PatATE: Asymmetric Transfer and Exhaust in two-stroke engines

Hello all.

Here is an unconventional two-stroke (the PatATE) in comparison to a conventional two-stroke:

The main difference is that the exhaust port is now a "hybrid" port (the 4): at the end of the expansion the valve 7 connects the hubrid port with the exhaust passageway 5, and the exhaust starts the conventional way.

Later the valve 7 opens the transfer passageway 6 and seals the exhaust passageway 5; the exhaust continues through auxiliary (and "lower") exhaust ports 8; the transfer uses the conventional (symmetrical) transfer port 9 and the hybrid port 4.

Later the piston closes the conventional transfer port 9 and the auxiliary exhaust ports 8. 

In the conventional two-stroke the exhaust remains open for several degrees (while the transfer is closed). In the PatATE the transfer continous (while the exhaust is closed). Regarding the operation of the engine, it looks like a significant difference.

Here are some more drawings / graphs:

For more: http://www.pattakon....takonPatATE.htm

Thanks

Manolis Pattakos

How is it actuated?

Hello Gruntguru.

Depending on the engine, the valve can be activated either mechanically (a cam lobe on the crankshaft and a linkage), or electro-magnetically (as shown in the last drawing of my previous post), or . . .

For high revving, an option is the use of a rotary valve for the control of the communication of the hybrid port(s) with the transfer and the exhaust passageways, say like:

There are two hybrid ports arranged anti-diametrically on the cylinder liner.

The timing plot reminds “4-stroke” timing plots:

In the specific case, but not necessarily, the rotary valve rotates at crankshaft speed:

More at http://www.pattakon....takonPatATE.htm

Thanks

Manolis Pattakos

Edited by Kelpiecross, 01 July 2017 - 03:14.

Hello Kelpiecross.

The rotary valve is not dealing with the sealing of the combustion chamber.

The piston with the piston rings that slide on the (stationary) cylinder liner do the sealing, as in the conventional two-strokes.

The rotary valve is arranged outside the cylinder liner, around the hybrid ports, and needs not tight sealing. 

It just controls the flow of the exiting gas and of the entering gas.

In the following animation the rotary valve and the casing are sliced to show how the exhaust and transfer passageways formed in the rotary valve cooperate with the hybrid ports to give asymmetrical exhaust and transfer:

And looking from the cylinder head (the exhaust passageways are shown by purple color):

If it is not yet clear how it works, please let me know to further explain.

Thanks

Manolis Pattakos

Edited by manolis, 01 July 2017 - 11:33.

Hello all.
 

Edited by manolis, 06 July 2017 - 04:41.

Hello all.

The following version of the PatATE seems the fittest for realization in metal.

It has the typical arrangement, with one only hybrid port “looking” forwards and one intake port at the back of the engine.

The width of the hybrid port along the periphery of the cylinder liner is nearly 180 degrees.

The duration of the hybrid port is 180 crankshaft degrees:

The port at the lower side of the (red) rotary valve, in cooperation with the (blue) piston, controls the intake port of the engine:

The rotary valve (red) rotates at crankshaft speed (1:1).

The exhaust port side:

The intake side:

The Exhaust Ports area and the Transfer Ports area versus the crank angle:

Is as in the version with the half-speed rotary valve and the “double” ports.

According this plot, the variable exhaust and intake of the conventional two-strokes seem as not necessary for the PatATE because it separates the cycle in a “4-stroke-like” way:

after the expansion it follows a rapid opening of the exhaust port and the pressure drops quickly,

then the transfer opens progressively with the exhaust still opening,

then the exhaust starts progressively to close with the transfer opening more and more,

then the exhaust closes with the transfer being still widely open,

finally the transfer closes and the compression starts.

Its operation is closer to a 4-stroke with extreme valve overlap (say, as the Cosworth DFV with the 116 crank degrees valve overlap) than to a conventional 2-stroke.

Thanks

Manolis Pattakos

Hello Kelpiecross.

Judjing from the extreme revs the 2-strokes can run, it seems the two stroke engines have plenty of time for the exhaust and the transfer.

Sometimes they have more time than what is really necessary.

For instance, the OS.18TZ miniature (model / RC) engine (more at http://www.pattakon....est-rcnitro.pdf ) makes its peak power at 30,500rpm and has the red line at 42,500rpm 

Spot on the height of the exhaust port.
It is about double as compared to the height of the transfer ports. 
Why? 
Because the conventional 2-stroke has to give time to the cylinder pressure to drop before the beginning of the transfer.

