From: Johnny <nwaero@northwest-aero.com>
Newsgroups: rec.aviation.homebuilt
Subject: Re: Chevy Vortec conversion specs?
Date: Wed, 07 Oct 1998 15:58:44 -0700

Owen Davies wrote:

> Among other things, Charles K. Scott wrote:
>
> >To the engine weight you must add the weight of the PSRU because in
> >order to produce over 200 horsepower it must be spun to over 4,000 rpm.
>
> Hmm.  Do you have any idea how fast it must be going to produce 150 hp?
> This sounds like it might be a reasonable substitute for the Wittman
> Buick/Olds/Rover direct-drive installation.

I was talking to a customer today who is just getting there Vortec
together and onto the airplane. They built one up from scratch, starting
with a junkyard engine. Not a cheap build, and has some really nice
components involved. Not Stock I guess you could say. There dyno time
has shown 272hp at 4700 rpm, and that's at 3000' MSL. Corrected to SL is
somewhere closer to 300hp. I will be putting the dyno reports on the
website as soon as I get them.

I think after seeing the chart you will see that the tuning emphisis was
put on optimising for a cruise rpm of somewhere around 4150 rpm. IOW,
torque curve is highest there, and we really don't care too much about
what it is below that.

Just for fun, you might extrapolate a bit. Let's say that this
particular engine is putting out 304 ft/lbs of torque at that rpm to
make the 272 hp. Theory might tell you that the same engine, optimized
for a lower rpm, would be able to make the same torque at that lower
rpm. Optimized simply means that the cam, intake manifold, intake ports,
valves, exhaust ports, and exhaust headers are all sized for that rpm.
If that were the case, it would appear that the same engine could make
about 162 hp at 2800 rpm. But even optimized, can that same engine
actually make the 304 ft/lbs at 2800 rpm? I think it's plenty doable.

[bold statement mode on]

Aircraft application for car engines allows those engines to exceed any
performance that they may have achieved in the cars they were originally
built for.

[bold statement mode off]

Huh? How can that be? Simple. And I think we have talked about this
before, but maybe not exactly in these terms.

In the car application the engine is required to have a bunch of
characteristics that it doesn't have to adhere to in the aircraft
application. The most important being width of powerband. In the car the
thing has to idle, come up off of idle with lots of torque, be smooth
down there in those basement rpms, yet still have enough flow to do
something when you drop down a gear to pass someone.

In the aircraft application, we don't care at all about lower midrange
torque. We can give that up to optimize for a very narrow powerband. The
same for the high end of the scale. Whatever we get above 5000 rpm is
purely coincidental. We can give that up too, and size things like
exhaust maybe a little smaller than would otherwise be 'normal' for the
same amount of cam in a car application. We are interested in what the
thing does from 4000-5000 rpm (typically) and that's it. What's more,
roller cam lift rates pretty much dictate an almost totally flat torque
curve for a narrow 1000 rpm band like that. What this means is that when
you crank some pitch into the prop and load the engine down from your
5000 rpm takeoff to your 4000 rpm cruise, you are not seeing much of a
power reduction, even though you've taken 1000 rpm out of it. This means
that even naturally aspirated engines can still have something at that
same 4000 rpm after taking a 30% power hit for altitude.

There are a bunch of other factors that are commonly missed when taking
that car engine comparison to the airplane. One is that we don't care
too much about emissions. Car manufacturers are strapped pretty tightly
to it, and through those straps have learned all about improved
emissions through improved fuel burn. This is a great bit of technology
to take advantage of, but in looking at that power curve again, we don't
have to worry if we're a little dirty on the low end where our somewhat
lumpy cam is going to make us fail at the sniffer station. Not to
mention that if you want you can still get real gas at the airport which
allows for some real efficient compression ratios to be used. Or, lots
of boost if you prefer that method.

Yet another area to get some extra out of is our dreaded open exhaust. I
prefer some sort of muffler myself, but most conversions flying are just
as open as it gets. It's hard to tune any better than a good set of
tuned open headers. Try this in your grocery getter sometime. I bet you
don't make it to the store more than once before the man gets rather
rude with you about it. Haven't seen that man pull up along side me
while flying yet. Maybe someday.

