2 stroke thread (with occasional F1 relevance!)

All that has to do with the power train, gearbox, clutch, fuels and lubricants, etc. Generally the mechanical side of Formula One.
manolis
107
Joined: 18 Mar 2014, 10:00

Re: 2 stroke thread (with occasional F1 relevance!)

Post

Hello Brian Coat.

You write:
“Thanks for the discussion.
No need to reply.”

Thanks for the proposal, but I will reply.


You also write:
“One error is the way you analyse the valve motion assuming it is sinusoidal. This assumes the valve deceleration is the same as the acceleration.”

The pure sinusoidal motion is the softest possible motion for a reciprocating mass and it minimizes the required forces (and stress), the relative friction and the reciprocation of power.

It was chosen for simplicity and for easy of calculations.

Actually it was the worst motion for my calculations if I wanted to exaggerate about the forces and power required for the motion of the valves.

On the other hand, the pure sinusoidal motion is quite close to the actual motion of the valves in high revving engines.

Here are several valve lift profiles of Ducati motorcycles:

Image

And here is the same plot with pure sinusoidal opening and closing curves for comparison:

Image

Compare the curve formed by the red circles with the actual cyan curve of the Ducati G-Corsa exhaust valve lift profile.

Then compare the curve formed by the yellow circles with the actual red curve of the Ducati 748 intake valve lift profile.


Your also write:
“For a real cam profile, deceleration is typically under 1/4 of acceleration. This means the spring calculation will be way out. “

Here are the valve lift profiles of the original Honda B16A2 VTEC engine (with the “wild” camlobes engaged, it is a, more or less, racing engine) and of the same engine modified to VVA-roller:

Image

Look at the original (red) curve of the intake valve lift profile. The valve lift is symmetrical about the full open position, which means that the deceleration of the intake valve a little before the full open position (at about 105 degrees after TDC) and the acceleration of the same valve a little after the full open position are about equal. The same is true for the acceleration a little after the valve opening and the deceleration a little before the valve closing (say at 0 and 205 degrees respectively).

For more: http://www.pattakon.com/pattakonVtec.htm and http://www.pattakon.com/pattakonRoller.htm

Youtube video at https://www.youtube.com/watch?v=-zzW8YkReLU

So, are we talking for same thing?

In the high revving engines (see what valve lift profiles the Duactis and the Honda VTEC are using) the shortage of time and the need for the maximum possible valve lift leaves no time to keep the valve fully open for several crank degrees.



You also write:
“The biggest error is assuming that the valve train friction loss is a (guessed) percentage of the k.e. being stored in the spring during each valve opening cycle.”

The calculations are not based on the valve springs.
They are based on the required energy for the opening and closing of the valve.
Either in desmodromic systems (wherein there are not restoring valve springs) or in conventional valve trains.



You also write:
“The biggest error is assuming that the valve train friction loss is a (guessed) percentage of the k.e. being stored in the spring during each valve opening cycle.”

Think simply and you will see how reasonably this assumption is.

You have two parts and a quantity of energy is reciprocating between them.
The taking and giving of a quantity of energy requires forces to act from the one part to the other and vice versa.
The more the required energy and the less the time provided for this taking and giving back of energy, the heavier the resulting forces.
The heavier the forces the higher the friction loss.

Even in the crankshaft – connecting rod – piston mechanism, the friction is a percentage of the inertia loads involved when running without load.

So what is more reasonable than assuming that when a quantity of energy passes from one part of a kinematic mechanism to another, the friction loss is a percentage of the reciprocating energy.


I asked it again in my last post:
Is there a racing engine having 17.7mm valve lift and 120gr “valve mass”, capable for 14,000rpm?
Do you have any diagrams of the valve lift profiles of it?
Do you have a dyno showing the FMEP of such an engine?




By the way, here is another version of the PatRoVa short stroke model engine:

Image

There are two differences as compared to the previous animations.

It has “funnel” exhaust ports (the port area at the outer side of each disk is substantially bigger than at the inner side of the disk).

It exploits the exhaust gas inertia to accelerate a fresh air stream and cool internally by it the exhaust port of the rotary valve as well as the exhaust passageways in the cylinder head.


