Ah, thanks. I wondered how people modified the adder.

Drive-ability of the car is good, and not far off from excellent.

If anything the car runs a little on the lean side but I should be able to take care of that with the fuel maps. I also think it may be a little too conservative with low rpm timing, as it will bog pretty bad from a standstill if the rpms aren't over 1000, but I'll start with fuel and see what gremlins remain.

I also use the ISC in my Mirage and it rev's high when cold (2k-2.5k RPM) but settles down just fine...although I have messed with my tables quite a lot so maybe I did reverse them in essence. I haven't cared to really mess with it yet this winter since I have zero motivation lol. It should be easy enough to find an electrical diagram of the ISC coils to make sure its safe to simply reverse them.

I have looked at the diagram and it is safe to reverse them.

http://www.stealth316.com/2-isc-iac.htm

Here is a screenshot of the 1g ISC motor wiring. So it looks like the ISC gets power to the center tap of the ISC coils. The ecu then provides a ground signal to the outer legs of each coil, of which there are 4 to move the rotor. By changing the 4 ISC wires you are simply changing the way the way the rotor moves with respect to how the ecu thinks it is moving it.

Thanks for the info 0x that was a big help!


So at the end of this adventure I have learned that there is a mistake in the labeling of pins 67 and 68 on the 1G DSM harness referenced here. They are backwards.

67 is ISC motor coil B2, 68 is ISC motor coil B1. The "A" coil is correct.

I swapped these two pins with each other, and re-loaded my original map.

Preliminarily idle function is normal, however the car has only seen one heat cycle and 10 minutes around my neighborhood since the change.

Credit goes to keydiver for finding the specific problem and suggesting to change only the wiring for the "B" coil. Thanks Jeff! {thumbup}

Glad that helped and glad we figured out the error in the diagram. Report back and let us know if everything is 100%.

Idle has been rock solid. I think it's safe to call that one solved. {thumbup}

General question. Let me start by saying I know nearly nothing about Hex, but I'm trying to understand.

I moved to the Tephra v7 ROM, and ported over the settings. Because I can't change the adder in EvoScan I downloaded a Hex editor, and I found address (I think it's an address??) 1513 on both the Tephra v7 ROM as well as the ROM I started with. 1513 because that was the location mentioned earlier in this thread.

My understanding is that the stock evo8 adder is 140. Now that I'm looking at Hex, I'm drawing the conclusion that 140 must be in decimal, because you're limited to 2 character expressions in this Hex editor I'm looking at.

My understanding also is that the correct adder for the 1g MAF is 64.

So on the Tephra ROM, at 1513, I have a Hex value of 8C. I used Google to verify that Hex 8C = Dec. 140, and it does. Great so far I think I'm tracking.

On the ROM I started with have a value of 64 but it's in the Hex address section. This equates to 100 decimal.

So if I'm understanding this so far, that was a mistake, and it should really be a 40 in the Hex address section, as Hex 40 = Dec. 64.

Does that fly?

Well, I can say for sure, that entering Hex 40 (decimal 64) in the adder in combination with the MAF scaling that others have figured out for a 1G is wrong. The car runs super lean. The STFT maxed out at a little over 20 at idle.

Hex 64 (decimal 100) adder ran well. With stock (450) injectors scaled at 395, idle was fine. I don't remember seeing any STFT compensation at idle or much at all at cruise. The car was lean up top however. I'll need to do some more logging with this configuration looking at AFR and Timing to determine if I run out of tune up top due to MAF scaling or not.

Hex 8C (decimal 140) that comes stock with the evo8 ecu also ran fine. It's obviously richer, and the fuel trims show it. I did not run like this for very long, because at the time I still thought a 40 decimal adder was right so I was more interested in figuring out how to change the adder, than paying attention to how the car behaved.

Some possibly false theory below, for discussion only. Don't treat this as "advice", I would just like to know what others think.

At the end of the day, I don't think there is a "right" adder. In your MAF scaling, you can only go as low as a 0, and as high as a 255. If you need a final value greater than 255, that's where the adder enables you.

Additionally, as a tuning tool, the adder will shift your MAF's influence on Load in a different way than changing the MAF size.

Changing the MAF size which will influence Load as a percentage evenly. An increase in the MAF size that yields a 5% greater Load value in the lower spectrum, should also yield a 5% greater Load value in the upper spectrum.

Changing the adder however, will yield a greater change in Load percentage in the lower spectrum than in the upper. Another way of putting this, is that the adder has a diminishing influence on the final value as we move up the spectrum.

For example. With raw MAF scaling of 100 units at 100Hz, 110 units at 200Hz, and 120 units at 300Hz;
With an adder of 40, at 100Hz your final unit value is 140, a 40% increase.
, at 200Hz your final unit value is 150, a 36% increase.
, at 300Hz your final unit value is 160, a 33.333...% increase.

If you change your adder from 40 to 50;
With an adder of 50, at 100Hz your final unit value is 150, a 50% increase over the base 100, and a 7% increase over our previous 140.
, at 200Hz your final unit value is 160, a 45.454...% increase over the base 110, and a 6.666...% increase over our previous 150.
, at 300Hz your final unit value is 170, a 41.666...% increase over the base 120, and a 5.882% increase over our previous 160.

I have confirmed that hooking up the cruise control clutch switch on a 91 DSM enables the use of stationary rev limit, valet mode, and no lift to shift (NLTS).

EDIT: pin 43 on the EVO harness.

Is this pin 43 you are talking about?

Yes sir, sorry I thought I wrote that.

Awesome!

Englishracing.net "Arron" and I Nick Carroll

Have got another 1g to evo ecu working!

But the car is not a 1g..... It's a Evolution II clone running a 6bolt and AWD

Special thanks to:

keydiver for the harness.
Arron at english
And DanL for original Info and research

4 bar, with speed density.

-Nick


http://www.cardomain.com/ride/2429199 <--- =Build

Great Job Nick and ER! :)

Thanks,

Michael

Works pretty good in fact :)