Technical Theory : Plenum Design

[ikessky]

I'm saving up and going for a TIG!

[bszopi]

I plan on someday getting at least a MIG, and perhaps even a TIG. But that's beside the point...

Was bored at work, so I had a guy measure the circumference of the 3400 port for me. Came out to 5.5". I also was talking to Michael (FFP) some, and he made the suggestion to just use round tubing of the same circum. and just hammer it into the flat oval shape at the port end. Made sense, but I wasn't sure how easy it would be.

Using my vast trig knowledge, I determined that a 1.75" ID has the same circum. Decided that a 1.875" OD would probably be the size I want to use. Then, I found a piece of 2" conduit and took it to the vice. Less than a minute later, I had a beautiful flat oval, and the conduit will be much thicker than the 16ga. tubing I plan on using. So going to the flat oval from a round tube is now the plan.

With that all figured out, can anyone suggest the bext plan to go to get 1.875", 16ga mandrel bends? I looked on Headers By Ed, but they seemed kinda expensive. Summit and Jeg's both had them at a more reasonable price. Tried looking on JC Whitney, but couldn't get the site to load all the way up. Any other places I should look?

[SappySE107]

http://www.stahlheaders.com/Frame%20Tubing.htm

Possibly? Im about to pass out but thats the site I have bookmarked for header tubing. I was looking for something cheap to make up a test set of headers and thats what I got so far. Sounds good so far I may have to mess around with that idea as well but a bit different. I need to flare it out or something on one end and oval on the other.

[bszopi]

Ok, so here are a few pictures of the conduit I messed with...





And here is an Excel file that allows for calculation of intake tuning RPM by entering plenum volume, runner radius (if not round, figure out the surface area, then determine radius of a circle with the same surface area) and runner length. The equation is based off the Helmholtz resonator equation, so it normally would result in a number in Hz. I converted from Hz to RPM by multiplying by 20 (assume 3 cycles/rev and 60sec/min).

http://www.60degreev6.com/hlperf/60V6%20Site%20Files/plenumtuningcalc.xls

[XLR8ING]

Brad, don't you know you're suppose to clean up your house before posting pictures all over the internet?!?!? :P

[weisinator]

Cool spreadsheet, it should get you in the ballpark.

One thing popped into my head: when the intake valve for a specified cylinder is closed, wouldn't that runner's volume effectively be added to the plenum's overall volume?

Given this piece of info, one could effectively change the tuning RPM of the plenum by using a longer duration cam (holding the valve open longer, reducing the plenum volume)

Edit: another thing popped into my head: two revolutions per four cycles. Effectively, the intake valve opens during 3/4 of one revolution of two revolutions, assuming a 270 degree valve open duration.

[SappySE107]

Well, you also have backflow from the pulse hitting the valve as its closed just after it closed, causing a wave to go back up the runner. Im pretty sure the closed off runners don't count as plenum area because they won't be feeding into the runner that does have the intake open.

You would have to figure in cam timing for the intake duration and when more than 1 intake valve is open, and for what percentage.

Id rather experiment with different lengths and plenum areas based on the spread sheet as a starting point.

[weisinator]

Like I said, it should get you in the ballpark and is quite simple. Simple is always good.

My idea came from people who make so-called "SQL" ported subwoofers. The enclosure starts off as a dual port subwoofer tuned to 60hz or so for SPL competitions/dbDrag/etc. When you plug one of the ports, the enclosure gets larger (the plugged port volume becomes part of the enclosure volume) and port area shrinks (with the same port length), yielding a deeper tuning frequency for "daily driving".

I wouldn't have even thought of it if Brad hadn't mentioned "hemholtz resonator".

[MasterJacko96]

that would be sweet if u made those i would be intrested in buying one

[bszopi]

If I make it, it will be a one off deal.

[Corpse]

Think of it like an exhaust manifold right, scavenging occurs between multiple cylinders, due to the fact that they all feed into a chamber which is similiar to a plenum although everything is in reverse.

So, what I believe about the plenum and runners is that the air wave after bouncing off the valve goes back against the plenum volume and tries to pressurize it, whereas, in turn, it reverses and forces air back down. Although it is hard to comprehend at the moment, I would asume that all runners would come into play as the total volume for scavenging.

It is hard to figure because, even the air just after the throttle body is moving inward into the plenum and that would mean that each cylinder would be trying to reverse the plenum's flow. I was always under the impression that each runner was only using the length of itself and the pulse stopped at the end. But, I always wondered, why would the pulse just reverse at the end of the runner, if there was enough force of the air rushing into and through the plenum to reverse the flow, it would kill the strength behind the pulse.

But, if two valves are opend at once?? I dunno

It's Saturday night and, I just got home and I'm a little inebriated.

-Dave

[bszopi]

Drunk intellectual posts are always fun... I know I used to do it every once in awhile myself and I think Ben does it constantly :P . The pulse coming back up makes sense and all, and you are wondering why it would just stop at the end? From what I have read when doing research awhile back on the same topic, you want there to be a "wall" at the opposite side of the runner so that air coming back up will reflect off and be sent back down. That is what I understood it as at least. I know they mentioned having a flat surface perpendicular to the runner opening and I'm almost positive it was for reflection purposes. Or I could just be . I'll have to see if I can find that again...

[bszopi]

This wasn't what I was looking for, but I'm glad I found it again...

http://www.grapeaperacing.com/GrapeApeRacing/tech/intaketerms.cfm

This site is full of very useful, technical theory articles on everything from intake and exhaust, to boost and valvetrain. I'm gonna try using some of his formulas as well and see what I come up with.

[RednBlack]

If you get some finalized prints on your project,send them to me.I can TIG weld aluminum. I used to build aluminum rotocast molds. I would love to take on this project!

[XLR8ING]

Just some things I've found in my research. That spot where the pulse changes directions is what took me forever to find, and I'd forgotten about it until you guys mentioned that.
The intake pulse changes direction again after it's a little ways out of the runner on a naturally aspiratted engine. The distance away from the runner that it changes direction is approx the same as the radius of the runner. When calculating runner length, this extra "imaginary length" needs to be added on for proper tuning.
Why it changes here, I've long forgotten.
On a boosted engine, you do want a flat surface to "bounce" the resonation off of, and length doesn't really matter, unless you're just anal about it!
Also, the pulses don't really follow the same rules as air flow, since they're travelling at the speed of sound, so air flow coming in the TB isn't what kills the pulse strength. What kills the strength is multiple pulse "reflections" before the valve opens again. That's why idealy a really long runner to get the first pulse bounced back in at exactly the time the valve opens would be optimal, but good friggen luck getting tubes that long all under the hood. Most "tuned" length runners use the 3rd or 4th bounce so you can use shorter tubes, and still get some benefits. With a one bounce intake runner you can get as much as 10% more air into the cylinder, and the percentage goes down with each bounce there after.
Oh Brad, you'll need to know the cam specs to make that intake properly as well. Basically just the stuff on the intake side of the cam is what you need.