In the animation the engine is shown at the “overlap”.


The exhaust is near to finish.

The exhaust port in the PatRoVa rotary valve has just started to bridge the exhaust passageways in the cylinder head with a port cut on the head cover (the blue part) providing fresh / ambient air (not air-fuel, just air).

The inertia of the exiting exhaust gas into the exhaust gas piping (not shown) creates a vacuum and suctions fresh air (through the exhaust port of the rotary valve) from the port on the “cylinder head cover”, scavenging the hot gas and cooling internally the exhaust port of the rotary valve, the exhaust passageways in the cylinder head and the rest exhaust piping.

In the specific design, the exhaust port of the rotary valve continues to bridge the “fresh air port” on the head cover with the exhaust for some 120 crank degrees after the overlap.


Note:
In the animation they are not shown the two side covers of the exhaust passageways. Without them, the exhaust gas exits from the sides of the cylinder head. With them, the exhaust gas exits from the two orthogonal holes at the “front” (shown at right) side of the engine (these orthogonal holes are arranged oppositely to the inlet port).


Thoughts?

Objections?

Thanks
Manolis Pattakos

manolis
107
Joined: 18 Mar 2014, 10:00

Re: 2 stroke thread (with occasional F1 relevance!)

Post

Hello all.

This stereoscopic animation of the short stroke PatRoVa model engine:

Image

Click http://www.pattakon.com/PatRoVa/PatRoVa ... nt_STE.gif if the animation is not clear or not in full size.

(more on how to see stereoscopically at http://www.pattakon.com/pattakonStereoscopy.htm )

is supportive to the animation in the previous post.

Thanks
Manolis Pattakos
.

J.A.W.
109
Joined: 01 Sep 2014, 05:10
Location: Altair IV.

Re: 2 stroke thread (with occasional F1 relevance!)

Post

"New Engine Technology."

Is it 2T EGR?
Or what? https://www.youtube.com/watch?v=FwcCd9mAZeY
"Well, we knocked the bastard off!"

Ed Hilary on being 1st to top Mt Everest,
(& 1st to do a surface traverse across Antarctica,
in good Kiwi style - riding a Massey Ferguson farm
tractor - with a few extemporised mod's to hack the task).

User avatar
henry
324
Joined: 23 Feb 2004, 20:49
Location: England

Re: 2 stroke thread (with occasional F1 relevance!)

Post

manolis wrote:
By the way, here is another version of the PatRoVa short stroke model engine:

http://www.pattakon.com/PatRoVa/PatRoVa ... _ports.gif

There are two differences as compared to the previous animations.

It has “funnel” exhaust ports (the port area at the outer side of each disk is substantially bigger than at the inner side of the disk).

It exploits the exhaust gas inertia to accelerate a fresh air stream and cool internally by it the exhaust port of the rotary valve as well as the exhaust passageways in the cylinder head.


In the animation the engine is shown at the “overlap”.


The exhaust is near to finish.

The exhaust port in the PatRoVa rotary valve has just started to bridge the exhaust passageways in the cylinder head with a port cut on the head cover (the blue part) providing fresh / ambient air (not air-fuel, just air).

The inertia of the exiting exhaust gas into the exhaust gas piping (not shown) creates a vacuum and suctions fresh air (through the exhaust port of the rotary valve) from the port on the “cylinder head cover”, scavenging the hot gas and cooling internally the exhaust port of the rotary valve, the exhaust passageways in the cylinder head and the rest exhaust piping.

In the specific design, the exhaust port of the rotary valve continues to bridge the “fresh air port” on the head cover with the exhaust for some 120 crank degrees after the overlap.


Note:
In the animation they are not shown the two side covers of the exhaust passageways. Without them, the exhaust gas exits from the sides of the cylinder head. With them, the exhaust gas exits from the two orthogonal holes at the “front” (shown at right) side of the engine (these orthogonal holes are arranged oppositely to the inlet port).


Thoughts?

Objections?

Thanks
Manolis Pattakos
I can see how the exhaust inertia might entrain cooling air at high revs but at lower revs might the exhaust go back along the cooling air tract.

