REFLECTOR: FW: Brian's Turbo

Wed Jan 12 19:41:46 CST 2011

Lots of good analysis and observations in your post, Chuck -- it least most
everything you say makes good sense to me!

When trying to size and finalize the turbonormalizer for our IO-550N (8.5:1
compression ratio) we tried to do some modeling along the lines of what you
describe.  Because we wanted to get back in total power what little we
thought we would be losing because of the more restricted exhaust flow and
higher exhaust back pressures, even for take-off at sea level we generally
use about 30.5" of MAP.  In our case, that requires our wastegate to be
about 40% closed for take-off, and a compressor outlet temperature already
more than ten degrees above ambient.

Once up to 15,000 ft with an OAT of ~6 deg F our model shows the compressor
outlet air temperature up to about 137 deg F, which we expect our
intercoolers to bring back down to about 51 deg F.  I believe our wastegate
gets to about 70% closed at 15,000 ft in order to maintain our WOT ~30.5"
MAP.  Closing the wastegate to about 90% still gives us 30.5" MAP at 25,000
ft.  

Our modeled 137 deg F compressor outlet temperature at 15,000 ft and WOT
seems to be inline with your estimate of up to a 45 deg F temperature rise.
Unfortunately, we lost our extra A/D instrumentation channels for
intercooler input and output temperature, so we haven't yet been able to
verify the above modeled compressor outlet air temperature gains, or the
expected efficiency of the intercooler in lowering them back down.

I'm surprised by your being able to produce 45" MAPs, unless you're saying
that occurred with your wastegate partially, or fully closed.  I've never
seen over about 37" MAP, and that was when I forgot (missed it on the T/O
checklist!) and left the wastegate fully closed for take-off, rather than
setting it at our normal 40% closed position.  Also, CHTs in the high 400's
sounds very high to me, and almost a certain recipe for detonation -- did
you run your engine for very long in that condition, and when you were
there, did you notice any signs of detonation?  It sounds like you must have
something like an 360 cu. in. displacement engine, correct?  Why doesn't
your blow-off valve limit your MAP's to something closer to 35" if you're
only operating turbonormalized?

Concerning the electronic ignition, our Lightspeed, at least, never advanced
the timing above ~22 deg's at WOT, however we have seen it closer to 25
degrees, especially at high power LOP conditions.  I'm not sure how it knows
we're LOP (if it does), because it is only monitoring RPM and MAP.

Thanks for the good info -- we appreciate being able to compare notes on
these things...

Bob Jackson

N2XF

  _____  

From: reflector-bounces at tvbf.org [mailto:reflector-bounces at tvbf.org] On
Behalf Of Chuck Harbert
Sent: Wednesday, January 12, 2011 4:37 PM
To: Velocity Aircraft Owners and Builders list
Subject: Re: REFLECTOR: FW: Brian's Turbo

Actually, it does slightly increase the detonation risk, unless you can cool
(intercooler) the incoming charge from the compression heating at higher
altitudes. Roughly, for every lb of pressure increase, you add 11 degF
temperature increase (or about 6 degF for 1" of manifold pressure). Since
the lapse rate is about -3-4 degF per 1,000', you get a net temp increase of
2-3 degF/1,000'. At 15,000', you've lost about half of sea level pressure
(or -1" of mp/1,000'), so to normalize you need to compress the incoming
(cold) air about 15", or a net temp increase of 30-45 degF, which the
intercooler should handle. 

Turbocharged engines use extra fuel to cool the incoming charge and to keep
the CHT's down, especially in climb. Lycoming and Continental also call for
reduced ignition advance (25deg  BTDC to 20 max) on a much lower compression
ratio engine (7.0-7.5 vs normal 8.5-8.7). The 10.5 CR engine is a high
performance version that can detonate much more easily than a lower CR
engine, especially we boosted. Detonation is a very bad thing which can
quickly destroy an engine.

Also, the thinner air at higher altitude does not cool the engine (nor the
incoming heated charge through the intercooler) as well as sea level air.
Finally, for aftermarket installations, you must be careful to avoid high
CHT's from inadequate waste gate sizing. If you are using a TO-6, it will
produce a lot of air volume which is controlled by the waste gate. Too
small, and you will back up (pressurize) the hot exhaust gases in the
exhaust manifold, plus you could overboost (over turbo normalized) at higher
rpms. The spec on TC's is no more that 1-2" pressure differential on the
turbine side vs. the compressor side, and just using a blow off valve (aka
PRV) to control pressure on the intake side is not a viable solution. You
need to make sure that the waste gate is large enough to not make any boost
at full throttle sea level rpm (for turbo normalization).

I recently found this out on my turbo installation when the pressure in the
exhaust manifold was 45" and the CHT's were in the high 400's above 2,500
rpm. I am signicantly increasing the size of my waste gate. Another problem
I had was a fuel flow restriction through my aftermarket Fram racing fuel
filter which I subsequently found out was rated for 12-14 GPH when I needed
over 20 GPH. I don't know what you're running for an ignition, but make sure
it is designed for referencing MAP (manifold absolute pressure), so it will
not over advance the engine. The same is true for the fuel delivery system. 

