Knurled wrote: I'm going to say calibration error... looking at the post, there's no way any engine is going to run at 3" of vacuum at WOT and low engine speed. I don't care what kind of turbo is on it.
That's exactly what I think.
It's very likely that two of the same sensors are running with the same calibration, which would give the same results under the same conditions. Doesn't mean the numbers are correct- just that they are the same.
I'm thinking maybe the sensors are plumbed wrong, or there's bad math going on. Or maybe it's a relative measurement where it's reporting pressure in the intake tract relative to the intake manifold, mixed with some sensors reading absolute, some gage, and maybe one in inches Hg and and another in psi thrown in to really mess things up.
The one constant of the universe is that it's impossible to have 30 pounds of vacuum. Argue all you want, but it can't and doesn't happen without the above said problems making it appear that way.
In reply to yamaha:
Ah bike engine, makes sense now!
As for the saab alfa fight? Pure gold.
Alfadriver is 100% right on all that stuff. It's not just an issue of using psi vs inhg, it's an impossible scenario with either unit. The only way you have 30 anything at idle is kPa (kilopascals).
alfadriver wrote:Knurled wrote: I'm going to say calibration error... looking at the post, there's no way any engine is going to run at 3" of vacuum at WOT and low engine speed. I don't care what kind of turbo is on it.That's exactly what I think. It's very likely that two of the same sensors are running with the same calibration, which would give the same results under the same conditions. Doesn't mean the numbers are correct- just that they are the same.
A thought.
We're roughly 800ft above sea level, and atmospheric pressure here (according to the map sensor in every engine controller I've played with calibrated in kilopascals) is 92-94KPa, approximately, depending on the weather.
94kpa is 13.6psi. If 14.7psi is assumed to be atmospheric, then I'd have to make 1.1psi boost to equal "atmospheric" on a GOOD day.
Just another reason why boost pressure is not really important, and absolute manifold pressure is what really counts.
I am curious to see what NordicSaab's pressure readings are KOEO. That is the true baseline, boost pressure is the pressure readinga above THAT. I have a feeling that they would be in the neighborhood of "-1.6psi".
(*and yes I broke out the royal "we", dangit!)
In reply to Knurled:
I don't get the impression he will be coming back. Just trying to teach the- "does the data make sense?" thing. Have this discussion with modelers all the time, and reminding them that in spite of model, the car is always right.
The kPa confusion kind of makes sense, but 45 psi of boost would end up being 145 kPa, and not a whole lot of boost. And I don't doubt the car is powerful and fast.
Seems that the slope, offset, and quad that is being used is just wrong. And now that it's calibrated around the wrong answer, well, it's done.
Ayup. I went for many years with my injector modeling completely wrong on my RX-7's computer. I knew it was wrong, but I didn't care because the computer gave the engine the fuel that it needed. I had to fudge it once when I had a hardware failure and I just sort of kept tuning around that fudge. So what if it was showing up to 130% duty cycle? Ir worked just fine...
Until I changed hardware and injectors. Then it was just all FUBAR and I had to start all over again from scratch instead of just saying "okay computer, you don't have 680cc injectors anymore, you have 1000cc now" and the computer going "Ryokai!" (japanese cars...) and everything is fine. When there's a baked in .5ms pulsewidth error, the whole fuel map is FUBAR because it is nonlinear.
One problem the DSM guy might be having is that he may have all his sensors in the charge piping, and he isn't actually reading manifold pressure but pressure somewhere between turbo and throttle plate.
That and his complete cluelessness about his units of measurement and the concept of how air works 
You'll need to log in to post.
