wankel rotary project

hello all....didnt know where to post this so threw it in here...if needs be higher powers can move it

in my second year in college and been asked to come up with a small project, the problem is i dont do small lol

ive my head set of a wankel rotary project and with a partner we have done some solidworks drawings and threw together a model with experimental dimensions

now this will be just a demonstration model that we plan to be turned by hand to demonstrate the wankel engine concept, we have as of now not included any sort of apex seals but then can be added for realism.

from doing some research there is a few different ideas on the shapes and various names for geometries that i hope some can clear up

we gleened most of our information from the PDF of kenneth weston chapter 7 rotary engines and also the yamamoto pdf.

the formula used in our project for the housing shape was given with x and y co-ord in the these formula

x = e*cos(3*§)+R*cos(§)

y = e*sin(3*§)+R*sin(§)

eccentricy (e)
rotor radius (R)
§ is varied from 0 to 360 degrees to locate the tip of the rotor as far as im aware.

we were unsure as to what dimensions to use so wee looked in the yamamoto book for some numbers and it had a few examples so we scaled them to out size, weight and machine capabilities limitations.

using excel i calculated 360 X and Y co-ordinates to do up my CNC code as the housing will need to be done on a subprogram and called upon in the main program. the machine used in my college is a HASS CNC mill....not sure on spec of it right now but it is something i can confirm.

we have chosen our gears and have hence designed around these components

here is a few pictures of a early model in solidworks. the solidworks package will not allow us to mate the correct parts in order to virtually test out the mechanism so i am weary of our design...we are running a eccentricity of 15mm and a rotor radius of 70mm. at the side the rounded edges on the rotor are crashing as they are just arcs taken from a point.some guidance in this area would be welcomes greatly.

the gears we have chosen are 48 tooth 1.5 mod internal gear and a 32 tooth 1.5 mod spur gear....both have teeth 10mm wide.

for some reason i do not have the most recent pictures to hand of the working drawings and solidworks models but in these the stationary gear is missing a 10mm hub on the front side which will be modified to be held by the front housing on some way. the centre of the stationary gear will also need to be machined ut as the centre shaft will be running in centre with it and also flush with front cover.

http://i188.photobucket.com/albums/z...ps12ef6f3b.jpg

http://i188.photobucket.com/albums/z...ps27a9cbba.jpg

all comments and criticisms welcome and encouraged.

thanks
leigh

Kind of a cool project. Maybe get on ebay and find a small motor with gear deduction so you can talk to people while it slowly rotates.

Hi Leigh,
you can use the well proofed epitrochoid-parameters of the 294 ccm engines:
R= 71mm e= 11,6 mm a=0,5 mm
I’ve made a Excel-sheet to check the parameters:
In C2: 11,6 (e); in C3: 71 (R)
A5 – A605: 0 – 600
B5: =PI()/100*A5
B6: =PI()/100*A6
and so on to B605
C5: =$C$2*COS(B5)+$C$3*COS(B5/3)
C6: =$C$2*COS(B6)+$C$3*COS(B6/3)
and so on to C605
D5: =$C$2*SIN(B5)+$C$3*SIN(B5/3)
D6: =$C$2*SIN(B6)+$C$3*SIN(B6/3)
and so on to D605
marked C5 to D605 -> Point (XY) Diagramm gives the shape of the epitrochoid
column C5 - C605 is X; column D5 - D605 is Y
[a] is the parallel distance to the epitrochoid (because of the apex seal radius)

..Nipsel..

Quote:

---

Originally Posted by Nipsel

Hi Leigh,
you can use the well proofed epitrochoid-parameters of the 294 ccm engines:
R= 71mm e= 11,6 mm a=0,5 mm
I’ve made a Excel-sheet to check the parameters:
In C2: 11,6 (e); in C3: 71 (R)
A5 – A605: 0 – 600
B5: =PI()/100*A5
B6: =PI()/100*A6
and so on to B605
C5: =$C$2*COS(B5)+$C$3*COS(B5/3)
C6: =$C$2*COS(B6)+$C$3*COS(B6/3)
and so on to C605
D5: =$C$2*SIN(B5)+$C$3*SIN(B5/3)
D6: =$C$2*SIN(B6)+$C$3*SIN(B6/3)
and so on to D605
marked C5 to D605 -> Point (XY) Diagramm gives the shape of the epitrochoid
column C5 - C605 is X; column D5 - D605 is Y
[a] is the parallel distance to the epitrochoid (because of the apex seal radius)

..Nipsel..

