From: Epijk@aol.com
Message-ID: <d.157d2656.284807e8@aol.com>
Date: Thu, 31 May 2001 16:47:36 EDT
Subject: Re: PSRU's and alternative engines
To: lancair.list@olsusa.com
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Reply-To: lancair.list@olsusa.com

          <<<<<<<<<<<<<<<<--->>>>>>>>>>>>>>>>
          <<  Lancair Builders' Mail List  >>
          <<<<<<<<<<<<<<<<--->>>>>>>>>>>>>>>> 
>>
In a message dated 5/30/2001 4:09:30 PM Pacific Daylight Time, Eric Ahlstrom 
at StarAerospace@aol.com writes:

<< Last time I checked, the 2001 Oshkosh race was still a few months away.  
If we offset by one year, we get last year's results as the "nearly won". >>

I hate it when my keyboard does stuff like that to me. Of course, you're 
right. Indexing back one year aligns the claimed results with reality, which 
is the Factory won in 2000 and Engineair won in 1999.

<< Let's be totally clear about last year's Dayton to Oshkosh 
race...........On the remaining leg a factory stock Lancair IVP finished 
first overall at 319 mph average equipped with a full interior and powered by 
a TSIO-550;  all stock.>>

In the interest of the total clarity which you seek:

(1) The 2000 race was planned as a two-leg race, so that participants with 
limited fuel capacity could run flat-out on BOTH legs. The cancellation of 
the first leg for weather has little bearing on the point you make, except 
perhaps on reliability issues.

(2) I am sure you already understand that evaluating an engine installation 
solely on the basis of a race like Kittyhawk-Dayton-Oshkosh is potentially 
specious reasoning. There are numerous factors other than the engine which 
enter into a free-form race, not the least of which is the optimization of 
the vertical course profile, which attempts to balance time lost in climb 
against higher airspeeds at altitude, headwinds (even using up-to-the-minute 
winds aloft data), and the optimal recovery of the altitude potential energy.

The Venture has a very low-drag airframe, and can take good advantage of fast 
max-power cruise at low altitude and avoid the loss of a climb into the 
flight levels. (I had the privilege of picking Jim Griswold's brain about the 
clever ways he designed low drag into the Venture airframe.) 

Both the factory L4 and the Venture you mentioned had an advantage: Oren and 
Lee are good pilots who are apparently able to fly a straight line course, 
follow a planned vertical profile, avoid inadvertent entry into a TCA (aka 
"Class Bravo Airspace"), and cross Aurora on a heading somewhere between 307 
(MGY to ARR) and 358 (ARR to OSH).

BTW, If you believe the TSIO-550 in the factory airplane was STOCK, then I've 
got some beachfront property in which you might be interested. ( I have seen 
that engine and know the guys who built it.)

(3) You attempted to marginalize the 1999 race that Engineair won. However, 
if you examine the vast database, I think you'll see that the 1999 margin of 
victory (in seconds) was comparable to the 2000 race. So if the 1999 win was 
"...marginal at best...", then by your own criteria, the 2000 differences 
were "marginal" as well.

(4) The problems you mentioned regarding the Orenda installation are 
completely tangential. "Temps-in-the-red" are symptomatic of problems with 
the heat transfer characteristics of the installation, which are a major 
challenge in a liquid-cooled aircraft engine installation. There is a tough 
tradeoff between good heat exchanger performance and cooling drag, without 
even mentioning external constraints on the installation, such as a 
requirement that it be completely "firewall forward". (For more data on 
Orenda installations, check out our website at www.epi-eng.com) 

<< As for helical gears being inferior to straight cut, every turboprop 
engine 
on the market uses a reduction unit.  Almost all use helical gears.  They are 
nothing new and helical has proven it's superiority in load carrying on 
everything from turboprops to the prop reduction units on aircraft carriers 
(yes, the 100 rpm, hundreds of thousand ft-lb. props pushing the ship). >>

(1) Comparing the PSRU in an aircraft turboprop or in a ship to that of an 
aircraft piston engine is simply another example of reaching a false 
conclusion by making superficial comparisons without considering the various 
technical constraints each application must solve. The design constraints of 
each system you mention are quite different. It's a lot like comparing apples 
to oranges. 

(2) I'm not clear on your criteria for inclusion in the category of "almost 
all", but the Garrett TPE-331 constitutes a large proportion of the installed 
turboprop population. The TPE-331 that I'm familiar with have a two-stage 
PSRU consisting of (a) a small SPUR gear on the turbine shaft driving a large 
SPUR gear on an intermediate shaft which then (b) drives a planetary final 
reduction which is also implemented with SPUR gears. (I don't know firsthand 
what's in the PT-6 PSRU.)

