Ford Focus Forum banner

1 - 16 of 16 Posts

·
Registered
Joined
·
11,123 Posts
Discussion Starter · #1 ·
Somebody sent me an email yesterday, they have a Superformance S1 running with a Zetec and quad throttle bodies, using a Pectel in Alpha N mode. Superformance doesn't know how to tune the Pectel, and he was bringing it to another tuner. He wanted some advice on tuning the Pectel, and I ended up writing a huge thing again. It has a lot of good info, and I wanted to share here.

I'll take questions if anybody wants to ask some tough questions, but please don't ask things like "what's a Pectel".


Well, looks like you've got a big job ahead of you.

Who have you been working with at Pectel? The guys in California? I have to caution you, that myself and others have found they really don't know as much as you would think dealers should. I had a problem with my basic Zetec Turbo setup, for which they claimed the T2 was a "plug and play" setup. I had tuning issues, and they couldn't even begin to help me out. I ended up contacting Pectel in the UK, and it turns out I uncovered some problems with the software. I believe I was the first one to even try to use the two "modules". It's been about 10 months now, and still no fix. The modules aren't likely to be causing your problem, and you can run with them disabled, as most people are. That's what I'm doing now as I await a fix.

As for RPM-motor-sports, who referred you to them? I looked at the web page, and didn't really see anything that impressed me. Do you know somebody that has used them? I mean, maybe they are good, but I don't really see any customer cars that show they know how to tune an engine, or anything like that.

There are a lot of shops out there that really don't know how to tune engines. They usually manage to get them running well enough to get down the 1/4 mile at full throttle, but really don't know enough to make a car run right at part throttle and different weather conditions. When I first got my system, I went to XXXXXXX XXXXXXXXXX, and paid $300 for about 2 hours of dyno time. When I left, my car ran worse than when I brought it in, and actually would have died if I'd kept using their map.

I did a lot of tuning myself on the road with just the narrowband O2, and just my wits. It worked out well, part throttle drivability was much better than the dyno tune. But I knew that I had no clue about full throttle mixture. It's easy to get the part throttle stuff with a narrowband, because you're shooting for 14.7:1 anyway, which narrowbands are very good at indicating. But at full power, you need to know the actual A/F.

I ended up buying myself a Motec wideband O2, and finishing my mapping like that. It was pricey, but I figured a couple hours on the dyno and it's paid for itself. I like tuning on the road better than the dyno, because it REALLY simulates real road loads, airflow, engine temps, etc... But in a fast car, it can get pretty hairy. You basically have to drive down the road, holding the car from accelerating with the brakes. With a powerful car, it's a bit un-nerving.

The real KEY to tuning A/F, is you have to take the car to EVERY load site, and tune that A/F. Every single site. It's actually not that bad, takes about an hour. This CAN be done on a dyno, but NOT a Dynojet. A dynojet lets the engine accelerate slowly, holding it back with the weight of the rollers. But it's ALWAYS accelerating. What you need is to hold it at one rpm, control the throttle to hit a load site, and tune that site, move up to the next load site, tune it, move on... until you are done a column of loads @ one rpm. Then you move to the next rpm column, and do all the loads. You can't do this on a Dynojet. What most tuners do, is floor the car, let it accelerate to max rpm. Look at the A/F trace, and then fiddle with the settings. Run it again, over and over. It's just not efficient. Something like a Mustang dyno can hold the engine at a given rpm, and let you stay in a cell until you get the A/F right. You can also do this on the street using the brakes as I mentioned.

The other problem with most shops, is that they don't have a good understanding of the physics involved with air density changes from temperature and pressure. That's what the guy I went to really screwed up. He didn't have a clear understanding of the effects of air temperature. The car ran okay in his hot dyno room, but soon as I took it out into the cool air, it was pretty ugly. What I ended up doing was just inputing corrections for air pressure and density that are based on the physical models of air.

Now, what about timing? Well, you have to start somewhere, and in my case I'm still running on "best guess" timing. I have a good ear for detonation, and haven't had any problems. But I'm probably not getting as much power as I could. Eventually I'm going to go to a Mustang Dyno, and tune the timing. It's the same deal as the A/F. Most tuners just run it up at full power, and tune the top power timing. What you need to do, is hold it at each [email protected] cell, and play with the timing until the dyno shows maximum power. You really can't do this on the road, because it's basically impossible to feel small differences in power.