---

i'm a bit of noob when it comes to excel but have managed to make up a spread sheet to define the shape of the housing for CNC code purposes.

this was done using the formula i found above and the dimensions are basically home made to a shape that looks like it would show best the irregular geometry of the housing.

i took X and Y co-ordinates ranging from 0 degree to 360 degree and fed them into the graph on the sheet

if you have any dimensions or link to other sueful information that you may think is helpful that would be greatly appreciated as was your post already .

http://i188.photobucket.com/albums/z...ps112417c3.png

Quote:

---

Originally Posted by stevehuckss396

Kind of a cool project. Maybe get on ebay and find a small motor with gear deduction so you can talk to people while it slowly rotates.

---

yes thats the idea i had but was told to make something that is only a mechanism or linkage to show the motion....

thank you for your input it is greatly appreciated.

Hello Leigh,
the best "information" written in english language is the "bible" of NSU-Wankel engines: "Rotary Engine" by Kenichi Yamamoto.
A lot of literature is written in german because of the inventor of the NSU-Wankel rotary-engine Dipl.-Ing. Hanns Dieter Paschke of NSU Motorenwerke AG.
The basic of most of the NSU-Wankel engines are the NSU-engines and the 'Fichtel & Sachs' engines.
The parameters of the KM914: R=71; e=11,6; a=0,5; B=68,2 mm, Norton has changed sometime to: R=69; e=11,6; a=2,5; B=68,2 mm.
On page 20 of the '1971 Yamamoto' you can find the missing parameter for the rotor: "R-2e"

..Nipsel..

sizes as of yet are not finalized but are subject to the internal gear size that i have ordered for the project being 100mm overall diameter that i am challenege'd with designing a interference fit to press the gear into the rotor.

and what are the "a" and "b" parameters you write of?

i was under the assumption the housing shape was described by only eccentricity and rotor depth.

Hi,

Looking at the rotor: where are the combustion cavities? Most (all?) Wankel engines have combustion cavities, see for example the topmost picture on the right on the English Wikipedia page Wankel engine - Wikipedia, the free encyclopedia. I do not think you have enough flesh between your internal gear and the outer edge of the rotor to accommodate those.

Also, did I get that right that you plan to machine the rotor from Aluminum and the the internal gear is steel? In that case, a press fit will not work. If you press the cold gear into the cold rotor, the gear will come loose in operation of the engine as AL expands more under heat than steel. And if you press hot, then that little bit of AL material you have between internal gear and outside of the rotor will break once things cool down.

I'm working (on and off) on a little Wankel project of my own. I decided to make the rotor of mild steel, and cut the gear myself: use my Tormach and a 1/16" extra long cutter to mill away most of the material, and either (original plan) make my own single-tooth hob, or (more likely now) rely on a hardened spur gear, lapping paste, and elbow grease to bring the gear into its final shape. in order to do so I had to use fairly coarse teeth: 21 in the internal gear, and 14 in the spur gear IIRC, module 1.5. The gear is about 4 mm wide. I'm sure that, with this type of construction, this will be a very noisy engine at the end, and probably not long-lived, but it will hopefully work. If it does, I could try heat treating a rotor made of appropriate steel.

(If you decide to go this route, keep in mind that the tooth geometry for internal and spur gears meshing at the required 2/3 to 1 ratio is different from what would normally be prescribed for gears with that number of teeth.)

Stephan

thanks for your reply....keep in mind that this will never be a running engine and is for display and demonstration purposed only.

but i am a bit puzzled as you mention the geometry of the flanks....what is their largest radius. if i cannot find that out i will have to create an almost straight edge flank so as not to foul the housing.....

im looking for a nice smooth rotation and dont want the rotor to strike the housing and bind.

the rotor design is not yet finalized and is greatly dependant on the rotor gear as you have said....i will have to therefore machine down the gears outer diameter or increase the dimensions....i have set up the excel sheet so that i generate the subroutine and modify the numbers by just changing one cell number to update the sizes.

solidworks will not allow us to rotate the mechanism to flag up any potential fouling problems so im looking for experienced members to flag up any they might see that i may be overlooking....

can i ask what formula you have used to generate the housing shape for your project?

thanks you for taking the time to reply to me.

leigh

Hi Leigh,
i hope i understand right,
your "starting point" is your choosen gear
in this case you need to know the distance of the axis of your gears
-> this will give you the eccentricity!
then you can choose the Radius ..usefull will be the R/e of ~7
->with 15 mm distance of your gearaxis you shold choose ~105 mm Radius
(this are the parameters of the Mazda engines)

