer. And yes, the iac was wired straight. The ecm wanted the iac closed: the datalogs showed the iac closing and it was stalling. If we flipped the wires around, I don't think it would've been able to control idle, it would've shot through the roof after startup.

I don't remember what the advance was doing, but we had a pretty flat curve, it should've been steady around 10-15 degrees at idle. don't remember.

For what it's worth, if he were to do it again, he'd go with megasquirt. Mostly the same features as the holley with 1/5 the cost, if you don't mind soldering of course. Right now though, I'm having a hardtime trusting any aftermarket computer.

oh: we did figure out how to stop it from stalling. we just blocked the throttle open enough so that the engine still runs when the iac is fully closed. It's not a great solution, because then the idle's high until it warms up, but it works.

Might try advancing the timing by 5-7 degrees just below idle RPM. That way if the RPM falls and it starts to stall the timing will advance and generate more torque to bring the RPM back up.

Can't you change the cold idle settings to make it idle normally even when cold?
n/m... it's a Holley.. my fault...