New VVT and VI Control Theory using common ECMs

[robertisaar]

now, this came to me as i was about to fall asleep, but here it is: since the 3900 doesn't use an EGR of any type(IIRC), how about using the CCP output in place of it? it's PWM and should be easy enough to massage into what we would need...

and for the variable intake: any one of the EGR outputs. just have to set up the EGR solenoid combination table so that the rest of the solenoids aren't attempted to be used. then at specific RPM/kPa, it will open the short runners...

sound plausible?

[mickaz]

you see fucked up thing when you sleep... But it makes some sense.but we need a test mule.

[robertisaar]

Joseph! We needs a tester!

[Joseph Upson]

Thought I posted it here before leaving home the last visit but I guess not. I did 3 test runs with the GTech module installed before taking down the old turbo outfit for a baseline to compare to after the revamp and also to see what effect advancing and retarding the camshaft to full stop would do. For starters I watched a video of track testing of the module to see how accurate it is conducted by the creator and by an unassociated customer who compared the Gtech results to the Dyno results for his car and the numbers were staggeringly close to actual test results, so it is a very good measure of performance.

I did 3 runs, 2 with the cam fully advanced yielding an untuned best of 202 hp @ 4305 rpm and 250 lb/ft @ 4108 rpm to the wheels on 7 psi. It's actually more because I under estimated the vehicle weight by more than 200 lbs when I programmed the module. The GTech also does net readings which are typically lower than Dyno readings. As you can see by the rpm peaks the engine was limited from its maximum power levels which should have been around 2000 rpm higher as a result of the cam position.

With the cam fully retarded power was miserable, 186hp @ 5388 rpm and 197 lb/ft @ 4729 rpm. So as CNCguy suspected sometime ago there is a happy medium neither of which lies at the positive and negative extreme. I would say the best you could get out of VVT without the OE PCM is to do what I did, set the cam stop to the proper install degree angle for the cam which is 112 deg for my cam and allow it to advance from full advance or near full advance to pick up a little extra bottom end than what you would have if you locked the cam at the designated lobe centerline. Note also that the non VVT engines usually produce more power than their VVT equivalent such as in the LS series motors. In other words anything more than a simple on off cam retard would be a waste of your time. The 3500 VVT and non VVT don't count because they have different strokes.

I expect to see numbers above 300 at the wheels with the cam at the proper angle especially since I measured 20 deg advanced which has to kill peak power possibly to the tune of 1000 rpm loss per 10 deg advance which makes sense given the corralation when you add 2000 rpm to the hp peak.

As for the intake, just install the non variable version and be done with it.

Desktop Dyno is also close to the output numbers when you look at the corrasponding rpm points although it shows the peaks arriving at a higher point. I'm sure if it had been tuned it would have put down a bit more but I recall being a bit conservative up top with the spark table.

[Joseph Upson]

Thought I'd let you guys get a peak at the wheels, 18x8 and 18x9. The car will be getting a gloss black paint job once the engine is done. I figured if I'm going to do it I may as well finish the forged build with lower compression than I intended and get the car painted last. I can come back later with a thinner Cometic head gasket to get my compression up close to 11:1 and will not have to remove the engine from the car to do it. It will be lowered about an inch or so and I will be keeping the front mount intercooler.

[robertisaar]

i would MUCH rather just keep it the way the factory indended, and use the PWM capability of the CCP driver to hold it in it's best range based on RPM. i would also like to at least have on/off control of the intake valve as well(i believe the factory control is PWM according to some documents i've run across).

if GM can pull 240/240 with it(and a nice flat torque curve), i'd like to think i could do the same without having to worry about emissions.

[Joseph Upson]

i would MUCH rather just keep it the way the factory indended, and use the PWM capability of the CCP driver to hold it in it's best range based on RPM. i would also like to at least have on/off control of the intake valve as well(i believe the factory control is PWM according to some documents i've run across).

if GM can pull 240/240 with it(and a nice flat torque curve), i'd like to think i could do the same without having to worry about emissions.

Impossible without reading the stock crank position ring and cam position sensor. You would need to know those values at all times not to mention that oil temperature and viscousity will vary having an effect also as well as any change to the cam specs. Without the stock PCM there's no simple way to hold a cam position between the minimum and maximum stop. The stock cam will provide the best output when degreed in as I've mentioned already the non VVT cam install has more potential than the VVT cam which relies on a wide LSA for its variability to work properly and surrenders power as a result.

[robertisaar]

i'm thinking if there are a few free bytes of RAM, a little bit(probably not so little) of code patching can be used to read the existing crank and cam sensors calculated locations to account for variability that would occur with different oil viscosity, pressure and temps...

[Joseph Upson]

i'm thinking if there are a few free bytes of RAM, a little bit(probably not so little) of code patching can be used to read the existing crank and cam sensors calculated locations to account for variability that would occur with different oil viscosity, pressure and temps...

The broad torque curve is partly the result of the wide LSA, given my experience with the engine, if I were going to put the kind of effort you are considering into it, I would switch to the 3.6 DOHC and install the complete system with stock PCM.

[robertisaar]

to me: coding is a lot less expensive, and is something i can work on in short bursts if necessary, not something i could do with the 3.6.... which BTW, if the 3.6 wasn't so rare and prohibitively expensive for now, i would likely do the same thing...

but for now, my brain matter outnumbers my cash by a LOOOOOOOOOOOOOONG shot. i may as well use it to my advantage.

[90BlueZ]

Why are you guy so addicted to the stock ECM ?

Megasquirt 3 + MS3X can control VVT and a SHITLOAD of other thing super easily.

[robertisaar]

least amount of added-on and unnecessary stuff possible = least amount of things that can possibly go wrong.

the MS brings up an interesting idea... could it reliably control the VVT and VI functions as described here? and how much work would be required to do so? i'm not looking for it to control fuel or spark in any manner, just being able to reference the cam sensor, crank sensor(s) and whatever else is necessary to dial in an exact amount of advance/retard consistantly.

[Joseph Upson]

Why are you guy so addicted to the stock ECM ?

Megasquirt 3 + MS3X can control VVT and a SHITLOAD of other thing super easily.

It's not so much an addiction to the stock ECM as it is a proven time after time relaible piece of equipment. If I'm driving down the road 500 miles from home (which I am) and my Mega Squirt burns out, I can't reach under the seat in an instant and pull out a new one, neither can I stroll over to the nearest salvage yard and purchase one for as little as $15, transfer the chip and keep moving. I have actually had this happen with a stock ECM which I smelled burning and did just as I stated above and kept rolling after a brief stop. Reliability is extremely important for me because most of the time I'm hundreds of miles away from my major repair tools and need flexibility of easily obtainable replacement parts for a daily driven car.

The Mega Squirt sounds great especially if it REALLY can control VVT as you've stated but it's not an acceptable risk in my situation nor do I believe VVT on a cam in block engine is worth it. The DOHC engine is an entirely different story.