Pushing some new intake manifolds to the limit!
#331
Account Disabled
Join Date: Nov 2006
Location: Akron, Ohio
Posts: 174
Likes: 0
Received 0 Likes
on
0 Posts
Complex control of fueling? I'm not talking about carburators here,, and decent ECU ought to be able to tailor each cylinder individually. It isn't difficult... all you need is a few more O2 sensor bungs in the appropriate spot to be able to check each cylinder individually.... WAY back in the day my FSAE team did this to try and validate and correlate some results we got from a flow bench. FLuent CFD, and FloWorks CFD in Solidworks. Heck, for your "hig rpm bike motors, and custom EFI and manifolds, , ( let me guess,,, recently finished a FSAE car??) even a piggyback like a Dynojet Power commander can be unlocked to control each injector on its own unique map..
By the way,, the helmholtz doesn't apply to just high rpm engines.. in the late 50's and early 60's chrysler had used this in there v-8's wedge motors... (real low rpm tuned runners, and yes, they were feakishly long! the carbs hung off the opposite bank while the runners criss crossed... thus the name cross ram..)
If i recall correctly chevy's 69 Z-28 camaro with the trans-am spec 302 also did this
way back in the day... and now, I don't believe any modern auto manufacturer today doesn't do this, regardless of the model..
But I'm not try to argue, i agree that boost can mask some design flaws in an intake, but not eleminate them completely. Sometimes it can magnify the differences. beyond a few simple hand calculations to determine optimum runner length for a given rpm there are terrific software programs out there that can take this a step further, Ricardo wave and GT power are two off the top of my head.....
By the way,, the helmholtz doesn't apply to just high rpm engines.. in the late 50's and early 60's chrysler had used this in there v-8's wedge motors... (real low rpm tuned runners, and yes, they were feakishly long! the carbs hung off the opposite bank while the runners criss crossed... thus the name cross ram..)
If i recall correctly chevy's 69 Z-28 camaro with the trans-am spec 302 also did this
way back in the day... and now, I don't believe any modern auto manufacturer today doesn't do this, regardless of the model..
But I'm not try to argue, i agree that boost can mask some design flaws in an intake, but not eleminate them completely. Sometimes it can magnify the differences. beyond a few simple hand calculations to determine optimum runner length for a given rpm there are terrific software programs out there that can take this a step further, Ricardo wave and GT power are two off the top of my head.....
I was on an FSAE team too, used a motec m4 the first year with a crappy designed manifold and had to tailor each cylinder. The engines I was talking about were not those though. Restricted FSAE manifold show much different trends than unrestricted ones. You dont find a lot of street cars of race cars running with a wideband or egt sensors in each exhaust runner though, those are used mainly for R&D and development work. I dont think any of this is difficult, but Im just trying to provide some information since the topic came up. It obvious that Helmholtz dosent apply to just high RPM engines, that was not what I meant, and I feel that you misunderstood what I was saying. A small change in runner length (1.5 inch like you stated) can greatly change the RPM in which the the effects take place. I guess what I was trying to say is that change may put the useable effect outside of these engines (4G63) RPM range. Sorry, Its possible that we are saying the same thing, but not understanding what each other is saying, I find that happens a lot in engineering.
#332
Evolving Member
Join Date: Mar 2007
Location: In front of a Catia screen
Posts: 272
Likes: 0
Received 0 Likes
on
0 Posts
and for Mr. Deiwert, I think we are on the same page and saying the exact same thing, I just think it sounds better the way I say it,,, or atleast it does in my head, before i type it! Ha! Kidding of course,,, sort of...
Last edited by 4G63DSM; Jan 11, 2008 at 12:04 PM.
#333
Account Disabled
Join Date: Nov 2006
Location: Akron, Ohio
Posts: 174
Likes: 0
Received 0 Likes
on
0 Posts
#334
Evolved Member
iTrader: (2)
Still have the Crate 71 Hemi Cuda. My big bro still restores Mopars and has a blown 71 Hemi Cuda Indy 528.
Very first pass of him driving the new 'blowfish' at 9.58:
http://videos.streetfire.net/player....3-E51AE05076CB
Very first pass of him driving the new 'blowfish' at 9.58:
http://videos.streetfire.net/player....3-E51AE05076CB
Last edited by 9sec9; Jan 11, 2008 at 12:41 PM.
#335
Evolving Member
Join Date: Mar 2007
Location: In front of a Catia screen
Posts: 272
Likes: 0
Received 0 Likes
on
0 Posts
#336
Account Disabled
Join Date: Nov 2006
Location: Akron, Ohio
Posts: 174
Likes: 0
Received 0 Likes
on
0 Posts
#337
Registered User
Join Date: Aug 2002
Location: The big wide world
Posts: 128
Likes: 0
Received 0 Likes
on
0 Posts
I would love to know how you got an AMS race manifold. The last time I looked we sold one that Paul has now and we have one on our drag car. I could be mistaken but I think you have a standard or twin rail unit...not the larger plenum drag version. we have sold so few and I don't think any of them have made it over seas. A picture would tell us for sure.
Eric
Eric
My mistake it is a twin rail as we refered to as the race intake .
Sorry for the mis information if it caused any problem .
