Turbo plenum design

[must learn more]

What are the design criteria for a turbo plenum? specifically small capacity 4 cylinder "blow through", I did read that the plenum should be the same volume as 100% of the engine capacity, surely there's a need to be a bit more specific than this?

[526FIREHAWK]

Ken Duttwieller says that there is not alot of intake tuning that can be done on a turbo application, So what would work best for a NA motor should work well with the turbo. But Im no expert, Im shure theres someone here that can give you a better answer.

[must learn more]

Any thoughts anyone?

[SWR]

I've used anything from about 47% of engine volume up to 1.3 times the volume. The smaller percentages seens to give a slight increase in responsiveness,the bigger ones gets more sluggish (due to volume after the butterfly) but give more power.

Except if you run like I do on my bike, 4 throttles between the head and the plenum, then the bigger plenum had the best response and good power. You don't want turbo lag on a bike. Period.

It's all about what application you will use it for. Most intakes I make I generally go for good distribution of air,making sure I either have dynamic airflow to all cylinders (tricky) or fairly static air (simpler) to all.

And tuning...you can totally alter a turbo engine by different tuning,optimum lenghts/dia's become atleast as important as in a NA-engine,if not more so... that is,if you're after really broad power ranges and really high HP,all in the same package. Most people seem to just make lots of high-end power with no bottom end,or just the opposite. At least here where I live.

[Silverback]

Seems like in most cases you’re best off with as much plenum as you can tolerate and skip or not worry about the runners. If you want some killer examples take a look at what the formula cars (and race cars in general) were running when they were running big boost… basically a big box plenum with no runners.

[JustinOlson]

What do you guys think about these dual plenum inlet manifolds? They use a tappered cone to distribute air to the main plenum. Personally I think that any gain see from better distrubition will be lost to the more surface area and restriction. What are your guys thoughts?





Does raising the velocity stack off of the plenum floor have a large impact on power?

[SWR]

Personally I think that any gain see from better distrubition will be lost to the more surface area and restriction. What are your guys thoughts?

There is no restriction there if it's properly made... I've made dozens of intakes that way,and the "slit" between the feeder tank and the plenum itself has the same area as (or slightly larger than) the piping from the IC. It just redistributes,it doesn't restrict,even though it seems to do so at first glance.

Does raising the velocity stack off of the plenum floor have a large impact on power?

In my opinion,only if you have a distribution problem without them and/or if you need them to get proper tuning (correct intake runner lenght). I've tried making funnels in the base plate for the intake (i.e. no velocity stack at all) and just like in this photo,and any disadvantages with the one or the other was most times traced to making the intake runner too short or too long for the rpm used...

[JustinOlson]

How does the dual plenum perform compared to a more traditional tappered single plenum like this:



Here is some flow testing that was performed on the above manifold:

Virtual Works
--------------------------------------------------------
Cylinder 1 - 334.2
Cylinder 2 - 333.9
Cylinder 3 - 334.7
Cylinder 4 - 334.1
Cylinder 5 - 334.6
Cylinder 6 - 334.5

Average - 334.33
Variance cylinder to cylinder - 0.2%
Efficiency - 92%

My question after seeing these flow numbers is there really a need for a dual plenum manifold? What is to be gained by doing so if you can get such consistent flow cylinder to cylinder with a tappered single plenum?

http://www.horsepowerfreaks.com/miscell ... nBench.jpg

http://www.horsepowerfreaks.com/miscell ... w_mini.jpg

[blownzoom440]

my brother builds turbo hayabusas making from 600 to 700hp from 1300cc's.in the plenum[pic to come later] there are short stacks that work better.[dyno proven]the boost range is from 10 to 35lbs injected intercooled[some air to air/some air to water]it is a single plenum design about 1.5 of engine volume.[1300cc's]there are fuel injectors above the velocity stacks for added fuel.[piggyback injectors.
http://i27.photobucket.com/albums/c192/ ... urbobk.jpg
the plenum gose above the throttlebodies then directly into the head.
he has used years ago a drawthru turbo kit[mr turbo for kawasaki's] and the plenum was small[about 500cc's] then direct into the head.it worked ok but a lot of heat from the turbo with no intercooler.

[MadBill]

Those flow numbers <EDIT> (in Justinolson's post) look wonderfully consistent, and judging by the appearance, that plenum should certainly do its job. However, individual flow is not the issue for a plenum, it is the possibility of "charge robbing", where one runner steals from another, or if the total demand starves one or more cylinders.

For example:
I no longer have the data sheet, but in 1984 I was working on a high compression TBI natural gas 2.5 L. 4 cylinder engine. We used tuned runners and, to keep the response crisp, a small diameter plenum. Proportionally, it looked about like the above pic would if the plenum was a constant diameter the size of the small end. We verified each port's flow with and without the manifold, to ensure no restriction from the long TPI-like tubes.

With 17:1 CR, the engine knocked at WOT, even on 130 octane CNG, but #1 & 2 were significantly worse than #3 and especially #4. We then did WOT motoring tests, looking at peak pressures for each cylinder, and the curves generated showed two tuning peaks at the predicted RPMs, pretty much resembling a torque curve (no surprise), but the pressures marched steadily lower, dropping almost 15% by #4, obviously due to the inadequate plenum area...

[cencodsm]

Plenum volume = (2.57) X (displacement)

No, I am not kidding. Go smaller if a proper plenum won't possibly shoehorn into the bay/between the frame spars... or whatever. If you want to try working with little plenums, be prepared to prototype 'till the end of your days.

I would say the same thing about the "velocity stacks". The runners really don't belong inside the plenum but sometimes space constraints necessitate it.

And the last equation:

velocity stacks + small plenum = disaster

[JustinOlson]

So is there any advantage raise the velocity stacks off the plenum floor to get them out of the boundary layer at the plenum base?

[cencodsm]

Yes, If you don't know what you are doing when developing the plenum-runner intersect geometry then a velocity stack can save you from yourself. That is how I see it.

[wbclassics]

Is there a benefit to flow testing a turbocharged plenum / intake / head combination differently than you would a conventional NA head and intake combo?

I noticed that VW plenum in the photos above was being sucked through. Why not reverse the flow of the bench, and blow into the plenum at the plenum inlet with the plenum affixed to the cylinder head? This would necessitate standing the plenum and head on end (vertically), or using an elbow to turn the air into the plenum from the flow bench.

I would imagine that a greater understanding of the cylinder head, intake and plenum could be obtained by trying to bring the flow testing to more closely resemble the actual function of the parts when they are in use on the motor.

Kai

[cencodsm]

Since we all put so much faith in static flow testing on the bench, here are some CFD static flow tests:

http://www.gt40s.com/ubbthreads/showfla ... &fpart=all