Author Topic: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine  (Read 84279 times)

Offline lohring

  • Full Member
  • ****
  • Posts: 117
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #330 on: February 25, 2019, 02:35:17 PM »
I think harmonic is a misnomer.  The formula refers to wave lengths.  The first "harmonic" is the length needed to allow for the reflected pressure of a full wave to return to the intake port.

Lohring Miller

Offline dieselpilot

  • Full Member
  • ****
  • Posts: 280
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #331 on: February 25, 2019, 04:17:23 PM »
Do you know how these factors are derived, or a source? Google turned up something, but it was conflicting and no mention of how it was derived.

Offline lohring

  • Full Member
  • ****
  • Posts: 117
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #332 on: February 26, 2019, 06:32:14 PM »
I'm not sure how they were derived.  My source is a restricted two stroke forum.  I've tested the 3rd harmonic length in a simulation and it helped the power some.  Serious tuners have dyno tested intake tuning on a variety of engines.  The problem is it only helps a little over a restricted rpm range.

Lohring Miller

Online MJM460

  • Full Member
  • ****
  • Posts: 862
  • Melbourne, Australia
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #333 on: February 27, 2019, 02:45:13 AM »
I am normally another silent onlooker, but I am really enjoying this thread and in total admiration of the amazing work being done.  But perhaps I can contribute to the vibration question.

The motion of a violin string and similar systems is often referred to as simple harmonic motion.  The equations of motion are simple enough to be derived in junior physics classes, and all the solutions which have stationary nodes at the fixed points at the end of the string occur, and are simple multiples of the first natural frequency.  This the rich sound of a stringed instrument.  The vibration is at right angles to the length of the string.

In a long pipe with many supports and bends, like in a petroleum refinery, there are many such natural frequencies for each pipe, and they are related to the different lengths between the fixed support points, so are generally not simple multiples of each other.  So just different natural frequencies rather than simple harmonics.  The equations of motion are no longer simple, and the natural frequencies come from a form of finite element analysis, which is definitely not simple.  However computer programs are used to calculate the natural frequencies, or the frequencies at which any motion will tend to be amplified by resonance, and are used extensively where there is a regular forcing vibration, for example in reciprocating compressor piping.

Inside the pipe, either a gas pipe in a refinery or an engine exhaust pipe, there is longitudinal motion of the gas column, which is seen as pressure pulsation, and also experiences resonance at certain frequencies where a standing wave occurs.  These natural frequencies are again not simple multiples of each other but relate to different features of the pipe which give rise to a reflection, such as a bend or a change of diameter.  Some of the resulting standing waves result in a low pressure at the pipe end, helpful for an engine exhaust, while others result in a high pressure at the end and not so helpful.  The frequencies require a finite element analysis of the gas system, and are not simply derived.

The trick is to design the exhaust pipe so that a standing wave will occur with a low pressure at the engine exhaust port, at a frequency corresponding to the engine speed where the power boost is required.

In addition, the resonance can be very strong over a very narrow frequency range, or less strong but significant over a wider frequency range.  A sharp mountain peak compared with a rolling hill if you like.  For an engine, you want the rolling hill, so there is assistance over a wider rpm range, and you don’t want one of the patterns with a high pressure at the exhaust port.

There is no simple formula to predict the frequencies, it is again done by computer with a finite element technique.  Unfortunately not a programme most of us would have access to unless we are in a formula one team, or a University fluids mechanics department.  And it may be just as “easy” to find the frequencies by trial and error. 

The effectiveness of tuned pipes is well known.  I assume that the typical conical features, used in a tuned pipe for a single cylinder engine, by providing a gradually changing diameter, contribute to making the resonance less pronounced, but effective over a wider rpm range.

I hope that is a little help to understanding what is going on.

MJM460
 
The more I learn, the more I find that I still have to learn!

Offline dieselpilot

  • Full Member
  • ****
  • Posts: 280
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #334 on: February 27, 2019, 02:00:42 PM »
The theory is not in question, just how it was boiled down to a simple factor. Tuned pipe doesn't work in that exact manner, but close enough for general discussion. It appears the 108K factor is simply the resonant frequency of the length of the intake with some corrections and is not the fundamental, but the 2nd harmonic already. I don't understand why the 3, and 4th don't follow the expected pattern as Admiral_dk pointed out already. Google turned up several series of these factors and two completely unrelated methods of intake tuning.  Tuning to the fundamental gives the largest increase, so the 3rd or 4th give smaller improvements. Chances of randomly picking a short length (for 18kRPM) and being close to the 3rd,4th harmonic are pretty good. So, unless the initial length tuned it with negative results, you may not see much improvement. Not knowing what correction factors are included, it would be easier to dyno with a telescopic intake....

Offline lohring

  • Full Member
  • ****
  • Posts: 117
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #335 on: February 27, 2019, 03:56:41 PM »
The simulation program I use, EngMod2T, divides the engine and its pipes into small volumes and calculates the changes between these volumes from basic principles.  It doesn't simulate combustion or actual scavenging flow but is accurate in modeling pressure, flow velocity, and mixture purity through an engine.  Combustion and scavenging are determined by empirical models backed by lots of real world testing.  This engine model was developed first by Gordon Blair at the Queens University of Belfast and has been refined ever since. 

