Question on Flowbench design

[rob vine]

at the 97/98 pri show smokey had his smoke-tron dyno there , was there when putting on a demo , it sure was noisy ,, quadrant scientific had a flow bench that mounted 2 heads and an intake ,dont know how well they preformed or if he ever sold any , i remember having the litature on it , pretty cool looking unit

[David Redszus]

Header suction can have the air in the intake port moving well over 100 ft/sec even before the piston starts to move down. Air has inertia, so if you wait for the piston pull, you might as well forget about getting any kind of VE.

Perhaps the true function of valve overlap and the behavior of air flow during overlap needs to be examined correctly.

The primary function of valve overlap is not cylinder filling. It is the scavenging of burned gases.
Burned gases that are not removed affect trapped charge purity. We might increase trapped air mass, but it doesn't help of 10% of the trapped gas is hot exhaust gas.

The particle velocity of air during overlap is entirely due to pressure ratios composed of inlet pressure, cylinder pressure and exhaust pressure.

Gas particle velocity as a function of pressure ratio Pi/Pe, when:

Pi/Pe.......m/s
1.1......23.78
1.05....12.14
1.0.... .0.0
.95.....-12.67
.90.....-25.92
When particle velocity numbers are positive, gas flow is inward, when negative flow direction is reversed.

Cylinder pressure ratios will change with piston position before or after TDC and flow will respond accordingly.
Unless offset by port pressures which also vary with crankangle.

Whichever the direction or velocity of gas flow, it must flow through a very small orifice which is the lesser valve curtain area of either the inlet or exhaust valve.

[Erland Cox]

David, have you looked at any cylinder pressure graphs showing pressures during overlap?
If so, what did they show?
What is the inlet port velocity of a race engine at TDC where the piston is momentarilly stopped?

Erland

[BigBro74]

Header suction can have the air in the intake port moving well over 100 ft/sec even before the piston starts to move down.
Air has inertia, so if you wait for the piston pull, you might as well forget about getting any kind of VE.

Even Briggs and Stratton lawn mower engines use overlap.[/quote]

This is the real value of the overlap as I see it. No, the majority of air/fuel mass is not inducted during this time, BUT, IMO it is the effect of giving the charge a good bit of inertia into the cylinder to begin with that is the big positive aspect of the "5th cycle"

As stated above, waiting long enough for piston pull is really missing the bus on a NA engine....

people will argue both sides of this forever but this is definitely how my picture the overlap period works, an exhaust driven INTAKE event. J

[MadBill]

2X. In effect, it is transferring exhaust energy from the latter part of the power stroke over to 'kickstart' the intake flow many degrees in advance of any piston-induced activity. Without it, cylinder filling could never reach the same levels of V.E.

[randy331]

Personally I don't see how VE at 50*-60* ABDC (IVC) can be significantly changed with more/less overlap.

I've seen small valves with steep seats, with horrible low lift flow, and wider than conventional LSA, beat up on much better low lift flowing set ups.
All a bunch of overlap, and overlap flow will do is hurt the BSFC #s.

Randy

[David Redszus]

David, have you looked at any cylinder pressure graphs showing pressures during overlap?
If so, what did they show?
What is the inlet port velocity of a race engine at TDC where the piston is momentarilly stopped?
Erland

Erland,
that is what the above pressure ratios were based on. There are three possible pressure ratios that exist during overlap.
One, is inlet pressure to cylinder pressure, Pi/Pc.
Two, is cylinder pressure to exhaust pressure, Pc/Pe.
Three, is inlet pressure to exhaust pressure, Pi/Pe.

Typically, most discussions involve option three and ignore the other possibilities. But actual pressure traces and simulations show the existence of all three.

At IVO, Pi/Pc < 1, and Pc/Pe >1. Typically, cylinder pressure is greater than either inlet or exhaust pressures. This indicates reversionary flow which is quite normal and evidenced by deposits on the backside of inlet valves. Forced induction being the exception.

At TDC, with little or no piston motion, the pressure ratio could be either Pi/Pe >1, Pc/Pi >1, Pc/Pe = 1, or Pe/Pi <1.

At EVC, we would see Pi/Pc >1, and cylinder filling does occur.

The above posted particle velocities are based on Blair's data.

A problem that often presents is reversionary flow of exhaust gases back into the inlet at IVO, which are then pulled back into the cylinder along with fresh charge. While the cylinder may be fully charged, it now has reduced charge purity.

The volumetric improvement due to overlap is mostly due to the head start that the inlet valve lift curve has so that higher valve lift occurs closer to peak piston air demand.

Overlap pressure ratios mostly affect charge purity.

Clint may have presented some pressure data showing cylinder pressure, inlet pressure and exhaust pressure during overlap.

[MadBill]

Personally I don't see how VE at 50*-60* ABDC (IVC) can be significantly changed with more/less overlap.

