Subject: Thought you might get a kick out of this (review of Klaus's system)
From: Paul
Date: 4/16/2001, 10:21 PM

Hans Conser wrote:

From: "John Smith" 


I?ve been flying with a Plasma II Capacitive Discharge Ignition system from
Klaus Savier?s Lightspeed Engineering on our Glasair for two months now
(almost 100 hours), and since there are no comments I could find in this
newsgroup searching with GOOGLE, I thought I?d write a review of it for the
newsgroup.

Here?s the short story -

PROs:  Better than a magneto (hotter spark, more spark energy, automatic
timing advance, easier starting, nothing much to break); German engineering
(looks well thought out and built); price is right ($800); Kit comes
virtually complete (you supply vac hose and tee-fitting); Hall Effect
magneto replacement unit much better and much easier to install and time
than the old crank bolt kludge; spark plug wires are the best; no EMI at
high frequencies (NAV, COM, & higher); uses automotive spark plugs (cheaper
to replace).

CONs:  German engineering <grin> (?we can do no wrong? - if it breaks, it?s
probably your fault); really bad EMI (ignition noise) at low frequencies
(ADF, LORAN, Stormscope); installation manual is a hodge-podge and doesn?t
always explain the rationale behind doing something his way; light ?off?
(instead of ?on?) to set timing is backwards from convention; there is no
?Operator?s Manual? per se; stinky replacement policy on bad parts means AOG
while waiting for parts to arrive; should have ?beeper? on firewall forward
side of unit to alert you when ignition is hot for safety reasons.  (He says
?no.?)

There - that?s the short story.  Here?s the really long story.

History

I?ve always wanted to do away with the impulse magneto - one look at how it
works, and you can see why there?s a repetitive AD on Bendix impulse mags.
(They are destined to break!)  I don?t mind servicing non-impulse mags every
500 hours - after all, they have a good track record, and have proven their
reliability over lots of years and lots of hours and lots of bizarre flight
conditions (p-static, lightning, etc.) and I intend to keep one in the
airplane.  But I sure didn?t like that impulse mag.  Besides, they make
starting more problematic when it?s cold out.  So after a few hundred hours
of playing it safe with a conventional mag, getting most of the bugs out of
our Glasair, I was on the lookout for a replacement for the impulse mag.

I came across a system that was nothing more than a 1960 style automotive
high voltage coil triggered by Hall effect sensors in a cast magneto-like
housing.  This is what?s called an ?inductive? ignition system.  It didn?t
have any fancy timing, just the promise of 50,000 volts, and was fully
shielded.  (I?ve since learned that there?s more to ignition systems than
just volts.)  I ran it for 1000 hours, mostly trouble free.  Initially, it?s
high voltage proved too much for a standard aircraft harness (burned through
to the shield), and we had to replace the harness with solid core, spiral
wound racecar wires with braid soldered to special washers on the outside
for shielding.  Once the braid was over the wires, this was a very quiet
system - absolutely no ignition noise in the ADF.

Funny thing about this system was that the burn pattern on the pistons was
always bigger on the plugs driven by the magneto instead of the plugs driven
by the inductive ignition system.  I also had the ?feeling? that the spark
from this system snuffed out about halfway through combustion.  Turns out
that it?s not only voltage you need to make a good spark under pressure, but
a lot of energy to keep that spark going.  (Mine wasn?t.)  An inductive
system, due to physical limits of storing energy in an inductor, can?t
supply the energy needed.  But a capacitor can.  When a seal started leaking
oil on the inductive system, I started looking for a replacement ignition
system.

Back to the Future

I?ve always been partial to Klaus Savier?s Lightspeed Ignition system -
perhaps because he was one of the first to pioneer such a thing and write
about it in the literature.  (The fact that he wins a lot of races
reinforces the feeling that he might know what he?s doing.)  I?ve followed
the development of his system from back in 1990, when he first wrote about a
manually controlled (timing-wise) system he built.  I wrote him then
pointing out the obvious that pilot controlled timing was sure to cause an
accident (ever forget to turn carb heat off during a go-around or forget to
richen the mixture?), and that microprocessors were too finicky to use in
critical aircraft systems.  The result came a year later, when I received a
brochure with a note on it for the first Lightspeed system featuring
automatic timing using discrete components instead of a uP.

I didn?t jump then because we were still building, and I wanted to hear from
others about how ignition noise affected a Stormscope, how reliable the unit
was, etc.  Also, the unit back then required you to kludge up a crank bolt
trigger system, which I didn?t like.

Fast Forward

Now, 11 years later, Klaus? system has matured a lot.  He?s come up with a
timing pick off unit that replaces your magneto, which makes installation
and the setting of timing a no-brainer - almost.  [Aside - it?d be really
cool, if, through the use of Surface Mount Technology, the CDI unit could be
made to fit entirely within the Hall Effect unit!]  There is a minor nit.
There?s a green light on the back of the timing unit that you use to set
timing.  (Not sure if it?s LED or incandescent.  It looks incandescent.  If
so, not sure how long it?s going to last.)  The strange thing is that for
the Plasma I ignition system (the fancy, expensive, multi-spark unit), the
timing is set when the light comes on, just like with a buzz-box for
magnetos.  However, for the Plasma II (the less expensive, single spark
system), the timing is set when the light just goes off.  The latter isn?t
an ergonomic design, and is going to confuse a lot of people unless they
read the instructions carefully.  (And what about those unfamiliar with the
unit doing maintenance who just assume?)