Following on from adding ports to the exhaust side of the valve, is it possible, or worthwhile, to arrange the inlet porting to stratify the charge either by different timing in the two disks, or perhaps porting in the periphery to allow more than one inlet tract in the valve?
Fortune favours the prepared; she has no favourites and takes no sides.
Truth is confirmed by inspection and delay; falsehood by haste and uncertainty : Tacitus

manolis
107
Joined: 18 Mar 2014, 10:00

Re: 2 stroke thread (with occasional F1 relevance!)

Post

Hello J.A.W.

It is not an EGR.

What it does is to fill (starting from the top end of it) the transfer pipe with clean air and the crankcase with mixture (the clean air “intake” is controlled by the recesses, at the height of the wrist pin on the piston (case of the engine in the video), the suction of mixture into the crankcase is controlled conventionally.

During the first part of the transfer, the clean air in the transfer port scavenges the cylinder.

Then the transfer and scavenging proceed normally (the mixture from the crankcase makes the rest scavenging).

This way the quantity of unburned fuel in the exhaust reduces.

The problem is that the mixture entering into the cylinder has plenty of time (till the end of the exhaust) to escape unburned to the exhaust.

Things would improve if they could close much earlier the exhaust (asymmetric transfer, PatAT / PatATi, more at http://www.pattakon.com/pattakonPatAT.htm )

Thanks
Manolis Pattakos

manolis
107
Joined: 18 Mar 2014, 10:00

Re: 2 stroke thread (with occasional F1 relevance!)

Post

Hello Henry.

You write:
“I can see how the exhaust inertia might entrain cooling air at high revs but at lower revs might the exhaust go back along the cooling air tract.”


Imagine an RC engine (say for model airplanes) having an aluminum PatRoVa rotary valve on the cylinder head.
In such a case the internal cooling is vital; but the engine never runs at low revs.

In the typical case (rotary valve made of, say, steel) the internal cooling is just an option.

In any case, with a reed valve on the “clean air port” of the cylinder head, the exhaust gas is prevented from “going back”.



You also write:
“Following on from adding ports to the exhaust side of the valve, is it possible, or worthwhile, to arrange the inlet porting to stratify the charge either by different timing in the two disks, or perhaps porting in the periphery to allow more than one inlet tract in the valve?”


The simplest way is to isolate, inside the cylinder head, the space around the one disk from the space around the other disk of the PatRoVa rotary valve, and use two throttle valves.
The ports on the one disk can provide substantially different timing and overlap than the other disk (the one disk will be the “wild”, the other disk will be the “mild”).
Depending on the conditions (revs, load) the engine can run with the one only throttle valve open, or with both throttle valves open.
I.e. depending on the conditions the “breathing” is controlled either by the wild disk, or by the mild disk or by both disks of the PatRoVa rotary valve.

Image

(stereoscopic animation; for instructions on how to “look” at it: http://www.pattakon.com/pattakonStereoscopy.htm )

Spot on the width of the “bridge” between the exhaust and the intake ports on the PatRoVa rotary valve in the inner side of the two disk.
For instance, the one disk can provide a lot of overlap, while the overlap in the other disk is zero. If you want operation with zero overlap, open the respective throttle valve, if you want a lot of overlap open the other throttle valve, if you want more power open both throttle valves.


It is a kind of built-in “variable timing”.
Considering the high flow capacity of the rotary valve, this “variable timing” is different than having different cam lobes for the two intake poppet valves of a conventional engine (see the two “low-rpm” valve lift profiles (the green curves at right) of the Honda VTEC in the plot in the first post of this page and think what if they where controlled by independent throttle valves . . . ).

By the way, this option is not applicable in the typical rotary valves (Cross, Aspin, Bishop-Cross).

Thanks
Manolis Pattakos

manolis
107
Joined: 18 Mar 2014, 10:00

Re: 2 stroke thread (with occasional F1 relevance!)

Post

Hello all.

