Douglas Fir wrote:
Everyone
Turbo compound project. Report #2
With the turbo 13B engine safely in my workshop I am starting to plan the
research project for turbo compounding it. After rebuilding the engine with
new seals, and putting Tracy's EC-2 unit on it. I'll leave the engine in
its stock condition (yes running on gasoline) with only the alternator and
water pump, connected during primary testing.
The first dyno testing (after I assemble the dyno) will be to determine
baseline BSFC data for this particular engine at different power settings.
I already have a small 25 channel data logger made by PICO in England that
interfaces to the PC through the RS-232 port.
The second phase of testing will be to design and make a turbo compound
system using the standard turbo and see if I can document an improved BSFC
with the same test variable loads.
Please jump in with your thoughts and correct me if I am wrong or help fill
in the ? marks.
IMHO:
It is safe assumption to say that the majority of homebuilt aircraft owners
are very concerned with improving fuel economy.
A standard turbo is sized to provide boost and waste the exhaust gas in the
high MAP situations and hopefully before the engine starts knocking.
Common aircraft economy cruise ranges are from 55% to 75% power.
At these power settings not a lot (if any ?) exhaust energy leaves via the
waste gate. We need to find a way to measure the actual torque real time
that is leaving the turbo shaft and would like to hear any ideas. Turbo
tach and torque sensor. transducer?
Stealing energy from the standard turbo/compressor, diverting it through a
CVT and delivering it to the e-shaft via a belt drive will become a
lose/lose/gain? situation. In short many variables come into play at almost
real time.
Lets say we steal 20% of the energy from turbo and send it to the e-shaft.
We still have 80% to use for the boost compressor though. When lowering the
boost by 20% it lowers the air/fuel amount but perhaps not by 20%?. For a
moment we could say the Hp is also lowered and assuming we didn't touch the
C/S prop setting, the engine rpm and hence turbo torque would go down.
However simultaneously we are applying 90% of this 'borrowed' torque (10%
loss through the CVT) thus adding a boost to the e-shaft. Logic would
dictate we are borrowing from Paul with interest to pay Peter. We are in a
losing proposition and it would seem that our system would rapidly
degenerate into becoming an engine brake as the rpm decreased even more. Of
course rpm, Hp. and boost curves are not perfectly linear. We may find a
sweet spot. But likely not.
Our experiment will fail. Fine but we expect that. Now we look at the data
and see where we failed the least. Even with the standard turbo there is
bound some turbo range that is a little less bitter in defeat. We then know
what to improve. For starters I would think we would have to get the CVT
operating well over 95% efficiency.
Moving on:
Logically the best rpm to transfer power to the e-shaft would be at the rpm
or manifold settings that causes the turbo waste gate to open. High power
setting? Not so good for economy though because of the high drag effect on
our airplane at high speeds. Forget for a moment we may be above 10K' where
the thrust/drag ratio becomes more beneficial to BSFC in some planes.
Thus:
The ideal situation will be to use an over sized turbine that has energy to
waste so to speak. One that is specifically designed to take advantage of
the rotary engine's dynamic boost. In short a very efficient turbo motor.
The unit will be determined by size, weight, performance. It might? need
moveable vanes in the intake to improve the compressor's compression ratio
at lower speeds or keep a boost setting through a wider (slower) RPM range.
With the oversized turbo we will at least have more wasted energy to play
with. We can then have our cake and perhaps nibble it too. Turbo
compounding is not the technology to let us go fast and sip fuel. At best it
will let us have our full boost climb out and full throttle dashes with a
small addition of torque to the eshaft.
I am betting it is somewhere in the cruise ranges that this idea will
shine. With a CVT we can dial in what percent of the turbo motor power goes
to the compressor or e-shaft. No more waste gate 'wasting' an available
resource. Best 55% cruise might have only a small air boost with most of
the energy going through the turbo compound? Don't know. Have to test it.
Bound to be some surprises. At what point does a low BSFC happen in your
airplane with this device? That will depend upon your altitude, power
setting, type of airplane, prop and rpm setting.
For me that life saver cruise speed could be lower than 50% power when
encountering headwinds more than halfway into a 2000 mile leg over cold
waters.
With some success and a lot more failures we can make an educated guess on
how to size the turbo-compound system for the 20B, which is the engine I'll
eventually use.
Doug in Japan
Doug, I got the gear catalog from QTC Gears and I suspect it is the same
as the KHK catalog. http://www.qtcgears.com/Q410/QTC/Q410P344.htm
I found a pinion/shaft and gear that will give us a pitch line velocity
of 15,500 ft/minute at 100K RPM which is an improvement over my first and
second iterations. Since the V22 Osprey gear box is running at 25,000 ft/minute...
15K ft/min is a rather low number. (See this pdf for more info on high
speed gears. http://gltrs.grc.nasa.gov/reports/2002/TM-2002-211336.pdf )
The KHK part number is SSGS1.5-10 ten teeth, PD is 15 mm or .59 inches.
It was driving me crazy trying to figure out a first stage planetary configuration
using this pinion/shaft and off the shelf ground precision planet gears so I
wrote this QB program to help me out. The last column checks to
see if a three planet configuration is even possible. If the last column is
an integer it is possible to build a high reduction ratio 3 planet
planetary using the planets with that number of teeth.
