Best way to flow exhaust ports?

[GordonE]

Hi!
Im about to try to test flow backwards through my pitot bench. I did a test the other day, but I have a rather weak design in that direction.
First of all I have a valve that sits in line with the flow to the pitot. I havnt fitted a flow straightener on that side of the pitot yet.
The valve creates massive turbulence upstream(no problem when sucking) of the valve. I think this complicates the readings. A better design would be to vent/bleed the flow insted of restricting it. Any thaughts of this? I thaught of this before, but were told that motors could explode if they didnt have enough load. Eaven though they would have load enough I think.

Secondly I have six motors that are massivly strangled at lower flows. This puts a terribly high heat load on everything, and worst of I think that the air temperature drops off at higher flow. This should not give good linearity.

What air temperature should I have for the correction formula? If it is the temperature of the flowing air which I suspect, I got a problem with my current design.

I have seen very few/none descriptions of how to make good exhaust flow tests on the pitot forum. I guess that this is the Akilles heal of the pitot bench...

If anyone wonders how my bench look, look in the Readers Flwbench projects area.

Thanks Gordon

[GordonE]

Today I flowed exhaust for the first time.
I just flipped the pitot tube 180 degrees and calibrated the flow against the usual calibration plate.

I now bypass the flow instead of restricting it from my nine motors. It works good, but it gets a little hot.
But it seems to be no problem, I tested the calibration once again after I had tested the head. Still spot on!

Pitot rules

[bruce]

[GordonE]

[bruce]

[thomasvaught-1]

Flowing heads backward is a good way to check for reversion in a port, that is the ONLY reason to flow a port backward.

Ford for years had mostly "suck only" benches (using a flow nozzle array) in Engineering.

What they did was bolt a cylinder on the combustion chamber side of the head and then the "sucked" through the exhaust port flange. VERY ACCURATE Deal was you needed adapters for a lot of head flanges vs the normal "two direction" bench.

Smokey Yunick, (I personally knew the man), used a block "Torque Plate" mounted on a special rectangular open hole bench to simulate the bores of the cylinders. He then mounted the head on the torque plate, hooked up the manifold or header and then sucked on the end of the header with his "one direction bench" Smokey, RIP, was NOT a stupid man. A picture of the set-up is in his Power Secrets book.

Tom Vaught

It took me years to get Bruce away from that PITOT crap.

[Tony]

Sucking on the exhaust side is o/k for getting some reliable flow numbers, but the port opening is then completely blocked from any type probing, or experimentation. All rather inconvenient.

I take the view (right or wrong) that exhaust flow is going to be turbulent anyway, a lot of high pressure burning fury trying to escape. The same logic applies to blow measurements of mufflers, intercoolers and pipework.

I use the blow test hole on my bench, feeding straight up into the normal cylinder head adaptor, and just accept that the flow is going to be a complete turbulent mess, at any point along the entire exhaust flow path.

You can use flow straighteners and all sorts of fancy tech to get laminar flow up to the head of the exhaust valve, then it turns to chaos trying to squeeze past the rather unaerodynamic exhaust valve seat the wrong way.
And that is going to happen whichever side of the system (blowing or sucking) is doing the work moving the air.

In my bench the suction side has a fairly generous settling chamber to give the measurement orifice some chance of working.
On the blow side, the single blower exit goes through about two feet of straight pipe before hitting the head adaptor vertically.

Even though flow is turbulent, it is always turbulent, and the blow side measurements still seem to be fairly consistent, which is all I can really ever hope for.

[johno]

Hi there guys, i have been modifying my own, and friends cylinder heads, for a few years now and, up until recently, have always had my flow-testing done on a SF-110. I decided it was time to buy/build my own flow bench, but due to the expense of a new fixed-orifice style bench, decided the pitot suited my budget! My questions are; if both types are repeatable and sensitive enough to show small changes in flow, why would it matter what style of bench your using? Would'nt it be more important that the SAME bench was used rather than comparing readings between the two types? Why does the PITOT bench get such a bad rap? ???

[bruce]

[Tony]

[johno]

Hi Tony, thanks for the info, but i was under the impression that all of those corrections are taken care of when re-calibrating the bench before flow-testing using the calibration orifice Audie supplied with the unit?! My pitot set-up is 'by-directional' and does'nt require 'flipping' of the pick-up tube for exhaust readings and i was told that by having the vac motors at the very end of the flow "circuit" , did'nt require corrections to flow figures like an SF-110 does. Have i been mis-informed? The 'dark side' is looking more interesting by the day but i just can't get over the speed of the Flow-Quick and i find orifice benches very slow and cumbersom to operate in comparison. You have got me thinking i might be greedy and have both!! :;):

[johno]

I have a brand new outa the box set of DART PRO-1 alloy heads (2.05/1.60 valves, 215cc intake and 64cc chambers) and wanted to know if anyone has any tips/results on porting them as i have'nt done these heads before?! On initial inspection it would seem they don't need a lot of work, just some light blending of the lip under the valve-seat insert, a reduction in guide-boss diameter and some port-mouth matching. Any info would be greatly appreciated!!!!!!! ???

[jfholm]

I had a FlowQuik and do not find it much faster if any than my orifice bench. I am actually using the same head adaptor that I was using with the FlowQuik so mounting time is the same. I am using the electronic unit similar to Bruce's that he sells. Once mount the time is the same - open to lift, run up to depression and hit F1. I really don't find much difference in time.

Now on the heat correction, I was under the same impression as you. Think about this though. Yes your vacuum is on the end, but when you are doing the intake you are bringing in cooler air. When you do exhaust the air is preheated by your vacuum isn't it? Or you may have a different setup than me.

John

[jfholm]

[johno]

I wipped up a radiused flow guide similar to the one you showed me john, then i flow tested the DART PRO-1 head and got 160.2 cfm @ 10" on the pitot bench (did'nt quite have enough pull for a 28" test!!). I was surprised how much they flowed outa-the-box as the heads were quoted by DART at; 276cfm @ 28" / 164.9cfm @ 10" at 0.600". Of course, i tried to pick the worst looking port to flow-test and start from there but was thrilled with the improvement in readings and accuracy the radiused flow guide gave me!! Thanks heaps mate!!!! Can you elborate on which section of the head you are refering to as the 'CSA'?