exhaust

Evening Trevor, having a bit of a nightmare with the routing of the pipe now ive tried to adjust it a bit..  may have to have another re think  ... anyway, quick question about deflected waves back into the header. Do these waves attach themselves to the outer edges of the pipe or do they travel back through the middle thought occured to me that some of  returning wave might hit the divergent cone and deflect back again towards the rear cone and not doing fully what is required.. just bouncing back and forth. 

Dan, sent pm with e-mail adress..thanks

shot you an email nigel, i forgot to say that in the file ,of the two groups of cones drawn , the left group is for higher torque and the right is high power!! wow , lol, careful no wheel spinning .

Hello Nigel,
If the pipe, and in particular the diffuser, is designed correctly then when it works in its optimum cycle then there is the least amount of time available. That point is flat out at peak power rpm and the revs stabilised at their maximum. That reality is not often met on the types of circuits Bantams race at, and flash peak readings won`t do it! It also pre-supposes that as a rider you are not shutting the throttle, braking, changing gear and other sundry off peak activities. For to do so changes the dynamics of the pipe-engine relationship and precise dimensioning goes out the window. The first parameters to alter are power and temperature. When these two change their relationship with the static pipe dimensions it is never for the better! The quantity of exhaust gas generated is proportional to the power produced, so immediately the pipe is not conforming to the engines changing needs. The same applies to temperature, power may start to rise but pipe temperature lags behind and it takes a while for the pipe to begin to resonate as your calculations say that it should? Formulas gain no respect from reality!
If you were able to see inside the pipe, chaos would reign. Pressure waves would be bouncing around travelling through each other and at differing speeds, duration and energy levels. Some will come from the rear cone and some from the piston face when the exhaust port is partially or completely closed. Out going waves generated at peak power may propagate at sonic level but their return will be sub-sonic. One reason why mean levels are used in calculations. There is too much turbulence inside for waves to attach to the walls, boundary layers may form, be stripped away and re-form. One feature that is not often reported on is that of returning waves being themselves reflected off the diffuser wall as it tapers down to the header! When you consider that our pipes are little more than hollow, profiled tubes it seems astonishing what goes on in there. Whilst all of this is occurring the slug of hot exhaust gas rushing from the cylinder progresses on its way to the atmosphere expanding both linearly and radially. It is the relative diameters and cone lengths that control that expansion rate and serve to maximise its expulsion from the end of the silencer. So in essence two things are required, getting rid of unwanted spent gas and clearing the pipe in each cycle. The second is to harness reflecting wave energy to augment basic engine power.

Adding a little more nourishment to the food for thought, I`ve done some enquiring and have been offered alternative numbers that may further question the speed of sound issue. The quote was 800*c at the cylinder and 150 at the silencer exit. These give an average of 475*c and then a speed of sound of 431m/s.
Oh, and just to confuse still more , add a few hundred more revs and you can make yet another nice new pipe!

Trevor

That sounds very positive, Trevor!

Icarus One -- of the CB ign and 8,400 revs with wind behind -- went better in the cold rain.

That´s a subjective measurement -- of course. Fact is it would out-accelerate a good Senior´s rider´s bike in cool rain, when on a hot-sunny day I´d be slipping the clutch longer and only staying with him. It was rather depressing to hear some of the Seniors talking about their 10,000 to 12,000 rpm (some water cooled) motors and one or two speaking of higher revs than that, yet it was refreshing to know that Icarus One actually beat some of these higher-revving motors when in Inter events -- I guess that could be attributed to a good torque curve and using clutch-slip as a torque converter.

In that same vein -- and referring to your remark on turbochargers -- when I was with a Zambian (African) transport company we had 250 horsepower Mack trucks, with turbochargers, hill-climing better than 400 hp naturally-aspirated Volov trucks. It was obvious that the Mack´s 27% torque back-up versus the Volvos´ 12% ...made all the difference...
Of course, gear-ratio matching, along with numbrer of gears, came into it...

