Complements of WOT-TECH on the 60 Degree V6 Forum.

Wow, those put up some big numbers relative to things like the Ford 4.6 2V PI heads.

The intakes are literally double the flow of any stock Fiero Iron Duke casting.

[This message has been edited by KurtAKX (edited 03-29-2010).]

Nice numbers. Good to know my .610'' lift isn't a waste of effort.

quote Originally posted by joshua riedl: Nice numbers. Good to know my .610'' lift isn't a waste of effort.

According to Sappy who generated the test results the 3900 heads have the same exhaust port dimensions as the 3500 heads except for the larger valve so a mild clean up and additional clearancing might improve on those numbers some.

[This message has been edited by Joseph Upson (edited 03-29-2010).]

Up through about .300", these are no better than the iron Vortec 350 heads. But being a smaller bore, these are impressive. These intake numbers are about what you'd expect with an LS4. Which, coincidentally, has very similar displacement per cylinder.

The 3900 has the same bore and valve sizes as the LS1 and LS6. These flow numbers seem to be slightly higher than stock flow numbers I've seen for either of these LS engines. May have to keep my eyes peeled for an '06 wbody with this engine - would be a great swap into a fiero.

Cheers

[This message has been edited by TiredGXP (edited 03-30-2010).]

quote Originally posted by Isolde: Up through about .300", these are no better than the iron Vortec 350 heads. But being a smaller bore, these are impressive. These intake numbers are about what you'd expect with an LS4. Which, coincidentally, has very similar displacement per cylinder.

And if I only stick it in half way, mine is no bigger than yours!

quote Originally posted by joshua riedl: And if I only stick it in half way, mine is no bigger than yours!

Why would you put this innuendo into this thread? Some people can do it and it's funny, but this didn't work.

I was trying to figure out why you posted what you did? Don't try and take the high road when what you posted added as much to this thread as what I posted. Maybe you need to reread wht you posted and ask yourself how other people would take it?

Fair enough. Mine is telling that these stack up well when compared to some well-known heads that are widely regarded as decent-performing heads for the price. The poster you should question is whomever threw in the miserable 4.6L Ford Modular V8 heads. Those heads are universally known to have inadequate intake port flow, stock or ported. Let alone the pre-PI versions. The Mustang guys get good results from them anyway, but that's off topic.
Besides, mine was relevant. Innuendo was not.

[This message has been edited by Isolde (edited 03-30-2010).]

Ok, so what is a good cylinder head so I can compare to those?

I'm not clear what you're wanting to know. These 3900 heads far surpass the Buick 3800 series 2 heads, to the point I'd give up thinking about the series 2 except that it's cheaper and easier to get. Seeing these numbers, I'm wondering how a 3900 would do with a cam matching the specs of Thunder Racing's popular TR220 cam for the LSx V8s. I'd expect about 290 rwhp, manual trans, assuming headers and a tune, of course.

I guess I'm just reading you wrong or you are shining me on. You used the phrase "no better than," opposed to "as good as." Then you went on to talk about other poor cylinder heads. But now you say that the 3900 is good. I was just wondering what you considered a good head to compare to the 3900.

I see. Okay, I'm first realize that bore diameter is huge when it comes to airflow. The Vortec 350 is the standard reference head for all GM lovers, but it has 4" bores. The 3900 has smaller bores, and less displacement per cylinder. So while the 3900 low-lift flow numbers are nearly the same as the Vortec numbers, the fact of what's under the heads makes a major difference.
The stock LS4 heads are great for what they are, they put to shame everything that ever came before them, but the aftermarket managed to improve on them. They found small improvements by porting, and large improvements by designing entirely new heads. Now, no company is about to design new heads for the 3900, it just wouldn't sell well enough.
But look at the flow versus displacement-per-cylinder, the 3900 is way beyond virtually every 2-valve V6 known to man.
If you want to compare these heads to AFR's LS1 heads, these aren't close. Won't be even with good porting. What these should be compared to is stock LS4 heads. Because these far surpass stupid Ford 4.6 2-valve heads, even the ported "PI" versions.

There's probably no real benefit to comparing them to heads from engines they can't be used on or anything other than a V6. They were designed from features found on the LS1 and LS6 cylinderhead, mostly LS1. Considering the 2006 SEMA show HO version that according to GM has improved cylinder head flow, you can get more out of them. The most imporant note is that GM did a good job on them at the starting line to a point where they can be left as is and still be appreciated.

[This message has been edited by Joseph Upson (edited 03-31-2010).]