As we go up, we loose manifold pressure. However, I don't recall the
last time my car was sucking in air that was -40F either. In terms of
density, it certainly doesn't hurt, especially if you are planning to
boost to compensate for the altitude. You intercooler doesn't have to be
anything like the car one to get reasonable intake temps, which means
that turbo sizing can be such that your boost source is more efficient
-> less power used to develope the same amount of actual boost (actual
meaning after you take density into account).

As for how slow can you turn it and still get the 150hp; Ya, I think you
could get 150hp at prop speed, but you are carrying quite a bit of
weight around for that kind of power. The advantage comes when you get
the real power that's available out of the weight that you are carrying
around. The added weight of the PSRU is out weighed by the increase in
power that comes with the increase in RPM that the prop won't allow
otherwise. The bigger the engine, and the more the power, the less of a
percentage the PSRU becomes and the more the hp is that can be gained
per pound of reduction drive weight. You could probably say the same
thing for the dollars to hp ratio.

-j-

From: Johnny <nwaero@northwest-aero.com>
Subject: Re: Fuel injection
Date: 15 Aug 1998
Newsgroups: rec.aviation.homebuilt

Howdy,

ras wrote:

> This is a pretty general question, but what are the issues involved in
> using a modern fuel-injection auto engine in a homebuilt (as opposed to a
> carburated engine)?

Too many issues to cover in a reasonable length post here. Main ones are
redundancy and cost. Neither of which can't be overcome, but it ain't as
easy as just slapping a Holley on it.

> I've watched a couple of the Oshkosh auto conversion
> forum videos and they all seemed to be afraid of fuel injection, even to
> the point of retrofitting a carb to a fuel injection intake manifold (which
> didn't work).

In my opinion, that is just ridiculous. People fear what they don't
know. Probably rightly so in aviation, but it just shows a lack of
willingness to learn what you need to, to get the job done. To run a
carb because that's what you know AND that's what the engine came with
is understandable. To basterdize an EFI engine out of ignorance is just
lame.

> I'm looking at the case of having the whole junked car to
> work with, so the EFI unit and other computers would be available.
> (Obviously it has to have a reduction drive; I'm concerned with the
> fuel/ignition issues.)

If you want to use the stock EFI system you are going to have to fake a
few things. The stock computers are going to expect some things that you
probably aren't going to want to run on your airplane. The biggest of
those will probably be O2 sensors and an EGR system. The missing EGR
will cause the stock computer to generate some error codes, but that
doesn't matter. That's just to tell the tech what's wrong when he puts
the scanner on it during diagnosis. The intake charge will be denser
without the EGR, but the knock sensors will probably take care of that
for you.

The lack of the O2 sensors is another matter. This will not only cause
error codes, but they do actually modify the mixture as you drive, and
having them missing will do a number of different things, depending on
which computer you are using. The latest of GM computers copes with this
fairly well with it defaulting to the base lookup table (open loop
operation), if there is no or eronious O2 sensor input for a long enough
period of time. They typically run open loop at high power settings
anyway.

You should probably retain the knock sensors. Most of the newer systems
are setup to rely on them fairly heavily, and yes, they do work.

The biggest obstacle will be redundancy. There is really no way around
it with the stock boxes other than to just run 2 of them. This is a pain
because you are going to be firing a common set of injectors and a
common set of coils. I do it by having a second set of all of the input
sensors (map, tp, iat, ct, cmp, etc.), and then a simple diode interface
on the output side feeding the injectors and coils. Then the only switch
you have is to power EITHER of the boxes, like an SPDT with no center
position. The diode interface can cause you some grief sometimes with
the coil drivers, and can effect the timing as well. However, you can
lie to the computer by changing it's reference as to the cam position
sensor. If you wanted to get more sophisticated with the output
interface, you can add your own little driver box that uses the output
signals from the computers as just that, a signal only. Then your
interface is the actual driver for the coils and injectors. It's a
pretty simple circuit using one chip per output, and one transistor or
HGTP, depending on what your driving.

If it gets really bad, and you can't get the stock computers to do what
you want, you can always go to programmable aftermarket computers. If
you don't want the hassle of 2 computers, you could choose to run only
one.

