Ok, primary tubes (vice loop) will be shorter. Yes, with OAT
around
70
and
higher my static is 5200-5300 rpm. As the OAT drops the rpm
goes
up.
Around 50 deg I can get 5400-5500 and around 40F I can get 5800
rpm
static
runup. Yes, I've checked the prop with a digitial prop tach and
the
numbers
are correct. So it appears that my longer tubes (about 2"
longer
than
Tracy's) for a total 1/2 length of 21" moves my peak torque
point
down
the
rpm curve at bit.
I wish it were true indication of power, but it appears on a
cold
morning I
get just enough more power to put the prop into a stall
condition
(which
of
course then takes less power to spin it even faster and that
enable
the
engine to produce a bit more HP which spins it faster - well, up
to
about
5800-5900 rpm. Then apparently even the drag of the stalled
prop
matches
the HP being produced.
My airstrip is 2300 ft long with trees and electric wires off
one
end
and
trees and golf course off the other, ponds on both sides of its
35
foot
width (a couple of folks have scared the ducks and geese in it).
Some
folks
refuse to land on it, but once you get use to it, its not bad.
Oh, by the way, that brazing rod did an excellent job of putting
additional
material on the edge where I had the hole break-out, I'll send a
photo.
Ed
OK Ed thanks. So you are using a dynamic chamber. Is that correct?
With a total length of 42 inches? I would love to see a dyno curve
on
your
engine.
You have really good street porting as I recall. You should try 17
inches
with a simple plenum chamber and you should be able to give
Tracy a nice headwind. You don't really need all the torque at
1500
RPM.
Paul Lamar
Yes, after seeing your Sony Mavicas, I had to have one. Excellent
for
this
sort of thing.
Regarding dynamic chamber, its more like a direct connection between
the
two
rotor's secondary ports and between their primary ports with no
chamber
in
between. I calculated the pulse travel time at various rpms in
order to
arrive at the length of the tubes to support the dynamic charging
effect.
Had to make some assumptions about timing and port opening and
closings
as
well as the effects of different manifold pressure/density on speed
of
pulse
(basically the speed of sound) While the length obviously changes
with
the
rpm, my calculations indicated 25" for 5000 and 21" for 6000 rpm
(distance
from port on housing to throttle body). My engine performance would
appear
to indicate that my 21" distance gives me a boost in the mid 5000
range,
so
shorter would indeed move it up the rpm band.
However, because my aircraft is a bit on the heavy side and I fly in
and
out
of grass strips, I prefer to have the additional power (if any)
provided
by
the dynamic charging effect for the take-off regime. I will shorten
the
length this time to bring it down to 19" from 21" as clearly
spinning a
stall prop is now doing me much good on take off {:>) By the way,
that is
total length from port to throttle body, the actually tube length is
more
like 9" the remainder in the lower manifold and the throttle body
housing.
By the way, pulled the manifold tubes out of the oven this morning
after
8
hours of cooking and indeed they do appear to have hardened. The
tubes
have
a distinct sheen rather than a dull color and they respond to a tap
on
the
tubes with a ring rather than a dull thud. FWIW
Ed
Sounds like a dynamic chamber. If so it sounds like it is tuned for
2000-3000 RPM.
A simple tuned manifold with all four runners going into a 160 cubic
inch
plenum would be
about 17 inches from plenum to port at 6000 RPM and 21 inches at 5000
RPM.
That
will also give good BSFC and torque at 5000 RPM.
A true dynamic chamber operating at its best
at 5000 RPM would be half that length or around 10-11 inches chamber
to
port.
Here is a torque & HP curve on a stock dynamic chamber with runner
lengths
around
21 inches chamber to port. My guess on the exact length although I
don't
have
one here to measure. Note the torque hump at 2000-3000 RPM indicating
the
dynamic chamber is working at the 2000-3000 RPM range. I have other
charts
that correlate best torque with best BSFC. Same for a piston engine.
Tracy had this torque & HP curve in his book for some strange reason.
This gave the false impression that it was the thing to have on
your aircraft engine. Apparently he missed the significance of
the hump in the torque curve at 2000-3000 RPM. I am also to partially
to blame as it took me years to figure it out as I was confused by
Mazda SAE papers on the subject.
