200 CFM Research

[jsa]

[jfholm]

[jsa]

[jfholm]

Good idea John about putting the data into EAP as it has a provision for turbocharged engines or roots superchargers. I may have to have you or Tom give me some information about turbos. I will let you know what I need. That will be interesting and I had not thought of that, thanks.

John

[200cfm]

Thanks guys for the discussions. I want have the new heads ready til later in the year. Busy with home improvements and gardening. And new short block is still in the shop. John, let me know what data you might need for EPA. I have Dyno Sim and it has a menu for superchargers and turbos but the turbo selections are limited versions and no twin turbo combo designs. Have plans to upgrade the turbos to a S trim wheel also.

SuperChevy show is in town this week. (tom)

[emarsh]

[emarsh]

After reading some more posts discussing porting for turbocharged engines I thought I'd do some speculating.

The reason that higher intake port velocities are desirable is to continue to fill the cylinder after the piston has changed directions, increasing VE. In a supercharged engine the additional pressure of the intake charge does the same thing to a certain extent, though I guess that in this case port flow becomes more important because a restrictive port will cause a greater pressure differential between the inlet side of the head and cylinder which would require longer cam timing to let the cylinder side "catch up" . I guess that as is the case with a non-supercharged engine there is a "balancing point." How boost affects that point is an interesting topic for research. Someone also stated that big ports will kill off-boost performance. My response to that would be "so what?" When a turbocharged engine is off boost it's a pig and the relative effect of port velocity is probably negligable. Especially with a big turbo the trick is just to get on boost ASAP. For drag racing this means loading the converter, blipping the throttle rapidly or using a two-step. There have been a couple of times when I've launched my bike without boost and it just bogs until the boost comes on.

As for velocity not going up as pressure goes up, I'm finding that to be somewhat confusing. With more boost a greater volume of gas is being moved but not necessarly at a higher speed. However, as was pointed out, because it has more mass it has more inertia and as such there may be something like an equivelance between the two.

Finally, there is the question of how much porting really helps a turbocharged engine. Intuitively I'd think that porting would be a good thing, but I guess that it's necessary to determine how much greater the flow requirements are as boost goes up. This I don't have a clue on but I think it's a good topic for research.

[jfholm]

I ran the turbo tests in EAP. I set it up with 7.5 compression and twin turbos. It did make a difference to have bigger better flowing heads, but not nearly as much as I thought. With my SBC with 194 cc intake heads flowing a max of 263 cfm the peak torque was 745 ft lbs at 4000 rpm. The peak HP was 629 HP at 5500 rpm.

2nd test the only change I made was lowered the intake volume to 165 cc but left the flow the same as I just wanted to see the effects of smaller port. Now I got 735 ft lbs tq at 4000 rpm and 612 HP at 5000 rpm.

Then I modeled the engine with stock SBC heads that had a 165 cc intakes and a realistic head flow that was 203 cfm instead of 263 cfm. Now max torque was same as big heads 745 ft lbs at 4000 rpm but max HP was down to 605 HP at 5000 rpm.

One thing I did notice was mach # on small heads was .405 mach at peak torque and .557 mach at peak HP of 5000 rpm - with the big heads peak torque mach was .367 and peak hp was around .497 mach.

now from my Engine Analyzer manual:
MACH#:
The 1979 SAE paper "An analysis of the volumetric efficiency characteristics of 4-stroke cycle engines using the mean inlet mach number, MIM", 790484 by Fukutani and Watanabe is the basis of this calculation. It was an extension of the Mach Index characteristics first identified in the 1940s by C.F. Taylor and co-workers from MIT, "The Internal Combustion Engine In Theory and Practice", C.F. Taylor, 1985, MIT Press.

These papers state that an engine's air flow potential per cycle (volumetric efficiency) depends on its average intake flow coefficient, intake valve flow area, cylinder size, speed of sound in air and RPM. These five terms are combined into one value called the Mach Index, called Mach # by the Engine Analyzer Pro. In simple terms Mach # relates the average velocity of the intake charge past the valve to the speed of sound. The speed of sound is theoretically the maximum velocity possible past the valve, which would give a Mach# of 1.0. A Mach# of .4 states the average velocity is only 40% of the maximum possible velocity.

Taylor's work showed good correlation between volumetric efficiency and Mach# for several engines with conservative cam timing. The correlation showed that volumetric efficiency (and therefore power) would start to drop sharply when Mach# increased above aproximately .55. However, more recent studies show poor correlation if intake cam duration increases significantly. The 1979 paper includes a correction for intake duration; the higher the intake duration, the lower the Mach#, and the higher the RPM for peak volumetric efficiency.

General "rules of thumb" concerning the Mach# include:
1-Peak volumetric efficiency should occur in the range of .3 to .5 Mach# with no tuning effects.
2-Volumetric efficiency drops rapidly in the range of .6 - .8 Mach#"

btw: You can download the complete Engine Analyzer Manual and Port Flow Analyzer manuals from the Performance Trends web site. The are very informative readings.