I think the reason so many dyno tuners around here don't know what they're doing is because most of the time, they're tuning drag cars. All they have to do is get the car to start, idle, and then launch at full power and down the track they go. I don't think that's what you want, so you really have to watch what you do, who does your tuning.

Another thing is, a lot of these guys mostly tune chips for factory computers. In most cases, the factory part throttle tune is already good, all the air temp factors are well accounted for, and all they have to do is modify the full power settings. A lot of them just really don't know how to tune an engine from scratch, which is what you need with the Pectel.

I've never found any really good books on tuning. There is one that's decent, written by Dave Walker who is a tech editor for Car and Car Conversions magazing in the UK. You might know that magazine, if you're a fan of Caterham type cars, since it's full of cars just like that. The book is called Engine Management: Optimizing Carburettors, Fuel Injection and Ignition Systems and it's available at Amazon.com if you're interested in reading up. It's not quite a "step by step guide", but it's a really good primer. From there, you really need an understanding of the physics of air, because he doesn't get into that as much as I'd like.

I actually learned most of what I know now, from reading Ford training manuals for engine calibration engineers. It's really a shame they don't publish them, because it's the best source I've found, by far. It's interesting, because inside the factory computers, it's basically running on a physical model of air. You can actually get the engines running pretty well in most cases just by inputting some information on the engine, and some tables like Airflow Meter Calibrations, and it does the rest. It would just need fine tuning from there. It's just interesting that you could run an engine purely based on accurate phsyical models of air.

That pretty much confirmed what I had done to get my car running, by using some of the laws to fill in the tables like air temp correction, etc.

The air temp and water temp sensors I'm using are the ones they sent with the T2 Focus plug'n'play kit. I'm not sure exactly which they are, but I'd think you had the same ones. What is the problem with the ones you're using?

So, I really don't have much info I can pass on to those tuners, I'm basically more concerned with warning you to watch what they're doing. In my case, I didn't know much when I went to the dyno, and the guy was a shyster. He BS'ed me to keep me on his dyno. When the car didn't run right after, he blamed it all on me. The fact is, my car was detonating HEAVILY when I left his shop and would have blown if I kept running it, I fixed it just by putting the accurate physical models back in. I've been running now for almost a year, 9psi boost on a stock block and internals, all through the winter, summmer, and about 300 miles on a road course so far. The car starts and runs well, idles better than stock (I'm the only one I've found to actually use the closed loop idle). I'm also getting around 26mpg, which I find is a good indication also, of how well the engine is running. If you're using stupid amounts of gas, it's not running right.

The only real info I could pass on to those guys is the two modules which do not work. The first is closed loop narrowband oxygen sensor feedback. It has an error where sometimes it defaults to minus 100% fuel. It's kind of an intermittent thing. Pectel has admitted to a software error, but have not fixed it yet. You have to watch the values of the Lambda correction. When it enters closed loop mode, it should start at 0% correction, and slowly adjust whichever way it needs to go to hit 14.7:1. When the error occurs, it starts at -100%, pulls 50% of the fuel, and the car barely runs. It knows it's lean, and it starts adjusting up but it takes a long time to get close to 0%. There's really nothing you can do, if it is giving you this problem, just disable the feature. I doubt it's an issue for you, since I doubt you're running O2 sensors.

The second feature which doesn't work is Decel Fuel Cutoff. The idea is, when you're coasting down in a gear, it turns the injectors right off. Most stock cars do this, it saves fuel. The problem is when you step on the gas and the fuel comes back on, it has a period where it needs to inject more than the necessary fuel to get the manifold walls wet.

Intake manifolds always have "puddles" of fuel which stick to them, right under the injector. In normal circumstances, the fuel is being sucked off the puddle as fast as it is being added by the injector. Anytime you quickly change the fueling however, the puddle needs to chance size. This is why you used to have throttle pumps on carbs. It's also the reason for the "Transient Fuel" correction in the Pectel. When you step on the gas, and the injectors spray more fuel, the puddle grows. There is a delay in the fuel that the engine gets, as the puddle grows to the new size. That's why you inject more fuel for a split second when you step on the gas. That's transient fuel correction, it overfuels the motor, so that it gets the right fuel as the puddle grows (this all takes split seconds).

