SSR and Coanda effect

[jfholm]

I was thinking about making me a device like in the attached picture to test the air flow in a port. I have just let one string flow in a port while I was flowing the head and it is very interesting. The device I want to make will have multiple strings of different colors shaped closely to the port I am flowing - then I will hold it up to the port and see where the strings go. The strings will be long enough to go into the cylinder and I will have to make a flow cylinder that is clear to see all angles.

John

[slracer]

John, that is a good idea. Make each string so it can be individually changed in length during the test as a long string will go through a turbulent zone and the downstream drag will straighten it back out. Making it shorter will "find" the problem area you wouldn't see with only a long string. FYI In wind tunnel testing, we didn't call them "strings", the preferred term was "frog whiskers". It sounded so much more "technical"! :p

There seems to be a little misunderstanding of what the Coanda effect actually is. The air will stick to whatever it hits in most cases, be it the wall or a valve, etc. The speed of that air and the angle of impact will determine how turbulent the resulting collision is and where the air goes next. The Coanda flow is the part where it follows the surface where you don't expect it to. In 200cfm's case, when it gets to the edge of the valve, it will continue to follow the valve over the seating area and actually on to the face (a little) traversing an angle more than 90* rather than continue off as Newton intended (undisturbed in a straight line). There are definite limits as to how fast the air can go before it "believes" Mr. Newton more than it does Mr. Coanda. The viscosity of the fluid (it's "stickeyness") also plays a part in determining that limiting speed. As with everything Mother Nature does "for" us, there is no free lunch! Somewhere in all of this, you pay for what happens, but sometimes you get primerib for the price of a hamburger. Our job is to determine IF this is one of those times! This is ALMOST like being back at work and "arguing" with the experts about which one is right. A lot of theories and only one answer (but which one?). -- Doug

[jfholm]

Doug,
being as how you have wind tunnel experience, is it true an egg is more aerodynamic big end forward? If that is true, then with my pear shaped body I could sit on your motorcycle backwards butt first and go fast as heck and set a new record

John

[slracer]

John, the coefficient of drag is only half the story, the other half is the cross sectional area. How big are you around?

:p -- Doug

[bruce]

[jfholm]

[blaktopr]

UGH. I hate trying to put my head back into something. This topic is very good. I have separation in three different ports, but have an answer for these that I have to still try and fix before moving on. It is with speeds and pressure differentials in the port. But this is not what I want to add.

It's what I see in wet testing to give out to add to your info and give more to think about. How the air exits around the valve in response to any changes alters the "weather" in the chamber. At every point, lets say at every degree point, 360 degrees around the valve, has a different speed. How it recovers in the chamber and or how it gets "thrown" in the chamber creates it's own weather pattern. This weather pattern has it's own profile of pressure differentials. These differentials (not completely proven though) may effect the SSR's ability to keep the air attached and effects where the liquid goes, in the port and in the chamber. The liquid properties from the "weather" has been my main focus lately.
Here are some observations on shallow valve angle heads.

1. equal short and long side flow ( string follows centerline to valve) no turbulence, equal distribution of liquid converging to wall of cylinder by exhaust valve. Area of highest pressure with least speed between IN and EX valve with small vortex there.

2. NO tubulence, BUT long side favored. This is when a port still goes up in flow but at smaller incriments. This happens at or below the convergence point but offers no signs of turbulence with the manometer or noise. This phenomenom is interesting because I seen this in a few heads and even found others to have the same thing. But they thought the port was still good and not turbulent. But flow is separated. Fluid exits primarily at the long side and travels the chamber/ wall in the same direction around the cylinder. Looks like swirl. But because of the high speeds of the wall side of SSR, sucks the liquid back in and chucks it on the cylinder wall at the shrouded part of the IN valve. Port flows air dry, but is DISCUSTING wet.
3. Big throat/big valve. Moves air, no turbulence, looks great on the bench. BUT same as #2
4. In cases 3&4, the vortex will happen right behind the plug as the high speed air passes over it. This vortex is super close to the plug. Must be why companies like MSD are in business.

Let me know if any of this helps or want any more info,

Chris

[slracer]

Chris, What do the head and top of piston look like? You say shallow valve angles. I have a small hemi with quite a bit of popup on the piston (12:1 CR). I'm trying to figure out how that would influence the wet flow and how I could see it. Any thoughts? -- Doug

[blaktopr]

Doug, most of the stuff I am working on are 14 degree Pontiac castings. Especially the old stuff. (gotta crawl first) So far unfortunately I personally have not seen entire back to back flow data to running engine signs. What I can tell you that there are similarities on my old motor of where the burn is and is not. What I see on the bench represents what does not get entirely burned. Basically, the outer edge of the quench pads are the darkest getting lighter toward the center. You can also make out the vortex between the two valves. I have been looking at piston and chamber pics all over and speculating what is happening "on the bench" Plus, I am figuring that when in a running motor and velocities are higher, the #2 and 3 phenomenom is happening even though at 28" static and peak lift, there are no signs. Thats why I can't wait to wet test at >28" The fact that a port does not "back up" and lose flow, but does separate off the SSR at higher air speeds interests me. I'll be getting to your post Doug.

