3900 Info

[Joseph Upson]

I mentioned in another thread the possibility that the forged floating pistons from the 3.6 DOHC engine might possibly work in the 3500 since the 1.3 mm larger stroke and the 10.2:1 compression could mean they didn't change the piston pin location causing the higher compression, anyway I acquired a burnt 3900 today for a good price and on breaking it down discovered that it has floating MAHL pistons like the 3.6L although I don't know if the pistons for the 3.6L are made by the same company.

[Guest]

Hey Joseph, send me an email as I dont' have it anymore. Shaun41178@hotmail.com

YOU still have that fiero in Jax?

[427v8]

for anyone interested, I think I came up with a non destructive way to lockout the cam to disable the VVT.

It woould take a little trial and error to figure out where to put the cam but thats always the case I guess.

Anyway, why not just cut a couple pieces of aluminum to block the vanes on the phaser so they can't move, 1 on either side and it's not going anywhere.
Make them different sizes to adjust the cam timing.

cheap effective and won't hurt anything.

[Joseph Upson]

for anyone interested, I think I came up with a non destructive way to lockout the cam to disable the VVT.

It woould take a little trial and error to figure out where to put the cam but thats always the case I guess.

Anyway, why not just cut a couple pieces of aluminum to block the vanes on the phaser so they can't move, 1 on either side and it's not going anywhere.
Make them different sizes to adjust the cam timing.

cheap effective and won't hurt anything.

Me and CNC have been chatting back and forth about the subject for a while via PM, and we have arrived at several good ideas so in the future I'll post here for all to see what we have come up with.

The easiest solution to what you mentioned above is this; remove the spring from the pressure deactivated lock inside the actuator and install a still rod so that it can't be released when the oil pressure reaches the appropriate level, then remark the timing sprocket for the best cam angle. If you can't get within 2 degrees of that mark due to the sprocket teeth not lining up with the chain at the exact location then manipulating the vanes might be in order. There is a pressure relief valve built into the VVT actuator so it shouldn't cause a problem.

I discussed with CNC the possibility of an on off application of the VVT with a limited range, coincidently I suggested to him the exact same range he had already suggested earlier in this thread. +10 to -5 I believe it was.

I have also located a cam manufacturer that will resize the journals on the non VVT cam so that it will fit the 3900 properly for ~$127.

Schneider racing has agreed to mark the cam sprocket for the best location with the OE cam with a regrind.

[RacerX11]

I mentioned in another thread the possibility that the forged floating pistons from the 3.6 DOHC engine......

...discovered that it has floating MAHL pistons like the 3.6L although I don't know if the pistons for the 3.6L are made by the same company.

The 3.6 pistons are made by Mahle, but they are cast, not forged as the GM literature states. The pictures of the 3900 pistons look to be cast as well.

[Joseph Upson]

The Dodge 318 and 360 have a 3.313 stroke and 3.910 and 4.040" bore, due to the popularity of those engines I suspected forged castings would be an easy find at a good price and sure enough they are for as low as $52.99ea at summit, unfortunately to use the piston/rod combo at a minimum the 3900 would have to be bored which is not a problem and the crank would have to be turned down due to the Dodge rods smaller diameter. I'm not sure about the rod width however Dodge also has a 3.9L that might be even more favorable, I just couldn't readily find the specs on it.

318= 3.910 bore, 3.313 stroke, .984 pin, 2.125 rod big end, 6.123" rod.
3.9L Dodge, not sure but I believe the bore is 3.9"

If it's possible it would cost close to a custom set of forged pistons except in the end you have increased displacement and depending on deck clearance possibly a little offset grind for even more displacement since the crank would need machining any way.

GM mentioned enhancements that would allow future increase in displacement for the 3900, why wait for them.

[Nightingale]

Whats availible to tune the 3.9L PCMs? I am pretty sure they are using CAN on those PCM's and with added VVT wasn't sure what Tuner software would work. I am starting to think that with want I want to do it might just be easier to do a complete 3.9L driveline and computer swap and aftermarket guages. Atleast then if I can find a tuning solution I could keep the vvt and manifold solenoid functioning properly.

[Joseph Upson]

I'm using the 727 or 730 ECM with modified TGP programming for my setup to be tuned with Zeitronix wideband O2 datalogger and Ostrich realtime emulator.

I'm going to give the VVT cam a try after a regrind and use a basic stamp to read engine rpm and pulse the modulator linearly with rpm for a rudimentary approach. As long as the cam is retarded to the maximum effective point at wide open throttle I'm good to go.

Below are 2 screen shots, one with the VVT cam at 0deg compared to -15deg, and a single graph of the stock engine on 7psi, with 72% turbo efficiency and 80% intercooler efficiency because I'll be using both an intercooler and water/meth injection.

Note in the comparative graph how close the peak power and torque output is to the 3900s actual rating. I'm using 3500 head flow numbers and that probably has a lot to do with why the HP is down.