Sort of reminds of Cadwell and BRC using the club circuit where the hill followed the hairpin directly ....
I can still smell the burning clutches....

Cheers!

JayBee.

Thanks for the info Trevor, thought perhaps the waves were  a little more disaplined and returning waves could be harnessed and returned  unmolested via an internal tube back into the header...  .. perhaps not.

Dan got the e-mail thanks.

My header pipe into the start of the next section. When i attached it to the rest of the exhaust i had problems with getting the fairing around it and could see a better way, so ive got another plan !!

Looking good Nigel,
Very nice multi section joint to form the lower curve , you're getting good at this. It must be a lot of work but I suspect there is also a lot of personal satisfaction gained from what you have achieved so far. An object lesson for us all. You may worry about the pipe interfering with the fairing, but there will be more performance gained with the correct pipe than is lost from an incorrect faring. Could be simpler to mod that and re-glass it?
Bit worried about what else I see. Is that an empty donut box and a beer glass?

Cheers! Trevor

My thanks to Mike Powell for the info on Cadwell... I had thought the practice day was a sort unofficial Bantam race day....?? Like what Mallory used to be....????

Sorry to get that bit of slopp in the way of the technical stuff. As usual -- my brain is all out of sync..

At the time of my making many exhaust pipes it probably wasn´t much different. But back to that point of engine running with a cold pipe: Icarus-One always sounded better when started in my home´s snobb residential area at around six in the morning. Its crackle was staccato --- like the continual popping of a champagne cork multiplied a thousand times in volume and time. That is, compared with how it sounded when warmed up at a hot, clammy pits, Icarus-1 always had a harsh crackling sound to its tail-end outlet in the cold Thames-side morning air.

As a point of interest here -- I always felt that this sound suggested a wastage of energy rarther than smart tuning?

Our second house at Barking had a footpath at the end of our lawn just right for starting Icarus-1 and then getting him quickly hidden.

Of course, you clever ones out there will tell me I was hearing reflections from the surrounding houses whereas at that wretched Snetterton circuit -- for example -- the listener is in an almost open-field.

All very subjective -- of course -- and not much use in pipe building.

Speaking of well maintained work-kitchens, I unfortunately killed the kids bugderigar, Joey, by doing a fibreglass tank repair in our Barking kitchen. I was making a new pipe. in the garage, at the time and having cleaned the
the fiberglass crack with some sort of solvents had set all heating appliances on full -- in the kitchen -- to complete the drying out process and then -- in between bouts of pipe welding in the garage -- popped back into the kitchen to apply layers of filberglass...

I´d forgotten that Joey´s cage had been put in the kitchen out of Tibby´s way when the two kids went out with their Mum. Joey will be safe, from Tibby, there ...

Imagine my horror when floating on the joy of having made a fair job of repairing the tank and good headway on the pipe I saw Joey´s feet sticking up from the bottom of his cage which was hanging on a neon-light fitting which had drooping plastic covers....

Remembereing that brandy once brought a canary back to life after it had had a heart attack I found only Glen Fiddick in the cabinet and tried giving Joey sniffs followed by a drop at a time on his upturned beak and then standing him upright on his stiff, little legs. All to no avail -- he kept falling over! Having a nightmare time to face I took a swig -- or two -- of GF and when the three came in I told of Joey having a heart attack which Glen Fiddick failed to cure.....

But back to the technical bit -- heat, humidity and the real training of reflecting pressure waves....


.... and whatabout more torque versus horsepower....?????

Cheers!