> Secondly, has anyone considered the 3.4L V-8 that Ford's putting in the
> late-model Taurus?  It's a nice small V-8 (fits in the Taurus sideways) and
> looks fairly light (should be anyway; it's half plastic).   It's rated at
> 235 hp @  6100 rpm.  If you derated it to say 4000 or 4500 rpm would you
> see 180 hp?

Might get that at 4500, I haven't seen the dyno results for it. If you
choose to use one of these, or the 3.9L version, I'll build (and sell)
you a drive for it, if you send me the engine. I have been interested in
doing something with this series of engines, but I can't afford to go
out on a limb and spec build drives and engines for every new one that
the big three come out with. Initial research of mine would lead me to
believe that this all aluminum series of engines from Ford could be a
good solution. I haven't been able to get decent price and availability
out of Ford for buying these engines on a pallet to offer as a regular
'catalog' item. But don't let that stop anyone from getting low milage
ones out of the yards to use. If I have the engine here, I can make the
pieces required for mounting my PSRU's to it.

> It has a 10:1 compression ratio so I'm sure it burns high
> octane; could that be lowered easily without running afoul of the EFI
> system?

10:1 sounds pretty good to me. Efficient, good power, and the engine is
designed to be that compression and still burn that swamp water they
sell at the pump these days. Why change it? If there was real fuel out
there still, and 100ll is pretty real, I'm sure it could use a higher
compression ratio yet. The more you squeeze it before you light it, the
more bang you get back out of your fuel that you have to buy and carry.

Above all, don't let the "no'ers" get to you. Do your homework, and run
what you are comfortable with. Hell, they've been telling me for years
now that none of this will ever work. It can't be done. It'll never hold
up. Now that I'm doing it, and helping others to do it, those doofasses
have just kind of melted back into the armchairs that they came from.
I'm sure they'll pop up again every time there is an accident involving
something non-conventional, but their voices are becoming fainter and
fainter all the time.

With all that said, my last tidbit of advice would be:
Make sure you really want to go this route before you start. It's not
for everybody. If you do want to go this way, be prepared to do your
homework and plenty of problem solving.

-j- (I guess you could apply that advice to pretty much any apect of
building your own aircraft)

From: Johnny <nwaero@northwest-aero.com>
Newsgroups: rec.aviation.homebuilt
Subject: Re: Prop doubles as flywheel?
Date: Wed, 21 Oct 1998 00:33:43 -0700

RA247 wrote:

> I can certainly see how you could get rid of the heavy flywheel in a direct
> drive setup.  However, could you do the same with a belted or chain reduction
> drive?  I would guess the intense tortional vibration would be devestating on
> the chain.  Maybe the belt would be more tolerant.  Comments?

[very large can of worms opened]

In a typical auto conversion using a belt reduction there are actually 5
flywheels and 2 elastomeric and/or viscous fluid harmonic dampers. No
wonder they're so smooth.

Using my own company's drive configuration as an example, and starting
from the accessory end of the crankshaft, the major players in spinning
mass (flywheels) are:

1) Harmonic damper or balancer.
This little jewel is around 7 inches in diameter, at least 4 pounds,
with the bulk of the mass being at the outer diameter. It uses either an
elastomer or a viscous fluid to turn the "ringing" caused by crankshaft
flex/whip into heat. It doubles as a flywheel on that end of the
crankshaft. In some configurations it is also used as a source of
external crankshaft balance weight.
-- Required, don't leave home with out it --

2) Flywheel / Ring gear.
The 'real' flywheel. My aluminum versions are around 4 pounds, and
depending on the engine, 12-14 inches in diameter. Again, the bulk of
the mass is in the outer portion to capitalize on the weight carried.
Steel ring gear is most outboard station. Acting like any flywheel, this
mass smooths out the firing pulses present in the crankshaft by
absorbing and releasing energy. It has to be present anyway to provide a
ring gear for the starter. In some engines it also provides an external
crankshaft balance weight. Direct drive aircraft engines with electric
start carry one of these too.
-- Your not going to hand prop a reduction powerplant, might as well
capitalize on the mass that supports the ring gear --

3) Spacer / Lower sprocket / Bearing support / Front roller bearing.
While not really considered a flywheel, the combined mass of these
components directly attached to the crankshaft does provide some more
flywheel action that otherwsie would not be there. This rotating
assembly is present for other reasons, but drag racers would cringe at
the added spinning mass it provides.

