True throat/window, low port heads

[blaktopr]

200, good points. I have yet to add the intake because of the port entrance shape, did not want to grind on it until I get close with the head. Even though the #7 test shows better overall avarage, I like where #9 was going. The turbulence is murdering the wet flow, period. #7 flow was very unstable regardless what the tests show. At lower lifts, introducing a pitot tube or liquid into the runner put the port into turbulence. So dry air and without anyway disturbing the entrance air has a later stall but I think on a running motor, it would stall. The wet flow shows at turbulence how bad the burn path would be and the "stacking" of fuel in one part of the cylinder. I want to get rid of it and if I can, maybe a test #9 would dominate test #7. It's probably the discharge giving me this problem. The chambers were allready relieved by the previous porters so there is no way for me to put anything back to keep the air from slowing down so much. Once the valve clears the seat and approaches the deck (.500 is at the deck) it just goes to crap. Something may need to hold the air back so to speak, but I'm not sure how to go about it. If I try sinking the valve and working the seat into the chamber, the flow curve would go down because of the window area less at all lift incriments. Not to mention that it may still go turbulent but at higher lifts. Wether it does or not, I would need to run more lift which means a cam change to a roller. I will make a try at machining a smaller valve to seat on the 55 degree cut and see what happens with the discharge. I will make an assumption now and post to see if I was right. Flow curve will go down, and once the valve approaches the deck, turbulence will still happen because of the larger area in the discharge area from the smaller valve diameter.
Chris.

[blaktopr]

Seeing where the air is going doing wet tests as the port approaches turbulence, the faster the air gets in the port, the larger the speed diferential at the exit on the side of the exhaust, creating a higher pressure front, forcing the air to go toward the cylinder wall side and separate at the same time because of the different speeds in the port and chamber. My .02$ I wonder if I am right.

[blaktopr]

Tried a smaller valve with a 55 seat and 45 backcut. Valve size is 2.03. It did what I thought it would do. It approached 80 percent, (253 cfm) and went turbulent. Entire flow curve was down and looked to start to catch up at .475 and did not drop off as bad but stayed level. Wet flow droplets were a fraction smaller and hit the cylinder a little lower from the chamber. When turbulence did happen the wet flow was worse than with the larger valves with 45 seat. The fluid actually climbed the walls and went back to the valve !

[blaktopr]

Did some reading and going to chew on for a little while the use and testing with the smaller valve with this port. Maybe something to test in the car. Port with 1 valve size and transitions to match and then another. See how some theories stack up in regards to the ram effects due to the different seat throat sizes. I feel like this now ??? I guess if I am to learn, I must swap the heads out to see where the results are and cross reference to the bench results.

[200cfm]

We salute the effort and focus. Hang in there! You are ahead of us because you can make wet flowing and observe flowing changes with the valve changes, so that is very promising. Time at the track dyno will be the final verdict. What grinder and valve resurfacer did you get? I hope to do some stone practice this week.

[blaktopr]

200, The tools are both B&D. 11/16 Super Surface valve refacer, and the seat grinder. I am very pleased with the valver refacer. Once I cleaned and lubed the parts, machine runs great. Plus it came with the cabinet and some stones for both machines. The collet is very good, have not checked runout though. Got them from 2 separate people both in PA. Total for both tools was 276$ but I had to drive 12 hours which included dropping off my snowmobile that I sold.

Made a little headway today. Redid the seat with 5 angles. 55 seat with a 68 and 80 bottom cut. 45 to slightly blended 30 top cuts. ( can't kill the opportunity for other valves) This was for the 2.03 valve. Slightly widened the apex of the ssr only, not the whole area. Still narrower on the roof. I also took a little clay away from the tail behind the guide. Turbulence still happens but at higher lift now. Port can go to .550 to .575. It also doesn't take much to reattach the flow so once I test a larger valve with 45 seat, I will put on the intake and see if the port stays calm. Wet flow looks very good even now if it goes turbulent. The flow finds an easier time to reattach itself while using the water probe (1/16 tube) and the fluid position in the runner has an effect on the attachment of the air over the ssr. Droplet size is the smallest I have seen yet When turbulent, droplet size stays small, has less accumulation at the walls and has a smaller area of high pressure in the cylinder. I have to try to get a better look at the chamber and do an imprint test next time.

