3900 Info

[CNCguy]

Here are some shots of the timing chain cover from the 3900 compared to the Gen3 cover. As you can see the center is raised quite a bit to clear the VVT unit, the water pump has been relocated, the 3 bolts that are used for motor mounts on some engines has been reduced from M12 to M8, 2 other bolts were changed from M10 to M8 and the water passages have been rerouted around front of cover instead of over the top.

Top view of new vs. old


Side view of new cover


Red dots show the M12 holes that changed to M8

[Joseph Upson]

If I recall correctly and I believe I do, that bolt hole change is probably back to what they were on the 2.8L because when I installed the Fiero 2.8L intake and heads on a 3100 I had to increase the diameter of the same three holes in the Fiero timing cover and water pump so that I could bolt it down.

KEEP THE PICTURES COMING DON'T BE SHY, MORE, MORE, MORE.

[wasas9]

mmm. yummy....

[MidnightriderZ24]

does it look like the timing covers are interchangable at all?

[CNCguy]

does it look like the timing covers are interchangable at all?

They are interchangable.... BUT... the old style cover will not clear the VVT hydraulic actuator on the front of the cam. Even without the actuator, the end of the cam extends about 3/4" from the block. I am hoping to build something in the very near future to eliminate the VVT for use in applications that are space restricted.

[Superdave]

Actually, we mounted a 3500 head to the 3900 engine and it appeared that the holes were in the correct locations. The problem is that the combustion chamber is not centered with the bore. GM moved the cylinders out 1.5mm on each side, making the block and heads wider. Whether it would be capable of running that way, I do not know. You would want to keep the 3900 top end anyways.... much bigger ports and valves.

If you put a 3400 or 3500 head gasket on the 3900 head you could see any possible interfearance issues. 1.5mm isn't much.

[Fiero11]

I wounder if the 3.9 heads with the 3.9 intake will mount to the old version of the 3.5 That could really help. The head/intake and intake/block interfaces would have to be modified.

[Joseph Upson]

I wounder if the 3.9 heads with the 3.9 intake will mount to the old version of the 3.5 That could really help. The head/intake and intake/block interfaces would have to be modified.

Pistons would probably have to have reliefs also for the larger valves, I can't find where I wrote the info down but I just performed a rough measurement of the valve to piston clearance at top dead center and I believe it was about .140 int .133 ex so with the new valve location and size that might be a problem for the 3500 and since it was stated that the valves are further apart they may even be to close to the cylinder wall.

[CNCguy]

since it was stated that the valves are further apart they may even be to close to the cylinder wall.

That was a concern that Ben had when we were checking it out. Only way to know for certain would be to drop a head on a 3400 and see where everything is. I still haven't found my valve spring compressor... when I do, I will put a 3900 on the 3400 block that I have laying here and let you know what I find out.

[CNCguy]

I put the 3900 head gasket on an old 3100 block and you can easily see the amount GM shifted the bores. All the holes line up correctly. The 99MM bore gasket looks huge compared to the 89MM block. Later this weekend, I plan to throw the 3900 head on the 3100 and check the valve to cylinder wall clearance as well as check the valve to valve clearance and piston clearance on the 3900.

[Spankdamonkey]

Hmm....just a quick question here, did GM change the locations of the coolant passages on the 3900 heads since it utilizies the U-flow design? I would think so, but can't be sure.

[CNCguy]

Hmm....just a quick question here, did GM change the locations of the coolant passages on the 3900 heads since it utilizies the U-flow design? I would think so, but can't be sure.

The 3900 heads do not have the coolant passages to the LIM, instead the coolant exits through the end of the head, eliminating the common LUM leak issue. Since the R&L heads are symetrical, the head gaskets were designed different to direct the flow of coolant. The 3900 uses RH & LH head gaskets and the gaskets have small holes all over except the hole at the rear of the engine, it is larger to force the water to flow to that end of the engine. If you put the gaskets on the wrong side, the rear of the engine would overheat.

Basically, the head gaskets act as orifices to control coolant flow throughout the entire engine.

[427v8]

Well I am pretty excited about the 3900.
It seems like it would be easy to use the VVT, DOD and Variable length manifold with just some simple switches.

GM made it completely transparent to the driver, but I think it would be cooler if you had a switch at say 2/3 throttle that engaged all 6 clyinders, talk about your kick in the pants!

A little more complicated would be using the VVT, but it could be as simple as measureing engine RPM and switching it on and off maybe controlling the intake the same way. A PIC MCU or maybe the megasquirt could do it

Either way I really want one even if I have to use a manual switch to do the deed.

[Fiero11]

There are many of the old style 94mm 3.5L engines out. Can you put a head gasket from a 94mm 3.5L on the 3.9 head? I would like to see if the 3.9L valves would interfere with the 3.5L cylinder walls. Since the valves are angled the block could be knotched for small interferances if needed. 3.9L heads and intake on a 3.5L engine could be a real performance improvement. The intake/head intersection would need some machining due to the 1.5mm offset of the 3.9L block. I can send you the 3.5L gasket or money to buy one if you need one. Thanks.

[smilinguy99]

I'm really curious about the operation of the VVT. It only uses two wires to control it. If anyone has any info on it's electrical operation, I would appreciate the input.

I had a chat with the supervising engineer of the VVT and he said the thing is controlled with pulse width modulation. Basically it uses oil pressure to actually move the cam back and forth, but the oil pressure is turned off and on by an electrical valve controlled by a stream of off and on pulses that vary in their ratio of OFF times to ON times. It is ultimately controlled by a map in the ECM. Kind of like a fuel injector is controlled.

The gizmo has a LOT of authority over cam timing. The cam timing can be varied by a total of 27 camshaft degrees in relation to the crank.