Fundamentally all that is required from the exhaust pipe is low pressure around bdc and a high pressure from transfer closing up to exhaust closing. Throw in a suitably sized hole for the gas to escape out of the rear and that's about it. As with all things Bantam, its not that simple!
If the exhaust is proportioned correctly then when it functions in its phase of maximum efficiency there is the least amount of time available for each cycle. That is at full noise at sustained peak revs, anything either side of this means losses. The better the pipe works then the sooner and deeper the case will go sub-zero. But you have to be careful with wide diffuser angles or you may create too large a depression in the cylinder around bdc. A big, fully maxed out exhaust port in a 64x58 engine has the potential to enable a lot of fresh charge to short circuit and disappear down the pipe. Similarly, at lower revs and torque the returning exhaust pulse arrives too early when the transfer ports are still open. Cylinder gas is pushed into the crankcase, pre-heating what is there and badly affecting the subsequent combustion phase, and torque goes walk about! The higher the transfer ports the greater will be the effect. Poor torque production can also be very prevalent in barrels that exhibit poor directional control over the in going scavenge streams. A double whammy like this really drags the power band down to unrealistic levels that a meagre three gears will not adequately cope with. Consequently, lap times will suffer with sluggish acceleration off the corners and reduced top speed.
Pressure in the cylinder just at transfer port opening is largely dependent on the extractive action of the exhaust system. However there are also heat energy losses to factor in that are mutually inter-reactive with pressure. Crankcase pressure will be lower than that in the cylinder at transfer closing. This is because the kinetic energy of the transfer flow converts to potential energy. Kinetic Energy = (mass x velocity^2)/2
so the significant value here is velocity not mass. Pressure differential does not equate to flow but to acceleration. At transfer port opening flow is zero and inertia very high. Mixture is then accelerated and reaches a maximum at bdc and will continue as long as cylinder pressure is lower than flow pressure. Any system involving gas flow will carry static pressure. Once inside the cylinder flow slows and this pressure can be used to manipulate scavenge flow streams forming a column at the cylinder rear wall.
Total Pressure = Velocity Pressure + Static Pressure, so there is a lot to play for here. At the equalling of the transfer duct/cylinder pressure, flow will have reached its maximum velocity. The longer transfer ducts in the long stroke/rod engines will have more trouble in getting that column moving and we need to get that mass into the cylinder in a short, given time frame. Rising pressure in the cylinder eventually overcomes transfer flow until it stops or the transfer ports close.
Complicating this scenario is that what we really need is mass transfer, which is.. velocity x mixture density x duct area, and again this is maximal at bdc. If all of that wasn`t complication enough, velocity is inversely proportional to duct area! So a short column and smaller duct area, for a given pressure ratio, will get the mass flow up to an optimum in the quickest time and therefore flow more gas. It must always be borne in mind that the transfer cycle is the briefest of all the physical timings. To make things worse is that the flow is interrupted by the piston for all but the few degrees around bdc. So when the ports are fully open they just have to flow damned well, and in all of the correct directions!
It is at this point that control over flowing mixture must be considered. Minimal flow and turbulence losses in the transfer ducts and port windows are the result of correct profiling and exit angles. Taking care over these will always make the power. This control comes from the geometry and profile of the ducts from case entry to the port window`s axial and radial efflux angles. It should be pointed out that with Bantam stroke lengths varying from 50mm to 58mm adjustments will have to be made to these angles to suit these extremes. Varying piston crown radii will demand similarly suitable adjustments. It may be worth considering the fact that transfer ducts are open permanently, the ports are not!