4) The Belt.
This is the other elastomeric damper. It achieves this in several ways.
The surface of the belt isn't really pliable like a rubberband since
it's corded, but there is an overall 'cusioning factor' to the surface
of it when compared to something more rigid like steel. The belt isn't
really 'stretchy' because it uses Kevlar cords for structure, but it
does offer some damping in the same way that a piece of Kevlar rope
would if you strung it between you and a friend. Every time you tugged
and then let off, there is a bit of distance that has to be drawn up
before it's actually drawn tight. Also, like a chain drive, there is
some inertia that makes the belt want to be the shape of a circle when
it's running. This is where the 'slack' is between pulses. Like the
harmonic damper on the crankshaft, any energy taken out by the belt is
dumped in the form of heat loss.
-- I seriously believe it's important to have a flywheel on boths ends
of the belt --

5) Upper spocket / Propshaft.
This is, in addition to the propller, the flywheel on the other end of
the belt. On my 2:1 drive, the outer surface of the sprocket is
approximately 10 inches in diameter and 5 inches deep. Not trivial in
the flywheel department. This is about equal to the 'real' flywheel on
the crankshaft in spinning mass, but it's on the other end of the damper
(the belt) that's in between.
-- Did I mention that I seriously believe it's important to have a
flywheel on boths ends of the belt? --


6) Prop hub / Propeller.
Like with any prop type powerplant, this is the big flywheel. However,
it is in addition to all the previously mentioned flywheels, and it too
is on the other side of the damper. The reduction ratio and the damper
in between makes a bunch of difference on how both sides see the world.
8 (or 6) power pulses per prop revolution means the prop doesn't see
near the accel/decel that it does in a direct drive 2 bangs per turn
situation. So in power supply terms, the prop is still a big 'filter
capacitor', but with a rectifier, pre-filter, and filter choke before
it.

As for how the energy gets from the combustion process to the prop,
think of it as a dead blow hammer with neoprene inserts.

Hmmm, Molt's dry coupler with a Fiat's elastomeric joint??? Well, maybe
my analogy was a littel too literal. ;)

-j-

From: Johnny <nwaero@northwest-aero.com>
Newsgroups: rec.aviation.homebuilt
Subject: Re: 13B Engine and Belt vs. Gear
Date: Wed, 21 Oct 1998 23:45:26 -0700

George Dennis wrote:
>Instead of using this exact ratio
> many engineers would use a driving gear of 97 teeth instead of 96, and a
> driven gear of 24 teeth making a ratio of 4.04166...  Now as the number
> of teeth do not have a common divisor, each tooth of one gear will mesh
> with all of the mating teeth one after the other, instead of meshing
> with the same teeth continually.  The use of the "hunting tooth" is
> based on the theory that the wear is so distributed that all of the
> teeth will eventually be worn to some indefinite, but comparatively true
> shape. This method of gear design is used in industry such as driving
> conveyor belts etc. where an exact ratio is not critical.
>
> An automotive engine crankshaft to camshaft ratio must have an exact
> ratio of 2:1, where the camshaft runs at one half crankshaft speed. (a
> "hunting tooth" design is not feasible when exact ratios are required).

Hey George, you forgot about the idler gear in between. It can have any
number of teeth that's convenient for that pitch.

> For simplicity picture one tooth on the belt moving in rectilinear
> motion and approaching a tooth on the pulley. It impact engages the
> pulley tooth and changes direction to a momentary radial motion thus
> causing vibration. Now parallelism also holds true for belt pulleys.
> Note that belt pulleys, for the most part have flanges on both sides.
> The flanges are to keep the belt from sliding off the pulley. You
> mentioned that the dynamics of even teeth on a belt drive lap fits the
> system  and reduces noise and vibration.  I would say that the belt is
> abrasive and smooths the surface of the pulley grooves, but I would
> hesitate to agree that it reduces vibration.  A lapping process removes
> material.