Edited by manolis, 10 July 2017 - 05:52.

Hello Kelpiecross.

The 90 degrees exhaust opening before the BDC is for a racing engine wherein the power output is above all.

Here is an Opposed Piston PatATE:

In case of divided load (Portable Flyers with two intermeshed counter-rotating propellers, Electric Power sets with two counter-rotating generators, Marine Outboard engines driving two counter-rotating screws etc) the synchronization gearing (not shown) runs unloaded and the basis of the engine is perfectly rid of vibrations. 

The compact combustion chamber is shared between the two opposed pistons (the instant pressure on the two piston crowns is the same). 

The scavenging is not of the through or uniflow type. 

Thanks

Manolis Pattakos

Edited by manolis, 10 July 2017 - 15:04.

But I don't think that with having the exhaust opening 100 degrees before BDC you can expect a very good TE - even at 90BDC (which I think your engine is) it must giving away a lot of TE through insufficient expansion.

Hello Kelpiecross.

Regarding the power and torque gain (as well as the BTE (Brake Thermal Efficiency) gain) when the expansion extends:

Here is the pressure (and the torque) versus the crankshaft angle of a typical 4-stroke engine:

It is from Taylor’s book “The Internal Combustion Engine in Theory and Practice”.

After the closing of the ports a 2-stroke gives a quite similar plot.

Spot on the pressure at 90 crank degrees. It is about 1/6 of the peak pressure.

Spot on the torque at 90 degrees.

While at 90 degrees after the TDC the pressure is several times lower than the peak pressure, at the same 90 degrees the eccentricity of the connecting rod from the crankshaft axis maximizes, explaining how, with six times lower pressure, the torque at 90 degrees is only two times lower than the peak torque.

The mechanical energy provided by the engine during the rotation of the crankshaft from an angle f to an angle f+df equals to the torque (at the angle f) times the angle differential df.

The total mechanical energy provided to the crankshaft from 0 degrees to 180 degrees equals to the area underside the torque curve, which is about equal to the torque at 90 degrees times 180 degrees (according the plot of Taylor, the torque at 90 degrees is about the mean torque during the expansion stroke).

About one quarter of this mechanical energy is consumed during the compression stroke (torque curve before the TDC).

According the previous, extending the expansion from 80 degrees after the TDC (RD350LC) to 90 degrees after the TDC (PatATE), another 6% of mechanical energy arrives to the crankshaft during the expansion stroke.

And because the compression is already "paid", this 6% is "clean", which means the mechanical energy to the crankshaft (as well as the power of the engine) increases by some 8%.

An 8% increase of the power and of the torque (and of the mileage etc) cannot be considered as insignificant.

As for the above rough calculations, they favor the conventional, not the PatATE.

Regarding the exhaust opening and the transfer closing:

It is not the valve that opens the hybrid port.

Several degrees before the piston starts opening the hybrid port, the rotary valve has uncovered the outer side of the hybrid port.

See the following animation.

Spot on the hybrid port and on the rotary valve when the piston is around the TDC: the piston keeps the hybrid port closed, however the rotary valve keeps rotating to get at the right angle / position when the piston will start opening the hybrid port.

Quote from http://www.pattakon....takonPatATE.htm (near the end of the web page):

Rate of Exhaust Opening and of Transfer Closing

In the following graph it is shown the Exhaust Port area and the Transfer Port area versus the crank angle:

In the following animation it can be seen the position of the rotary valve of the PatATE just before the opening of the hybrid port by the piston.

During the compression and the expansion the rotary valve keeps rotating.

At the beginning of the exhaust the outer side of the hybrid port is fully "uncovered" by the rotary valve, so that the complete hybrid port is dedicated to the exhaust.

The rate of the exhaust opening is about double as compared to the rate of the exhaust opening in a similar conventional 2-stroke.

According the Yamaha RD350LC port-map (click http://www.pattakon....LC_port_map.jpg to download), the exhaust port extends on the periphery of the cylinder liner for some 80 degrees while the hybrid port of a similar PatATE extends along the periphery of the cylinder liner for nearly 180 degrees.

The graph at http://www.pattakon....0LC_porting.png shows the difference.

With similar timing with a conventional, the blow down of the PatATE is substantially faster.