Here is an illustration out of a Mazda SAE paper on
the subject of series dynamic chambers probably used
on the RX8 engine. This had me confused for years as the diagram was
mis
labeled in the SAE paper. Then one day it dawned on me the
illustration
was mis-labeled. I fixed it in this version of the illustration
using Linux Paint.
Paul Lamar
Paul, Perhaps I have a tuned manifold rather than a dynamic chamber
then,
as I actually do not have a chamber. The the 21" you cite above as
tuned
for 5000 rpm is in line with what I calculated for the time it took a
pulse
generated by a closing port on one rotor to transverse the tube and
arrive
at the other rotor as its port was opening. Now that is based on the
assumption that the difference between ports is only 60Deg (or in other
words the pulse does not miss a port in transit).
So perhaps I should refrain from referring to my setup as a "dynamic
chamber". I would guess my my torque peaks in the 5000 rpm range rather
than the 3000 rpm range based on the rpm increase from 5200 to 5800 on
cold
mornings
I've got one of the 2nd generation NA manifolds with the dynamic
chamber
and the valve that switches between the long and short path (shown in
your
second diagram), so I'll go measure the paths and see what length they
are
and the correlation to the HP humps in the curve.
There is no question about the two humps in the torque curve and
interestingly it appears the short runner helps the HP at 6000 rpm quite
a
bit. Looks like it increase the HP from 140 to 160 HP or by 14% - pretty
significant. So, I may have to get out the hacksaw and do some
trimming.
However, its always easier to cut off tubing length than add it on {:>)
Ed
Dynamic chambers helps at 6000 RPM in comparison to the old untuned four
barrel carb set up.
So far we have no comparisons with a simple tuned manifold other than
the HP curve from the peripheral port Lemans engine. That engine
generated 120 HP per rotor or 240 HP at 6000 RPM for a two rotor
with a simple tuned runner length. I am working hard on the dyno and
I am making progress so we should have some numbers to add that data
and to what Al Geitzen has recently uploaded.
The fundamental characteristic of the single dynamic chamber is it works
at only two RPMs. If it works at 6000 RPM it will work again at 12,000
RPM.
Extreme example. If it works at 3000 RPM it will work again at
6000 RPM but not as well as a simple tuned manifold IMHO.
Ideal for our use would be a Lemans type variable length set up
where we would have one length for take off RPMs and another for
top speed and a third for cruise RPM for best BSFC. Not all that
hard to do as we are only talking about a four or five inch
change in length. This will become more important as the 2.85:1
and 3.12:1 gear box and the higher max RPM RX8 engine come into use.
I think Mazda has already found that out with the RX8 engine.
Basically the poor torque characteristics of the rotary engine
make it a better aircraft engine than a car engine.
IMHO Mazda will eventually address that issue with e
super charger to enhance the torque down low.
BTW Ed how far down your airstrip do the RPMs finally peak
if at all? What are the RPMs that you use to climb out
after you break ground?
Paul Lamar
It would be nice to try some different configuration on a dyno for
comparison. My current sliding tube manifold was fabricated with the idea
of trying different manifold lengths (can vary length by 2 1/2"). Its
currently set for the long position (21") but could be shortened to 18 1/2"
by simply opening a tube clamp and lowering the top part. I had envisioned
a high torque dc motor driving a screw to run the tubes in and out with a
two position swithc "Take-Off" and "Cruise". Had too many other variables
early on to play with changing its length, but now might be the time to do
so.
The RPM doesn't really peak until straight and level, max so far is 6100
rpm. I climb out at 5400-5500 rpm until level off then it accelerates to
6000 or so unless I throttle back. I think I may have some airflow
impediment at higher rpms. The reason I say that is while I can hit
6000-6100 rpm in level flight at 3500 MSL, lowering the nose into a slight
dive showing 200 mph IAS does not cause the engine to rev higher as I would
think it would with the prop/engine unloaded in the dive.
Ah, well, if we knew all the answers, it wouldn't be as fun {:>)
Ed
I have been thinking about it too. If you apply a point load on
telescopic tubing it tends to cock and jam. I am thinking about
a light weight integral annular hydraulic cylinder on each runner to
serves as actuators. I'll draw something up.
Paul Lamar
The AirCraft Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://home.earthlink.net/~rotaryeng/
Copyright 1998-2003 All world wide rights reserved.