Not yet, but I want to give it a shot. I noticed some discussion in the thread about MAF scaling ( https://www.evolutionm.net/forums/sh...maf+scaling%22 ) that involved using a GM MAF without needing a MAFT.

I'm willing to do the testing to see how viable an idea this really is. I'm currently using a MAFT Pro and GM MAF, but I can easily bypass the MAFT and feed the GM sensor directly into the ECU. I can take care of the mechanical/electrical and provide logs/readings/etc, I just need help figuring out the ECU side of it. For now, I'll use the MAFT Pro to send a constant 100kpa baro signal and a constant 70degF AIT signal. Let me know what you guys think.

 

Dam if you make it happen that will be a big step. Good luck

 

The GM MAF I have outputs about 2000Hz @ idle and about 15000Hz WOT @ redline. The first thing I need to know is if the ECU can even handle a 15000HZ signal. If not, then a MAFT would be required to rescale the signal before it even got to the ECU. Does anyone know what the frequency input limit is in the ECU?

 

The dsm-ecu forum on yahoo has some information on this. They also have set up the stock ecu to run MAP if you'd like that option as well (I believe its still rough however).

 

Bez has a thread on aktivematrix that explains how he modified the Evo ECU to run speed density, if you want to go that route.

Eric

 

The only way this will be viable is if I can scale the frequency down before it goes to the ECU. If anyone can build this circuit:



that would be a start.

 

You can speak to Craig Moates www.moates.net. He is always helping DIY people out and has helped out the honda world alot, producing some of the hardware for honda diy engine management

 

Damn, I would really like to do this, if it allows to VTA.

No, I am not wanting to do it because of being a ricer, as I already switched from an HKS to a Greddy to quiet it down more than the HKS.

I am getting tired of di(kin around with having to recirc. I have the whole Injen setup (Intake and UICP) and it doesn't line up exactly, and it shoves my intake into my top radiator pipe, pushing it up about 1/4"-3/8". It would be sooo nice to just eliminate the added pipework/clamps/connectors, then I would free up soo many options of what I can do in the engine bay...

 

Right now its not the easiest thing to do without external electronics. But it will eventually be possible when we have the ability to configure the input signal and associate it with the scaling.

The ECU probably won't handle the 2000hz-15000hz range that well because there are more than just the MAF scaling that are affected by those numbers, And from the looks of the values, if its linear, its not at the same slope so there's more to it than just a frequency divider.

3 things you have to take into account when removing the stock MAF

It contains 3 sensors, a barometric pressure sensor (Different than the MDP sensor on the intake manifold) an AIT sensor, and the Karmann frequency sensor.

In general, Hot wire, and Film sensors don't need the baro sensor, so its output signal can be simulated. The Ford sensor is analog from 0-5v, MOST GM sensors are frequency (the exception is for some of the 6 cylinder cars which used a Hitachi sensor, nearly identical to the ford one, and similar to the subaru)
Finally you need an AIT sensor (air temp) the baro sensor (tho no longer needed because the sensor compensates for tat) and the AIT sensor are used by the ECU, so you would need to scale the AIT accordingly, or find a sensor with a similar output (the stock karmann sensor does have a thermisistor, so replacing it with a ford thermesistor ait sensor will work without much if any modification to the signal)

On my car, I have a simple control box that converts the ford 0-5v signal to a Karmann square wave, it simulates the Baro sensor signal so the engine doesn't throw a cel code, and passes through the Ford AIT sensor 0-5v signal. The electronics are adjustable and the calibration is done within the sensor. This by far has always been the simplest and fastest solution as it "Just works out of the box"

The only problem is there is no such thing as a "plug and play" conversion. Though on my car I came real close by salvaging a MAF plug off a broken MAF sensor. So I can easily swap a MAF onto the car's intake pipe if needed.

There are advantages of using an external box vs an internal solution..

1) An external box doesn't affect your ability to add other piggyback devices if needed
2) an external box is easily adjustable using pots, trims, or programming
3) an external box allows you to swap back to a stock MAF in a parking lot should you have a hardware failure, or if it needs servicing

You need to be able to easily fine-tune the curve between the aftermarket MAF and ECU, it takes time, and would require TONS of reflashing to get it perfect therefore it is/was way more time consuming than me making some minor trim adjustments to the electronics.

If you swap sensors, for instance I can swap from a standard ford sensor, to a Granatelli sensor, and only make a few adjustments to compensate for the new curve in about 15-20 minutes. Doing this would take hours if you were in the ECU.

If you were intent on doing this all within the stock ECU, then a speed density solution might be a better choice. Although I don't like speed density because I am very familiar with its quirks and pitfalls, its a better option, and we already know it can and has been done. Unfortunately it suffers from the same pitfalls as doing any internal conversion, and that is you have to be VERY careful about anything you add to piggyback functionality on your ECU

 

My head hurts after reading that...

After about the third time, I get it!

 

Revive this thread!!!

I'd love to tune on Speed Density with only JDM MAP and IAT sensor!

 

Another thing to mention is Mellon's testing with the GM/MAFT is that the readings are not correct with temperature changes. I myself have built a voltage to frequency converter to run a ford 0-5v hotwire. Anyone can throw in a resistor to set the AIT and Baro. The issue with that, is what Mellon found out first hand with a bunch of logs. The GM sensor compensates for baro/ait different from our ecu. MalibuJack also touched on this some and he has much more real life experience than I do with it. Now, in ecuflash you can set those tables to compensate differently. Allowing you to ditch the resistors all together and never even change any of the Baro and AIT wires. (you would have to get the sensors back on the car without the rest of the maf, but you could just run aftermarket sensors anyway)

Now, we would need the algorithms that the ecu uses to figure out 2-byte load via baro, AIT, and the frequency from the maf. Then find out how much the gm/ford sensors compensates for baro/AIT and change the tables in ECUflash to match.

I was very much into this in the past, put due to personal problems I have not been able to work on these things. Maybe in a couple weeks I will be able to focus on this and some other projects that people have asked for.

Something else to add: there is a much easier way to go from the GM frequency range to our range. I saw an IC somewhere in the past to cut the frequency of the square waves in half. Then just run a second chip to half it again and you are in our ecu's frequency range. Where I saw it I have no idea..... I think it was a DSM page on a different forum.

 

It's worth mentioning that the detailed testing I did was with several different GM MAF's and MAF-T boxes on my Stealth. I haven't done the extensive testing on the EVO yet.

The problem I had was that the AFR would change with ambient temperature changes which was unacceptable to me.

I haven't found a need to go with a larger MAS on my car yet

 

I've found the same as Mellon. I was using ECU+'s GM MAS translation and it worked well, but my guess is that the ECU still needs to see the REAL baro and AIT to adjust its IPW according to the fuel maps. I think the the ECU likely scales the fuel tables via the AIT and possibly the baro. Simulating these signals didn't seem to work. I had to retune before every track day at the track because altitude and temperature changes varied my AFR drastically. (plus the GM doesn't like water/meth injection to be reverted back through it when the throttle snaps closed)