Porting basics

[OuiBJammin]

Bought TZ cylinders that were junked cheap,read up the best I could pre-internet.

Went in the Navy,spent years in engineering schools and built turbines,gas motors and diesels for money after that.

Built two strokes for me.

I learned where useful information is to be found and that I now know half as much as most of the people posting on the internet.

Still don't know what I was doing wrong,I learned what doesn't work.

[tfaith08]

Porting for flow into the cylinders is at it's peak.  That's what I imagine 90% of all successful builders go for (and for good reason).

The areas that I feel are untouched or not fully exhausted is getting AFM to fill the underside of the piston during upstroke, scavenging that volume during transfer opening, filling the area below the transfers (which I've experimented with to very slight avail), and reducing reversion into the intake during downstroke. 

 

Passion appeared to have touched on some of those with his extra port leading into the secondary transfers, but his numbers showed nothing promising.

The guy that I mentioned earlier about the hotsaws told me that he'd experimented with styrofoam and cork stuffed under the piston and had gained a huge amount of bottom end and midrange power, but that it somewhat capped his high rpm power which is what hotsaws are all about.  He also said that he didn't experiment with the porting very much to try and compensate, so there's that as well.  Also, these engines see about 3 minutes of service life before being rebuilt, so there's that as well.

The things that I've tried with boyesen ports is making them bellmouthed on the intake side and making them reduce in taper and open with 90° edges into the transfers.  Basic fluid mechanics says that will allow flow in easier than letting it out.  Combining that with different angles did show differences in bottom end and midrange power.  Essentially, I found that the downward angle affected how the ports were fed at different rpm ranges.  A steeper angle helped lower end power, while a flatter angle helped midrange and maybe high rpm power. 

 

My theory for this is that the AFM will flow from where it is to the transfers at a somewhat static speed as the piston leaves TDC and approaches the transfer opening.  If that angle is steeper, the AFM from those ports will flow down more and end up directly below both ports at a later point, contributing to low end power.  A flatter angle will put that AFM directly below both ports sooner and will contribute to power in the higher rpm ranges.  This effect is also somewhat dependant on whether the lower section fo the sleeve has been removed.  The size of the boyesens is also extremely important here.  A larger boyesen will flow more and feed the transfers a little better, but it will also allow reversion back through themselves.  Also, a larger boyesen will not be affected by the angle as much. 

[trickedcarbine]

Passion appeared to have touched on some of those with his extra port leading into the secondary transfers, but his numbers showed nothing promising

Really? You sure about that? You keep speaking as though you got this figured out yet recently posted that you really haven't done much 2 stroke work let alone banshees. So come again those ports Jim did were about filling of the secondary transfers? Do you have one of these cylinders in your hand? And where are the numbers you're judging your assumption from?

Really look at those ports. Look how high they are and re think what you think they are doing.

 

My 12 old 12 port made almost 50 foot pounds before 9k and still made 84 hp at 9,600 with a decent over rev on shearers. Took some sorting out, but if that ain't promising on a motor with 192 exhaust I'd like to see what you're working on. Jim was a little quirky, but toward the end of his run he finally got those 12 ports to run the way he claimed. At least mine did/does.

 

Not bashing, just trying to get a guy that claims to be a heavy thinker to think a little outside he box. My apologies if that sounded brash.

 

Carry on.

[tfaith08]

I don't mean to sound as though I have it figured out by any means.  His 12 port setup was great in theory, but I never saw where they did that great although that was years ago.  I'd always heard reviews from both sides of the fence, but I base my thoughts on what I've seen in pictures.  I didn't research him heavily and heard nothing but bad about him.  That's the basis of my statement.  To hear that it did work out shows a solution to of some problems I've had for a long time.

My thoughts may also be biased becuase I've always been taught that after the transfers have been covered up, the volume within them will be drawn to a vacuum by the vacuum created by the piston on upstroke since the crankcase isn't at equilibrium.  Yes, the intake feeds that volume, but there is a vacuum within the transfer tunnels during upstroke which is where boyesens come in.  If Jim's ports didn't serve the same idea, what did they serve?