The TPE-330-10, as a specific example, produces 940 Shaft HP at 1591 prop 
RPM. If you do the arithmetic, you'll find that the output torque is over 
3100 lb-ft. However, the turbine shaft turns at about 41,000 RPM, so the 
torque the turbine produces at 940 SHP is slightly over 120 lb-ft. It is 
clear that, in this design, spur gears were the choice for such varying 
requirements as very high torque (in the output stage) and very high 
pitchline velocities (over 40,000 FPM in the first stage).

<< However, they (Engineair) have ignored the vast database of both piston 
and turbine aviation engine reduction unit design that goes back nearly to 
WWI. All of the big radials used reduction units, none were designed the way 
that the Engineair PSRU is. >>
and

<< As an engineer, it is not my job to invent everything myself, especially 
when there is a large database of applicable hardware already available.  For 
an offset reduction unit on a piston engine, I would look to the V-12 engines 
used in WWII. >>

(1) Thank you for the philosophical input on your approach to engineering. 
I'm not sure where you obtained the necessary knowledge about the EPI PRSU 
for you to make the claim that "...none were designed the way that the 
Engineair (actually the "EPI") PSRU is...", but here are some facts. Drawing 
upon the "vast database", the EPI PSRU borrows liberally from the PSRU 
technology found in the Rolls-Royce  Merlin, Griffon and Peregrine engines, 
the Allison V-1710, as well as the (more current) Continental GTSIO-520, all 
of which use stacked SPUR gears in the PSRU. You might also be interested to 
know that the Lycoming TIGO-480, which used helicals in the PSRU, proved to 
be rather fragile.

Of the "big radials" you mention, almost all P&W 985's and 1340's were direct 
drive. The P&W 1830 and 2000 used single stage planetary reductions 
implemented with SPUR gears; the P&W 2800 used a two-stage planetary 
implemented with SPUR gears; the P&W 4360 used a multistage planetary which 
used spur gears for one stage, while the other stage used helicals for the 
specific purpose of implementing a torque-sensor in the gearbox. The Wright 
1820 used a single-stage planetary and I'm almost certain it used spur gears. 

EPI chose not to use a planetary system because we didn't want the thrustline 
down low on the engine (on the crank centerline, as it is on a radial). As 
you know, lowering the thrustline with respect to the aircraft CG is 
destabilizing (see Perkins & Hage, Airplane Performance, Stability and Control
), and raising the engine up in the airframe to compensate has the potential 
for difficult lofting.

(2) Your assertion that "...helicals have proven their superiority..." has, 
again,  apparently been made without consulting the "vast database". 

Helicals are well-suited to a variety of applications, and their properties 
can be used to great advantage. For example, the Napier Sabre (WW2 
leading-edge piston engine) used helicals in the final stage of the PSRU to 
implement effective load sharing between the two crankshafts. However, the 
input gears to that PSRU were SPUR gears. The P&W 4360 used a helical gear in 
its PSRU to implement the integral torque sensor, but the input stage uses 
SPUR gears. FYI, The EPI PSRU uses one set of helicals for a special purpose. 

In addition to all the PSRU's I've listed, if you examine the accessory 
drives of almost all the WW2 engines, you'll find spur gears almost 
exclusively, especially in the extremely-high-load application of the 
supercharger drive units, having step-up ratios of 10:1 to 12:1. These 
relatively small gears carried several hundred HP driving the blowers and 
were subjected to unbelievable acceleration loads. For specific examples, see 
the P&W 1340, 1830,  2000, 2800, and 4360; Rolls Merlin & Griffon; Wright 
1820; and for a real surprise, check out the sleeve and accessory drive 
system on the Bristol Hercules radial (over 25 SPUR gears). 

<< As for helical gears being inferior to straight cut, .....>>

Upon a careful review of my posting on PSRU Design Issues, I don't think 
you'll find any claim that helical gears are inferior to straight cut (spur) 
gears. What I attempted was to refute a previous posting claiming the 
superiority of helicals over spur gears, and to counter rumor and speculation 
with fact. In so doing, I specifically referred to gears of similar design. 

<< ...Many of them (WW2 V-12's) dealt with far more severe conditions than we 
would need on a Lancair IVP sized powerplant. >>

As an engineer, I'm sure you understand the concept of scale. The  torsional 
excitation loadings imposed by the even-fire V-12's are actually LOWER than 
those of a V-8 when proportioned (scaled) to the engine mean torque. The 
other loads (mentioned in the previous posting) are proportionally similar 
when all the design parameters are taken into account. 

Jack Kane
EPI, Inc.
tech@epi-eng.com
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
LML website:   http://www.olsusa.com/Users/Mkaye/maillist.html
LML Builders' Bookstore:   http://www.buildersbooks.com/lancair

Please send your photos and drawings to marvkaye@olsusa.com.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>