Intake/ exhaust lift ratio theory

[strokerjlk]

You don't need more lift on the exhaust than on the inlet and you don't need more duration on the exhaust than the inlet, of a low speed engine.

   Partial QUOTE from uglyDougly

   Not true at all,it depends on the whole of the combination  especially the airflow numbers, low lift airflow ,compression and the engine configuration.

   For example some porters/shops (bc gerolamy,Branch) for example have low intake to exhaust flow and could use a little help in the exhaust department

be it lift,duration or centerline.

     S&S for example I believe have done their homework and the camshaft from top to bottom with a few exceptions IMO are configured to work with their heads as it should be,the whole proven performance thing 

   Lunati has a few grinds that would work well with alot of the ported heads out there,they have ground a few for me that look nothing like the shelf cams and there will be more.
       

Ok this is the meat of the other thread.
Can we discuss the theory here.
I think at this point we can all agree that combination of parts is critical.
Tq john , ugly dougly , WFO Larry and myself discussed this at great length.
Unfortunately , 2:00 am wake up calls thurs, and fri , then driving to cinci , having  1 beer  at the bar  I had a hard time storing parts of the conversations.
I think it is something that bears discussion.
Still trying to understand intake to exhaust flow rates, and how they pertain to port size.
intake length, how the exhaust pipe would differ?
That should be enought to get things started .
It really was nice discussing this face to face .

[Cracked Head]

It would be much easier to obtain one of the better engine Sim's. programs that displays pressures of  combustion chamber, intake, and exhaust tracts compared to degree of crank angle at various rpm points. Pressure volume loops are very informative also.Much of whats good happens during the overlap triangle,as does much of whats bad.As yourself having much knowledge in this field ,you realize that an entire book would be needed to cover this subject,and that in itself would be controversial.Your dyno has already told you much, you just have to relate the results to theory's.Good luck there.Many times what should in fact work will kick you, as you most likely know.I started brewing beer 5 years ago just to keep my sanity form engines that are supposedly so simple.I found out brewing is also complex, but the results are more fun.

[strokerjlk]

It would be much easier to obtain one of the better engine Sim's. programs that displays pressures of  combustion chamber, intake, and exhaust tracts compared to degree of crank angle at various rpm points. Pressure volume loops are very informative also.Much of whats good happens during the overlap triangle,as does much of whats bad.As yourself having much knowledge in this field ,you realize that an entire book would be needed to cover this subject,and that in itself would be controversial.Your dyno has already told you much, you just have to relate the results to theory's.Good luck there.Many times what should in fact work will kick you, as you most likely know.I started brewing beer 5 years ago just to keep my sanity form engines that are supposedly so simple.I found out brewing is also complex, but the results are more fun.

I suppose your right.
It's frustrating to see guys struggle .
I don't have hard fast answers other than "heads " when the questions start flying.
So that is frustrating to me , when results are not as expected.

[TorQuePimp]

   The majority of the dyno/engine sim programs are usefull but do not generate camshafts that work very well for the late model engines.

  I also believe that there are far too many cam choices available and while it gives customers/builders choices, very few seems to apply to alot of the combinations that people seem to want/becoming popular- based on the ability to run on pump gas,easy to tune,maximize output with little other than a cam change.

  There is alot to be gained by tailoring a cam to every combination stock or otherwise and 1/4  point of compression or 10-15 cfm can really change what cam you should be running.

  The cam I just did specifically to be a bolt in for the current 103 was to use the lobes that were available to me and get as close to what I could to what  my data was telling me to do for a cam,getting any closer or better numbers will require 2 custom lobes and another test.

   

[wfolarry]

http://speedtalk.com/forum/viewtopic.php?f=1&t=2183

[TorQuePimp]

  Cam King/Jones has it about right as usual

  Dynomation is helpful but doesnt seem to habla a harley too well.pipemax is one of the most useful tools I have ever used.