Here is a V-2 model (RC) engine:

Image

The animation is stereoscopic (for instructions on how to look at it:
http://www.pattakon.com/pattakonStereoscopy.htm )


Bore 24.8mm,
stroke 13mm (i.e. same bore to stroke ratio with the Ducati Panigale
1299 which has 116mm bore and 60.8mm stroke),
total capacity 12.5cc,
rev limit at 50,000rpm (21.7m/sec mean piston speed).

With the secondary balance web (slim orange) at the free end of the
crankshaft, the “vibration free quality” of this engine is the same
with the vibration free quality of the big Ducati Panigale.

The one connecting rod is red, the other is blue. They share the same crankpin.

There is one only, common for both cylinders, timing belt.

There is one only crankpin (it is of bigger diameter than in the
single cylinder).

The crankshaft is rotatably mounted on the casing (green part) by two roller bearings. The roller bearing beside the main balance web
(orange, with the crankpin on it) is the strong one.

The casing (the green part with the cooling fins) is a single piece part.

Thoughts?

Objections?

Thanks
Manolis Pattakos

manolis
107
Joined: 18 Mar 2014, 10:00

Re: 2 stroke thread (with occasional F1 relevance!)

Post

Hello.

Here is the transparent version of the previous V-2 PatRoVa model engine animation

Image

It unhides several details of the mechanism (spot on the two roller bearings, for instance).

Thanks
Manolis Pattakos

tok-tokkie
36
Joined: 08 Jun 2009, 16:21
Location: Cape Town

Re: 2 stroke thread (with occasional F1 relevance!)

Post

The ghost view seems to show ball bearings rather than needle bearings. They will give axial constraint to the crankshaft whereas most needle rollers do not.

What is the position with making this engine or the heads for the Panigale? I would love to see them made. This little engine running up to 50 000 RPM but even more exciting is the prospect of a Panigale putting out much more power at the crank.

I was flabbergasted by your first principal analysis of the power absorbed by the desmo valves at full chat (and valve engine would require much the same). I have owned plastic extruders with 50 kW electric motors so have a very vivid idea of the size of a 50 Kw electric motor. That a toothed timing belt can deliver that also surprises me. Those high revs really make the torque manageable compared to the 2880 or 1440 RPM electric motors I had with v-belt drives.

manolis
107
Joined: 18 Mar 2014, 10:00

Re: 2 stroke thread (with occasional F1 relevance!)

Post

Hello Tok-tokkie

You write:
“The ghost view seems to show ball bearings rather than needle bearings. They will give axial constraint to the crankshaft whereas most needle rollers do not.”


To keep the crankshaft at its position along the axial direction the ball roller bearings are preferable.

Imagine a propeller secured on the free end of the crankshaft (the left end in the animation).
Somebody has to “take” the thrust force created by the propeller during operation (it is the force that powers a model airplane).
The big ball roller bearing (the one closer to the crankpin) does this work too. Its outer ring abuts on the casing, while on its inner ring it abuts the big balance web of the crankshaft.
The radial force on the big ball roller bearing (due to combustion and to inertia) is dozens of times heavier than the axial (or thrust) force from the propeller. This makes the ball roller bearing a good solution. In comparison, while a needle roller bearing can take heavier radial load, is not good in taking axial load.

The other ball roller bearing (the small one) keeps the control at the opposite axial direction. Its outer ring abuts on the casing while on its inner ring it abuts the crankshaft (say, by the small balance web at left). Without such “control” the crankshaft would slip to the “right” because nobody stops it from doing so.


If instead of the roller bearing, plain bearings were used (say as in the Ducati Panigale), then additional thrust bearings are required to keep axially the crankshaft.


You also write:
“What is the position with making this engine or the heads for the Panigale? I would love to see them made. This little engine running up to 50 000 RPM but even more exciting is the prospect of a Panigale putting out much more power at the crank”

We are looking for a Ducati Panigale with broken heads.
The idea is to modify it to PatRoVa and put the owner to test it on the road, side by side with the stock Panigales.

What such a modification takes?
The substitution of the cylinder heads and of the cylinder liners (in the PatRoVa it is preferable (and not difficult in manufacturing) to combine in a single piece the cylinder head (excluding the cylinder head cover) with the cylinder liner).