Exhaust gas energy is significantly affected by compression ratio and ignition timing and as such the pipe can only deal with what is provided. This amount of energy obviously varies along with temperature over the range of rpm that the engine is operated at. It is reasonable to conclude that pipes resonate less well outside of the 180*ideal and don`t respond well outside of the effective power band. There being little energy to work with. Out of the best power range, but not necessarily the very base of the band, a pumping action helps. When the engine is fully in the best power band at optimum revs it is no longer a pump but a resonator! The pipe is doing more work than the case. Two stroke development history has shown that...... Smaller case volume = good pump/poor resonator, larger case volume = good resonator/poor pump. pumping action is only effective for the period of crank rotation from inlet closing to just after transfer opening, The pipe should then take over the action. One really big problem with a large case volume is that the pulse signal influencing the carb jets is so low that clean carburation becomes difficult and pick up to useable revs is sluggish. This then is made immeasurably worse by our paltry 3 gears, where being able to shift down a couple of cogs would get revs up and into the power band, for the same road speed.
Knowing that the average Bantam, large or small, produces around 22-25hp, how large a case volume does the engine need to breathe from?
At bdc the case pressure will be what has been designed into the engine geometry, with variation of stroke , rod length, crankshaft diameter and so on. As the piston rises case pressure falls and gas velocities fade. For reed valve engines this signals the first stirrings early inlet opening. With atmospheric pressure on one side of the reeds, lower pressure (around .75bar) on the other and the added influence of pipe action, the reeds begin to open. This will take place around 15* abdc. and the reeds will remain open for possibly another 250*. Piston port engines will have to wait much longer for their inlets to open, and much later for it to close.
It may seem counter intuitive, but with the reeds open before transfer closure there is a clear passage open all the way from the carb bell mouth to the silencer outlet!

A few musings to mull over?

Trevor

Blimey,  some info to digest there Trevor  (sainsbury cream and jam doughnuts 4 for £2 , how can you resist  )Finally got my pipe together, the snowy one. Funny how it looks alot better before welding than after, (looks like im going to have to  paint it )also you wonder how it took so long and why it seemed so difficult to get the right flowing shape.. or is it just me! Still at least my next one , Trevors pipe, should be easier. will post picture when ive sorted the muffler bit.
  Got my trailer sorted now so hopefully im ready for a test day 

Wow Trevor!  What a contribution! I agree with Nigel....

Must read thro´ it again....

The article is really good in the respect of highlighting all the little bits that depend on other little bits for the plot to work well. I still come back to welding rugosities, inside the pipe,  maybe disturbing the flow which was the reason I always cleaned up the inside of each weld before going to attaching  the next piece...

A lot of work....

The matching of all the little bits & their functions is, I guess, why it is called "Tuning" ....

I never did any -- my poor excuse being I was too busy with my job, family plus wife´s car and house maintenance -- and I wished I had had... There was barely time to go and collect the Icarus-1 engine from my mentor and engine-tuner before making time to go racing.  With Icarus-2, The Wobblyeman (Colin Aldridge) did the ports and whilst many Bantam Club members derided Wobbley´s tuning (and bike preparation) at least Icraus-2 would rev to 11,500 rpm -- yet not too regularly -- I was still using BSA pistons which had a habit of `stretching´ such that the Bump-Clearance went to nil,  and minus nil, before disintegration.      

Wonderful to see and hear that the the Bantam-Racers´ role-call is increasing. If I happened to win the Lotto (Lottery in Blighty) I´d spend some on Bantam Racing and would consider obtaining a (forged) racer´s licence....
I wonder if the scrutineers check on identity and age of riders??

Cheers!

v



I have  had to use the first part of the  header pipe from an original bantam ( crome bit) as i cant make anything, with the equiptment i have, to produce a machined type attachment  and ive made the rest of the exhaust fit to it via springs to make it more quicker to remove. (Picture doesnt  show spring set up..) im thinking i can reduce the header diameter by use of an insert for the smaller header diameter of Trevors pipe design. i have managed to obtain a cylinder head from an old rotax kart to fit, which  has more fin area and thicker fins than the standard head i have at present. Im trying to get the cc up to 23cc (as Mick stated in the snowy thread) and the stud holes need tickling out a bit to fit on the bantam barrel.. The end can is a shortend aprilia rs 125 and it seems to keep the noise down very well. I shortend and removed a fair bit of the internal wadding and pushed it over the stinger in an attenpt to keep the wieght down so as not to stress the end of the exhaust.