Bottom line is about the only off the shelf high precision planet gear we can use is the
KHK MSGA1.5-50. A 50 tooth, 1.5 module gear with a PD of 75 mm or 2.952 inches.
this gives us a 12:1 reduction for the first stage or at 100,000 RPM turbo charger
input 8,300 RPM output and in to the CVT. This is of course going to require a custom
ring gear with 110 teeth and a PD of 165 mm or 6.496 inches. I'll do a 2D drawing
shortly.
Now I need help from you oil engineers. I am looking for the specs
on synthetic oil and how it relates to standard gear oils. I suspect
synthetic oil will work much better at these high pitch line velocities
than traditional oils. There is such a thing as Kinematic Velocity
and I need to know what that is for say, Mobile One.
Paul Lamar
The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://home.earthlink.net/~rotaryeng/
Copyright 1998-2004 All world wide rights reserved.
CLS
'mod 1.5
OPEN "planet.dat" FOR OUTPUT AS #1
PinGearNoT = 10
PRINT #1, "PinGearNoT ="; PinGearNoT
DiaPin = 1.5 * PinGearNoT
PRINT #1, "DiaPin ="; DiaPin
PRINT #1, "DiaPin Inches ="; DiaPin / 25.4
'PRINT #1, "DiaPin Inches/2 ="; (DiaPin / 25.4) / 2
PRINT #1,
PRINT #1, "Planet Gear Ring Gear Dia Ring Ring Gear "
PRINT #1, "No Teeth No Teeth Gear Dia. Dia. Inches Ratio 3 Planets"
FOR PlanGearNoT = 35 TO 70 STEP 1
RingGearNoT = PinGearNoT + 2 * PlanGearNoT
PRINT #1, USING "###"; PlanGearNoT;
PRINT #1, USING "##############"; RingGearNoT;
DiaRing = 1.5 * RingGearNoT
PRINT #1, USING "##############"; DiaRing;
PRINT #1, USING "#########.####"; DiaRing / 25.4;
PRINT #1, USING "#########.####"; RingGearNoT / PinGearNoT + 1;
PRINT #1, USING "#########.####"; (RingGearNoT + PinGearNoT) / 3
NEXT
CLOSE
SHELL "More < planet.dat"
END
'PRINT
'PlanGearNoT = (RingGearNoT - PinGearNoT) / 2
'PRINT "PlanGearNoT ="; PlanGearNoT
'PlanGearDia = PlanGearNoT * 1.5
'PRINT "PlanGearDia = "; PlanGearDia
'PRINT "PlanGearDia Inches ="; PlanGearDia / 25.4
'PRINT
'TotalRadius = PlanGearDia / 25.4
'PRINT "TotalRadius="; (PitDiaPin / 25.4) / 2 + PlanGearDia / 25.4
'PRINT "Gear Ratio ="; RingGearNoT / PinGearNoT + 1
PinGearNoT = 10
DiaPin = 15
DiaPin Inches = .5905512
Planet Gear Ring Gear Dia Ring Ring Gear
No Teeth No Teeth Gear Dia. Dia. Inches Ratio 3 Planets
35 80 120 4.7244 9.0000 30.0000
36 82 123 4.8425 9.2000 30.6667
37 84 126 4.9606 9.4000 31.3333
38 86 129 5.0787 9.6000 32.0000
39 88 132 5.1969 9.8000 32.6667
40 90 135 5.3150 10.0000 33.3333
41 92 138 5.4331 10.2000 34.0000
42 94 141 5.5512 10.4000 34.6667
43 96 144 5.6693 10.6000 35.3333
44 98 147 5.7874 10.8000 36.0000
45 100 150 5.9055 11.0000 36.6667
46 102 153 6.0236 11.2000 37.3333
47 104 156 6.1417 11.4000 38.0000
48 106 159 6.2598 11.6000 38.6667
49 108 162 6.3780 11.8000 39.3333
50 110 165 6.4961 12.0000 40.0000
51 112 168 6.6142 12.2000 40.6667
52 114 171 6.7323 12.4000 41.3333
53 116 174 6.8504 12.6000 42.0000
54 118 177 6.9685 12.8000 42.6667
55 120 180 7.0866 13.0000 43.3333
56 122 183 7.2047 13.2000 44.0000
57 124 186 7.3228 13.4000 44.6667
58 126 189 7.4409 13.6000 45.3333
59 128 192 7.5591 13.8000 46.0000
60 130 195 7.6772 14.0000 46.6667
61 132 198 7.7953 14.2000 47.3333
62 134 201 7.9134 14.4000 48.0000
63 136 204 8.0315 14.6000 48.6667
64 138 207 8.1496 14.8000 49.3333
65 140 210 8.2677 15.0000 50.0000
66 142 213 8.3858 15.2000 50.6667
67 144 216 8.5039 15.4000 51.3333
68 146 219 8.6220 15.6000 52.0000
69 148 222 8.7402 15.8000 52.6667
70 150 225 8.8583 16.0000 53.3333