The runner's velocities are where I think they should be. After all, this is my first head using all these tools rather than the flow numbers. No exact numbers to post yet but while probing still found that the fastest speeds in the port is at the ssr apex at about 400fps. The seat areas around the valve as much as I can probe range from around 290 to 350 fps. Checked before the seat and after and watch the rate the speeds fall going further away from the seat. Basically, this port with the 2.03 valve and 55 seat has the fastest speeds at the seat in relation to port centerlines and even againt the surfaces in some spots including the push pinch wall apexes. Also, with string in the runner, showed that the air follows the ssr better and produces a larger arc as it transitions to the valve still following the shape of the ssr. Many tests ago, the string was more straight as it found it's way to the valve. I'll try to post some pics.

Low lift flow did go down but overall all aspects of this particular port shows "me" some potential especially with testing big valve vs small, and proper speeds and wet flow characteristics through a port. Enough for now, here are this port's flow numbers at 170cc and 2.03 valve.

.200 126.7
.300 167.3
.400 226
.475 250.4
.500 255.3
.525 257
.550 257
.575 255
.600 255

Chris

[blaktopr]

By the way, a good friend of mine thinks I will not find anything better, says, "don't you think people allready tried this?" in relation to Pontiacs, and that more guys found gains using the "bigger" valve. (2.15-2.19 vs stock 2.11) Thinks I am wasting my time with these "small" valve tests in my search for trying to make a more efficient port and to learn the effects of applying more science in an intake tract system.
Chris.

[larrycavan]

[blaktopr]

[blaktopr]

wet test of this port

[blaktopr]

These means of documenting the testing is better, BUT I have to clean the seats a little more and try textureing the chamber. Not to mention looking more at where the fluid is stacking up. I was not sure to show this here or in the wetflow discussion.

Chamber vid 1 shows the stacking of fluid at the ssr of the chamber and the rotation of fluid and stacking behind the plug. It is just like what happens in the bed of a truck. High pressure behind the plug because the low pressure and high speeds passing over and around it. The fluid stays there and starts to rotate with the direction of the high speed air and starts the trend over producing the vortex. There is not enough energy to just introduce itself back into the airstream unless maybe more stacks up. OR Once a cirtain amount of fluid is built up in that vortex pocket, (it allows the boundary layer to ride across it, after the plug, not letting as much back in.

Chamber 2 is taken through a mirror. When the port is turbulent the fluid rotation is counterclockwise in relation to the screen. Once it goes back, you can see the normal clockwise rotation where the two fronts converge at the wall at the center of the exhaust valve. Notice when it goes turbulent what the fluid does on the cylinder wall in test 1? I think even when not turbulent and the stacking and clockwise vorticing behind the valve at the ssr is a result of the 55 seat. Air is rushing past, with particle flow, at high speeds. Wet heavy fluid clings on to the surface and moves across the chamber in the directions dependent of the direction of air below the chamber in the cylinder. Fluid that gets trapped in the chamber is held in from high pressure above the low pressure front created by the high airspeed exiting the seat.

Well something like that, just rambling off what my brain is telling me. My .02$ thats all. The port is more turbulent because of the dykem. You can see why I don't want a turbulent port. Also this shows where power may be the same turbulent or not, the BSFC #'s may be higher when turbulent because the fuel introduced is not being used as effectively. Now there has to be more fuel added to make up for the fuel not being burned or burned slowly. Now you have to increase your spark lead to start the process sooner because of how inneficient the burn is, just to make the power close to what it is supposed to be.

I wonder what texturing the chamber like in the runner can do in addition to working with the pressures that reside in the chamber/cylinder can lead to. I know that it is said sharp edges lead to hot spots, but this test shows what a smooth chamber does to the fluid once it is held there.

No expert here and not stating as fact but my ideas from observations in all my tests.

Chris.