For similar blow down, the PatATE needs substantially more conservative timing (which also means longer expansion, more power and torque, better fuel efficiency etc)..

End of Quote

If something is confusing, please let me know to further explain.

Thanks

Manolis Pattakos

A sealing nightmare with extra friction. I would not waste too much time on this one i feel. But it looks cool tho.

Sealing is low pressure only.

Friction area could be reduced substantially compared to the simplified model in the animation.

Hello MatsNorway.

You write:

“A sealing nightmare with extra friction. I would not waste too much time on this one i feel.”

No.

The sealing of the high pressure is realized exclusively by the piston rings sliding on the cylinder liner (i.e. as in any four-stroke or two-stroke conventional engine).

Only after the opening of the “hybrid port” by the piston, and only then, the working gas finds the exhaust wide open, it also finds the top end of the (red) rotary valve.

The main “duty” of the rotary valve is to prevent the slightly compressed, inside the crankcase, air or air fuel mixture to escape to the exhaust.

The rotary valve has a second “duty” to accomplish: to close the intake port on the casing in synchronization with the piston (providing, this way, a highly asymmetrical intake and ridding the engine from reed valves).

All you need is a small clearance between the cooperating surfaces (no mechanical friction involved).

In any case, the rotary valve does add some mechanical friction (cooperation of the two bevel gears, support of the rotary valve on the casing).

The question is whether the added mechanical friction justifies the radical change the PatATE design brings to the two-stroke engines.

Worth to mention here:

the basic two-stroke has substantially lower mechanical friction than a four-stroke,

and (as Gruntguru wrote):

“A lot of the mating surfaces shown in the animation could be removed, relieved or run with large clearances.”

Now compare the straight / simple way the PatATE operates (red and blue curves) reminding four-stroke engines:

with the mess / confusion / chaos during the gas exchange in a conventional like the Yamaha RD350LC (green and yellow curves) two-stroke.

In the conventional,

after the scavenging is over (scavenging: when transfer and exhaust are both open),

the exhaust keeps being wide open for several crankshaft degrees.

Can you see any reasoning behind it, except that keeping the exhaust port wide open after the closing of the transfer port is an unavoidable drawback / a heavy-side-effect of the simple 2-stroke design?

Thanks

Manolis Pattakos

Hello Kelpiecross.

Hello Kelpiecross.

It is easy to find who the third party is.

Strictly Technical:

.

how long did it take you to learn those cad drawings and anmation?

Hello Pierrre.

You write:

"how long did it take you to learn those cad drawings and anmation?"

Does it matter?

CAD is a tool, like a lathe or like a 3D-Printer.

More useful is to describe how a specific animation (say, the last animation in this page, at post# 13) was made.

A reader of the discussion may learn one thing or two.

At https://www.pattakon...atATE_Piere.dxf is the AutoCAD 2000 drawing (in dxf format) whereon the animation is based.

And here is the AutoLisp program made to create automatically the 40 slides of the animation of my last post:

;**********************

;you have ready the blocks: valve4 (sleeve valve), casing5 (casing), crankshaft4 ;(crankshaft), conrod (connecting rod) and piston (piston)

;before running the program, you open once the commands: rotate3d, insert (giving the ;right parameters) and render (giving also the basic parameters, like render to file and dimensions of the file).

;it is good to give at “Tools /Options” a higher (say 5) render smoothness.

;you also need to create a folder (c:/tempacad/) wherein the slide files will be stored.

;**********************

(defun c:anim (/ ste f r i d dd xx yy n)          

  (command "_.osnap" "" none "")

  (setq r 25.0)

  (setq f 0.0)

  (setq ste 9.0)

  (setq d 90.0)

  (setq n 0.0)

  (while (< f 360.0)   

      (command "_.erase" "all" "")

      (command "_.insert" "valve4" (list 0.0 0.0) ""  "" (/ f 1.0) "")

      (command "_.rotate3d" "all" "" (list 0.0 0.0) (list 100.0 0.0) 90.0 "")

      (command "_.insert" "casing5" (list 0.0 0.0) ""  "" 0.0 "")

      (command "_.insert" "crankshaft4" (list 0.0 0.0) ""  "" f "")

      (setq i (* (/ (+ f 180.0) -180.0) pi))

      (setq xx (* r (sin i)))

      (setq yy (* r (cos i)))

