Jump to content
HybridZ

Turbo Intake Plenum


MONZTER

Recommended Posts

by back corner you mean that area of very high pressure near the rear cylinder, right? That looks like the perfect place.

 

how far away from the exhaust manifold would this plenum be situated? perhaps you can modify the back corner so there is a flange on the underside?

 

Yep that is where I was thinking. Tim had some good ideas that might help.

Link to comment
Share on other sites

I kinda thought you might have already had that covered, just couldn't tell from the pics. Do you have an external boost controller, or do you plan on using the TEC3? Maybe leave yourself room to add one if you don't like the GPO control?

 

I plan on using the GPO control on the Tec 3. How are you running your vacuum lines? Are you tapped into the compressor or the plenum? How would the routing be different for a GPO control vs. an external? I have never looked into the external controller

Since I seem to be in the business of making your life more complicated tonight, what if you just used the diaphragm assembly from your TiAl and integrated the valve body part of it into the plenum? This might allow a bit more clearance (and would be slick as hell :flamedevi )

 

That’s a cool idea; I will look to see if I have room. The main problem is that the plenum will be made in two halves, surface machined in two operations normal to the seam line. This makes it difficult to get ports or details on the sides. I could always weld it in and blend it afterwards??

 

 

Oh - and have you considered either V-band or Wiggins clamps (stupidly expensive) for connecting the plenum to the throttle bodies instead of the bolt-on flanges? This is one thing I wish I could have done with mine - the bolts on the bottom side are a huge pain to get to. A nice quick release setup would have made the setup much more serviceable.

 

I am using Wiggens clamps everywhere else. I machine the flanges myself and only buy the clamshells (saves some money). That would make life easier. The thing I worry about is the expansion of the Wiggings clamps. They are designed to float and have about 1/4 inch gap between the flanges. I am not sure but I think when under boost the flanges might be pushed apart causing the plenum to have a lot of movement. Again maybe Tony D has some comments.

 

If you stay with the flanges, I had o-ring grooves cut in the throttle bodies so I wouldn't have to mess with gaskets...

 

Once again another good idea, do know how much money you are costing me today

:shock:

Link to comment
Share on other sites

I plan on using the GPO control on the Tec 3. How are you running your vacuum lines? Are you tapped into the compressor or the plenum? How would the routing be different for a GPO control vs. an external? I have never looked into the external controller

If you use an electronic external controller it will most likely have its own MAP sensor which will require a reference. I use the same plenum vacuum source that I use for the TEC3's MAP sensor for this. The other lines that actually go to either side of the wastegate diaphragm are taken from the pipe that runs between the intercooler and the plenum.

 

That’s a cool idea; I will look to see if I have room. The main problem is that the plenum will be made in two halves, surface machined in two operations normal to the seam line. This makes it difficult to get ports or details on the sides. I could always weld it in and blend it afterwards??

 

Hmmm... How much room do you think you'll have between the plenum and the firewall? I was originally thinking of locating the diaphragm assy on the back facing the firewall and creating a little chamber on the end for the valve. If that doesn't work, would there be room to extend the plenum a bit rearward and mount the diaphragm assy such that it faces the head, kind of like a seventh intake runner? That might be more conducive to your machining method (maybe?) Just throwing ideas out...

 

Also, were you planning on venting it to atmosphere or recirculating?

 

The thing I worry about is the expansion of the Wiggings clamps. They are designed to float and have about 1/4 inch gap between the flanges. I am not sure but I think when under boost the flanges might be pushed apart causing the plenum to have a lot of movement. Again maybe Tony D has some comments.

 

Maybe just some simple v-band clamps would work better here - they'd certainly be easier to implement. It would just be really nice to be able to do all of the fastening action from above, instead of having to fish around underneath. Speaking from experience here, and if I'm looking at your drawing right, your diffuser plenum is going to make access to the bottom bolts even harder...

 

 

Once again another good idea, do know how much money you are costing me today

:shock:

 

Sorry... :flamedevi:flamedevi:flamedevi:flamedevi

Link to comment
Share on other sites

by back corner you mean that area of very high pressure near the rear cylinder, right? That looks like the perfect place.

 

This got me thinking again (always dangerous)...