John

[emarsh]

John,

That's very interesting and something I want to investigate further. But the one thing that hasn't been discussed at all is the exhaust side. Can you model the effects of different exhaust flow rates, valve sizes and so forth for the exhaust? Also, on your previous post you didn't mention the amount of boost. Is that adjustable in your software and if so what effect does it have?

Finally, I want to ask what some might consider to be kind of a dumb question. You have to understand that it's been a couple decades since I was heavy into engine development and things have obviously changed since that time.

I see a lot of discussion of the port volume. Seems to me that port volume will vary significantly between heads and in some cases between different runners on the same head because of the port's length. How can you compare the volume of a SB Chevy and SB Ford, for example? Is the port volume a normalized number in some way?

[emarsh]

[jfholm]

[emarsh]

I've been thinking about the differences between porting for a turbocharged engine since I last posted and have come up with a few thoughts that I thought I'd toss out for feedback.

First of all velocity.

In a normally aspirated engine port velocity is important because up to a certain point the inertia of the charge can have a greater dynamic pressure than that within the cylinder as the piston moves back up. The value of this is that the closing point of the intake valve can be delayed long enough to trap that extra little bit of charge and when things are tuned correctly can push the VE above 100%. When the engine speed is outside of the tuned range the longer intake duration causes part of the charge to be forced back into the intake track causing all sorts of negative effects which essentially cause a lumpy idle and poor low speed performance.

I was so used to thinking about port velocity and trapping that extra little charge that I overlooked the obvious. In a turbocharged engine it's not necessary to try to trap that extra little bit of charge. Want more cylinder filling? Turn the screw up a bit more (i.e. increase the boost). In general there is no negative to reducing port velocity (and creating artificial, dynamic intake pressure) when you can achieve the same thing by increasing static pressure. As an added plus shorter duration intake cams can be used which will widen the power spread.

If you have looked at some dyno runs from turbocharged engines with moderately sized turbos you would see that this is backed up to a large extent. I've seen plenty of dyno charts where peak power is flat, extending all the way from 6000 rpms (or so) past 10,000. With a large turbo things are a little different though.

Something else that has not been brought up, and I don't know how relevant it is, is friction on the intake tract. A larger port will have reduced friction losses. How significant they are I don't know but it is something to think about.

Then on the other end of the intake cycle there's overlap. In a normally aspirated engine longer overlap not only helps to scavenge the cylinder, it also helps to get the intake charge moving sooner, assuming that the exhaust tuning is right.

On the turbo long overlap is a bad thing because the increased intake tract pressure causes more fuel/air to be forced unburned into the exhaust. In the best case this just wastes fuel. In the worse case the burning mixture shortens the life of the impellers. But then again with the added pressure there is a greater pressure differential between the intake tract and the now empty cylinder which should get the intake charge moving faster without the help of the reflected wave during the overlap period. Besides, turbos don't have reflected waves the way that an open exhaust does and so can't benefit from them. So there is a reduced need for overlap, though a certain amount is still necessary to scavenge the residual burned charge.

One really big question about the exhaust side on a turbo is how much back pressure there is. I haven't the faintest, though I would be very interested in knowing. Also, is there any benefit or disadvantage to kicking up exhaust valve size and flow? Again, I don't know but I think this is a good question and one that's worth looking into.

Well, that's it for now. I might be on track or I might be off the track but until I hear a compelling argument (or flow results) to the contrary I'm going to assume that this makes sense.

[jfholm]

Well the other night when I ran the comparison between stock heads and ported heads and I said it did not make as much of a difference as I thought, well I am embarrassed to say that I made some errors. When I ran the test on the stock heads I forgot to lower the compression down from 10.5. I corrected that error and made sure I was more accurate on everything. the results are below:

Stock SBC head no porting with 1.94 int valve and 1.5 exhaust valve: 165 cc intake ports

with turbo 7.5 to 1 CR without turbo (n/a) 9.7 to 1 CR
peak HP 592@5000 371@5000
average HP 380 254
peak TQ 727@4000 442@4000
average TQ 497 362

With my porting 193.5 cc intake port Dart Iron Eagle heads 2.05 int and 1.6 exhaust

with Turbo 7.5 to 1 CR without turbo n/a 9.7 to 1 CR
peak HP 657@5500 439@6000
average HP 413 286
peak TQ 764@4000 461@4500
average TQ 525 387

This is just a simulation in EAP, but I have had fairly good results with it over the years. Still the proof in the pudding would be on the dyno or on the drag strip. Unfortunately I don't have a dyno or the money to do real live testing.

John

[emarsh]

So proportional to the amount of horsepower being made the turbo did not respond as well to the better heads as the normally aspirated engine did. Interesting. I wonder how well the software models the real deal.

[jfholm]