Most pipe design formulas found on line are empirical.  They are a good starting point.  The simulation gives a more realistic version of what really happens.  I and others have confirmed this with dyno testing.  The intake tuning formulas are also empirical.  However, the calculated length gives more power in the simulation.  We also did some tests on two available isolation blocks.  The stock length was best until the power curve was shifted into higher rpms.  A shorter block gave a little better power there.

Lohring Miller

Offline strictlybusiness1

  • Full Member
  • ****
  • Posts: 228
  • ME., Tool Maker
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #336 on: February 27, 2019, 08:24:05 PM »
"Not knowing what correction factors are included, it would be easier to dyno with a telescopic intake...." do you have any practical ideas on how to make a telescoping intake track, considering what the track presently looks like?

JA

Offline dieselpilot

  • Full Member
  • ****
  • Posts: 280
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #337 on: February 28, 2019, 01:48:19 PM »
The types I've seen were adjustable length velocity stacks. They would be of limited range when the stack is short. Regarding intake tuning, inertial effects are independent of acoustic and probably more important at high speeds. I've not looked into how to tune for it.

Offline strictlybusiness1

  • Full Member
  • ****
  • Posts: 228
  • ME., Tool Maker
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #338 on: February 28, 2019, 08:57:37 PM »
I have varied the total length of the velocity stack from .600" to .900" on various bore sizes. All were shaped (bell mouth) to give the maximum flow. None showed any appreciable difference in performance on my dyno. I wonder if the reason is because of the small amount of difference or the dynos inability to read the difference.

Jim Allen

Offline lohring

  • Full Member
  • ****
  • Posts: 117
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #339 on: March 01, 2019, 02:01:25 PM »
The only dyno test we did on intake tuning involved two different isolator blocks on a modified Zenoah.  We always found that velocity stacks reduced power.  Below is the test.  Note that the M&D carb on the standard isolator block was as good as my carb on the short block.  Also note that it's hard to separate any differences from the scatter.  We mostly ran the stock isolator block as a result.

Lohring Miller

Offline dieselpilot

  • Full Member
  • ****
  • Posts: 280
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #340 on: March 01, 2019, 02:44:27 PM »
Any change in diameter in the intake causes a reflection. If the carb has the typical venturi, it comes into play. The bell shaped stack also influences the resonant frequency of the system. The smooth bore carb might make it easier to find a sweet spot, if one exists, in the range of manifold length you're willing to run.

Offline Admiral_dk

  • Full Member
  • ****
  • Posts: 1365
  • Søften - Denmark
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #341 on: March 01, 2019, 07:33:36 PM »
I wonder (without knowing) if the Helmholtz resonance frequency of the whole system has more influence then than the "outside" resonance.
The reason for my thinking are the fact that some Two-Stroke engines has shown that they will continue running even when they inlet valveing stops working at high rpm's, because the Helmholtz resonance is very strong at that frequency and if this is the case, it could be much stronger than the "outside" system .... :thinking:

Offline lohring

  • Full Member
  • ****
  • Posts: 117
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #342 on: March 02, 2019, 03:38:12 PM »
That's called a 24/7 intake system.  One proposal uses a reed valve for low rpm that swings out of the way at higher rpm.  One model boater started and ran a standard race engine with the intake valve removed just to prove it could be done.  See
context=C3bacfcdADOEgsToPDskLxCTbBGuDPWU-hTqHEvnrC  Motorcycle racers have finished races with a disk valve that failed open.  Below is an illustration of the principle.

Lohring Miller
« Last Edit: March 02, 2019, 03:47:29 PM by lohring »

Offline strictlybusiness1

  • Full Member
  • ****
  • Posts: 228
  • ME., Tool Maker
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #343 on: March 02, 2019, 05:06:12 PM »
This was originally discovered by go-kart racers who ran reed valve induction systems. If the reeds failed while the engine was running, it continued to run. I'm unaware if the engine could be easily started & throttled properly. I'll look into what was said on the TWO STROKE TECH site about this.

JA

Offline DRT

  • Full Member
  • ****
  • Posts: 21
Re: .90 cu in, 30,000 RPM, 7.2 HP custom built nitro engine
« Reply #344 on: March 03, 2019, 10:10:23 AM »
I have varied the total length of the velocity stack from .600" to .900" on various bore sizes. All were shaped (bell mouth) to give the maximum flow. None showed any appreciable difference in performance on my dyno. I wonder if the reason is because of the small amount of difference or the dynos inability to read the difference.

Jim Allen

It doesn't matter really James. Like your test stand shows. If you can turn the same load faster with the same reliability or better as the motor before it you have a more powerful motor and that would certainly show on the dyno. The stacks are beautiful to really make more power or better tunability with them, based on the manifold bore diameter and cyclic input velocities involve, the length of the stack, taper, and bell mouth that also needs to wrap around towards the back of the muzzle shapes have to be optimized by mathematical model or some other means of flow analysis.

@Lohring,
All you have to do is take the scatter plots and connect lines of best fit and then you can see better your data variations and comparisons. Excel should have this ability. Also are you checking seeing which regression most accurately fits you data points? You should also be able to toggle between the data points and lines of best fit. Power, Logarithmic, Cubic, etc all have a  linear regression, so there will be one that likely fits your data best.

BTW I have some electric motors and ideas for your boats I know you'd love.

I miss you guys and think about you often. I'm also finally at a lathe and mill where dreams can happen. Big things going on now and more to come in the near future. Happy 2019.

Thanks,
Hubert