 

How confident are we that there is really going to be ~80psi in that area? It occurred to me this morning that if that really is the case, I don't think you'll be able to find a spring stiff enough to hold the bypass valve shut in that location. Even the stiffest ones are only designed to hold a pressure differential of maybe 20psi at WOT. Maybe we should rethink this...

 

EDIT:

I just went back and looked through the analysis in this thread, and there are a few details of the simulation that aren't clear to me. Please nobody take this as being critical - the work so far is great, just trying to understand some things...

 

In the simulation, what did the total airflow through the plenum need to be to get the ~23psi at the inlet to the head? How well did this correlate to what we would expect for ~700hp (flywheel) worth of airflow? That IS the neighborhood I expect Jeff to be in, if not higher, btw.

 

What happens at the manifold/head interface in the sim? Is there any restriction modeled in at that point, or does it assume that the manifold is dumping to open air at that point?

 

Reason I'm asking is that the 80psi number seems a lot higher than I would have expected, and somebody else already mentioned the supersonic flow velocity at the ports. Just looking for a sanity check.

Link to comment
Share on other sites

What happens at the manifold/head interface in the sim? Is there any restriction modeled in at that point, or does it assume that the manifold is dumping to open air at that point?

 

 

It looks to me like the runners are dumping all at one time to open air. You can imagine how much air would have to be pushed through to generate pressure like that with open runners. This is just me guessing again, as I have very little understanding of the CFD software. Maybe TurboBlueStreak can chime in on the details.

Link to comment
Share on other sites

It looks to me like the runners are dumping all at one time to open air. You can imagine how much air would have to be pushed through to generate pressure like that with open runners. This is just me guessing again, as I have very little understanding of the CFD software. Maybe TurboBlueStreak can chime in on the details.

 

That's what I was thinking too. That would definitely explain the pressures in the plenum and the high port velocities. If that's the case, I believe it could be easily remedied by adding a known restriction at each port and adjusting until the flow numbers look realistic.

 

Like the Wiggins clamp solution, btw. Sorry about the added expense... :flamedevi:flamedevi:flamedevi:flamedevi

Link to comment
Share on other sites

That's what I was thinking too. That would definitely explain the pressures in the plenum and the high port velocities. If that's the case, I believe it could be easily remedied by adding a known restriction at each port and adjusting until the flow numbers look realistic.

 

Like the Wiggins clamp solution, btw. Sorry about the added expense... :flamedevi:flamedevi:flamedevi:flamedevi

 

Tim, that was great advice. After your comments about getting to the screws I went and looked at it. I would need like 10" long fingers to get to some of them. I thought at first no big deal because I will just bolt the plenum on before bolting on the manifold. Then I realized, how am I going to balance the TB's without removing the plenum. Its funny how you get deep into a project and so focused on the details that you miss the obvious. This is why I like this forum so much, great people with tons of real world experience.

As for the added expense, It may actually be wash. I had to shorten the length of the air horns to make room for the clamps. The material stock size went down by a 1/4 of thickness, and that's a lot of weight and alot of savings when your talking a hunk of metal that big.

Link to comment
Share on other sites

I think Tim made a good point about the pressures in that rear area of the plenum. Perhaps the best Idea is to find a manufacturer who makes bovs that all use the same flange and cover wider ranges, and see if you can distribute their bovs with your plenum for desired applications.

 

That way the design is known, and tested, and can be placed where it wont cause an issue.

 

I don't know where else you could put it since we don't know how much room you'd have in the engine bay yet, right?

 

EDIT:

 

or, you can just forget about the bov completely, and have the customer install mid pipes if they want, between the compressor outlet and plenum inlet.

lol... much more simple and cost effective, and easier to upgrade to if you already have a bov.

Link to comment
Share on other sites

Still have not figured out how to integrate the BOV besides welding it to the plenum

 

Is this the BOV you've got?

tial-hardware1144_m.jpg

 

It's at least pretty compact - you could probably just leave yourself a flat area ~2.5" round at the back of the plenum for welding the flange and mount the thing on top - there should be ample hood clearance back there.

Link to comment
Share on other sites

Both the HKS and SK have their BOV mounted to the front portion of their plenum. Both HKS versions have the plenum blowing off at the front portion...on Jeff's it would be above his inlet pipe, where the pressure looks to be relatively low. I'd not put it at back simply due to packaging issues.