I've seen small valves with steep seats, with horrible low lift flow, and wider than conventional LSA, beat up on much better low lift flowing set ups.
All a bunch of overlap, and overlap flow will do is hurt the BSFC #s.

Randy

It's pretty simple. If the valve opening and intake flow isn't started well before TDC, it leaves a deficit all the way to IVC. Think of it this way: You're lined up waiting for the tree to count down but your opponent in a similar car starts 100 ft. back and arrives at the line just as you launch...

[David Redszus]

Personally I don't see how VE at 50*-60* ABDC (IVC) can be significantly changed with more/less overlap.

I've seen small valves with steep seats, with horrible low lift flow, and wider than conventional LSA, beat up on much better low lift flowing set ups.
All a bunch of overlap, and overlap flow will do is hurt the BSFC #s.

Randy

It's pretty simple. If the valve opening and intake flow isn't started well before TDC, it leaves a deficit all the way to IVC. Think of it this way: You're lined up waiting for the tree to count down but your opponent in a similar car starts 100 ft. back and arrives at the line just as you launch...

But, unless the inlet pressure is substantially higher than cylinder or exhaust pressure, the flow is backwards from IVO to TDC. In order to reduce cylinder pressure, EVO must be early enough to allow blowdown to lower cylinder pressure.

Even with forced induction, we see evidence of reversionary inlet air flow.

[randy331]

Personally I don't see how VE at 50*-60* ABDC (IVC) can be significantly changed with more/less overlap.

I've seen small valves with steep seats, with horrible low lift flow, and wider than conventional LSA, beat up on much better low lift flowing set ups.
All a bunch of overlap, and overlap flow will do is hurt the BSFC #s.

Randy

It's pretty simple. If the valve opening and intake flow isn't started well before TDC, it leaves a deficit all the way to IVC. Think of it this way: You're lined up waiting for the tree to count down but your opponent in a similar car starts 100 ft. back and arrives at the line just as you launch...

I understand the need for "some" overlap and cross flow, but I think many focus there too much with the misconception that all you can get is always good.

I think it's wrong to believe that high cross flow at overlap will equal good cylinder filling at and after BDC. The cylinder can't be filled to %115 of swept volume when the piston is near TDC.

Oh,.. and the good low lift flow/lots of overlap car would need a 100' running head start on a good small valve/steep seat bad low lift flow car!! LOL

Randy

[MadBill]

Don't forget that in the 'tune range' of a good header the exhaust negative pulse around TDC can pull the cylinder pressure down by as much as 7 psi., a far higher pressure delta than will ever occur due to piston motion. The reversion David speaks of usually occurs at lower revs, 'off the cam'.

[twl]

Enough is enough, and too much is too much.
If we can evacuate the chamber and get flow started in 20 degrees, the there is no point in sucking mixture out the exhaust for 50 extra degrees BTDC for no good reason.
We do everything possible to foil the crossflow at overlap with our big valve crossflow hemi, because at TDC our chamber volume is only 35cc, and that is easily cleaned out and filled without any excess degrees of overlap needed. I think overlap is needed, and I also think many people are prone to over-do it..

[David Redszus]

Don't forget that in the 'tune range' of a good header the exhaust negative pulse around TDC can pull the cylinder pressure down by as much as 7 psi., a far higher pressure delta than will ever occur due to piston motion. The reversion David speaks of usually occurs at lower revs, 'off the cam'.

I would very much like to see data showing a negative exhaust pressure of 7 psi. If inlet pressure were atmospheric, we would have a Pe/Pi of .5 which would produce gas velocity close to choked flow, but it would go out the exhaust and not stay in the cylinder.

Piston motion will frequently approach a delta of 7psi, depending on the ability of the intake valve to keep up with piston air demand. At that point the critical pressure ratio is reached and flow is choked.

[MadBill]

Such data has been posted on S/T David, but I can't locate it at the moment. However Vannick's software, EngMod4T, shows exhaust port pressure at TDC ~ 10.5 psia, Vs. the inlet port's ~ 14 and ~6 Vs 11 respectively at EVC per: http://www.vannik.co.za/EngMod4T.htm and Dynomation produces very similar curves. EngMod4T also shows the highest inlet/cylinder pressure delta at ~30° BTDC.

[digger]

Such data has been posted on S/T David, but I can't locate it at the moment. However Vannick's software, EngMod4T, shows exhaust port pressure at TDC ~ 10.5 psia, Vs. the inlet port's ~ 14 and ~6 Vs 11 respectively at EVC per: http://www.vannik.co.za/EngMod4T.htm and Dynomation produces very similar curves. EngMod4T also shows the highest inlet/cylinder pressure delta at ~30° BTDC.

viewtopic.php?f=15&t=37398

viewtopic.php?t=31606