[uglyDougly]

When Roland and I did the separate inlet Buell project, I gave him my Dynomation dongle and he did hundreds of simulations.
  The first thing we did was establish a Dynomation model that matched the parts and then changed that to replicate the dyno results of our parts combination.
  One of the major mistakes others make when modelling H-D engines is to input the inlet length by measuring the length from the valve seat to the mouth of the carb.
That model is optimistic because it implies a long-ish intake tract.
  In fact the 'Y'manifold H-D has an inlet length that goes from the inlet valve seat to the significant cross-sectional area change. That is a couple/few inches.
  At that time Feasel had modelled his from the seat to the carb inlet but he didn't consider that the port length went from the seat to the closest part of the siamesed section.
  George Smith showed us the cam specs, port molds, and flow-sheets to Dave Feasels 8.09 second pro-stock at that time.
  We figured out the power that it needed to do 8 sec. flat. and properly modeled, with the correct inlet dimensions, and it was spot on.
  Basically, that confirmed that the method we were using to model the inlet was correct. Then we confirmed the same with Mike's separate inlet Buell using actual dyno results.
  When Mike rebuilt the engine with 1/8 in. greater bore (the 78.69 CI engine) he had the 0.650" lift cams
that Andrews had done to the same open/close specs as the previous 0.510" lift cams, (White Lightning on slightly wider centerlines) he did the Dynomation simulations.
By that time we were able to predict direction and magnitude of the changes pretty well with Dynomation.
  When Mike modelled the next build with the 0.650" lift cams, he also modelled it with the existing 0.510" lift exhaust cams that we had been using.
The Dynomation results were exactly the same. We discussed it and he put the 0.510" lift cams on the exhaust side.
  Some people in the know have said that I have the best collection of Ducati Desmo-Quatro (851-996 engine series) cam information in the world and that shows me that Ducati went from 140HP in 1992 to 190HP in 1999 with the same exhaust cam. That cam is the 'G' cam with limited lift and way too much duration for 10,500 RPM in 1992 to over 12,000 in '99 with larger exhaust valves.
  At the same time the 'factory' increased the lift and decreased the duration of the inlet cams (to support bigger inlet/exhaust valves and keep them from hitting one another.)
  That showed me that the exhaust doesn't need high lift and does need longer duration for high RPM.

  All of this means (to me) that all of the H-D cams have too much exhaust lift and duration. Outrageous, no?!!!
  I worked with Steve Vermeer of Induction Research and he always said that too much exhaust (Ex-to-In%, or exhaust valve size, or exhaust duration) will kill torque but you can make the peak HP.

  The place you need enough overlap is at the HP peak, every place below that you have too much. It hurts emissions and drivability, but the correct minimum overlap is necessary to maximize the HP.

  Sorry guys, if all of the cams available to you have too much exhaust and/or overlap for the RPM you run, you will never know how little you need. And you'll never know how much torque you are giving up.

  Roland did put the 0.510" lift inlets into his 'Y' manifold engine along with stock (non-White Lightning) exhausts and it made 90HP when fresh but the power dropped off as the valves got carboned up. Never got 90HP after fresh. All we can do is assume that the stock Buell exhaust cams may be marginal for anyting over 90HP.
  Nobody ever puts in inlet cams only and none of the cam grinders make/sell Twincam cams with stock exhaust lift and duration, so nobody else has actually tested these things.
  Also, we tuned for every change, and almost nobody does that either.

  When I wrote that thing on the 110 Kury thread I realized as I was wrapping it up the the key component was the part about the Chevy cams being the centerline for the Chevy engines and the H-D engines are all to the more exhaust, less RPM side of that centerline. (Just a reminder that the discipline of writing, is a learning process in itself.)

  So, since I have had the opportunity to test a lot of these things, and most people don't, pardon me if I have a little confidence in my conclusions.

  In God we trust, all others bring data!!! (quote; Ed Dahir, The Dyno Difference.)

   Doug

[uglyDougly]

  Thought I should re-post;

It does depend, so lets look at the numbers.

There are some basic differences in the total combination of the H-D vs the Chevy/Ford.

1st  Exhaust to inlet size. 
For a good set of heads on a Chevy with 1.600" ex valve and 2.060" inlet the ration is; 0.7767:1
For a ported H-D with 1.610" ex valve and 1.900" inlet the ratio is; 0.8474:1

2nd The Chevy racing engine has somewhere around 12-13" inlet tract that resonates like a Saxophone at the torque peak and that gets an extra 10% air into the cylinder, which, needs 10% more exhaust to clear the cylinder.

So, the 'classic' Chevy guy rule of thumb; no greater than 70% exhaust-to-inlet ratio (as tested on a steady state flow bench, Superflow 600) will really be about 64% for the H-D. That is for a7500-8500 RPM engine (remember this for later.)

3rd Most porters can get greater improvements with the exhaust than the inlet, so I imagine that the Ex-to-In ratios go much higher than 80% and I've seen 100% + at lower lifts.
Right there the rule is smashed. Ignored. 80% Ex-to-In is 23% too much exhaust. (If you believe the NASCAR guys.)

4th Then there is the RPM.

As the RPM increases, the piston goes down faster and the airflow tends to increase with RPM. At some point, of course, that effect is nullified by losses, velocity and acceleration of the inlet charge.

