Flex fuel

I need to setup a flex fuel setup on a car, I have access to the complete original ecu file and can change any calibration. Unfortunately the ecu processing capability is very little, so its not possible to repurpose something for the purpose.

So my plan is to use a piggyback chip to enrich the mixture or let me rather say increase the fuel flow based on ethanol content.

As its a direct injection engine, I can increase the flow by manipulating rail pressure, maf voltage (currently close to its limit already) or by manipulating the intake tenp, coolant temp etc.

my question is has someone done it this way or have suggestions?

I thought the easiest way would be the rail pressure, so I lower the pressure the ecu see, thus increase the flow, but trying to think what the negative effects might be?

after I have sorted out the running, I will get to cold start enrichment, etc also based on ethanol content.

i have quite a bit of experience with tuning on flex fuel, so that is not a problem.

What engine?

Doesn’t really matter, as I have a few I want to do.

The first application is an Mazda MPS, but will do Golf 5, 6, 7 GTI soon and a few others if my plan works out.

They are mostly all direct injection, I have to think about a solution on port injection as well.

There is a reason why few piggybacks are in use today. The BMS JB4 is basically the main one, for BMW and Infiniti applications. Tricking the ECU is difficult because it results in a lot of unexpected behavior like fault codes (which could be turned off) or inconsistent AFR and spark timing due to all the correction tables inside the ECU.

First there are the basics: do you have a fuel system capable of higher ethanol content? The MPS for example is notorious for having very little headroom in the high pressure fuel pump. The MZR engine for the MPS and the various 2.0T engines for the GTI both have fuel pump internals available. You need that first, as well as in -- tank fuel system upgrades. It's also common for the fuel system to have problems if you don't run E0/E10 once in a while.

Yeah you could trick some of the sensors (mass airflow or pressure sensors) but be careful because it will throw a code, or you will get something wrong and blow up an engine. What kind of R&D budget do you have? Better be able to afford a new motor or two. Mazda MZR engine is notorious for experiencing preignition. Throwing ethanol fuel in there and tricking sensors increases that risk. Remember in terms of non American car manufacturers' vehicles, there are very few flex fuel kits available for stock ECUs (Cobb Subaru Flex Fuel being the main one). If it were easy somebody would have done it years ago.

The next big problem is starting. You need some kind of map switching for the ECU. It's not going to start well on colder days. Cranking fuel doesn't normally rely on many sensors (pre set map in the ECU, crank cam and rail pressure sensor) so there's not much to trick.

As for the rail pressure, WOT runs are typically at max rail pressure anyway, and as we discussed there's limited headroom in these fuel systems in the first place. For port injection it's simpler because the rail pressure isn't so variable and fuel system upgrades are much easier.

Yes, I always upgrade the fuel system.

For example this Mps is stock engine and turbo, with just bolt-ons and we are already on 240kw and nearly 500nm, on an equivalent of about American 91 fuel. They do stock 150ish kw by us.

Our problem is the only way to get more power is with other fuel, the easiest fuel to adapt to and make sure therevare no customer error, is with flex fuel, we use it with GTRs, BRZ, etc and works great. Those ecus can process the extra calculations. Unfortunately the mps ecu cannot, so will have to do it externally.

I have changed the knock control a lot on the mps and don’t have the engine in danger at all. My experience will also prevent danger of damage.

As long as I simulate all possible mixtures, temperatures, etc on the dyno, I can make sure its safe.

the question was what is your suggested way or ideas of what to use. The fuel pressure under full throttle is not max that the pump can deliver, its most likely 60-70% currently.

In my mind if the ecu see less rail pressure it will increase the injectors duty cycle correct?

i can modify or disable any dtcs I need to.

the other way I thought of doing is increasing the maf housing size, rescale it and then have extra room to add the additional fuel. Go from the current 67mm to 73mm housing.