I raced supercharged drag boats in my early days, but I found out
turbocharging is much more complicated. When done right, it's awesome, but
if not, it can be very dangerous. Good luck.

Chuck H

From: Laurence Coen <mailto:lwcoen at hotmail.com>  

Sent: Wednesday, January 12, 2011 12:17 PM off valve 

To: bobj at jaxtechllc.com ; Velocity <mailto:reflector at tvbf.org>  Aircraft
Owners and Builders list 

Subject: Re: REFLECTOR: FW: Brian's Turbo

Bob,

10.5:1 compression is standard for a Franklin so turbo normalized does not
increase the detonation risk.

Larry Coen

N136LC

From: Bob Jackson <mailto:bobj at jaxtechllc.com>  (Jax Tech) 

Sent: Wednesday, January 12, 2011 8:24 AM

To: 'Velocity Aircraft <mailto:reflector at tvbf.org>  Owners and Builders
list' 

Subject: REFLECTOR: FW: Brian's Turbo

Looks like a very nice layout, Brian -- I like the compact, minimal
induction plumbing design achieved with the way you've located the
intercoolers on the bottom of the cowling near their cooling inlet airflow.
Are you concerned about having the turbo sort of 'trapped' in the relatively
dead airflow space between the engine and firewall and with transferring
it's heat out of the engine compartment efficiently?

I assume that since it's a single turbo, then there's a single wastegate
bypass type arrangement near where the exhaust from each side funnels into
the turbo, correct?  Any photos handy of the inside layout?  Are you
concerned about increased detonation risk with the high compression
cylinders?  How will you manage that risk?

Thanks for the info and photos,

Bob Jackson

N2XF

  _____  

From: Brian Michalk [mailto:michalk at awpi.com] 
Sent: Monday, January 10, 2011 12:22 AM
To: bobj at jaxtechllc.com; Velocity Aircraft Owners and Builders list
Subject: Re: REFLECTOR: Brian's Turbo

There's lots of information to address your questions.  I'll start on them,
and send them out as I get time.

1.	Basic engine design you started from 

1.	It's a stock Franklin.  6A-350-C1R, 10.5:1 compression.  I had the
idea that someday 100LL would go away, so with the turbo in there, I could
convert to low compression pistons and perhaps reprofile the cam to better
suit turbo operation. 

2.	Performance and functionality objectives for your improvements 

1.	Turbo normalize only.  I'm not looking to exceed TBO horsepower. 

3.	Single or dual turbo?  Piping layout for the turbo(s), wastegate and
exhaust pipes 

1.	Single turbo.  From memory, I believe it's a T06, Airresearch. 
2.	Intake air 

1.	armpit scoop from the pilot side 
2.	Bosch 3" throttle body 
3.	turbo situated slightly to the right side such that axle is pointing
right at the armpit scoop, and the compressed air exits center of the
firewall, pointing down 
4.	Intercoolers.  Two RJR (I think) motorcycle intercoolers welded to a
center plenum that goes to the turbo.  Air makes a right angle turn out each
side of the intercoolers towards the outside. 
5.	intake manifolds that pick up from the intercoolers near the
firewall an go the all of the cylinders aft.
The intercoolers get a fairing so it's all smoothed out.  Not shown in the
pics below.

On 01/07/2011 08:05 PM, Bob Jackson (Jax Tech) wrote: 

Brian,

Maybe during the break while you're waiting for parts you could take a
little time and post some info (and photos?) on your turbo work.  Unless of
course it's something you'd rather not share for business reasons, etc.
Here are some facets that at least I would like to know more about:

1.	Basic engine design you started from 
2.	Performance and functionality objectives for your improvements 
3.	Single or dual turbo?  Piping layout for the turbo(s), wastegate and
exhaust pipes 
4.	Inclusion of any induction over-pressure, or alternate air valves?
Do the intercoolers have special cooling air inputs? 
5.	Automated wastegate/throttle controller design concept and some
implementation details (the version you want to end up with), i.e., the
'linear stages', 'embedded controller', etc..  Does the design also involve
mixture control? 
6.	Primary electronic fuel injection system -- part of the starting
point, or something you designed?  More design detail? 
7.	Any details on the electronic ignition system, if other than
standard magneto 
8.	Overall engine cooling airflow concept and unique layout details 

As someone (Jim Agnew?) used to always say on the Reflector -- 'curious
minds need to know'!

Bob Jackson

N2XF

  _____  

From: reflector-bounces at tvbf.org [mailto:reflector-bounces at tvbf.org] On
Behalf Of Brian Michalk
Sent: Friday, January 07, 2011 5:54 PM
To: Velocity Aircraft Owners and Builders list
Subject: Re: REFLECTOR: Fixes for Stick Slop

Great questions.

