flowbench design - floaing depression

[FTCH]

Hello, I am new to your great forum, I would be happy to add my experiances. Do you have or do you want to have a more deicated thread in regards to overall cylinder head development. More so than flow testing, both subjects I am very interested in.

look forward to being a part
cheers

[106-1194218389]

Do a search and be sure you set the time frame to cover all older dates. There may already be one that fits your needs. If not then start one that fits your needs. And welcome aboard!

John

[thomasvaught-1]

Quote:

"Tom,

Please enlighten the board with the meaning behind your post of "zz"."

Larry, I wrote a post that I attached to this topic about some flow bench history.

Basically it was a quiz for the members.

After I thought about it I decided that it was not the right place for the quiz as the posts were going in a different direction.

I was able to delete the quiz but not every letter so I changed the last two letters to ZZ.

That is the reason for the ZZ post.

Tom V.

[larrycavan]

[49-1183904562]

[color=#000000][quote]I have to recall my previous post. I have been thinking about this and I am thinking differently now. After watching this video In Cylinder Video I am not sure the depression is super high at low lifts in an actual running engine. We need to remember it is at camshaft overlap when the valves are at low lift. Also the piston speed is slower than at higher valve lifts. On a Small Block Chevy the piston speed is usually fastest in the 75

[302ford]

Heello
Sorry was gone for so long.
I am in process of buillding the bench as described in go fast news.
I will keep u posted. Hope to test intake manifolds with this project also. I think it is is just another way of testing and hopefully will show promise.
regards.
Randy

[1bolt]

First off let me get out of the way that I'm a newb here and basically a newb to flow benching (about 6 months under my wing now), but I happen to think Vizard's theory about floating depression makes more sense than static testing at 28". Unless I'm missing some fundamental variable, math or scientific principle (which I admit is entirely possible, even likely). That said something rubs me the wrong way about testing a port with static depression, in a running engine the depression is not 28" for more than brief moments during the cycle...

This would be kinda like running your engine on a Dyno, measuring torque only at 3500 RPM's and trying to extrapolate the rest of the Dyno graph from the one number...

At higher depressions flow dynamics (to borrow Larryc's term) should be different (shouldn't they?) than at lower ones, currents, and velocities, flow reversals, all the things that make porting as much of an art as a science.

As I understand it when the air velocities are higher turbulence changes and those changes effect how the air interacts with the surfaces and shapes of the port... Why then would you purposefully flow your ports at a velocity that is much lower than it will see on the engine (at low lifts) and significantly higher (than at high lifts)

What if you spent hours playing with clay trying to figure out a horrible flow reversal between .150 and .250 at 28" only to flow it at 100" to 60" and find it doesn't even exist at those depressions?

Shouldn't we measure the flow of a port at any lift as close as possible to the actual depression that would be found in a running engine? Obviously we can't acount for every pulse, or variable. Certainly not even every variance in depression the combustion chamber sees as the VE changes with RPM and scavenging. But basic floating depression is an EASY one to add... And it's actually MORE simple to set up than static depression. If it then more closely resembles the conditions of a real engine, then In my mind it becomes more of a "Why Not?" than a "why?"

Just combine the ratiometric bench with the floating depression pressure concept (lets coin it FDp so I can shorten this already long post a little). In fact too many posts in this thread seem to be confusing David Vizards first flow bench, with the concept of a FDp bench. ANY bench can use FDp. It is not exclusive to his cheap first bench. (BTW anyone else think he could have just added a depression pressure U tube on the vac side of the orifice to get CFM numbers instead of that calibration plate dance?)

Isn't the electric motor variability the same in either case? I don't know how you guys make your motors run perfectly static test pressures with perfectly static line voltage, but A floating ratiometric bench compensates for the motors varying just like it does for barometric pressure or air temp.

If nothing else the simple fact that 28" was arrived at rather arbitrarily should at least have us asking if it's really the best way... Everyone loves Smokey but lets face it there's no magic in 28" it's convenient for comparing numbers in a magazine...

Are you leaving power on the table flowing .100" lift at 28" that you might have found flowing it at 120"?... Hard to know...
But Intuitively speaking, it's hard to imagine how 28" of test pressure can accurately show how it may flow on the engine pulling as much as 100 more inches, or 18 less.

[Tony]

Static depression gives you a fixed reference condition to measure flow, which is repeatable by adjusting the test pressure,

Floating depression changes with how your vacuum cleaner happens to be running on the particular day.

At higher depressions, the flow still follows the square law formula, just as your measurement orifice does.

And no, the electric motor variable is not the same in either case.
With a fixed depression measurement, you vary the motor speed or some kind of throttling device to generate the fixed test depression.

With floating depression, you just let the motor do it's own thing, and ASSUME it never changes...... ever.
Hardly the same thing.

Nobody passed a law that says all constant depression flow measurements have to be made at 28 inches. Many people here test at lower depressions, and some at even higher depressions.
It all depends on how big and powerful your flow bench is. But it really does not matter, except at higher depressions the numbers are larger, and therefore easier to measure and more sensitive to change.