 

Ideally you're venting the plenum first, but tapping the manifold downstream of the ITB's is going to give a vacuum signal to pop it open relatively quickly anyway (it does on my HKS!) on any sort of lift-throttle. That negative pressure bias on the diaphragm / spring balance will make it relieve quickly. I'm figuring the distributor won't be in the way, so putting it toward the front of the plenum and letting it occupy the former position of the distributor might make a lot of sense. (Jeff P was looking at his setup on Thursday and said similar things, as he now has no dizzy there, and his blowoff is soooo close to there already...only minor piping and it opens up more real estate underneath!)

 

The Wiggins Clamps will expand slightly,you can always use limiting straps, but I don't think you will need to do it---some nice bails and snick snick snick and the plenum is ready to come off. I like it!

 

As for mounting of the BOV, the SK plenum mounts the BOV similarly to the HKS, and to describe it roughly: Take the bottom half of your BOV in the above photos, and incorporate the entire lower body into the plenum. The SK uses the valve seat which is integral to the plenum, the HKS has a flat flange that the BOV sliding valve seals against, with the body bolting directly to the plenum. Integrate the lower valve seat into the plenum, and those vents, and you seriously reduce the mounting area you end up needing. Curiously, the SK valve looks almost identical to the one they used on their EFI conversions...on that one they had a cast J-Pipe, with the lower seat integrated into it. When I built my original HKS "Knockoff" box, I used one of the SK BOV's mounted directly to a machined flat spot on my plenum box. Towards the front face, near the distributor...

Link to comment
Share on other sites

Thanks Tony for the input, since I will have to make the weld fittings for the Wiggings clamps myself, I will just make them fit close together and no straps will be needed to keep them from moving.

 

Tim this it mounted in back on top, hood clearance is going to be close. I have it mounted on the V-band clamp/fitting that comes with the BOV. I like you ideas about integrating the bottom half into the plenum, but it will not be possible to machine it the way I have designed it to be made. I think I will have to resort to welding it on after the plenum is finished and installed in the engine compartment. I think this will be the safest bet:)

 

final_cnc_plenum.jpg

Link to comment
Share on other sites

Okay, I've been looking at this a bit more - that 80psi number was bugging me.

 

Jeff - I'm making a couple of assumptions, but I think they are on the conservative side, airflow-wise. I'm assuming the following:

The plenum inlet is 2.5", giving a cross-sectional area of 4.9in² = 0.034ft²

runners are 1.77" (45mm), giving a cross-sectional area of 2.46in² = 0.017ft².

 

Also, I'm assuming airflow at 1.6 cfm/hp, so 700hp ~=1120cfm.

 

So, looking at this diagram:

cfd_rev3_velocity_2.jpg

 

The simulation is showing in the neighborhood of:

16000 in/s velocity at the inlet

8000 in/s velocity at the ports

 

converting to ft/s gives:

1333 ft/s at the inlet

666 ft/s at the port

 

Using the simple relationship of Flow = (Cross Sectional Area) * (Velocity)

 

We get at the inlet:

Flow = 1333ft/s * 0.034ft² * 60 s/min = 2719 ft³/min

 

At the ports:

Flow = 666ft/s * 0.017ft²* 60 s/min = 679 ft³/min

 

From this, I'm thinking our assumption about the model dumping to free air is fairly likely to be correct. That's enough airflow for ~1700hp :shock:

 

Also, I'm noticing that the flow at the ports is not the flow at the inlet/6, but since it's kinda in the ballpark I think we can chalk that up to my guesses as to the inlet and runner sizes being a little off, combined with having to make a SWAG at the actual velocity at these points by guessing from the color code...

 

Anybody care to check my math? Corrections for actual inlet and port sizes, as well as the actual mean velocities would be more than welcome.

 

EDIT:

I just noticed my first wrong assumption - 45mm at the ports is most likely too big - I was thinking of a likely size for the throttle bodies, but it necks down after that.

1.5" (38mm) is probably a better guess. This would result in port flow of ~490cfm which jibes alot closer with the inlet flow number.

Link to comment
Share on other sites

Tim,

I PM'ed TurboBlueStreak who did the CFD work. Here is what he had to say: "Jeff as your question of how the cfd was run I ran the cfd with pressure openings. the runners were environmental and the inlet was 25psi. I'm going to review all my information and see if I can run the test in a better way. Hopefully we won't see extreme pressures or supersonic flow. "

So it seems were were correct with our assumption.

