It seems that at 6000 rpm, the rotor would
be doing 2000. that is 720,000 degrees of rotation per min or 12,000 per
second.
So if the exhaust port is 12 degrees of
rotation wide, it would require .001 seconds to open fully, followed by 50 more
degrees of rotation to squish the face to its lowest volume, that is another 4
ms, (a total of 5 ms) then as the volume is increased, there will be another 60
degrees or 5 ms before the next apex seal blows by and starts it all over
again.
On a graph, it should look like a
tremendous spike followed by a positive then a negative sine wave. The whole
thing would be 10 ms.
(After thinking about it, maybe it would
be 9 ms with half of the spike on both ends??)
The spike would occur while both faces are
open to the port so the last portion of the negative sine wave would overlap
this spike.
I am having trouble getting my head around
what is actually happening. Then to try and figure out what would help
scavenge, and then figure out how to do that, and then figure out how to build
an exhaust that will do it…
I am getting a headache and going to
bed! :>)
Good Luck with this project!
Bill B
From: Rotary motors in aircraft
[mailto:flyrotary@lancaironline.net] On
Behalf Of Mark Steitle
Sent: Saturday, November 20, 2010
9:54 PM
To: Rotary
motors in aircraft
Subject: [FlyRotary] Re: Exhaust
Calcuations
Bill,
That's one reason why I asked the question about the rotary being more
like a 2-stroke when it comes to exhaust design. The other is that there
is no exhaust valve to open and close. Instead, there is a revolving door
that passes across the exhaust port, never really closing. So, I took
Ed's recommendation and ordered the book on how to hot rod the rotary engine.
There's a chapter on exhaust design that I'm hoping will explain all this
in terms that I can understand.
On Sat, Nov 20, 2010 at 8:42 PM, Bill Bradburry <bbradburry@bellsouth.net> wrote:
Here is where I show my lack of understanding….
Isn’t 40 cu in the size of one rotor and shouldn’t this
calculation be based on the rotor faces instead of the entire
rotor? With a recip engine the exhaust valve on a cylinder
opens once every other revolution of the crankshaft, but with the rotary, a
rotor face exhausts every revolution of the e-shaft. So there are twice
as many pulses in a rotary header pipe as in a recip header.
So can the calc for a recip be used for a rotary without some
modifications?
Bill B
From: Rotary motors in aircraft
[mailto:flyrotary@lancaironline.net]
On Behalf Of Ed Anderson
Sent: Saturday, November 20, 2010
12:21 PM
To: Rotary
motors in aircraft
Subject: [FlyRotary] Re: Exhaust
Calcuations
I did some
algebraic manipulations to find HPL based on a selected HPD (tube diameter) and
got an answer. However, I am somewhat skeptical of it although it
does tend to fit with the general theory of larger dia pipes = shorter lengths.
By arbitrary
selection an diameter - you automatically select the rpm
factor in the equation (but, its hidden in the HPD value). The way you
check what your new rpm sweet spot with the arbitrary diameter is to vary
the rpm in the first set of equations until you get the length determined by
manipulated equation.
So the
revised equation for 1.8" dia tubes gave 27.58" inches for the
HPL. You vary the rpm value in the first equations until you get an HPL
of 27.58" and you get an rpm of approx 7900 - so I agree, probably too
high for your configuration.
Sent: Saturday,
November 20, 2010 11:18 AM
Subject: [FlyRotary] Re:
Exhaust Calcuations
Ed,
OK, that answers
my question. By going with 1-7/8" OD pipe (.049 wall), it would be
tuned for about 7800 rpm. Obviously, this is too high for our purposes.
I'll go with 1-5/8 (.063 wall) to shoot for 6700 rpm peak power.
This is typically the rpm I see on ground roll and early climb before I
start adjusting the prop down to a lower rpm to transition to cruise climb.
Have you used the
header designer tool available from Burns Stainless? You need to know all
the timing numbers, rpm, etc. I may give it a try and see what I come up
with. I have a test p-port housing that I made up to see if the tube
could be welded to the inner liner. So, I can use that with a spare rotor
to verify the EVO and EVC numbers.