large inlet valve size vs small port?

[ampor]

Hello

Trying to learn a bit more about head flow, and wondering what is the effect of installing a larger intake valve without enlarging the port. I'm aware of previous info posted here about having the valve throat end up at 86-88% of the inlet valve diameter, but what happens if there are constraints that prevent opening the throat? What to expect if your valve throat is only 80% of inlet valve?

Thanks

[Mignon]

It's not exactly complicated. You should be able to answer this for yourself. A bigger valve will show flow advantages at low and medium lift up until the point at which the port size becomes the limiting factor but you can't get more peak flow than the port will supply regardless of how big the valve is so there's not much point fitting big valves if you can't also size the ports to suit.

[Stan Weiss]

^^^^^^^^^ All true.

If your head has some kind of valve lift limit and flows less at max valve lift than with the valve installed upside down. You might try seeing what a larger valve will do.

Stan

[ampor]

"You should be able to answer this for yourself. A bigger valve will show flow advantages at low and medium lift up until the point at which the port size becomes the limiting factor but you can't get more peak flow than the port will supply regardless of how big the valve is so there's not much point fitting big valves if you can't also size the ports to suit."

I knew an unchanged port would limit peak flow and higher rpm power; what I didn't know was how the larger valve would affect flow and pressure recovery and possibly reduce torque. I have never seen a back to back test on a mod like this. I'm looking to increase torque as much as possible under 3500 rpms. We have an old, heavy van that is underpowered, and we're looking to boost torque at cruising speed which is typically 3K rpm.

Looks like I'll give it a try, thanks for the answer.

[ampor]

Stan said: "If your head has some kind of valve lift limit and flows less at max valve lift than with the valve installed upside down. You might try seeing what a larger valve will do."

Don't quite know what you're saying here.

We don't want to change the cam, or increase the engine size. Just looking for a low budget band aid for our old van.

Thanks

[Inline__engine]

im curious as to why people think low lift flow as indicated on a flow bench gives more torque below the torque peak....

[tapkaJohnD]

Flow through a port is limited by Pouiseuilles Law, that says that, given all else being equal, flow is proportional to the FOURTH POWER of the radius. Doubling the radius, if that were possible, puts flow up SIXTEEN times (2x2x2x2) So any increase in port size is worthwhile, if allowed in the regs.

But that relies on laminar flow. Around a valve the flow is turbulent, and there flow varies as the SQUARE of the radius. So doubling the radius would only get four times the flow. (the space around a valve is an annular orifice, but approximates to a circular hole)

So on the physics, as far as we understand it, a bigger port is better.
John

[227bhp]

Stan Weiss said:

If your head has some kind of valve lift limit and flows less at max valve lift than with the valve installed upside down.

Stan

What do you mean by that?

[227bhp]

tapkaJohnD said:

Flow through a port is limited by Pouiseuilles Law, that says that, given all else being equal, flow is proportional to the FOURTH POWER of the radius. Doubling the radius, if that were possible, puts flow up SIXTEEN times (2x2x2x2) So any increase in port size is worthwhile, if allowed in the regs.

But that relies on laminar flow. Around a valve the flow is turbulent, and there flow varies as the SQUARE of the radius. So doubling the radius would only get four times the flow. (the space around a valve is an annular orifice, but approximates to a circular hole)

So on the physics, as far as we understand it, a bigger port is better.
John

Ports aren't always round, especially when they have been reworked so that rule is generally useless, CSA (cross sectional area) is generally a more accurate way of measuring them.
A bigger port is not better due to something called 'velocity'.

[GreenV8S]

227bhp said:

What do you mean by that?

I took it to mean with the valve literally inserted into the valve guide upside down, so just the valve stem is in the port, simulating a valve with infinite lift.

[227bhp]

GreenV8S said:

227bhp said:

What do you mean by that?

I took it to mean with the valve literally inserted into the valve guide upside down, so just the valve stem is in the port, simulating a valve with infinite lift.

I see, that would be BPF - bare port flow, or at least BPF + valve stem.