 

Here are the correct sizes of the model -

Runners at the head = 36mm ID

inlet into the plenum = 70mm ID

 

I will post anything new he runs

Link to comment
Share on other sites

Well, we have had some good comments about the excessive velocity and pressure numbers on the CFD plots. TurboBlueStreak was able to find a better way of testing it with more realistic numbers. We used a Excel sheet for turbo matching to calculate some flow numbers for different boost pressure and RPM. TBS says "all the test were done ignoring atmosphere pressure so inlet and outlet pressures where = to turbo outlet pressure the flow inlet was cubic feet per minute converted to cubic inches per second."

 

Here is the Excel sheet showing the flow numbers. The test number on the sheet relate to the CFD results below: Click on the picture to enlarge it from my gallery-

 

 

turbo_matching.jpg

 

Here are the new results as well as some videos

 

test#1 1.5psi boost 2500 RPM 127 CFM

Test1_Presure.JPG

 

Test1_Velocity.JPG

 

Test #3 5psi boost 3500RPM 214CFM

Test3_Presure.JPG

 

Test3_Velocity.JPG

 

 

Test #7 20 psi boost 5500 RPM 581 CFM

Test7_Presure.JPG

 

 

 

Test7_Velocity.JPG

 

Test #9 20psi boost 7500 RPM 793 CFM

Test9_Presure.JPG

 

Test9_Velocity.JPG

 

Some videos of test #9

 

http://album.hybridz.org/data/500/Test_9_velocity.avi

 

http://album.hybridz.org/data/500/Test_9_velocity_b.avi

 

http://album.hybridz.org/data/500/Test_9_velocity_c.avi

 

So now you can see that as the flow and velocity go up, there is some crazy swirling going on in the plenum. It looks like 1 swirl is feeding each bank of runners. It also looks like the runners 1 and 2 are not getting balanced flow. Maybe a baffle inside will help, or maybe something in the plenum to disrupt the vortices.

 

I hope you find the new information and test data interesting. A big thanks again to TurboBlueStreak for his time in the CFD analysis.

 

So any new comments? Should I be worried about the results as they are?

 

Jeff

Link to comment
Share on other sites

wow.. I can't make a comment without making this preface: I took one look at that spreadsheet, and thought to myself, "Gee, that's GREAT data!! But its still WAY over *my* head, heh.." and skipped right past it. I am beginning to be able to understand what I am seeing in the flow and pressure graphics; the numbers will be examined later.

 

Secondly, I must say that at least as far as I can see (with my limited understanding :) ) this intake looks GREAT by these standards. My "short-bus" interpretation is "Yah, he is probably right to point out the major flaw that is left; but he is hunting on the ragged edge of his intake design, and intake design is hunting on the ragged edge of performance anyhow..."

 

Thirdly, a question of scale, in a sense.

 

Once the pressures equalize to the "blue" zone (it seems to occur slightly earlier on the rearward cylinders) are they actually equal, or close, for the rest of the path into the cylinder head? (Given: all intake ports and combustion chambers on this hypothetical "ideal engine" are identical) Is this "blue zone" a wide and open-ended range, and there may in fact be slight definition and differentiation beyond the lines shown in the graph, or is it somehow "magically, just, all the same after that" ???

Link to comment
Share on other sites

This is one great exercise. I will be fun following it through fab and testing. One mistake on the spread sheet is the pressure drop though the intercooler. As the density and flow rate go up the pressure drop will increase. It will probably max out at about 1 psi. I really do not think this will affect you modeling calcs.

 

Here is a calculator for all of the rest of us for compressible air flow though pipe.

http://www.pipeflowcalculations.com/airflow/index.htm

 

The pressure is absolute pressure so add 14.7 to your boost pressure.

Link to comment
Share on other sites

Wow. CFD is scary cool. What's even scarier is that old HKS Type 1 with the baffling that looks eerily like it would counteract the vorticies.

I'm not sure where I'm working next week, but now I'm really curious as to how those HKS Plenums would test out...and how HKS came up with the designs they did when they put them into production!

 

I'm going to go back and watch those videos again....

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...