Actually, when the pulleys are aluminum, they have to be hard anodized
to last. The belt doesn't wear the hard coat. The belt itself has to
wear instead, but it's not visible in the 500 hour life of the belt. I
say 500 simply because I recomend replacement of the belt every 500
hours or 5 years which ever comes first. However, I do get calls that go
like this:

Customer: I'd like to buy a replacement belt for my Dave Blanton
redrive.

Me: Not a problem, I have those on the shelf.

Customer: Thank you... you're simply awesome.

Me: I know. BTW, how many hours do you have on the current belt?

Customer: It's the original belt. It has over 1000 hours on it now. It
still looks brand new, but I figured while I had the cowling off this
winter, I'd replace it seeing as how it's 8 years old now.

Me: Good Idea. You're simply awesome.

Customer: I know.

As a side note, the HTD profile belt is mucho smoother running than the
old Gilmer. The full radius belt tooth profile makes for a smoother and
quieter engage/disengage process with the pulley. I used to run the
Gilmer's on supercharger drives. They'd really howl nice when you get
the blower up there around 10,000 rpm.

-j-

From: Johnny <nwaero@northwest-aero.com>
Newsgroups: rec.aviation.homebuilt
Subject: Re: Auto engine application
Date: Fri, 11 Sep 1998 01:03:24 -0700

jkahn wrote:

> > > There is a fly in the ointment, a Canadian club, trying to save money by
> > > running Mogas, broke a crank from detonation. Part of their logic was
> > > that they got away with Mogas in the aviation engine. The difference was
> > > 5:1 compression in the aviation engine vs. 9:1 in the auto engine. The
> > > failure was expertly researched and detonation was the culpret.
>
> That's my club.  The poor old Auster (English airplane... looks like
> a T-cart) sits engineless in a hangar.  The engine (a Blanton
> conversion - don't get me started) resides in a shed.  The club
> directors haven't quite decided what to do with it.  In addition to
> breaking a crank, one of the plates for the cog belt had a big crack
> in it.  Needless to say we are another unimpressed Blanton customer.

This sounds like there was more going on than detonation. Usually,
extreme detonation would kill the piston(s) before it would break the
crankshaft. I'm not saying there wasn't any detonation occuring, I'm
just saying that it might be hard to say without a doubt that it was the
only cause of the broken crank.

The bigger picture gives more clues - a cracked drive plate, and the
fact that 9:1 + 92 octance + aluminum heads on a 3.8 Ford does usually
!= detonation. The pistons and rod bearings will not hide the effects of
detonation. If it was occuring in a magnatude great enough to break the
crank the pistons effected will be peened on top (if not holes broken in
them), and the rod bearings on those same cylinders will be hammered
out.

This still doesn't really explain the cracked drive plate. Of course it
could have already been cracked, just not visibly. But for the sake of
discussion, lets assume that it was fine when it was born. I just can't
see the engine having detonation to the point of breaking the
crankshaft, and the drive plate, without someone noticing it. Sure, with
short open exhaust it's not going to be something you would hear that
well, but there would have to be other detectable signs of somthing not
quite right if there were enough forces in action to bust those parts.

I am suspecting that maybe there was some harmonic action taking place
between the prop, the drive, and the crank. Let me guess... you guys had
a 2 bladed prop on there? Maybe a 2 bladed metal prop? I don't know, I'm
just guessing. It would be interesting to hear what prop was on there
though. I might like to put it on the 'avoid it' list. Or maybe get one
and give it a beating on the test stand.

In any event, the powerplant as a whole should have been smooth
throughout the entire operating range. It's possible that there was a
vibration on some order that wasn't felt in the cockpit. But from what I
have seen with belt drives, and I've seen one or two of them, if you
have an engine/prop combination that isn't compatible (and there is such
a thing), you will know it. Either by feel, or by sound, or both.

In investigating the detonation problem, the first question I would ask
is, "duh, what was your total ignition advance?" I know it's an obvious
question, but many people don't understand that you have to set the
timing by the total advance, at whatever your max operating rpm is going
to be. On the 3.8 ford, 30 degrees BTDC is plenty. The chev can use a
little more, but the gains are minimal. Another area I would poke around
in a bit would be "what was the size of the prop and the ratio of the
drive?" Here I am trying to get an idea of how loaded the engine was at
how low of an rpm. I don't build a 1.6:1 drive (that's like a hint).
It's possible that the advance curve in the dizzy was a little too fast
for a 1.6 drive and a big prop. Again, I don't know anything about the
installation, that's why I'm asking.