Anyway, back to decel fuel cuttoff.... The cuttoff works fine, it shuts the injectors off. The problem is when you get back on the gas, it does a period of overfueling, to grow that puddle again. The software error makes the overfuel correction stick. It just gets stuck on, and the Pectel overfuels the motor.

Same deal, Pectel hasn't fixed the software yet, and there's nothing you can do. Just turn the feature off, disable it. Both of these two errors are not seen by most people, because they just aren't using them anyway. As far as I know, I'm the first to try, and I discovered the problem.

One thing you need to ask these guys, is have they ever tuned an "Alpha-N" system. That's what you've got. Throttle bodies, with no MAP sensor. It's the trickiest system of all, Mass Air is the easiest, then Speed Density which I'm running, then Alpha N. If they don't know what you're talking about, they probably won't be successful tuning it.

The last thing I want to say is, that I'm trying to get started in the tuning business. I don't have a dyno, but I do a pretty good job despite that. If you're interested, I can help tune your car before, during, or after you visit these guys. If they leave you with a car that only runs well at full throttle, you need my help. If you're interested, let me know.
 

·
Registered
Joined
·
2,181 Posts
Nice write up Mustang...granted most of it was like 5mi over my head, its good to know that theres someone with ridiculous amounts of knowledge I'm goning to be attemptin to tap into over then next however long befreo I go Turbo. I had no idea it was that involved and complicated to tune an engine. I mean I know it had to be tough, but I figured as long as it came with a decent manual or I could find a book, its just a computer program and a data sheet...but I see I have ALOT to learn about engine management tuning.

So with that said...how do you like the Pectel what do you like, what don't you like? Besides the software errors what would you change? On a scale of 1-10, how difficult would it be for someone that has only a "basic: knowledge of engines (ME) to learn how to effectively operate the T2 system? Thats it for now...thanks man

CCC
 

·
Senior TEAM Member
Joined
·
4,821 Posts
QUOTE P-51,The real KEY to tuning A/F, is you have to take the car to EVERY load site, This CAN be done on a dyno, but NOT a Dynojet. A dynojet lets the engine accelerate slowly, holding it back with the weight of the rollers. But it's ALWAYS accelerating. . You can't do this on a Dynojet.

*NOT TRUE ,NOT TRUE AT ALL,This can be done on a dyno jet and we can do it we have LOAD capibilities but we have found they are not needed and it made NO differance in the type of tuning ,Much of this work like he said is allready done with doing a chip in the stock ECU

P you made a good point you were driving down the road with your foot on the brake and mashing the gas boost comes up and you are under a harder than normal load at that RPM doing your maps ,,At what point in your daily/race driving are you going to be doing the break and the gas to the point that you were to tune the car at every RPM

Yes i know what you are saying as for tuning the car that way but to say it HAS to be done and that it CANNOT be done on a DYNOJET is silly

We tune many standalones from Pectel to Edelbrocks and many other things at about 100 a year and we rarely use the LOAD FUNCTION any more and i am talking 200 to 700HP street cruisers,I have NEVER used it on my ZX3/SVT and i will guarantee you they drive better than new as im sure yours does as well

Tom
 

·
Registered
Joined
·
11,123 Posts
Discussion Starter · #6 ·
Real question: Can you quickly describe the differences between MAF, SD, and Alpha-N? Doesn't have to be too fancy. Thanks.
Ok, the goal of any fuel induction system is to measure the mass of air that is coming into the engine, and then inject the correct mass of fuel to hit the target Air/Fuel mass ratio.

A Mass AirFlow system, has an airflow sensor that DIRECTLY measure the mass of air coming into the system.

Really all the computer has to do, is read the signal from the MAF sensor, and inject the right mass of fuel. But it's a little more complicated than that. The EEC actually takes the mass airflow, then compares that to the displacement and rpm of the engine. It uses that to determine the load the engine is under. If it's only getting 50% of the theoretical maximum airflow, then it knows it's only under 50% load.