A four-in-line (it needs only one cylinder head) with crankpins at 0, 90, 90 and 0 degrees for motoGP against the Yamaha crossplane crankshaft is an interesting (but expensive) project, too.

In the meantime, a good quality model PatRoVa engine can show several of the advantages at way lower cost.
The CAD drawings for the single cylinder PatRoVa model engine and for the V-2 PatRoVa model engine (of the “ghost” animations) animations are available for anybody interested to make a prototype.



You also write:
“I was flabbergasted by your first principal analysis of the power absorbed by the desmo valves at full chat (and valve engine would require much the same). I have owned plastic extruders with 50 kW electric motors so have a very vivid idea of the size of a 50 Kw electric motor. That a toothed timing belt can deliver that also surprises me. Those high revs really make the torque manageable compared to the 2880 or 1440 RPM electric motors I had with v-belt drives.”

Imagine the case the Panigale had to use the belts / pulleys used in your plastic extruders.

As you write, it is the high revving that makes manageable the transfer of power.

The comparison of the gearbox of an Diesel truck making 200PS at 1,500 rpm with the gearbox of the Ducati Panigale 1299 (which makes some 200 PS at 10,500rpm) says it all.

Thanks
Manolis Pattakos

manolis
107
Joined: 18 Mar 2014, 10:00

Re: 2 stroke thread (with occasional F1 relevance!)

Post

Hello all.


Image


PatRoVa sealing versus Wankel sealing


Let's compare the sealing (and so the gas leakage) of a Wankel Rotary RC engine (say, the 49PI of the OS) with the single cylinder short stroke PatRoVa:

According OS, their 49PI Wankel rotary has:
4.97cc capacity (per chamber),
1.1hp/17,000rpm,
practical range: 2,500-18,000rpm,
(weight: 450gr).

According OS manuals / drawings, the height of the rotor (piston) along the axis of the power shaft is estimated at 15mm, while the total periphery of the triangular rotor is estimated at 120mm (40mm between each pair of apexes).

So, each chamber of the model Wankel rotary uses for its sealing two apex seals and two pairs of flat surfaces (the one on the side of the rotor, the other on the casing) each having 40mm length. So, besides the two apex seals, each chamber of the above Wankel model engine leaks from a periphery of 40+40=80mm. The leakage through these two long slits (80mm total length) is limited using a small clearance between the sides of the rotor and the flat surface of the “side covers” of the combustion chamber.


The PatRoVa short stroke model engine has:
6.28cc capacity,
13mm stroke,
and 24.8mm bore.
The periphery of each window of the PatRoVa model engine is 30mm, which means the total “sealing” periphery is 30+30=60mm for a 6.28cc capacity.


At the same revs, say 15,000rpm of the crankshaft and of the power shaft, the time interval for leakage is 50% longer in the case of the Wankel: while the reciprocating piston of the PatRoVa needs 180 crank degrees to go from the maximum volume in the combustion chamber to the minimum volume in the combustion chamber (actually from the BDC to the TDC), the power shaft of the Wankel needs to rotate for 270 degrees in order the volume in the combustion chamber to go from its maximum to its minimum.


The PatRoVa can feature a several times smaller clearance between the inner flat surfaces of the two disks and the lips of the chamber ports as compared to the required / necessary clearance between the flat sides of the rotor and the flat side covers of the casing.
Think why.
For the sake of the calculations let’s suppose the PatRoVa uses the same clearance with the Wankel rotary.

The absolute leakage in the Wankel model engine through the side slits is (80mm/60mm)*1.5=2 , i.e. it is double than in the PatRoVa model engine (the 1.5 is due to the 50% longer time the high pressure is maintained into the Wankel’s combustion chamber).

The relative leakage is even worse: 2*(6.28cc/4.97cc)=2.5. This means that if a percentage of 25% of the charge in a combustion chamber of the Wankel model engine leaks from the sides of the rotor, this percentage drops to 10% in the case of the PatRoVa.

(From another viewpoint: for the same percentage of leaked gas, the PatRoVa can run 2.5 times slower than the Wankel model. For instance, the PatRoVa running at 1,000rpm has the same leakage with the Wankel running at 2,500rpm (the lower practical rpm according OS)).