On the exhaust side, it is generally accepted that 50% of the exhaust mass  needs to be removed by BDC. Low lift, high pressure. (It's why we love the sound of these things!!!)

What changes with RPM between EVO (exhaust valve opening) and BDC is time. It isn't the piston pushing exhaust out because the piston is increasing the cylinder volume between EVO and BDC.
By the time the piston is pushing exhaust out at it's (the pistons) highest velocity, (around 72 degrees BTC) the contents of the cylinder is low in pressure (relative to the start of the exhaust cycle) and high in temperature, not very much mass there.

After TDC, the lift is low and extracting the last bit of residual charge from the cylinder is critical as residual exhaust has a big negative effect on torque (think,EGR.)

Basically, this means that the exhaust cycle is much more time dependent than the inlet cycle.
Because of that, engines need longer exhaust duration as the peak RPM goes up. The cam that works in an 8000 RPM engine is over-exhausted for a lower RPM engine with exactly the same port/valves and even more over-exhausted for an engine with a higher Ex-to-In ratio.

Over exhausting costs torque.
Still, airflow tends to increase with RPM, while exhaust gets less and less time and needs duration to keep up.

5th The one point I haven't made is that most of the H-D cams are car cams, or cams that have worked well in Chevy V-8s. That is the centerline for these cams and the statement 'it depends on the whole of the combination  especially the airflow numbers' works around that (Chevy) centerline.
All of the H-D combinations are on one side of that. I don't think any even sneak to the Chevy center.

Actually, the only example of a Chevy-like H-D, that I know of, is Roland's tuned inlet, separate inlet Buell and it made 128.9HP and 103.8 lb-ft of torque from 78.69 CI displacement. That had (actually it still has, it's the Buell Brothers Bonneville engine) 0.650" lift inlets and 0.510" lift exhaust cams.

That's what I'm sayin', and that's why a generalization is appropriate.

[TorQuePimp]

  I think we are going to have to disagree to a point and man did you really type all that crap out ?

[uglyDougly]

So, I type it in Notepad and copy-and-paste into the reply.

  Of course you can disagree, but what data do you have that disproves it.

  The scientific method isn't about proving ones theories but going thru the disproofs until you can no longer find a reason why it isn't true.

  I come from outside the H-D culture, just as I come from outside the US V-8 culture. From my perspective, culture drives a lot of what should be science.


  The Chevrolet racing engine is the most highly developed racing engine in history, but their culture keeps the same cam and designs every other element to work (and they do work really well) with that cam.

  When I started at Kuryakyn I told myself that everyone knows more about Harleys than I do, but after looking at H-D
cams I realized that they are Chevy cams and then began to question the configurations.
  Soon I saw that the Ex-to-In ratios were too high (for NASCAR) and the RPM was lower, both factors implying that they needed less exhaust cam.
  Initially, Roland was a bit reluctant to go too low on the exhaust lift (re; TC4G cam) but by the time he asked for the TC26G he left the 0.495" exhaust
of the 'drop-in' TC2 and had the inlet increased to 0.575".

  At that point I asked myself; 'if we can make 120HP from a 95CI with 0.495" exhaust lift, why do we any more?'

  You wouldn't use that to save on valve spring costs. (you could use 1/2 of a set of high lift valve springs and use the stockers on the exhaust.)

  But what nagged me was Vermeer's opinion that 'you can make the HP, but you give away torque with too much exhaust.'
  So, why would you?

  Doug

[wfolarry]

Same can be said about the pipe. Wrong pipe no power. Seems Mike spent some time developing a new exhaust.
As far as the Chevy crowd goes a lot of guys will tell you that exhaust flow doesn't matter. The latest trend is the 1.60 exhaust valve they're putting in just about everything with no loss of power.
As far as too much lift goes I can see where it might not be needed but I can't see how it would hurt you. I think the cam timing is a bigger player.

[TorQuePimp]

 Making a blanket statement that no harley engine will need more exhaust lift or possibly more duration than the intake I have to disagree.

  I dont come from a HD  background and what I do and have done is pretty varied.

  Far too many variables to just say that "one way" works.

I would agree that in almost no case would you have to chevyize a cam for a harley,(and yet they are still out there)

[uglyDougly]

  Totally agree, more than necessary Ex lift won't hurt anything, but it will give you a little more area-under-the-curve at the critical low lifts. Not significant though.

  Most important is the  Ex. duration.