Once it knows that, it will look on an RPM vs. Load table, and pick the target A/F ratio. At 50% load, the target is probably 14.7:1.

From that, it calculates how long it needs to open the injectors to inject enough fuel to hit the 14.7:1 ratio. They programmed the EEC with a cell that tells it what size the injectors are. You can change injectors, and just tell it the new size, and you don't need to touch anything else. It will calculate the time based on the size.

Speed density does not have the a MAF sensor. It basically calculates the mass airflow based on the pressure in the manifold, the temperature of the incoming air, the engine rpm, and the engine size. Some systems actually run off a volumetric efficiency map. You input the size of the engine, then you input the a 3D map of volumetric efficiency for all RPM and LOAD values.

It calculates the mass of air, based on the VE, the RPM, and the air density in the manifold (which is determined from pressure and temperature). Then, it determines how long to open the injectors to hit the target A/F ratio. These systems can be easier to get started a map. Because you can input the injector size, engine size, and take a guess at the VE table. That's probably enough to get the engine started, then you need to measure the resultant A/F, and adjust the VE table until you hit the target. Basically what you are doing is telling the computer "Oops, that VE guess was wrong, it's actually this". Once that VE table is filled in however, you could change the target A/F values, without adjusting the VE tables. This is nice, because you can directly change the A/F "I want to hit 13.0 instead of 13.5", without having to measure your A/F again. It will calculate it right, because it is actually calculating how much air it's getting.

But, any time you change the VE of the engine, like cams or headwork, you have to redo the VE table. A MAF system actually measures the airflow, so it compensates automatically. A Speed Density system calculates the mass from a VE table, so if you change the VE, you have to change the table.

Some systems like the Pectel, just skip a lot of that. Motorsports systems are harder to program, anytime you want to change anything you have to measure the A/F, because they don't actually calculate the mass of air. This makes the systems faster, because a whole round of calculations are skipped. It takes more work to tune, but requires a smaller/lighter/cheaper computer to do the same job.

The computer doesn't even care about the air mass. All it does is it looks at load in the manifold, and the rpm. Then the tuner DIRECTLY fills in an injector open time on an RPM vs. Load map.

Now, you're faced with the prospect that any time you change the VE, OR any time you change the target AF, you have to retune that table with an A/F sensor.

It can get to be a lot of work, and is one of the reasons I've never played with my Cam Gears. Just too much work at this point. If I wanted to tune them, I'd have to redo the injection time table every time I moved the cam gears.

But like I said, the computer can be smaller, lighter and cheaper. The offset is more work.

Now, Alpha N is the monster of all systems.

This is used when you really don't have ANY intake manifold, so you can't measure air pressure at all.

This system is very much like a speed density system, but instead of measuring load from the MAF sensor, you are measuring load from the angle of the throttle plates.

Luckily for us, air is an ideal gas, and it obeys a simple set of rules. If you measure the air temp, and the barometric pressure, you pretty much know the mass of air that is waiting to get into the throttles.

For any given throttle angle, and a known baro pressure and temp, you'll always get the same airflow. So you tune the injection map much the same way as the Pectel speed density system.

You look at the throttle angle, instead of the MAP sensor. And you tune the engine with an A/F meter, and fill in an injection time map that is based on Throttle Angle vs. RPM.

But these maps are impossible to begin to guess at, just so you can get the car running.

See, on my Speed Density map, across any given pressure line, the amount of injection time pretty much follows a curve that looks like the VE of the engine, which in turn looks an aweful lot like the torque curve of an engine. It's pretty easy to see what's going on, because your injection time for a given manifold pressure, looks like a torque curve across rpm.

But Alpha-N isn't like that. See, at low rpm, it might only take 1/4 throttle to be at full load. So you need tons of fuel at 1/4 throttle at 2000rpm. But you don't need much fuel at 1/4 throttle at 6000rpm.

At 2000 rpm, you could be at full load at 1/4 throttle, and at all throttle settings above that, the load is still full... so the injection time is flat. At 4000 rpm, full load might be at 1/2 throttle, and everything above that is flat.