And no flow bench can EVER duplicate the vastly complex thermodynamic effects, extreme pulsing, and exhaust reversion present in an engine running under load. If you think a floating depression measurement duplicates all that you are quite mistaken.

If you are happy with your floating depression bench, that is all that matters. Good luck to you.

[1bolt]

[bruce]

Make note that they are talking about measuring flow in a pipe which you will find on the forum here has been discussed in great detail about why it is not a good idea to do.

A well designed orifice style flowbench will have plenums to allow the flow to settle down before it goes thorough the measuring orifice and as such is not effected by bench/pipe turbulence.

[49-1183904562]

[color=#000000]1bolt,

First off do you have a name? I get wiggy sometimes in these posts and I have an issue thinking about it when I am posting to an object and not a person. Yes it is a personal problem and I should leave it at home or get some medication but the PTS forum is a great friendly place and being congenial is easier with a name. thanks

OK, Since you are a newbie per say I would suggest you read everything by Charles Taylor and Prof Goron Blair first. Then spend allot of time experimenting on your bench then you can read Vizards theories. Please do not get me wrong here but the education and experience between the gap is mind blowing once you start studying about these guy

[thomasvaught-1]

[color=#000000]While I agree with the vast part of your reply Rick, I have to make a correction/modification to your post:

"During the carburetor wars Holly and Carter both had advertised CFM these carbs were flowed at 40

[49-1183904562]

[color=#000000]Thomas;

It has been a while for me I worked for Federal Mogal in the late 80's early 90

[1bolt]

Sure Rick my name's Simon I usually put my name in my sig which I usually add after I post a few times to a forum, but haven't yet.

I think we're kind of sliding off the floating depression thing a little so let me take the car analogy a little further.

If you drove a race car through a straight tunnel at 28 MPH and 120MPH you would expect the car to behave basically the same... it goes through straight and predictable. You could use a simple formula to predict how long it took to get through the tunnel. We'll call it the "straight tunnel square root formula"..

Now try the same thing in a curved tunnel... the car hugs the inside wall perfectly at 28 MPH but it slams into the outside wall at 120 MPH...

Now obviously you can not use the straight tunnel square root formula to accurately predict how the car goes through a curved tunnel.

I bet not a single person reading this would consider using that forumla for a curved tunnel... Right? Not only does it need to model momentum, but all sorts of things have to be added, like the suspension, the friction of the road surface, how sharp the curve is... The Square root formula doesn't do curves.

Yet, some here are pretty comfortable using the square root formula -- Which is based on air flowing through a sharp edged orifice, or smooth straight pipe! -- To predict that the flow patterns in a curved port are the same no matter what speed the air is moving...

Air has mass, it has momentum, it does not behave the same way at different speeds, and it certainly can not be predicted accurately by a formula that is only meant to predict flow through a sharp orifice or a straight pipe...

I'm not even going to get into the fact that a port is not just a curve but a very convoluted shape with multiple pinch points (venturi), compound curves, bumps, a valve stem splitting down the center and a valve head restricting flow. AND an irregular cast surface that CERTAINLY is not the fine square edge of a "sharp" edge orifice; or smooth walled pipe.

The argument is that there's no reason to flow using Floating depression, because the square root formula predicts the same corrected CFM no matter what depression is used... The formula was never meant to predict flow in complex shapes. It's invalid for curves, This also invalidates it as an argument against floating depression.

[bruce]

Lets see hands raised for those who built a bench with the orifice in the pipe?

The orifice does not "see" the flow that is bouncing around it is only seeing the resultant pressure from that flow. If it bounces around and loses flow the orifice will see that result. If you build a bench where the orifice is in a plenum the formula is not effected by that turbulence. Now stick the orifice in that pipe and the results are different and yes the Cd of the plate will change. I can measure the same orifice plate at various static pressures and end up with the calculated CFM's for that plate. I just did this with 3 plates that flowed 1" static, to prove my math I flowed the same plates at various static pressures and guess what? Those plates were dead nuts on, no hokus pokus wave a magical wand at them, they flowed what they said they'd flow.

Try this experiment, put an orifice plate on top of your bench with nothing around and measure it, now cup your hands around that hole at different distances away from the hole and measure it again, bet you see a difference in flow? You are changing the Cd and that Cd will change with the static pressure. Flow to an orifice plate comes from the side not directly into the hole if you change how the orifice plate "sees" that flow you will have all sorts of problems.

We are " pretty comfortable using the square root formula" because we know you need to have the measuring orifice in a plenum. The bench design on GFN uses the measuring device in a pipe and as such has the problems they talk about. There is no discussion as to what location nor size of hole the pickups should be? The orifice calibration disk they talk about has a rounded inlet on their holes? How do they control that when machining? They give you a rough idea of what the radius "should" be? A sharp edge orifice plate can be machined to very exact dimensions, no guessing it machines to a known Cd, I know this I have machined quite a few of them

In the end though it is what the person building the bench is wanting to achieve and the flowbench is really just a tool to use. I was always a pitot style bench guy, but now that I have built my PTS orifice bench, I'll never go back to flow in a pipe testing!

I'm so satisfied with my orifice bench that I went back to those people I know I sold pitot parts to and offered to send them my plans No charge to make the switch if they wanted to.