Congratulations Nigel ,
That is a nice pipe, no awkward kinked angles, lovely flowing lines, efficient looking silencer and its in the correct place , no heat being breathed into the engine, nearly as tasty as a box full of donuts! The alternative head sounds like a move in the right direction, I know we are talking about a different engine, but the 186 ran all day with 12.5;1 cr and never ever seized and would rev with ease to 10,000, the very low cr of the Snowy engine would suggest a heat dissipation problem. Looking at the photos it is easy to see the very extensive finning of the 186 head, taking the heat away in a fore-aft direction using maximum straight line air flow!

Well done, are you taking orders now?

Trevor

Yeah im open for business.... 4 cream jam doughnuts £3  plus p&p.. bottles of Peroni £2 plus p&p   yeah pipes not bad i suppose.. and i only cut one finger once  

Would the lower cr ratio also enable the engine to rev easier and able to run on unleaded fuel?  

Nigel,
I ran my road going D14/4 on a cr of 10.5:1, I had to junk the standard head and gasket then skim the barrel top to get there. I used a Todd head, squish of .8mm and pump gas, it was mildly tuned and was about 5-6mph up on standard, 8:1 on an alloy barrel with avgas doesn`t seem to equate. A low comp allows you to rev freely, but revving doesn`t necessarily equate to power, you may get more bangs, but, they`ll be smaller bangs! The area where you definitely will lose out with a low comp is in torque at the base of the useful power band. This will be felt in poor acceleration up towards the peak. The more the combustion gas is compressed the more energy can be extracted from it, and energy is power, but you also get more heat, its a fine balance. If you can use avgas it makes no sense not to, you won`t detto on it at 8:1? Unleaded is fine if you adjust some of the settings, run a lot of fuel through and use the low cr. Unleaded supermarket pump gas can be a little erratic in quality however!
Steve flits between avgas and unleaded on his RS Honda, he fits a lower comp head insert and ramps up the carb jet sizes and then adjusts them back on the day, and maybe goes to a colder plug, usually ign timing stays constant for both. The unleaded is usually high octane race gas, of the proscribed type for the day. As such is perhaps not allowed for Bantams, but the principle is there.

Ok, you win, you have the pipe, I`ll take the beer and donuts!

Trevor

If I knew how to attach a PDF file to this form.
I have a very informative report that consists of 16 pages from a very old magazine, that has been sent to me.
In the science of expansion chambers, its called '' Do you really want to know about expansion chambers'' by Gordon Jennings.
I can attach to e-mails, but to this forum is beyond me. If someone may be interested I can put in an e-mail to you or if someone can tell me how to attach to this for all to read.

Yes Peter, I too wish I knew how to master the black art of posting a pic Ed`s life would be a little more peaceful!

I found the article by Googling, gordon jennings on exhausts, and then clicked on 3cyl. com second one down..... I think?

Thanks for that Peter, interesting but dated.

Trevor

this web page is interesting , http://www.dragonfly75.com/motorbike/ECtheory.html

heres a link to one of the pdf files,  which Pete refers to, about Gordon Jennings and his exhaust theory's    

https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0CCUQFjAA&url=http%3A%2F%2F3cyl.com%2Fmraxl%2Fmanuals%2Fjennings%2Fchambers.pdf&ei=uldZU9LuHeqm0QWO4IDoDg&usg=AFQjCNGIWD2LgOJ5oJlFHzx956kPpdlPRQ

Pete had another pdf file about tuning, but i cant seem to get the link to open so ill have to omit that one. 

Dan, tried to get the dxf files to print, but i couldnt get it to give me what i expected, would only print what showed up on the screen. Got a nice full black page with outlines of cones in red  

Nice one Nigel -- nice looking pipe.

They are getting longer because of the silencers and I wonder about the Waggle-Factor. Awfully embarrassing when the pipe comes adrift -- and when you happen to be up front it is most aggravating...

All the best!
Cheers1