But there is more:

The PatRoVa has the same bore to stroke ratio with the Ducati Panigale 1299 and even freer breathing (the ratio of the total chamber port area to the piston area is bigger than in the Panigale).
The Panigale (1300cc, two cylinders in Vee 90 degrees) makes its peak power at 10,500rpm (21,3m/sec mean piston speed). Reasonably the PatRoVa model engine will make some 4.7hp at 50,000rpm (21.7m/sec mean piston speed).

At 50,000rpm the time for leakage is 50,000/17,000 = 2.94 times less than at 17,000rp (where the Wankel model engine makes its peak power).

And because 2.94*2.5=7.35, if at 17,000rpm the 15% of the charge in a combustion chamber of the Wankel model engine leaks from the side “slits”, this percentage will drop to only 2% in the case of the PatRoVa model engine.


And there is more:

While the width of the combustion chamber of the PatRoVa (i.e. the distance between the two disks of the rotary valve) is less than 9mm, the rotor height (along the rotation axis of the power shaft of the Wankel) is 15mm, i.e. the one width is 60% bigger than the other.

This means that the necessary clearances in the Wankel need to be some 60% bigger than the clearances in the cylinder head of the PatRoVa.
Actually they are way bigger because of the architecture of the Wankel (see how the two side plates are secured to each other) and of the big temperature differences along the parts / surfaces that participate in the side sealing of the Wankel.

To be noted: a double clearance allows a way more than double leakage.



According the previous analysis:

The sealing quality in the PatRoVa model engine is many times better than in the existing Wankel model engines.


Objections?

Thanks
Manolis Pattakos

tok-tokkie
36
Joined: 08 Jun 2009, 16:21
Location: Cape Town

Re: 2 stroke thread (with occasional F1 relevance!)

Post

How are you going to set the clearance between the valve rotor and the ports in the head? Particularly in the Panigale size engine. I thought of having a slight taper to the two sealing faces so you could shim the valve rotor up or down to adjust the clearance - until I realised that only works in line with the cylinder bore but makes no difference 90° to that line = across the bore. So I wondered if the the valve spool is to have three parts (left and right sealing flanges and the barrel piece between) so you can shim them to whatever clearance works best. I appreciate that the valve spool can float sideways so the clearance between the flanges and ports is self regulating. That construction also makes it much easier to grind and lap the sealing face.

manolis
107
Joined: 18 Mar 2014, 10:00

Re: 2 stroke thread (with occasional F1 relevance!)

Post

Hello Tok-Tokkie

You write:
“How are you going to set the clearance between the valve rotor and the ports in the head? Particularly in the Panigale size engine. I thought of having a slight taper to the two sealing faces so you could shim the valve rotor up or down to adjust the clearance - until I realised that only works in line with the cylinder bore but makes no difference 90° to that line = across the bore. So I wondered if the the valve spool is to have three parts (left and right sealing flanges and the barrel piece between) so you can shim them to whatever clearance works best. I appreciate that the valve spool can float sideways so the clearance between the flanges and ports is self regulating. That construction also makes it much easier to grind and lap the sealing face.”


There is a conical version of the PatRoVa rotary valve:

Image

Suppose the angle f is 180-1=179 degrees.


The following animation also helps:

Image

At the high pressure portion of the cycle the periphery of the chamber port turns red. In order to maximize the ratio of the chamber port area to the chamber port periphery (note: the chamber port periphery is wherein the leakage occurs), the chamber port covers some 40-45 degrees around the valve).


If you think about it, what really matters is only the clearance between the lower side of the rotary valve and the lips of the combustion chamber ports.

The conical version is functional:

on one hand, the force (due to the high pressure in the combustion chamber) on the rotary valve is quite small (think: tan(1degree)=0.0175) due to the small angle of the conical PatRoVa,

on the other hand, by varying (even “on the fly”) the displacement (ds in the drawing) of the rotary valve, the ideal clearance can be achieved / checked.


Actually this is a luxury (the manufacturing gets difficult, the quality of the working conical surfaces cannot be as good as with flat surfaces, a mechanism is required to displace the bearings of the rotary valve etc).