Now you're faced with the prospect of mapping an injection table that isn't nearly as intuitive. Also, anytime you change the engine VE, or the target A/F, or even the throttle bodies, you need to redo that map.

Unfortunately it's the only way to run an engine with individual throttle bodies and no manifold. Actually, I should say that. Somebody here showed one with a MAP sensor, but I don't really know how they did that. Anyway, the common way is just to run Alpha-N.

Now, that last thing to know about Alpha-N is that you really have to measure the barometric pressure well, and you really have to input a GOOD correction curve. If you get this wrong, your A/F will be off if you drive in the mountains. Since, the higher the outside pressure, the more air you will get in the engine for a given throttle opening. Again, luckily for us, air is easy to model, and obeys some simple rules. Suffice to say, if you use the theoretically perfect model, you would probably only need a slight deviation if you tested it in the real world.

I had no idea it was that involved and complicated to tune an engine. I mean I know it had to be tough, but I figured as long as it came with a decent manual or I could find a book, its just a computer program and a data sheet...but I see I have ALOT to learn about engine management tuning.
Personally, I don't find it is that complicated, it's just that nobody has ever written a really good book. An unfortunately there are a lot of tuners out there who don't know what they're doing, and they just muddy the waters.

They often do a really bad job, and just blame it on the customers, or the engine, or the injection system. The things coming out of the mouth of the tuner I went to was quite incredible, and it took me a while to realize that what I thought was true about engine management, was true. I was lucky to find the Ford manuals, because it basically re-confirmed what I had thought.

So with that said...how do you like the Pectel what do you like, what don't you like? Besides the software errors what would you change? On a scale of 1-10, how difficult would it be for someone that has only a "basic: knowledge of engines (ME) to learn how to effectively operate the T2 system? Thats it for now...thanks man
I find the Pectel system VERY easy to use. It's reasonably easy to install, and the software is pretty intuitive to use. Now, it could REALLY benefit from a manual, or even just some help screens. But they are just too lazy to do that.

I guess it's a testament to the easy of programming it, that I figured out how to do it without any help from Pectel, no manuals or documentation. Just a lot of menus that I explored, and tested what they did, compared to the physical theories and saw that they agreed. Then I knew what to do.

I think a decent job can be done by somebody with moderate knowledge, because a lot of the really hard work has been done in the base maps. The car starts and kinda runs out of the box. All you really need to do adjust the fuel tables, and the transient fuel injection tables. Everything else you can pretty much just turn off.

Now, if you want it to idle like stock, with closed loop idle. If you want good milage, run well in any weather, use oxygen sensor feedback, that's more complicated.

*NOT TRUE ,NOT TRUE AT ALL,This can be done on a dyno jet and we can do it we have LOAD capibilities but we have found they are not needed and it made NO differance in the type of tuning ,Much of this work like he said is allready done with doing a chip in the stock ECU
Really? I had no idea. The Dynojet can absorb load? How? Does it have an eddy current generator or what? Do all Dynojets? So you can actually hold an engine at one rpm at full throttle?

I thought the Dynojet was just a set of heavy rollers.

Yeah, I think the reason you haven't needed it was because most of the part throttle work is already done. You input a new MAF curve, and so the computer knows the real load, and it will get all the part throttle stuff right. You just have to tune the full power fuel and timing for maximum safe power. But that should be as easy as just changing the target A/F and the spark timing right?

P you made a good point you were driving down the road with your foot on the brake and mashing the gas boost comes up and you are under a harder than normal load at that RPM doing your maps ,,At what point in your daily/race driving are you going to be doing the break and the gas to the point that you were to tune the car at every RPM
This I don't really get. Tom, have you never floored your throttle at 2000rpm? Like, if you're cruising, and pull out to pass somebody without downshifting, you just punch it and go right?

Well, that load at 2000rpm is the SAME as if you were holding the engine with the brakes. This idea that the load would be higher is not true. It's the same load.

The reason we hold the brakes, is so that I can stay there long enough at 100% load at 2000rpm, long enough to tune that cell. Then I would do the 2500rpm cell, and the 3000rpm cell.