#338
EvoM Guru
iTrader: (6)
Just like in a n.a. engine however, improving VE in a turbo motor increases power and efficiency accordingly. This is because there is always 'boost' on the other side of the cylinder, pushing in the opposite direction . . .
#339
Evolved Member
iTrader: (8)
What kind of wideband are you using to measure individual cylinder A/F ratios on boosted motors? I didn't think most wideband sensors could handle the heat on the hot side of the turbo. Also, doesn't the manifold backpressure greatly impact the accuracy of the measurements?
As for a turbo masking the actual performance of the engine, I've kind of thought of the motor just being there to get the turbo going but the turbo is the important part of the equation. In other words, the turbo will dominate over just about everything. Once the turbo inlet velocity reaches the speed of sound, or near it, it's pretty hard to get any more air through the inlet and the only real power changes come from thermal and combustion efficiency. At lower boost levels, below the point of maxing out the turbo airflow, I would think you would see power improvements from VE improvements. But most of the fast setups run enough boost that the turbo reaches it's maximum airflow right around the same time they hit full boost. I know a lot of the SFWD guys take this approach and actually pressurize the wastegate with CO2 in the higher gears to basically lock the gate shut to get all the boost out of the turbo that it can possibly make.
It would be interesting to see if this holds true though. Compare two manifolds at moderate and maximum boost levels. Both manifolds having similar flow ratings and air distribution on a flow bench, but tuned to different RPM ranges. Something like a smaller cross-sectional area, short runner manifold (effectively tuned for an RPM above where the cams will allow peak power) compared to a manifold with appropriately sized runners with lengths tuned to an appropriate RPM for the cam timing events.
Speaking of CFD work though, I can't imagine how long it would take Fluent or any other FEM software to run a full 3D time-dependent analysis at just one RPM point for a complete manifold. Much less the time required to get enough RPM points to get a real idea of how it would perform. I would imagine even a 2D simplified model would still take a fairly long time.
As for a turbo masking the actual performance of the engine, I've kind of thought of the motor just being there to get the turbo going but the turbo is the important part of the equation. In other words, the turbo will dominate over just about everything. Once the turbo inlet velocity reaches the speed of sound, or near it, it's pretty hard to get any more air through the inlet and the only real power changes come from thermal and combustion efficiency. At lower boost levels, below the point of maxing out the turbo airflow, I would think you would see power improvements from VE improvements. But most of the fast setups run enough boost that the turbo reaches it's maximum airflow right around the same time they hit full boost. I know a lot of the SFWD guys take this approach and actually pressurize the wastegate with CO2 in the higher gears to basically lock the gate shut to get all the boost out of the turbo that it can possibly make.
It would be interesting to see if this holds true though. Compare two manifolds at moderate and maximum boost levels. Both manifolds having similar flow ratings and air distribution on a flow bench, but tuned to different RPM ranges. Something like a smaller cross-sectional area, short runner manifold (effectively tuned for an RPM above where the cams will allow peak power) compared to a manifold with appropriately sized runners with lengths tuned to an appropriate RPM for the cam timing events.
Speaking of CFD work though, I can't imagine how long it would take Fluent or any other FEM software to run a full 3D time-dependent analysis at just one RPM point for a complete manifold. Much less the time required to get enough RPM points to get a real idea of how it would perform. I would imagine even a 2D simplified model would still take a fairly long time.
#340
Registered User
Join Date: Aug 2002
Location: The big wide world
Posts: 128
Likes: 0
Received 0 Likes
on
0 Posts
Agreed. A flow-bench is of limited, residual value, only being able to expose relatively gross issues, such as what I illustrated previously.
Poor manifold design tends to be masked by a turbo in general, simply because a turbo produces enough power to mask numerous shortcomings and faults that would be readily exposed in a n. a. situation.
Just like in a n.a. engine however, improving VE in a turbo motor increases power and efficiency accordingly. This is because there is always 'boost' on the other side of the cylinder, pushing in the opposite direction . . .
Poor manifold design tends to be masked by a turbo in general, simply because a turbo produces enough power to mask numerous shortcomings and faults that would be readily exposed in a n. a. situation.
Just like in a n.a. engine however, improving VE in a turbo motor increases power and efficiency accordingly. This is because there is always 'boost' on the other side of the cylinder, pushing in the opposite direction . . .
A flow bench will and does allow you to see restriction if used correctly in most case's . Partly why we recommend the use of a cylinder head in the circuit . Also nobody draws attention to runner profile and how air travels down different shaped tube's .
A modern flow bench like the SF1020 will give good valid data that is backed up time and time again by on the track results . No right minded racer would be with out there flow bench data .
Lets hope the tests can show clean data acquisition.
#341
Evolving Member
Join Date: Mar 2007
Location: In front of a Catia screen
Posts: 272
Likes: 0
Received 0 Likes
on
0 Posts
Speaking of CFD work though, I can't imagine how long it would take Fluent or any other FEM software to run a full 3D time-dependent analysis at just one RPM point for a complete manifold. Much less the time required to get enough RPM points to get a real idea of how it would perform. I would imagine even a 2D simplified model would still take a fairly long time.
A good 3D CFD program like fluent takes a real long time too for just 2D static stuff!!