 

I didn't claim to be a heavy thinker, I said I ovethink things.  I've just had more experience with other cylinders and other types of engines than I have with a Banshee.  Lots of things cross over, some don't.  One of the things that does cross over is keeping the transfers fed at all times so that they are uncovered with as much pressure as possible.  Of all the thinking I've done on Jim's 12 port setup, they seem to target that idea more than anything.  If that's incorrect, I want to know the correct answer.  By definition, a forum is a place where ideas or views are exchanged.  I'm going to state my thoughts.  If it's misguided, state it and explain it.  If it's dead wrong, explain it.

 

I ported and worked on whatever I could get my hands on because I was out of work and trying to support my wife.  I didn't do it because I was trying to topple Redline and K&T.

[trickedcarbine]

My thoughts may also be biased becuase I've always been taught that after the transfers have been covered up, the volume within them will be drawn to a vacuum by the vacuum created by the piston on upstroke since the crankcase isn't at equilibrium. Yes, the intake feeds that volume, but there is a vacuum within the transfer tunnels during upstroke which is where boyesens come in. If Jim's ports didn't serve the same idea, what did they serve?

I didn't claim to be a heavy thinker, I said I ovethink things. I've just had more experience with other cylinders and other types of engines than I have with a Banshee. Lots of things cross over, some don't. One of the things that does cross over is keeping the transfers fed at all times so that they are uncovered with as much pressure as possible. Of all the thinking I've done on Jim's 12 port setup, they seem to target that idea more than anything. If that's incorrect, I want to know the correct answer.

50% more intake area. Equivalent of a third intake window. Piston goes up and pulls AFM straight through in to the cylinder right past the transfers as though it's not there creating additional crank case volume as though they are intake windows. All the while the secondary transfer does scavenge off the intake side Boyheson ports like normal, as well as when the piston is dropping it forces some volume back in to the transfer from the cylinder. (Edit:I see you touched on this part.^ )Internally his Boyhesons are pointed more at the ports he has machined in to the sleeve then upward in to the transfer. It took a lot of fine tuning to get it to not surge and fade and come back on. When he started doing them there were fellas not to stoked as they can be a cunt to get running right. There is definitely some port interference going on as the piston moves causing those extra intake windows to close and it starts to feed the transfers hard. It showed up as a goofy dip in the curve on dynos regardless of pipe/carb combos. Mine just took some compression and a fat pilot to needle transition.

 

Made that easier to read.

[registered user]

by vacuum you must meen the crankcase pressure has dropped below atmospheric. regarding engines i believe the definition of vaccum is simply any space or void below atmosheric pressure.

 

the exact point when the crankcase reaches negative pressure is hard to say and cant be generalized with all engines. you would have to look at the pipe and its suction ability, how well the exh port and transfers are designed. likely the case comp would be a factor as well. a well designed engine package would reach negative pressure in the case before a poorly designed engine. i know this because alot of poorly designed engines/pipes will allow exh gas to push back into the crankcase for more degrees of crank rotation after the transfers open, obviously this delays the case from reaching negative pressure

Edited October 6, 2015 by registered user

[ClaudeMachining]

No one is using software like, Fluent or Abaqus for flow analysys? I would like to have a 3d model of a cylinder to play with, even if my knowledge of pprting is at noob level.

 

Sorry for my bad english, i'm a french Cannuck!

[tfaith08]

by vacuum you must meen the crankcase pressure has dropped below atmospheric. regarding engines i believe the definition of vaccum is simply any space or void below atmosheric pressure.

 

the exact point when the crankcase reaches negative pressure is hard to say and cant be generalized with all engines. you would have to look at the pipe and its suction ability, how well the exh port and transfers are designed. likely the case comp would be a factor as well. a well designed engine package would reach negative pressure in the case before a poorly designed engine. i know this because alot of poorly designed engines/pipes will allow exh gas to push back into the crankcase for more degrees of crank rotation after the transfers open, obviously this delays the case from reaching negative pressure

 

Vacuum is a bad choice of wording, I should have used "area of lower pressure."  That would have been technically correct.  