So far, for me, my separation problems are comming from ( i believe) the area beyond the push pinch still becoming smaller. The speeds are still too high closer to the apex of the SSR. Not enough time and would be too much of a transition to get large enough at the apex. I gotta try to basically hand "CNC" the port to get the transition started early enough and with enough area to make the turn, keep turbulence from happening before or at the apex, and to keep the wet stuff from falling or transitioning somewhere else through some freaky pressure differentials through the rest of the port. Not to mention addressing the bias through different shapes.

I have an Edelrock head ported by a Pontiac vendor. 2.19 valve. Now we all know and read to keep velocity highest there. We are aware of the ram effects, pressure differential, etc. What I also see is the possible place where some of the HP is derived from. The head flows well, and you will find it on many engines. Head peaks at 318cfm. Get this, over 400 fps through the pinch! But no turbulence. Well the air/liquid just shoots past the apex and out the long side of the seat. This is creating the "swirl" effect around the chamber. Inersia (? spelling and useage, help me out guys) is keepin the mixture pinned to the outside ring with the highest concentration of mixture there and around the cylinder. Just like how I saw the piston in another motor with a different head with similar properties.

Sooooo, here is a little tidbit of info I found for you guys. SSSHHHHHHHHH. Come closer.......closer.....

If you have a string in the port that follows the port through the centerline and out through the seat with no fluttering. I mean, it looks as if it was a solid piece of wire molded to your port. And goes right out the SSR following even the bias. That has the BEST wet flow properties I seen. And what am I basing this all on ? The fact that recent publications back up cirtain findings and Darin Morgan's vids and pics come in really handy to learn from. There are many similarities between the PS head wet pattern and even a dated cast iron Pontiac head.

Ok now, Doug. I love this site and what disciplines people came from. All we need now is a retired NASA engineer. All I do is screw in light bulbs :p The hemi head chamber I feel will be a different animal. I can speculate a little and say that port bias, if any, and areas of highest speeds, will dictate the mixtures motion in the chamber. There was a link I couldn't open discussing the science behind air motions that (may) continue as the space it occupies becomes smaller. My thoughts were once the mixture spins from one enrgy source, that it would continue as the open space diminishes from the rising piston. With some friction losses of course. With that in mind, and bringing in some 2 stroke ideas, the piston design would dictate the chamber efficiency after the IN valve closes as it approaches TDC if there is some degree of swirl. This may show you ares of dark on the outside of the piston with the peak of the dome being the lightest. If not, then I would figure that depending on cam profile and dome shape/angle, the fuel can either become separated at the point where the quench pad meets the dome on the long and/or short side of the port creating two dark spots on each side of the dome with the sides and dome lightest. Or, the mixture may follow down the wall, across the quench pad and up the dome having the two mixture fronts collide right above the dome. This may create a burn profile with the top of the dome darkest and still dark from the IN and EX side of the piston, leaving the sides lighter. All speculation. If you have a head and a burn pattern, give me all head specs and I will try to speculate what it would be and then you can post a pic of the piston/chamber.

Back to Coanda effects. I did the test and put a sponge on it container mimicing a port wall. What you see with that is the same as wet flow in a port. The mixture rides the walls over the ssr. The tighter the SSR, and even if the port is taller to make up for area, the more liquid you will see. Friction then slows it down in addition to being inside the boundary layer. It rides the wall slowly until it makes it to the seat area. Then gets spit out.

I gotta stop this rambling on stuff!!!

But I feel like a celebrity now

I'm still trying to hash out the wetflow tools to get it to you guys. I'm getting there.

Chris.

[johno]

Hi there guys, im only new to the forum but after reading your posts about coanda effect, thought you may find some flow-tools, called "flow-balls' by 'jamison equipment', may come in handy when attempting to re-attach flow in specific areas. If you have'nt already heard/used them before, they're small metal balls of increasing diameter fixed onto long wire handles which are inserted into ports during flow-testing and 'induce' turbulence in specific areas, causing flow readings to change, indicating which sections of the port carry the most flow and which areas are dead! I have also experimented with (for all you disco freaks!!) a smoke machine, which i allowed to 'leach' into the intake ports in an attempt to get a visual "picture" of where the flow bias really is! VERY interesting i must say!! Got a bit bored playing with string only..... ???

[jfholm]

Sounds like you are going to be a good addition to the forum. We have discussed these things, but really appreciate you keeping them fresh with us. I am in the process of building a multiple string device with flourescent string to let them flow into the port. I will mount a series of UV LEDs in the flow bench pointing up into my clear head adaptor and also shine a black light into the port as it is flowing to see how they act. I am hoping to capture them with my video recorder.

John

[blaktopr]

I like the idea John. Just some info though. I am not sure how much more "light" you can get past the valve with led's opposed to straight black light in the cylinder. I couldn't see anything, but the string may show up better. Also if you were to try some dyes, once dry they do not show as well. A hylighter shows up under the blacklight, so if you want to disect the marker and use the strands from that?? I'm not sure. The light issue is why I want to try machining a valve out of clear acrylic and inserting the black light led's into it, through the stem. Might be something we can all try.