[Mignon]

tapkaJohnD said:

Flow through a port is limited by Pouiseuilles Law, that says that, given all else being equal, flow is proportional to the FOURTH POWER of the radius. Doubling the radius, if that were possible, puts flow up SIXTEEN times (2x2x2x2) So any increase in port size is worthwhile, if allowed in the regs.

But that relies on laminar flow. Around a valve the flow is turbulent, and there flow varies as the SQUARE of the radius. So doubling the radius would only get four times the flow. (the space around a valve is an annular orifice, but approximates to a circular hole)

So on the physics, as far as we understand it, a bigger port is better.
John

Utter b****llocks

[227bhp]

Well yeah, that as well. Here's me being all diplomatic and polite on a Monday

[GreenV8S]

ampor said:

Just looking for a low budget band aid for our old van.

Have you thought about adding a small turbo? You could limit the boost at whatever torque you think is sensible for the engine. Assuming you're only looking for modest gains from fettling the intake side you wouldn't be looking for high boost levels to achieve the same affect, and once you've got the charge in to the engine I don't see that it will make much difference whether you did it by demon tweaks on the induction, or just by raising the ambient pressure a bit.

[tapkaJohnD]

Please, 227bhp and mignon, refute physics!

For the Poiseuille flow in a rectangular port, the error is 13% for a square Xsection, if you use half the side as 'radius'
And of course, it's pulsatile flow, not continuous.

John

[227bhp]

What if it's oval or rectangular?

[Inline__engine]

tapkaJohnD said:

Please, 227bhp and mignon, refute physics!

For the Poiseuille flow in a rectangular port, the error is 13% for a square Xsection, if you use half the side as 'radius'
And of course, it's pulsatile flow, not continuous.

John

i know you mean well but you are mis-applying the equations. air flow in an engine is far too compressible and flow very far from being laminar so you need to use different physics if you want it to work properly.

Poiseuille is really good for modelling oil flow in the oil gallerys of an engine, blood flow in veins etc and things of that nature that are for the most part incompressible or low levels of compressibility and flow in the laminar regime rather than turbulent as determined by the Reynolds number.

if you don’t believe us find someone with a flow bench and do an experiment flow a 40mm pipe and 50mm pipe and you will see that airflow is basically proportion to cross sectional area provided you keep the entry and exit flow coefficients the same and of course the lengths the same.

[Mignon]

tapkaJohnD said:

Please, 227bhp and mignon, refute physics!
John

Physics isn't what needs refuting! What needs refuting is people latching onto something fancy sounding they came across on google, didn't understand in the first place and tried to apply it to a situation it's specifically not designed for.

If you'd done even the most miniscule amount of reading about Poiseuille's Law you'd have seen it's only applicable to 1) incompressible flow, 2) in thin tubes, 3) that are very long in proportion to their diameter and 4) where there is no acceleration of the fluid. So nothing remotely to do with gas flows in engines which as any fule do know always scale with cross sectional area.

[Mignon]

Oh and btw. The flow in a cylinder head port isn't even laminar either or even remotely close to it, not that Poiseuille's Law would apply even if it were. Flow is only laminar at very low Reynolds numbers, below 2100, and for something as non-viscous as air that only happens at miniscule flow rates, many orders of magnitude below the sort of flow numbers that occur in engines.

Edited by Mignon on Tuesday 14th March 06:56

[Mignon]

Inline__engine said:

im curious as to why people think low lift flow as indicated on a flow bench gives more torque below the torque peak....

Or how in god's name anyone was supposed to divine that the original question about "the effects of" was actually to do with increasing low rpm torque in old van engines.

Anyhoo. If the OP wants to increase power and driveability, even down to lowish rpm, then you want to try and maximise head flow with as little metal removal as possible to preserve port gas speed. The best way to do this on most heads is put a 30 degree backcut on the inlet valves to match the seat width in the head and mild porting work on the short side radius to straighten this as much as possible. However every head is different and we have no idea which one yours is.