[blaktopr]

Man, I tell ya. U-tube's vids come out so crappy compared to viewing on your own program on the computer.

[Moriniman]

You need to optimise the resolution, bitrate and compression so that Youtube doesn't make it worse. Do a google search on optimizing tips.

[larrycavan]

I've been looking at the wet test vids and pondering this whole wet flow thing. I dont' know. Another can of worms to be untangled.....

The events in a live engine will take place so much faster through the entire lift cycle that I just don't think steady state testing can come close to duplicating the actual event.

To begin with, on the initial start of the intake cycle, the exhaust valve is open and the piston is rising. That, in my mind, sets up an entirely different pattern than you're going to see on the bench.

You set the valve at a particular lift, run the bench for a bit, watch the pattern, move to the next lift, things change.....OK, fine....Trouble is, the liquid will probably behave very differently in the live event because it has mass and the time factor is so much accelerated.

I would think the combustion chamber is going to be in a steady state of turbulence from piston motion. Would piston motion not affect what happens to the fluid in the chamber.

In other words, we're not able to accurately simulate the condition in the chamber itself.

That doesnt' mean I think it's a waste of time. I'm saying there's a lot of considerations to ponder over to properly evaluate a port's live flow pattern. More so than not.....I think it's one big guessing game. I see usefulness in the testing but even with this additional test process, we're still very blind to the actual conditions in the chamber of a live engine.....We are not able to view the affect on the fluid from the compression event. Fluid that appears to be in a bad location may relocate on us during compression or vice versa....


JMO [for now...and subject to change]

Larry C

And Chris....just so there's no misunderstanding. I think your test processes are awesome. I read your posts, sit here and say to myself....Yep...that's what I'm thinking also....Please continue to post your efforts and results. I look forward to them.



Edited By larrycavan on 1227793525

[blaktopr]

All valid points, and more information for when checking the results once it is on a running engine. I have lookied at piston crowns and chambers throughout the web and my own motor. There are darker spots where this stacking occurs. Going to try to write my thoughts without being all over the place. First I will use 200cfm's chamber pic as a reference. The pic is on the bottom.

We know as the piston ascends, there is more pressure building. We know that as it approaches the squish band or pad in our case, mixture is forced toward the open part of the chamber. But I feel that because we still see similarities in piston/ chamber markings with the static tests, there still may me a swirling movement while the piston reaches tdc. both combined still produces high and low pressures inside the chamber. For this discussion, lets just call it "swirl". This swirl still will be moving at a direction that still may change, but the speed may increase as the volume becomes smaller as the piston approaches tdc. This swirl motion gives you the pressure differences during this event. The pressure differential may become greater because of the increase in speed of the swirl vs the diminishing area around it where higher pressures reside. Any dropped out fuel will still remain (to a cirtain extent) on the surface of the piston and chamber where the higher pressure is. When the piston reaches the deck, the swirl motion effects the trapped fuel on the chamber/piston suface by giving it more speed but still may follow the paths you see in the vids. There still may be the high pressure holding the fluid to the surfaces and in the center, is the high speed vortex.
Imagine the vortex as a rubber ball. As the piston gets closer to tdc that ball gets squeezed from two sides, chamber/piston. because of the higher pressure differential, the high pressure (lets say the boundary layer) becomes greater and the rubber ball gets smaller as it gets sqeezed more. Having nowhere else to go and the speed of the vortex "swirl" gets faster, but never touches the surfaces, but gives some energy to the high pressure area to continue the movement along the surfaces.
By looking at some static tests and chambers shows me the possible directions the fluid may take and where stacking may occur during the pressure differentials. My thinking is maybe why some of those attempts to alter the charge in the chamber did not work but chamber/piston top design does. Because of this possible theory, what you do on the surfaces does not effect much because the novement of the charge is mostly happening away from those surfaces.
Before I lose my thoughts, I will stop here. Take a look at 200cfm's chamber and reference it with the vid. The motion is the same but the amounts of stacking changes a little through some different lifts. I will try this weekend to do a whole lift range test and post.
Thanks,
Chris