So we are thinking to start with the good version of the PatRoVa (flat sealing surfaces) with a clearance of 0.01mm at each side of the combustion chamber. Only around the exhaust ports on the rotary valve (i.e. wherein it runs hot) the clearance will increase to 0.02mm between each chamber port lip and its respective disk.



If everything is OK, the next model will have a slightly smaller clearance. And so on until to find the optimum: lowest leakage without rubbing (at all operational conditions) between the sealing surfaces.

The good thing is that even if the rotary valve is blocked and stopped during operation, the timing chain will brake and this is all the problem.

The DLC coating on the sealing flat surfaces is a requirement for reliability and long life.



The Wankel model engine (which is rid of side seals and seals simply by keeping tiny the clearance between the flat surfaces of the rotor and the side flat covers ) shows the way.
Read the article at http://ludens.cl/aeromod/wankel/wankel.html but not take seriously the numbers given there, because they do not fit with reality: the inevitable temperature difference between the inlet and exhaust sides of the Wankel model engine (say 250 degrees Celsius) and the “significant” width of the rotor (some 15mm) talk for a 50 microns (0.05mm) clearance at each side of the rotor (i.e. 100 microns (0.1mm)) difference between the width of the rotor and the distance of the two flat side covers during operation, and not for 5 microns difference between the width of the rotor and the distance of the two flat side covers.


And talking for the Wankel model engine, here is party quote from the HMEM forum:

“In a modified to PatRoVa (according the CAD drawings provided so far) old C50 Honda engine (39mm bore, 41.4mm stroke, 49cc) the width of the combustion chamber equals to the height of the rotor (along the power shaft rotation axis) of the 4.97cc OS Wankel model engine (mentioned in previous posts), and the total length of the “slits” results the same, too.

Suppose the PatRoVa C50 runs at 2/3 of the revs of the Wankel model engine (so that the time for leakage is the same for the Wankel and for the PatRoVa).

The one engine is 50cc while the other is only 5cc (i.e. the one tenth).

The peak pressure in the cylinder is the same.

(The time for leakage is the same, too.)

So, both appear to have the same absolute leakage (i.e. quantity of gas leaked).

Decide which is the maximum permitted leakage for the OS Wankel 5cc engine (more leakage and the engine stops running).
25% limit?
50% limit?

Take the 50% case.
So, with 5cc capacity for each chamber of the Wankel, the 2.5cc leak from the sides of the rotor.
The absolute leakage is the same for the ten times bigger PatRoVa C50 engine, which means that from the 50cc of the charge only the 2.5cc leaks from the PatRoVa rotary valve, which is 5% of the total quantity of mixture in the cylinder.
Compare the 50% leakage in the Wankel model engine with the 5% in the PatRoVa C50.

More reasonable is a 10% leakage in the Wankel 5cc engine. In such a case the leakage (which is automatically recycled during the next combustion) from the rotary valve of the PatRoVa C50 becomes 1%. By the way, how much do you think is the conventional leakage from the sides of the piston to the crankcase?”


Thanks
Manolis Pattakos

J.A.W.
109
Joined: 01 Sep 2014, 05:10
Location: Altair IV.

Re: 2 stroke thread (with occasional F1 relevance!)

Post

Here's a new 2T video, of an interesting 'sliding cylinder' design, & built with input from Uniflow, a member here.

https://www.youtube.com/watch?v=2p0dyPu7Gcs
"Well, we knocked the bastard off!"

Ed Hilary on being 1st to top Mt Everest,
(& 1st to do a surface traverse across Antarctica,
in good Kiwi style - riding a Massey Ferguson farm
tractor - with a few extemporised mod's to hack the task).

manolis
107
Joined: 18 Mar 2014, 10:00

Re: 2 stroke thread (with occasional F1 relevance!)

Post

Hello J.A.W.

Congratulations to Ken & Brett Seeber (Perth, WA), and to Uniflow, for their running prototype.

The displacement of the ports (exhaust and transfer) along the cylinder axis is significant for combining extreme peak power with flatter torque curve and cleaner exhaust.

Thanks
Manolis Pattakos

Post Reply