I installed the turbo, and ran the engine with wastegate wide open.  The
performance was 50RPM static better than a carbureted/magneto Franklin with
identical propeller.  I am very happy with the performance.  First flight I
calculated a 20 degree climbout angle.  I was at 900 feet when the engine
failed, about 2500 feet from where I took off.  I don't know what that is in
FPM, but I got there stinking fast.  I kept pulling back on the stick to
keep from overspeeding the engine.  The plan was to increase prop once I
reached 1000 feet.

The initial configuration was a bit complex, so this second time I am
removing complexity and will add back those features later.
To control the wastegate and throttle, I built some linear stages, actuated
via an embedded controller.  This could be overriden by the pilot by using
the friction lock on the knobs.  These stages have been removed, and I'm
waiting on a new set of throttle/turbo cables from AS&S.

Since the primary fuel system is electronic injection, I fabricated from
scratch a completely redundant injection system.  It actually performed
quite well, but tuning on the ground is difficult.  I am removing the
automated actuating part of it, and am instead installing a needle valve
that is manually operated from the pilots position.
The original backup EFI is nice, because in an emergency, it monitored for a
power failure and engaged on that event, or a pushbutton.

At this time, I am gated on several things.  The cable install really needs
to happen before I mount the intercoolers, which prevents induction or cowl
work.
Wings are ready to mount, but we are waiting on primer.

I think three weeks after the parts are in that we'll be ready for an engine
start.  Optimistically, I'm saying March for another flight, but reasonably
would me more like June/July.

I checked the runout on the crank, and was very pleased to measure .0005"
TIR.  The manual says I'm allowed .008" so that's pretty good.

On 1/7/2011 4:17 PM, Bob Jackson wrote: 

Sorry, Brian!

I forgot about your first flight adventure.  Don't feel bad, almost all of
us have done something.  I had engine problems (turned out to be my own
engine control mis-management) and ran off the end of the runway when doing
simulated power off landings.  And my A/C partner collapsed the nose gear
(poor technique and a suspect weld) and slid down the runway a 1000 feet
grinding away the bottom of the nose!

I'm sure your fix would do the job, I'm just concerned about the small-area
steel-to-aluminum contact space.  But probably for no reason.

I remember you also had a pretty fancy turbo design.  How is that working
out?  When do you think you might get back into the air?  We fiddle-farted
around for ten years to complete, then another half-year off repair the hole
in the nose after the NG collapse.  Nobody likes the 'when are you going to
fly' question!

Thanks for all you do operating and maintaining the Reflector,

Bob

  _____  

From: reflector-bounces at tvbf.org [mailto:reflector-bounces at tvbf.org] On
Behalf Of Brian Michalk
Sent: Friday, January 07, 2011 4:20 PM
To: bobj at jaxtechllc.com; Velocity Aircraft Owners and Builders list
Subject: Re: REFLECTOR: Fixes for Stick Slop

I think it's safe to say I have about .05 hours of airtime on it.

That would be the first flight, and subsequent off field.  However, over
years of airplane work, it's been tight.

On 1/7/2011 12:35 PM, Bob Jackson (Jax Tech) wrote: 

Nice design and approach!  How many hours do you have on it?  

Don't you still have concerns about the effect on the softer aluminum of the
hard steel bolt?

  _____  

From: reflector-bounces at tvbf.org [mailto:reflector-bounces at tvbf.org] On
Behalf Of Brian Michalk
Sent: Friday, January 07, 2011 1:02 PM
To: Velocity Aircraft Owners and Builders list
Subject: Re: REFLECTOR: Fixes for Stick Slop

There has been little discussion on a fix.  My solution involves making a
saddle, I'm not sure if it's what the factory does or not.

I think the tubes are 1/2", so I bought some 1/2"ID, .25" wall aluminum
tube.
With about 2" of stock, I first machined a flat on opposing side, leaving
about 1/8" thickness.
Next, cut the tube in half along a plane parallel with the two flats.
Take the two halves, and place them flat to flat, drilling a hole for the
bolt.  The tubes can now be cradled in the saddle.

The flats help support the torque transmitted to the two tubes at right
angles.  For extra smoothness, use a thin teflon washer(or sheet) between
the flats.

On 1/7/2011 11:13 AM, Bob Jackson wrote: 

We've been flying for a couple of years and now have 300+ hours on the
plane.  When we started doing some formation flying recently I began to be
aware that we're developing stick slop -- it's not what you want flying
close formation, or when landing or anytime you need small corrections and
fine attitude control.

When we built the plane we were leery of the standard steel AN3 bolt that
joins the bottom of the aluminum stick stub and the aileron/elevator torque
tubes) -- even when improved with the U-shaped steel reinforcing collar that
the factory started shipping.

In our case, we know that nearly all of our stick slop comes from this
mechanical connection between the stick's torque tube and the aileron and
elevator torque tubes.  We've been on the Reflector for 12 years, but
haven't heard much, if any talk about this problem and improvements that
builders have made.

The question is:  do any of you have good fixes for this connection that
eliminates stick slop?

Thanks,

Bob Jackson

N2XF

_______________________________________________
To change your email address, visit
http://www.tvbf.org/mailman/listinfo/reflector