[onefastV6]

I was looking around and something has intrigued me, and may be a possible solution to eliminating the VVT setup. Apparently the 07 Impala is equipped with a 3.9L LZ8 engine that appears to be setup as displacement on demand rather than a VVT setup.

This is how it apparently works:
"Valve lifters are operated by the engine’s camshaft, and lift a pushrod that operates the valves in the cylinder head. In the 3.9L, the De-ac lifters are installed in cylinders 1, 3, and 5, while the remaining cylinders use conventional lifters. The hydraulically operated De-ac lifters have a spring-loaded locking pin actuated by oil pressure. For deactivation, hydraulic pressure dislodges the locking pin, collapsing the top portion of the lifter into the bottom and removing contact with the pushrod. The bottom of each De-Ac lifter rides up and down on the cam lobe but the top does not move the push rod. The valves do not operate and combustion in that cylinder stops. During reactivation, the oil pressure is removed, and the lifter locks at full length. The pushrods, and therefore the valves, operate normally."

So if I understand this correctly, if normal lifters were used in place of the De-Ac lifters, it would disable the displacement on demand.

There is more info located at GMpowertrain.com

[Joseph Upson]

We discussed this particular engine earlier in the thread, it's VVT also and just has the extra function of being able to shut down one bank of cylinders and is only found in one car if I recall correctly, the 07 Impala.

[Joseph Upson]

Cam Specs Are Incorrect

I just got a call from Delta camshafts regarding my request to regrind my 3900 cam to about 225/225 and decrease the LSA to about 116 down from the preivously estimated 118 degrees. I was told the cam in its stock grind has a 112 degree LSA, which means I have all sorts of things to change now regarding dyno simulations. The good thing about that is the cam can be degreed into its straight up position and the VVT mecanically locked into that position with metal pieces to hold it there so that it works like a non VVT engine. I settled on 220/220 .510 lift and 114 LSA.

Once I get the engine on the stand I'll start posting info on the swap later.

[IsaacHayes]

Nice! Getting closer!b

[CNCguy]

That's very interesting but I don't understand it. What year is your cam? I measured this cam several times because the large LSA seemed wrong. Wonder if this cam has a lobe that is out of position. If I can ever get some free time, I will measure some other cylinders to see if the numbers are the same.

[Joseph Upson]

That's very interesting but I don't understand it. What year is your cam? I measured this cam several times because the large LSA seemed wrong. Wonder if this cam has a lobe that is out of position. If I can ever get some free time, I will measure some other cylinders to see if the numbers are the same.

My cam is from an 06 Pontiac G6 however, GM parts direct shows an 06 cam and an 06-07 cam for the 3500 and 3900 so I don't know what's going on there.

I have an 07 3900 from a Chevy Uplander and its cam is listed for the Terraza and Uplander on GM parts direct from 06-07 but that doesn't tell me anything as far as whether there are two cam grinds for the 3900. Since you mention the possibility of a lobe moving, I'm afraid of using the 3900 cam I just had reground, my engine has about 3000 miles on it and loosing it to a camshaft lobe would be devistating.

I was wondering why no one mentioned it before but realized it was because probably everyone that has a 3900 for a swap has one from a car like my first 3.9. I noted on the Uplander engine I picked up yesterday two hoses going to the oil filter base which turns out is a sandwhich style oil cooler that uses the radiator fluid to keep cool.

[Joseph Upson]

I just performed an experiment with the stock 3900 camshaft. I recall reading a thread on a different forum where one of the members and owner of a G6 3900 with 6 speed manual tranny had an incident with bent valves and suspected his transmission might have had something to do with it regarding shifting problems.

I used my extra modulator with return spring and cam lock removed so that I could rotate it by hand. With the crank at TDC for cylinder #1 I got intake valve to piston contact with the modulator rotate towards the full lock position which is where it would be on initial startup of the engine. Although it is possible and likely my hand manipulation without a degree wheel may have been off a little but not enough to ignore the fact that this is entirely too close for comfort.

Here is what I believe happened to the guy that ended up with the bent valves, he probably had a higher than normal rev occur during a moment where the cam would have normally been retarded for more clearance and somehow it wasn't, like during a fit of rage due to difficulty shifting gears a momentary high rev from stomping the accelerator in anger and not enough cam retard leaving the piston and valve at their closest achievable distance (0 degrees retard) and contact due to the higher piston speed enabling it to catch up to the valve movement just enough to make contact or even worse due to valve float. It could also have been a PCM hiccup allowing the wrong combination to occur.

I'm going to block my cam into 0 degrees advance and retard from there especially since the specs are now increased to 220/220 and .510 lift which is sure to see valve contact in the stock locked position.

That cam spec should also function like a normal high performance cam at that setting and still leave me with high end VVT benefit.

I haven't spent much time with the electronic control for the VVT function but the Zeitronix WBO2 unit I have for tuning has an auxillary terminal that I can program to energize at a certain rpm and other sensor reading so I'll set it to to turn on a relay that will supply power to the cam retard module at about 4000 rpm to give me top end performance while I work on programming the basic stamp to send pulses to do it gradually with rpm increase.