Direct injection can't run more than about 40-45ish percent duty cycle (300-360ish crank degrees of injection). Can't spray during expansion stroke or exhaust stroke.

MZR engines are notorious for preignition (Google mazdaspeed 3 LSPI) and ethanol tends to preignite more than gasoline. Crank up the rail pressure and you will wear out the pump faster.

What is your R&D budget? First thing I would do is build a test bench to manipulate the signals (hardware in the loop system) so you don't risk blowing something up. You need to verify the impact of manipulating the signals on the outputs of the engine before you put it in a vehicle and risk blowing something. Or skip that and do it all on your own shop car (not customer car), with spare motor ready. Flex fuel direct injection is a lot more complicated than port injection.

I'm a little confused about your intentions or reasoning behind using a piggyback when it sounds like you have the ability to directly tune the stock ECU in these applications (in order to disable DTCs)? Is this because you can't do flex fuel with the stock ECU but you can reflash it? I can't say a piggyback won't work, but I personally wouldn't advise it. If you're planning on tricking the ECU into increasing fuel delivery based on E% then this is only a small part of the equation. You'd also want to manipulate ignition timing and probably boost pressure. Then there is the wholesale changes required for good cold start operation that Raymond has mentioned. Manipulating the fuel pressure signal alone isn't going to get you there unfortunately.

I also wouldn't recommend cheating the MAF signal output to achieve these aims. For a start your tuning requirements would be at odds with what could be achieved. For example to increase fuel flow you'd need to have the MAF over-report airflow. This is almost certainly going to cause issues under load where you'll potentially hit some kind of airflow limit, but the other aspect to consider is that over reporting airflow is likely to result in much lower ignition timing being delivered - Exactly the opposite of what you want. Then there's the issues around torque modelling, idle control and all of the other aspects that rely on a realistic MAF output for the ECU to do its job properly.

It's possible you may be able to get some kind of workable result but I'd approach this very carefully as the potential for failure is quite significant.

Only reason why I am doing it via piggyback, is due to the fact that the ECU is very basic and cannot handle anything else. I can modify essentiall anything in the ECU, so I am not limited in that way.

I have been thinking whether I should not try re-purpose something like air temp for the purpose, will go through the maps and see what I can find, can re-calibrate the fuel/timing trims, etc.

Was really hoping you had suggestions or knew of a known solution, I tend to always do the difficult things...

I am still a noob when it comes to actually tuning the car, only because of the application of principles to the actual software. But my world of cars/engines is VW and I do know that that the bosch ME7 is a very complex ECU and you should not need to run a piggy back ECU to get the results you are trying to obtain. I did not tune mine by myself, it was a collaborative effort between me and my tuner, but I run an E50 blend big turbo on stock ECU with Eurodyne on MK6 GTI. The only downside on MK 6/7 GTIs is Eurodyne doesnt offer Anti-Lag/2step. Im sure you are already aware of all this, just thought I would chime in! I know 0 about Mazda's though..

I had a MZR engine 10 years ago. Piggyback was the only option available then. Parts and tech at that time hit a huge wall when it came to fuel flow. This wasn't even using e85. The headroom wasn't there even on e-10 with bolt ons and a tune. A ton of these blew up and I like many others gave up as the support in the aftermarket wasn't there yet and they just weren't effective platforms for increasing power 50 - 100% over oem. Timebombs. Expensive timebombs.

I thought I’d just update the thread, I managed to get the car running till E95, didn’t go further.

makes good power, just too bad I maxed out the turbo, so could not lift the boost further than 1.4/1.5bar.

With the uprated fuel pumps, etc I had enough supply.

The stock engine and stock turbo made decent power, to the point that I am very happy with it. Its an daily drive and have done a low of kms since, many of which was on ethanol.

I will be leaving this car like that, the dip is wheelspin on the dyno.

I forgot to say, cold starts, timing addition, boost addition, etc are all sorted and based on ethanol content

Nice work!