fast951

Hi Geoffrey, A very good question. Based on the various stand alone systems I have worked with, you have a base ignition table, then you apply a boost compensation table on top of it. The end result is a unique ignition value at a predetermined boost value. I assume you are asking how this is done in the Motronic.

Not knowing how familiar everyone is with the Motronic, a quick background on how the Motronic works might be of some help. In order not to get tangled up in the details, I will keep it at a fairly high level, feel free to ask questions.

The stock 3.2L Motronic understands "load", it does not directly deals with boost as measured by a MAP sensor. Load is based on air flow as measured by the AFM (or MAF when we do the conversion). Since air flow is directly related to boost (in general more boost means more flow), then the Motronic indirectly deals with boost.

In the Motronic, you have various states of operation. Idle, Part Throttle (PT) or light load, then you have the Wide Open throttle (WOT) or heavy load. Based on Throttle Position Sensor (TPS) and Air flow (from AFM or MAF) the Motronic determines the state it needs to operate in. For each state Idle, PT and WOT there are fuel and ignition tables. Idle is RPM based, PT is RPM & Load based, WOT is RPM based. There are other compensation tables (head temp, intake temps, etc..) that get applied as well.

Idle and PT are tuned for good drivability, response, fuel economy and generally operate in closed loop.... To simplify things, we assume that while on boost you are operating in the WOT state. (In reality PT state can operate at very light load or low positive manifold pressure).

The beauty of a MAF based system, is that it measures the air mass as used by the engine. By converting the DME to use Air Mass, I can set my desired AFR and will be 95-98% on target no matter what the boost is. (Boost is directly related to air mass). Go from a small turbo to a larger one, you will flow more air (hopefully) at the same boost level, the AFR will remain the same. (You can't do this with MAP based system!).

Now that we covered fuel, let's look at ignition under load. The DME does not use a MAP sensor, for tuning I rely on data collected while on the dyno. Based on the gathered data, I modify the ignition values. Unless it's a race car, I set my goal to "safe power" vs. "the most power safely". Depending on the DME, we offer 2 sets of externally selectable tuning maps, say for pump gas or race gas.. To relate this to a standalone, let's assume the Motronic contains our base table.

As we all know, each engine is unique, yet to see 2 100% identical engines. To get the "most" power, each engine will have a very unique ignition and AFR values. Any changes to the engine parameters (turbo, boost pressure, boost control, etc..) require a unique tune. This is the same no matter what the engine management is (Motronic or Standalone).

One can assume, that for a particular engine, the Motronic will get you 95% on target. To compensate for various engine parameters, boost changes from the base tune.. we offer the PiggyBack to allow the user to bring it to 100% on target. With the PiggyBack the user can alter fuel and ignition. When using the PiggyBack, a MAP sensor can be used, giving the user control based on LOAD/RPM/BOOST.

In summary, for fueling, boost is captured via Air Mass which is understood by the DME. For ignition, the base DME contains ignition data based on a preset configuration and boost level. However when the MAF software in the DME and the PB are used together, you have the flexibility of altering ignition (and fuel) based on boost over what is already programmed in the DME.

We have some items to be released soon, that will change/enhance many of the existing Motronic features. More information will be made available at the time of the products' release.

This is probably much longer response than you asked for.

Geoffrey

" Based on the various stand alone systems I have worked with, you have a base ignition table, then you apply a boost compensation table on top of it. The end result is a unique ignition value at a predetermined boost value."

Most of the aftermarket engine management systems I use, the load of the engine is based on manifold pressure so there is a single table for ignition angle. The only time I may use multiple tables is if the engine has ITBs on it and therefore requires load to be set by Throttle Position with boost compensation. Not nearly as nice or as easy to program give the VE% changes that occur over the range of throttle blade position.

"The beauty of a MAF based system, is that it measures the air mass as used by the engine. By converting the DME to use Air Mass, I can set my desired AFR and will be 95-98% on target no matter what the boost is. (Boost is directly related to air mass). Go from a small turbo to a larger one, you will flow more air (hopefully) at the same boost level, the AFR will remain the same. (You can't do this with MAP based system!)."

Yes, MAF based systems are great for keeping engines operating at a consistent AFR, however, in your example, if you are flowing more air at the same boost level, your cylinder pressure will be higher and therefore the ignition timing requirements will be less and I don't see how you have compensated for that. I don't see how your MAF configuation can be better at controlling ignition angle for all engine operating ranges than a system that measures boost pressure and utilizes it for engine load calculations. As you know, ignition angle has probably the largest effect on engine output power.

DonE

Ahhh - drivability.

Assuming you've built a custom motor, drivability is by far the most daunting issue. There is no shop that can build you a custom motor with EFI conversion and nail drivability unless they keep it for weeks. Then the conversion will be cost prohibitive due to the time involved. You can not simulate the variables of temp, altitude, traffic, driving style, starts, volumetric effeciency, cruise (at varying speeds), etc with 2 or 3 hours on a dyno. You can provide a reasonable map, but it will need to be adjusted (that last 10% mentioned above) to take full advantage of the $4k ECU you just installed.

With that being said, the new owner needs to be able and confident enough to make the changes and understand that any change made has a ripple effect within the tune (the joys of relational databases).