The Epitrochoid housing with this parameters needs a rotor with sharp edges (like a knive)
in practical application the apex-seals have a radius
the housing is machined bigger so the center of the apexsealradius will move exactly on the epitrochoid curve
[a] is the parallel distance to the epitrochoid = apex sel radius
[B] is the width of the Rotor

..Nipsel..

your R/e of 70/15 = 4,66 is to low.. should be between 6 and 8

my R/e ratio is 70/11 = 6.36?

the rotor depth will have to be increase in order to correctly flatten the flanks enough to prevent fouling....as of yet we have no plans to add apex seals and are just planning on adding 1mm clearance between rotor corners and housing....we may modify in some seals for show if we have the time....


here are my gears

http://i188.photobucket.com/albums/z...psc02f5e1f.jpg

Quote:

---

Originally Posted by leeK9

hello all....
..snip..
...we are running a eccentricity of 15mm and a rotor radius of 70mm.
..snip..

thanks
leigh

---

Your gears will give you a axis distance of 12mm
( a= (m*(z2 - z1))/2 )

..Nipsel..

my apologies...that must have been a mistake...take a look at the excel spreadsheet where e is 11mm and R is 70mm

the dimensions are plotted onto the graph to show the housing shape

what is the term ''axis distance''?

and what bearing does it have on the rotation of the rotor?

do you have any idea what are the limiting dimensions on the rotor flanks?....as per the drawing i think that they are two rounded and will foul but these are the preliminary dimensions so we could get some drawings for a design report....we will be starting back in the next few weeks to make this so we will need to have our dimensions finalized on the next few weeks.

Hi Leigh,
sorry for my bad english..
"axis distance" is Center distance
List of gear nomenclature - Wikipedia, the free encyclopedia
The center distance of your gear and the eccentricity has to be the same dimension!!

The rotor - bearing of the Fichtel&Sachs 294 was "Needle roller bearing"
Needle roller bearing - Wikipedia, the free encyclopedia

The limiting dimension on the rotor flanks is described in the '81 Yamamoto:
X=($C$3*COS(2*B5))+(((3*($C$2^2))/(2*$C$3))*(COS(8*B5)-COS(4*B5)))-($C$2*((1-(9*($C$2^2)/($C$3^2))*((SIN(3*B5))^2))^0,5)*(COS(5*B5)+COS(B5) ))
Y=($C$3*SIN(2*B5))+(((3*($C$2^2))/(2*$C$3))*(SIN(8*B5)+SIN(4*B5)))-($C$2*((1-(9*($C$2^2)/($C$3^2))*((SIN(3*B5))^2))^0,5)*(SIN(5*B5)-SIN(B5)))
0°- 30° 30°- 90° 90° - 150° 150° - 210° 210° - 270° 270° - 330° 330° - 360°
C2=e; C3=R; B5=PI()/180*A5

..Nipsel..

oh yes i see....i never thought of the gear eccentricity

the inernal gear has a PCD of 72 and the stationary spur gear is PCD 48......PCD diameter from centre of teeth on both.....that would leave me with a distance offset of 12mm correct?

i need to further study the yamamoto book.

soo over the past few weeks i have been making progress on this project....bearing in mind i only get 3 hours class time a week to work on this

but the CNC code was written up and debugged to create the rotor

Attachment 227032

How do you draw the rotor, I'm always finding how to do the housing but never the rotor. I figured it's done by hand in cad, but I was wondering if there is an easier way like the spreadsheet for the housing.

The rotor itself is only governed by the rotor radius and is symetrical through 120 degress at each side,

the rotor flanks limiting radius is found by using an equation which i found to be simply unfathomable, so i used a paper model and took at estimate of the radius and used solidworks to confirm

depth and inside circular pockets are project specific

for example purposes i'm using a 170mm radius on a 84mm rotor radius and 12mm eccentricity

I can draw the rotors in cad fine, was just wondering if there was a less pain in the ass method lol.

Yamamoto's equation looks daunting, but it's really not. Just use minus where the plus/minus is. Using axes of symmetry you only have to plot from 0-30 degrees (then mirror, and array). I resolved the equation every 6 degrees in excel. I then drew it in autocad, which is enough resolution for a smooth running engine.

The rotor flank is NOT an arc. Nipsel wrote out the equation format to use in excel 1-17-2014 (super easy in solidworks too).

Example in Siemens NX9:
Attachment 404600
..Nipsel..