Two stroke engine simulation software.

[Ernie Phillips]

Anybody doing any computer analysis on our old piston port, air cooled engines?  Over the last year I've acquired a few 100-125 Sachs cylinders with very different porting execution (huge exhaust, hogged out transfers, boost ports, worm ports, mirror finish polishing on one and then ruff as a cob on others).  Without having to install and tune each configuration, I thought that inputting data into a computer simulation program and seeing how each change affected engine performance would give a way of determining which cylinder (or specific mod) has a chance of actually improving performance.  For example, do you get better results by improving port geometry (time-area, entry-exit, direction) or can you just leave the ports alone and build a better pipe to evacuate and recharge the cylinder?   Is it reasonable to expect that significant performance gains across the desired RPM range could be achieved with the proper port/pipe combinations beyond what is generally known (Cobb, Cranke, Monark GS, Popular Cycling, FMF, TX specials, Circle-F)?  I've looked at programs by TSR, MOTA, Bimotion.  (Disclaimer:  I know for a fact that the stock Sachs engines perform very well, have good reliability, and have won many races.  I'm just curious)

Ernie P.
Chattanooga, TN

[454MRW]

Not being specific to motorcycles, but having experience as a backyard mechanic port matching and smoothing ports in big block chevy engines, it is well known that highly polished ports do not work as well as smoothed non polished ports on the intake side, but polished ports work best on the exhaust side. The reason is because on polished ports the fuel in the air/fuel mixture tends to stick to the port walls and cause turbulence and draw the fuel along the port walls, but after combustion burning of the fuel, the polished surface flows best. I have read this in many publications from experts in the porting field regarding big block and small block chevy performance modifications. Mike

Michael R. Winter
I enjoy rebuilding and appreciating Pentons!
1974 250 HS Pentons-1980 KTM 175-400'S

[brian kirby]

Sounds like the Phillips Vintage Cycles skunk works has a winter project.

Brian

'73 Berkshire

[firstturn]

Ernie,
  I have seen many attempts at trying to resolve this age old delima.  In my book there are two separe issues to address.  

1.  Build a flow bench for a two stroke cylinder and try and flow it with and without the chamber.

2,  Build a load dyno to bolt up the engine, carb and pipe to do load testing with simple digial/memory graph hardware so that you don't have to blow an engine test.

  Study the graphs and put your best combination in the bike and then test using gearing to set the bike up for the best performance based on the type of racing you are doing.
  If I remember correctly we have an ole National Caliber Kart Racer(Nelson Lingle) that may have a engine dyno and a lot of knowledge.

Just my thoughts.

Ron Carbaugh

[john durrill]

Ernie,
I used the Bimotion software several years back. They had a demo
 program you could down load and run.I tried it out with the ports on my sons 75 175 KTM and his pipe measurements just to see how close it was. It was in the ball park BUT the pipe that came stock was different than what we got for measurements from the software.
I think there are too many variables for any software model to give us accurate data.
 I asked a friend who is a good tuner and builds good pipes for road racing  about the software and he was of the opinion that is just was not possible to use it except as a starting point . Then you would need to do a lot of testing and cut an try to find what would work best on your particular engine.
 Gordon Jennings Book will get you just as close and give you more insight on the interaction of  the cylinder ports and pipe design i think.
 Ron gives good advise if you want to get the best out of a design.
I bounce all kinds of Hair Brain ideas off him on a regular basis []
We are going through this right now with an Iron Barrel 125 engine. Trying to just get the best out of the design with out sacrificing reliability or moving the power up into a narrow band.
 I have asked for advice from several good engine builders but have found out unless you have already worked with that design ( specific cylinder in this case) enough to  " know " it you can not offer reliable advise on specific changes. Just some things to watch out for  [B)]
 I can send you a short cut to the Jennings book on line and have 2 other articles , one by EC Bert and I think the other by Harry Taylor that can help you.
 Port width and shape is a biggie. The ports shape , type of ring's you have to work with and pistons design , limits what you can do safely. Get that all maxed out for an engine and.............
 You get into raising  and in  the case of the intake lowering port openings . The transfers work together with the exhaust. I am told there is a narrow window ( Mathematical relationship)  for change  between the exhaust height and the transfer height. Get outside that window and you go backwards. You can go back and change the pipe and or the duration of the exhaust or transfer and jump forward sometimes. Sometimes you have to start all over with a new liner or cylinder.
 Ron if i am a off center with any of this please jump in.
Ernie i will gladly pass along anything i learn from my project. The goal we are shooting for is get the most out of that engine with out moving the power band up more than 500 RPM or making the engine unreliable ( a 2 day trials type engine ). Just make it work the best it can with the iron cylinder. Its a bit different than building one for CC or MX.
I know others on here know a lot more than i do about this type of thing. All i could do is pass along what i" think " is accurate [:I]
 John D.

[brian kirby]

I am by no means an engine wizard but what I know about them and what I know about computer simulations I tend to agree with John. There are no doubt simulations produced in house at companies like KTM, Honda, etc that can be extremely accurate, but those are surely built for a specific engine. I think anything available to the public at a price that an individual can purchase will not have the fidelity to be of much use beyond getting you to a starting point. I would think for a simulation to be accurate enough to do meaningful testing all on computer it would have to be written for a specific engine and not a generic program.

I know this is not the answer Ernie wants, but I think the most practical way to test would be with two bikes on the track at the same time, one your test mule and one your baseline bike. It will be time consuming, but you get direct results in the environment the machine will have to perform in, so it is by nature definitive.