I would estimate that it took me over a month of driving around town to properly tune my car for drivability (again, that last 10%). I probably have over 75 different maps saved trying out subtle combinations, trying to find the best. Oh ya, then forget to turn on the temp correction table after all this and you'll have to start over (kind of). I recently had someone ask me for a copy of my map so they could try it. I tried to be as diplomatic as possible when I said no. He could have bought it for $500 and that would have been cheap.

One other comment about EFI. If you are chasing a HP number, then you've missed the point and wasting time/money. Those of you who know me, I personally think dyno runs for HP are always highly suspect and I take the numbers with a grain of salt. In my opinion, dynos are tuning aids. If you tune the car to it's peak potential - drivability, acceleration, longevity, usability - the HP and torque will come on their own and the overall value (smile factor) will be greater.

As a side note, I am looking at a custom tune for my 05 Powerstroke diesel. For $450 I can get three complete, proven maps and the hardware to install them. I consider that a steal...

Geoffrey

Don, I beg to differ on the length of time required to properly setup the car. Yes, the 2-3hour dyno time will only get you 90% of the way through, and of course you need cold weather for cold starting, however, if the shop has done a number of these conversions, they will be familiar with the engine and the ECU and will have the compensation tables set properly. I've done over 100 of these in many, many different configurations and mine always start when reaching in the window and twisting the key (even the high strung race cars) and they all drive cleanly from 1krpm. I can complete the drivability in a few days of work.

fast951

In your example, you are setting the ignition values in a single table where the compensation table has no effect. The bottom line is getting the correct ignition value no matter if it's hard coded or calculated. There are various methods for doing things, the end result is what is important. Saying one method is better or easier than the other depends on who you ask.

I never said one method is better than the other, I have no intention to get into what is the best approach debate. You asked how I did it, i explained.

As I mentioned in my earlier post, any changes in engine characteristics will require re-tuning to get the most out of it. This is the same with Motronic or Standalone. For a standalone, changing the flow at the same boost level (using a different turbo) requires you retune Fuel and Ignition. With a MAF system, which as you agree is great for keeping correct AFR, you just need to retune ignition. When using the Motronic, the user can do this via the PB which is also based on boost.

There are other variables to deal with, a MAP based system must compensate for barometric pressure. Each system has its pros and cons. There is no one unique solution for every application.

fast951

I strongly agree. The hardware used is as effective as the person programming it. People with a standalone rely on the experienced tuner to get them up and running, money well spent. Experience backed with positive results matters.

I like to Motronic, as Bosch and the factory spent 1000s of hours taking care of the last 10% which is the most time consuming.

Geoffrey

Fair enough, I was just curious about how ignition angle was calculated in your revised motronic system, knowing quite a bit about how motronic works.

fast951

To be fair, as I write this, many changes are being made to the Motronic code. I have an idea of what you are thinking, I just can't elaborate on what is getting changed yet!

Geoffrey

Cool.

DonE

We'll agree to disagree.

PorschePhD

I originally thought the same thing, but when it came down to it two cars with the same equipment still will behave differently. Your cam timing may be what you think it is yet may still differ from mine. Intake flows will vary, so on and so forth. The electronic verses mechanical timing will always be different. There are just to many deltas involved to send one file and say here ya go. Not to mention what if the wiring is wrong and the system will not run. What if a sensor is bad. Unless you have a scope you can not read it and most do not want to spend 3K on a scope. So you see there are so many things involved than just making the program work for all. Man, I wish we could. We tried believe me!

onboost

Don,

Those moments are truly priceless!!!

Paul

nathanUK '81 930 G50

Geoffrey, I think Don meant that the first ever initial setup of any ECU to a DIY'er would take weeks.

I don't think he was saying that someone who has successfully used an identical ECU on a 930 before with a similair build would take weeks to set it up.

911rudy

Stephen,
Thanks for the response. I think most of us on this forum would like to go EFI but there are so many bad stories from the guys that went first. I know that there could never be a single code that would work for all 930's out of the box but I was wondering if there could be a general code so that with the laptop hook up and some mechanical skills there could be a base line code to work from. The supplier, your company, would have to recieve from the customer certain parameters that you noted in your post, and then a base line code could be established and you would have checked all of the sensors and injectors for operability on a sample composite 930. From that start, it would seem that a guy with above average skills could take the base line and adjust it to fit his particular car. You mentioned the o'scope to test the injectors being $3500, but you could check out all of the pieces before you sent them out. It seems like we are just one "Geek" away from doing this. Some guy that you feed beer and chicken necks to to work on this project. C'mone Stephen, advertise for him!

m42racer

Interesting discussion.

I think it all depends what you consider " done right". It is not uncommon for the final tune to take as long as Don is suggesting. Different engine configurations, climate and other factors all dictate the time involved here. Some functions like enrichment, idle, cold start, Boost control, etc can all take alot of time. Add in special features like Cam control, E throttle and others all take lots of time. How one can say it can be done and done right as quickly as Geoffery suugest, I wonder. I know in my case, I had a car that was on the East Coast and tuned on the West Coast. The cold start feature could not be tuned for the colder climate on the West Coast. Although the software allowed for scaling, the final test had to include cold weather to see if it was right. They hired a refrigerated container and put my car inside of it and lowered the temperature to East Coast temperatures. This all took time. Not 4 days.

Is it possible we are all talking about different conditions and requirements. I think so.