  More than necessary Ex. duration gives you more overlap (narrow L.C.), or earlier EVO (wider L.C.) and that gives up low-end torque, or both. Neither case helps low speed running.

  That Chevy 1.6" Ex. valve is usually paired with a 2.060" in. valve with tuned inlet length (remember? an extra 10% VE) and it isn't a 6200 RPM engine.

  Because H-Ds don't have much inlet resonance (they do have it but it's high frequency and low amplitude) all the VE (torque) curve is due to the exhaust resonance.

  The reason I make those blanket statements is because H-Ds are all on the same side of the Chevies. (from an exhaust flow, Ex. valve size, RPM perspective) and it makes some people think about it.
  Ask yourself why and give me the examples so I can learn something.

   Doug

[Cracked Head]

Are we to the controversial parts of the debate now.Much of whats believed to be new theory is actually old and vise-verse.Sometimes people have differing thoughts, but all that matters is the final results and the rest is only academic.

[wfolarry]

That's my problem. I'm a Chevy guy. 

[1FSTRK]

strokerjlk
Your pm box is full

[Admiral Akbar]

At that point I asked myself; 'if we can make 120HP from a 95CI with 0.495" exhaust lift, why do we any more?'

Cuz most Harley exhaust systems suck?     They are designed for looks and sound..

Max

[uglyDougly]

 <Cuz most Harley exhaust systems suck? >

   Yeah, lift won't fix that but duration might. Headquarters?

   Doug

[strokerjlk]

2nd The Chevy racing engine has somewhere around 12-13" inlet tract that resonates like a Saxophone at the torque peak and that gets an extra 10% air into the cylinder, which, needs 10% more exhaust to clear the cylinder.

So, the 'classic' Chevy guy rule of thumb; no greater than 70% exhaust-to-inlet ratio (as tested on a steady state flow bench, Superflow 600) will really be about 64% for the H-D. That is for a7500-8500 RPM engine (remember this for later.)

3rd Most porters can get greater improvements with the exhaust than the inlet, so I imagine that the Ex-to-In ratios go much higher than 80% and I've seen 100% + at lower lifts.Right there the rule is smashed. Ignored. 80% Ex-to-In is 23% too much exhaust. (If you believe the NASCAR guys.)

I've seen 100% + at lower lifts

what was the result? low tq at low rpms?

64% for the H-D

THANKS

[TorQuePimp]

  You can throw anything that nascar does out the window,along with nhra/ihra prostock and comp eliminator those are about as purpose built an engine as it gets.

  Very narrow power window.

  I would also argue for larger valves for most harley combinations for other reasons than just airflow.

[hrdtail78]

At that point I asked myself; 'if we can make 120HP from a 95CI with 0.495" exhaust lift, why do we any more?'

Cuz most Harley exhaust systems suck?     They are designed for looks and sound..

Max

Plus the fact of.  If you really want a proper system for theory.  You have to get the front head pipe not to bend 90 degrees, 3 inch's away from the valve.  Frame mods?

[Admiral Akbar]

<Cuz most Harley exhaust systems suck? >

   Yeah, lift won't fix that but duration might. Headquarters?

   Doug

Seems to me that as long as you can increase flow with lift go for it.. If you need more exhaust flow that is. HQ cams have way too much exhaust duration IMO.. You could practically take the duration between HQ and WT cams and figure where to be.. 

Max

[Admiral Akbar]

Plus the fact of.  If you really want a proper system for theory.  You have to get the front head pipe not to bend 90 degrees, 3 inch's away from the valve.  Frame mods?

Good point..

[jam65]

Not a real frame hugger here.More along the lines of function than looks.We tried to keep the sweep open as much as possible.

[attachment removed after 60 days by system]

[Powerglides]

Very interesting thread. I am not a cam designer, but like everyone, I have my favourites as players in the overall combination of any particular build. It is interesting that it only appears to be MR and HQ cams that have the low lift exhausts. I haven't had the opportunity to try any of the Kury cams, but I do use the HQ ones.
This is a 2012 Canbus Dyna I tuned last week with a HQ113 build. HQ heads (1.900/1.625), HQ-575 cams, 10.5 compression, Supermeg 26 discs closed cap, HPI 55mm TB with 1.660 intake runners, SE Heavy Breather.
HQ-575 specs are IN 20/40 240 .575  EX 52/24 256 .530. The lobes are also asymetric.
This combo is designed for early torque rather than big horsepower, and whilst I accept that the cams are only one factor in this, it doesn't appear that the low lift, long duration exhaust is hurting anything at lower rpm. At small throttle openings on the street, the response is instant.

[attachment removed after 60 days by system]