Upson, the second part of what you just typed is true, but the first part is not. Noone has a vague clue what the 3900 is capable of, or what we should expect of it if we modify it. That's the point of putting the heads on the flow bench. Now we know the 3900 is a lot like 3/4 of the LS4, so we can expect the results to be 3/4 of whatever the LS4 guys get. That's helpful. There is the benefit of comparing these. Now do you see?

quote Originally posted by Isolde: Upson, the second part of what you just typed is true, but the first part is not. Noone has a vague clue what the 3900 is capable of, or what we should expect of it if we modify it. That's the point of putting the heads on the flow bench. Now we know the 3900 is a lot like 3/4 of the LS4, so we can expect the results to be 3/4 of whatever the LS4 guys get. That's helpful. There is the benefit of comparing these. Now do you see?

I'm not hypothesizing in the above. In order to get 270 horsepower out of the High output 3900 GM built it was specifically stated that increased cylinder head flow among other things was part of what they did to achieve that. So unless the production heads and the ones used to generate the above flow sheet are now being made to the exact same specifications as those used on the HO motor, one can reasonably conclude that there is more left in the heads with a little work unless GM lied. GM litereature also states specifically that the LS1 and LS6 cylinder head characteristics were used in the design of the 3900 cylinder head and points particularly to the LS1 valves and combustion design similarities. They make no mention of the LS4. Everything I typed in that post can be supported with GM documentation.

quote Originally posted by Joseph Upson: I'm not hypothesizing in the above. In order to get 270 horsepower out of the High output 3900 GM built it was specifically stated that increased cylinder head flow among other things was part of what they did to achieve that. So unless the production heads and the ones used to generate the above flow sheet are now being made to the exact same specifications as those used on the HO motor, one can reasonably conclude that there is more left in the heads with a little work unless GM lied. GM litereature also states specifically that the LS1 and LS6 cylinder head characteristics were used in the design of the 3900 cylinder head and points particularly to the LS1 valves and combustion design similarities. They make no mention of the LS4. Everything I typed in that post can be supported with GM documentation.

I'm not refuting any of that. But you're missing the point. First, the LS4 head IS the LS6 head, but without the hollow stems for the same-dimension intake valves, and the sodium-filled exhaust valves. The trick valves were only used in the LS6, but the LS4 valves don't affect flow. The LS4 having smaller bores hurt flow.
GM didn't need to have the 3900 heads Extrude-Honed, they could have just installed a decent cam. But GM stupidly insists that all their engines have at least 20 degrees of non-overlap at .050", so GM installing a decent cam will never happen. Those days are 40 years past. If GM didn't want to change the cam specs or add boost, their only option was Extrude-Hone.
But since enthusiasts aren't limited to GM cams, we can do better. This being the case, the factory HO prototype isn't of any use to any of us. Not as something we can get from the salvage yard, because we can't, and not as something we should reference, because it's not a good way for enthusiasts to go.

[This message has been edited by Isolde (edited 03-31-2010).]

quote Originally posted by Isolde: I'm not refuting any of that. But you're missing the point. First, the LS4 head IS the LS6 head, but without the hollow stems for the same-dimension intake valves, and the sodium-filled exhaust valves. The trick valves were only used in the LS6, but the LS4 valves don't affect flow. The LS4 having smaller bores hurt flow. GM didn't need to have the 3900 heads Extrude-Honed, they could have just installed a decent cam. But GM stupidly insists that all their engines have at least 20 degrees of non-overlap at .050", so GM installing a decent cam will never happen. Those days are 40 years past. If GM didn't want to change the cam specs or add boost, their only option was Extrude-Hone. But since enthusiasts aren't limited to GM cams, we can do better. This being the case, the factory HO prototype isn't of any use to any of us. Not as something we can get from the salvage yard, because we can't, and not as something we should reference, because it's not a good way for enthusiasts to go.

Not missing a point, I just see none in comparing V6 attributes to V8 attributes regarding cylinder heads given the number of V6 options to compare them to in a V6 swap decision. The HO engine is referenced as proof of an example to show that the flow numbers posted above can be improved on, GM also had the camshaft reground, and installed LS6 valve springs. You don't need a production HO motor to achieve those benefits. I had my camshaft reground and I installed LS6 valve springs and as an enthusiast I believe referencing that motor was a very good way to go especially considering my camshaft probably has a lot more duration and a narrower LSA than what they settled for. I'd be very surprised to find that port work on the exhaust side wouldn't improve flow. At some point I'm sure there will be another flow chart for a ported set as soon as Sappy gets around to it. In the meantime it's nice to know that these are probably the best flowing stock 2 valve Domestic V6 heads in production and that's the most important point in this thread for those considering a V6 upgrade. A 40 hp difference with just a .1L increase in displacement compared to a naturally aspirated 3800 is pretty good. Heck the 3500 only falls short of the 3800 by 10 lb/ft.

cool.