Yet another area of discussion would be all about the cooling system,
and also the intake air temp and routing. Both of these things can be
big players in the detonation game even when all else is 'normal'.

And finally;
How did the expert determine that detonation was the singular cause of
crankshaft failure??? How did the expert account for the broken drive
plate??? Were both breaks tested to determine age of origin??? What made
this installation different than the countless (meaning I haven't
counted them) Blanton conversions that are flying on autogas without
broken cranks or detonation problems???

Please understand that this type of inquiry is part of what I do for a
living (if you could call it that). It's not meant to be anything more
or less than that. Let's try and stick with the facts pertaining to this
failure.

-j-

From: highflyer <highflyer@alt.net>
Newsgroups: rec.aviation.homebuilt
Subject: Re: Planetary Gear Reduction - Comments?
Date: Tue, 20 Oct 1998 08:36:50 -0500

RA247 wrote:
>
> I'm wondering why you don't too much in the way of planetary gearing used in
> reduction units.  Parts are easily salvaged from automatic transmissions and
> the are relatively simple and compact.  Am I missing something?

Probably.  Gearing down an engine for aviation use is a very different
problem than gearing down an engine in a car.  The lightweight need
of the airplane generally argues against the inclusion of a heavy
flywheel that adds a lot of weight and does nothing except filter
out the pulsations in the power supplied by the crankshaft.  Since
the propellor on the airplane does the identical job at least as
effectively, most aircraft engines just go ahead and use the prop
for a flywheel.

What does this have to do with gearing?  When you set up a gear train
it is essential that the gears be always loaded, preferably in the
same direction.  Even in a manual transmission this requirement is
met, even for reverse gear.  With an aircraft engine, the load on
the propellor varies and can reverse easily and often.  At the same
time, the strongly varying torque from the crankshaft encourages the
gear train to unload and pick the load up again, sometimes many times
per second.  The release and reload causes extreme wear on the face
of the gear teeth along the "line of contact" of the tooth.  The
result, is a grossly shortened gear life expectancy.

In spite of the fact that aircraft gear reduction systems are way
overdesigned to keep the loads down well below the loadings on the
gear trains used in automobiles, it is often expected to see service
lifetimes of around six hundred hours for an aircraft gearbox.  To
put this in perspective the gears in the transmission of a 300
horse power Corvette are about three quarters of an inch wide to
transmit three hundred horsepower.  The transmissions rarely wear
out.  The gears used in the GO-300 Continental, a 175 horsepower
engine are about two inches wide and last about six hundred hours.

By contrast, a good cogged belt reduction unit wastes very little
power and the flexibility of the belt absorbs the variations in
engine torque with only a moderate temperature rise, easily dealt
with by allowing some airflow over the belt and the reduction unit.

hf

From: highflyer <highflyer@alt.net>
Newsgroups: rec.aviation.homebuilt
Subject: Re: Crazy Idea #6 (Chevy-Apache)
Date: Wed, 16 Sep 1998 09:38:40 -0500

Dan Nafe wrote:

> Anybody have ideas on where full-feathering props could be procured that
> will work with a PSRU?

It is not so much a question of finding a prop which will work with
a PSRU, as it is finding a PSRU which provides the appropriate
plumbing connections for the props.  Generally, you need drilling
in the prop shaft that will allow the hydraulic pressure to be
applied to the prop for those purposes.  Most prop governors do
incorporate an oil pump to get the oil pressures into the right
range, and the governor could me mounted in the prop shaft housing
in such a way as to run at prop RPM rather than engine RPM.  That
would allow you to use a standard prop governor.  You could feed
the governor supply from either engine oil, like is done with the
aircraft engines, or from a separate oil source reserved for
propellor control.

Get a good textbook on Hydraulic props and prop controls.  It should
help you see what you have to do to modify the PSRU prop shaft to
accomodate a controllable or feathering prop.