 

50% more intake area. Equivalent of a third intake window. Piston goes up and pulls AFM straight through in to the cylinder right past the transfers as though it's not there creating additional crank case volume as though they are intake windows. All the while the secondary transfer does scavenge off the intake side Boyheson ports like normal, as well as when the piston is dropping it forces some volume back in to the transfer from the cylinder. (Edit:I see you touched on this part.^ )Internally his Boyhesons are pointed more at the ports he has machined in to the sleeve then upward in to the transfer. It took a lot of fine tuning to get it to not surge and fade and come back on. When he started doing them there were fellas not to stoked as they can be a cunt to get running right. There is definitely some port interference going on as the piston moves causing those extra intake windows to close and it starts to feed the transfers hard. It showed up as a goofy dip in the curve on dynos regardless of pipe/carb combos. Mine just took some compression and a fat pilot to needle transition.

 

Made that easier to read.

I can see a lot of that taking place.  I wonder the exact reason for the dip in the curve...

[trickedcarbine]

I'll see if I can find old charts. Almost all of them had it and some folks just couldn't get it to iron out.

[OuiBJammin]

Corkster,CFD designed for engines is pretty pricy and there's really not enough money in the ATV game to justify buying it.A lot of time to set up a run too as there are a lot of parameters necessary for accurate simulation,more than the runtime on a decent home machine likely.

 

There are free CFD programs not specifically designed to be used on engines that may be useful if you are up to the task http://www.openfoam.org/features/

 

 

There are free programs for some of the aspects of porting and stuff like EngMod2T http://vannik.co.za/EngMod2T.htmthat's $400 but they don't address radial or axial angles, interior radii etc just exit area and chordal widths.

 

The best thing about ENGMOD is that it is updated regularly as more info is shared back from users and as newer technology becomes available,most of the rest are legacy programs based almost entirely on Jennings,Blair and Bell stuff that is either years or decades out of date.

[ClaudeMachining]

I'm using abaqus already. But like you said, for an engine, that would be very complex and over my head. I'm pretty decent at fea but noob at cfd. Anyway, if i had an engine already done (3d wise) i woul like too play with it for fun.

 

Another option would be to give that as a student project (msc or phd) at a university neer you.

 

Sorry for my bad english, i'm a french Cannuck!

[tfaith08]

I've modeled some cylinders in the past and ran quick simulations.  Unless you know exactly what to look for, all you see is that something is different; you don't really know what is going on when something does change.

I got lost in the data of it.  Trying to measure all the ports and input that to the sim and making changes and trying to implement that into the actual port became more of a nightmare than it was worth.

[ClaudeMachining]

I've modeled some cylinders in the past and ran quick simulations. Unless you know exactly what to look for, all you see is that something is different; you don't really know what is going on when something does change.

 

I got lost in the data of it. Trying to measure all the ports and input that to the sim and making changes and trying to implement that into the actual port became more of a nightmare than it was worth.

Yeah, i can easily imagine the mess!

 

Sorry for my bad english, i'm a french Cannuck!

[sleeper06]

When I first started porting I bought I bought a few diffrent software options, I too got lost in the data and numbers . I stick to my style now and use what I know

[tfaith08]

I found it best to focus on macro changes and find a very simple, easy palce to ride and test. 

 

I ported my wife's blaster and had a little track out behind our house that was maybe 1/8 mile long with 2 turns.  I knew what I could do in which gear and would annotate changes in power delivery and predictability from that.  If it previously bogged in 3rd around a curve and now almost pulls, it was a good change for example. 

Smaller changes that will add up to a big change are very hard to notice.  That's where a dyno comes in.  As such, I focused on port shaping and an even surface over anything else once I had my numbers where I wanted them.  I believe the blaster ended up at 192/130/128/128.  Fwiw, I've never been able to notice the difference in 2 degrees on any port.  A dyno would show it, but nothing that I would feel.