      (setq dd (sqrt (- (* d d) (* xx xx))))

      (command "_.insert" "conrod" (list xx yy) ""  "" (* (/ 180.0 pi) (atan (/ xx dd))) "")

      (command "_.insert" "piston" (list 0.0 (+ yy dd)) ""  "" 0.0 "")

      (setq n (+ n 1.0))

      (if (= n 1.0) (c:render "c:/tempacad/1" () () ""))

      (if (= n 2.0) (c:render "c:/tempacad/2" () () ""))

      (if (= n 3.0) (c:render "c:/tempacad/3" () () ""))

      (if (= n 4.0) (c:render "c:/tempacad/4" () () ""))

      (if (= n 5.0) (c:render "c:/tempacad/5" () () ""))

      (if (= n 6.0) (c:render "c:/tempacad/6" () () ""))

      (if (= n 7.0) (c:render "c:/tempacad/7" () () ""))

      (if (= n 8.0) (c:render "c:/tempacad/8" () () ""))

      (if (= n 9.0) (c:render "c:/tempacad/9" () () ""))

      (if (= n 10.0) (c:render "c:/tempacad/10" () () ""))

      (if (= n 11.0) (c:render "c:/tempacad/11" () () ""))

      (if (= n 12.0) (c:render "c:/tempacad/12" () () ""))

      (if (= n 13.0) (c:render "c:/tempacad/13" () () ""))

      (if (= n 14.0) (c:render "c:/tempacad/14" () () ""))

      (if (= n 15.0) (c:render "c:/tempacad/15" () () ""))

      (if (= n 16.0) (c:render "c:/tempacad/16" () () ""))

      (if (= n 17.0) (c:render "c:/tempacad/17" () () ""))

      (if (= n 18.0) (c:render "c:/tempacad/18" () () ""))

      (if (= n 19.0) (c:render "c:/tempacad/19" () () ""))

      (if (= n 20.0) (c:render "c:/tempacad/20" () () ""))

      (if (= n 21.0) (c:render "c:/tempacad/21" () () ""))

      (if (= n 22.0) (c:render "c:/tempacad/22" () () ""))

      (if (= n 23.0) (c:render "c:/tempacad/23" () () ""))

      (if (= n 24.0) (c:render "c:/tempacad/24" () () ""))

      (if (= n 25.0) (c:render "c:/tempacad/25" () () ""))

      (if (= n 26.0) (c:render "c:/tempacad/26" () () ""))

      (if (= n 27.0) (c:render "c:/tempacad/27" () () ""))

      (if (= n 28.0) (c:render "c:/tempacad/28" () () ""))

      (if (= n 29.0) (c:render "c:/tempacad/29" () () ""))

      (if (= n 30.0) (c:render "c:/tempacad/30" () () ""))

      (if (= n 31.0) (c:render "c:/tempacad/31" () () ""))

      (if (= n 32.0) (c:render "c:/tempacad/32" () () ""))

      (if (= n 33.0) (c:render "c:/tempacad/33" () () ""))

      (if (= n 34.0) (c:render "c:/tempacad/34" () () ""))

      (if (= n 35.0) (c:render "c:/tempacad/35" () () ""))

      (if (= n 36.0) (c:render "c:/tempacad/36" () () ""))

      (if (= n 37.0) (c:render "c:/tempacad/37" () () ""))

      (if (= n 38.0) (c:render "c:/tempacad/38" () () ""))

      (if (= n 39.0) (c:render "c:/tempacad/39" () () ""))

      (if (= n 40.0) (c:render "c:/tempacad/40" () () ""))

      (setq f (+ f ste))

  )

)

The bold letters show the five different commands the above AutoLisp program uses.

After opening the CAD drawing, the AutoLisp editor opens and the above AutoLisp program is loaded.

Giving the command “anim” they are created the 40 slides of the animation (it takes 2 minutes or so).

The time consuming work is to make (and correct / refine) the basic drawing, and then to make the AutoLisp program (which is nothing but how the parts relate with each other and where they go as the crankshaft turns).

Finally the 40 “bmp” files (created by the AutoCAD) turn to GIF files, and these files are combined using the old Microsoft GIF Animator to the animated drawing you see.

If there are questions, please do not hesitate to ask.

Thanks

Manolis Pattakos   

Edited by manolis, 15 May 2020 - 05:31.

Ha, most *blank* web pages have more code than that!