Brian

'73 Berkshire

[firstturn]

Let's get a base line on this discussion....there are probably no wrong answers and some ways of developement meets the needs of certain indiviuals.  Like Brian says about two bikes is OK if you have money and a lot of time...mostly a lot of time.  What John is saying has worked for a lot of people who have the background and do take the time to read articles and books written by notable people.
  The items I discussed do not break the bank and can be used to test while at home/shop and then test at the track.  I think one thing that Ernie was saying was Make It Easy on Me where I don't have to put wing nuts on the engine bolts to change and test and change and test.
  I am happy to see this discussion and I should add that Ernie and his crew went from blown engines and busted shift keys to really winning a lot of GOLD this year.  Ernie has several things on his side.  He doesn't give up and he knows how to test and double check his work.  The other thing is Ernie grew up in a area that was tough racing and to be a Pro you were a excellent rider....I know I raced in that area it it had some great rider in LA(Lower Alabama).

Ron Carbaugh

[linglewn]

Ron,
Thanks for the compliment. I wish I had all the answers, but I don't.

Ernie,
We modified hundreds of racing engines back in the 6o's and dynoed each modification to determine the results of all changes, and then raced the ones that ran the best and trashed the rest. Important factors were port timing, port shape, crankcase volume, port transfer passage shape and volume, direction of flow of the transfer ports into the combustion chamber, combustion pressure, and of course ignition timing. We were looking for maximum horsepower, so usually ended up with high ported engines with little low-end torque.

We also did a lot of dyno testing with special instrumentation installed on test engines. We measured fuel flow, air flow into the carbs, crankcase pressure, combustion pressure vs crank angle, and (the most interesting) the gas flow direction in the transfer passages vs crank angle. The tests showed that a good running engine had enough combustion pressure at the time when the transfer ports opened to force combustion gas into the transfers and pressurize the crankcase. The colume of gas in the transfer passages then acted like a spring to force the cooler crankcase gasses into the combustion chamber. The direction of flow into the combustion chamber seemed to be important in making an engine run smooth at full power. If an engine runs rough, it doesn't matter how much power it makes because it won't last long.  All of this was done on reed valve engines with a fairly short stroke. Add piston port, rotary valve, long stroke variables and you may get different results. Of course the expansion chamber has to be designed to match the rpm range of the proting.

As far as a computer simulation for all the variables of a 2-stroke engine, I have not heard of one. I know that wonderful things can be done by computer simulation, but I'm not sure there has been enough interest in the older engines to justify the expense and effort to produce one that could give reliable information.


Nelson Lingle
73 Jackpiner
71 DKW 125

[Ernie Phillips]

Gentlemen,  Thank you all for your well thought out comments.  Hopefully others will chime with their tuning experiences.

Mike,  Sounds like I don't need mirror finish.  Thanks.

Ron, "Wing nuts on the engine bolts" – now there is an idea that fits right in with my LA up bringin':D.

Nelson, I didn't know you were such a serious racer and engine builder.  We need to get together some time. I'd like to hear more.

John,  I'd like to see Harry Taylor tuning info.  Thanks for your continuing interest in always wanting to make it "just a little bit better.".

Brian,  It's all your fault.  Until I rode the SoCal Berkie, I was satisfied.  Now I'm not.





Ernie P.
Chattanooga, TN

[Ernie Phillips]

Too many choices!


Ernie P.
Chattanooga, TN

[brian kirby]

I posted these pictures of my SoCal Berkie ported cylinder a while back but since this topic came up I will putn them in here too so we can compare.







Only the bottom 2/3 of the transfers look like they were ground, the top 1/3 and the transfer windows look stock. There is minor polishing of the exhaust port towards the exit but the port height looks stock because there is some casting texture left right at the liner intersection although I admit I did not measure them. I have a 125 cylinder that was on the bike when I got it that is ported exactly like this 100 cylinder.

Brian

'73 Berkshire

[brian kirby]

It looks like the transfers at the base gasket area have been ground much "wider" on my cylinder than any of the ones Ernie posted.

Brian

'73 Berkshire

[firstturn]

Ernie,
  Nelson has a lot of history with two strokes.  He and his buddy, when he lived in Texas, were known as the guys that came up with BIG NEW IDEAS for karting engines and clutches.  Some rumors of NASA connections....and the rumors were true as I was fortunate enough to have some early teflon coated engine parts done a NASA.  Nelson may want to elaborate as some people think of karting as a kids game, but this was serious business and there were a lot more injuries than people were aware of from those early days.

Ron Carbaugh

[linglewn]

Ron, you really have a great memory and help to bring back some fond memories of my own. My buddy, Jim Akkerman, and I worked together as NASA engineers. As such, we really enjoyed experimenting and developing new ideas. Jim was the mastermind behind most of the development work and the owner of Akkerman Engineering. We developed and produced the full circle crank for McCulloch and West Bend kart engines. We were also the first to develop a functional "slippy" clutch for use with the Margay gearbox. We raced most of the enduro classes sanctioned by the International Kart Federation (IDF), and were fortunate enough to take home more than our share of awards-including several National Championships.

The accidents you mentioned are not-so-fond memories. In 1970 at Dallas International Raceway, Jim and I both won our first one-hour race and then got directly into a kart in a different class for another one-hour event. We both crashed on the same turn about halfway through the race. The results were a broken back for me and a broken leg for Jim. It was a long ride back to Houston.

That's probably more than anyone really wants to know, but thanks for jogging my memory.



Nelson Lingle
73 Jackpiner
71 DKW 125

[Ernie Phillips]

Nelson, Fascinating story!  Thanks.  I hope is wasn't two best friends and racing buddies that knocked each other out.  If so, it really was a  l o n g  drive back to Houston .  -EP