So now as I'm driving, if I punch it at 2000rpm, the fuel is right, and it stays right as the car accelerates to 2500 rpm, 3000rpm. The injection calculation passes through all those cells, and the values have to be right. And the only way to get them right, is to tune the car holding it in that cell as you tune the A/F.

We tune many standalones from Pectel to Edelbrocks and many other things at about 100 a year and we rarely use the LOAD FUNCTION any more and i am talking 200 to 700HP street cruisers,I have NEVER used it on my ZX3/SVT and i will guarantee you they drive better than new as im sure yours does as well
Tom, so how are you tuning these?

The load vs. rpm map on most systems is about 10x10. So 100 cells. On mine I have 500,1000,1500,2000.... up to 8000rpm, on the bottom. On the side I have 2psi, 4psi, 6psi, etc...

I think mine is actually 16x16, so 256 cells.

If you only run at full power on the dynojet, you only tune the top cells, as the engine accelerates through the rpm at maximum load.

What do you do to fill in the middle cells?

Like, if you have a brand new table, full of Zeros. You run it at full power and enter values on the top at full power. What do you put at 50% load at 2500rpm?

I'm guessing you're using a speed density system, that has a Volumetric efficiency table right? And the VE is pretty much constant at all loads for a given rpm, so you just fill in the same VE at all load sites within a given RPM table?

Is that true?

If so, you're lucky. Like I mentioned early, some systems, and the pectel is one of them, it's not like that. You have to actually tell it the time, because it doesn't know VE. So you have to actually GO TO that cell, to input a correct time to hit the A/F you want.
 

·
Registered
Joined
·
556 Posts
Wow-this is a great thread, thanks P I'm learning a lot

I've been thinking about a few things for either going turbo or all motor & how I wanted to acomplish either...now I really want a pectel. Too bad you're not closer, I'd love to work & learn with you in person. Keep it comin'
 

·
Registered
Joined
·
2,181 Posts
P-51...I think you should put together a little how-to supplement guide with some of the stuff you learned...you know a quick help or cheat sheet to the basics, like running correct fueling in simple "x#" of steps...then sell it!!!

I'm interested to know about your background in the sciences and what you do for a career. I'm a second year Physics student, and trying to learn more about what I drive from a physical science standpoint and I know you seem to have alot of that under your belt. Thanks.

CCC
 

·
Registered
Joined
·
4,983 Posts
P-51...I think you should put together a little how-to supplement guide with some of the stuff you learned...you know a quick help or cheat sheet to the basics, like running correct fueling in simple "x#" of steps...then sell it!!!
I second the notion! I would pay money to have all of P51's tuning discoveries in my hands.
 

·
Registered
Joined
·
717 Posts
Nice writeup, Rob. I think this info is good enough to be bordering on too valuable to leave on the forums. We might want to cull this into a separate web page and host it someplace (maybe wrcfan can donate some space on FJ; if not, I'll put it up on my site).
 

·
Driver > Car
Joined
·
23,988 Posts
Or cut the information out of here, post it in another thread, and make it a sticky.
 

·
Registered
Joined
·
2,181 Posts
P-51...I think you should put together a little how-to supplement guide with some of the stuff you learned...you know a quick help or cheat sheet to the basics, like running correct fueling in simple "x#" of steps...then sell it!!!
I second the notion! I would pay money to have all of P51's tuning discoveries in my hands.
Exactly what I was thinking!!!

CCC
 

·
Registered
Joined
·
11,123 Posts
Discussion Starter · #15 ·
I was thinking at some point about making a "booklet" for sale for something reasonable like $20. Specifically for tuning the T2. But then I heard that Pectel was doing one... silly me, I keep forgetting they don't care too much about T2 customers.
 

·
Registered
Joined
·
2,181 Posts
I was thinking at some point about making a "booklet" for sale for something reasonable like $20. Specifically for tuning the T2. But then I heard that Pectel was doing one... silly me, I keep forgetting they don't care too much about T2 customers.
[censored] you could charge like $100 and I'd still probably buy yours before Pectels...unless its really that good!!! BTW..did they announce a timeframe for this "lil booklet"???

CCC
 
1 - 16 of 16 Posts
Top