Then I read a KITPLANES article about a guy who installed a dual system on
his DragonFly, and that got me going.  I started searching for comments from
you all on the Plasma II, but found none.  For some reason, everybody I?ve
talk to has chosen the Jeff Rose (Electroair) system.  Not sure why that is
(I think it was price - Rose?s system was cheaper until recently.  I?ll
start asking to be sure.), but I never looked into Rose?s system because 1)
I talked to him at OSH years ago and asked about compatibility between his
system and Stormscope.  He was quite frank and told me that if you had a
Stormscope, there?d be too much noise from his system.  2) I searched and
searched the web but couldn?t find a web site for his product (there is an
indirect page now www.fly-gbi.com/eis.htm)  3)  I?m not comfortable with
microprocessors in unhardened boxes - we get some weird p-static build up in
IMC in winter in snow (draw 5 kV sparks off the windshield) and I don?t need
a uP hiccuping in that environment (Klaus says a Glasair with his system
took a lightning strike and kept working) and finally 4) I know a few people
who installed an Electroair system.   One blew a piston in his Eze and had
to make a forced landing.  (A vacuum line broke or the manifold pressure
sensor failed and Rose?s system went to full advance instead of full retard.
Bad design, which has since been fixed.)  Others complained of cheapo things
like using thumbtacks to terminate the spark plug wires when using aircraft
spark plugs.  [This just in:  Last week I met a Glasair owner who told me he
?s had 5 of Jeff?s Hall Effect sensors fail on his aircraft - one during a
Reno air race.]  So I just didn?t have that ?warm fuzzy? feeling about the
Electroair system.  I called Klaus, who does sound very knowledgeable, spoke
with him a few times, faxed a bunch of questions (got some answers - more on
that later), and e-mailed him some more before buying.

Missing Specs

This is probably par for the course for the homebuilt market, but there are
a few important specifications missing on these units, like operating
temperature range, max altitude, humidity, etc. etc.  Well, these aren?t TSO
?d units (and don?t cost as much as TSO?d units).  Klaus had some anecdotal
numbers that seemed satisfactory, and has a lot of units flying, so I
decided to take the plunge.

Installation

The unit arrived within two days of my order.  It seems well built, using
quality components and lots of shielded interconnects and just about
everything you need is supplied.  (You have to supply vacuum hose and a tee
fitting for plumbing manifold pressure to the CDI box.)  The spark plug
wires are MSD brand 8.5 mm solid core spiral wound wires, and as I later had
to pull the boots off to install a braided shield over them, I had
opportunity to see how they were assembled.  Klaus uses a special MSD
crimping tool - these puppies aren?t going anywhere.  Note that the wires
are pre-assembled, and pretty much relegate you to a top plug installation,
although I suppose if you specify it, he?ll give you enough wire to do a
bottom installation.  (You cut to fit, and then send the wires back to him
for crimping.)  I would have favored a bottom plug installation because the
bottom plugs are the ones that tend to foul, but Klaus says the engine
responds better to timing changes on the top plugs.  The installation
certainly is cleaner using the top plugs only (on a Lyc O-320), and since
the spark plug wires are shorter, ignition noise will be less - an important
consideration.

Here?s an odd contrast: Even though the parts are well built, the manual, if
you can call it that (3 pages plus a note in the spark plug package, plus
whatever you have if you downloaded a manual from his web site (same, but
different)), it somewhat lacking.  The instructions are kind of scattered,
and not as clear as I like.  Also, there are a lot of ?orders? in the manual
(?Don?t do this . . .?) that would be easier to obey if the rationale behind
the order was explained in one sentence.  This is especially true for
homebuilders, since we can?t seem to resist tinkering with stuff.

German Engineering

Here?s a funny story that?s stereotypical of Germans (which is why it?s
funny) - the Plasma manual says that it?s important that you thoroughly
understand how the unit works.  One thing I didn?t understand is why the
unit can retard timing to 15 degrees BDC.  (Like, does it retard to 15
degrees on takeoff?  If so, then I don?t need to have the mixture set at
full rich for SL takeoff.)  So I wrote, ?Klaus, I don?t understand why the
timing retards to 15 degrees.?  The answer back?  ?That?s okay, I do.?
<laugh>

Instead of answering my question, he referred me to a classic textbook on
Internal Combustion Engines.  I ended up spending $100 on two MIT textbooks
on engines to educate myself on timing curves and detonation.  An
interesting education, and I?m glad I bought the books, but I would have
liked a direct answer to my question anyway.

Another stereotypical German trait is their self-confidence.  It can be
comforting, but also can be a problem.  (Glad I didn?t buy an MT prop.)
This is also reflected throughout the manual.  Over and over again it says
that if there?s a failure anywhere, it?s probably your fault.  (Bad wiring,
bad connection, etc.)   I hope I never have to make a warranty claim.
Whoops!  Too late!

Bad Part

This is unfortunate: Already one of the connectors on one of the coils has
failed.  (Separated.  Appears the connector wasn?t swaged on properly at the
manufacturer in - where? - Germany!  Not Klaus? fault.)  It?s more of a
nuisance, since it only manifests itself when I pull a spark plug wire off.
I took a picture of it, and sent it to Klaus.  I?m to send the coil back for
?failure analysis? and then he?ll send me a new one.  That means the plane
will be down for a week.  I don?t understand why he can?t send me a new one
now, and then when I get it, I send him the old one.  If the failure is my
fault, he already has a credit card number to bill us.

Results

The system works pretty much as advertised, although I don?t seem to be
getting the 10% increase in fuel economy that everyone else claims to get.
But then, those numbers are advertised on Klaus? web site for the fancier,
more expensive, Plasma I multi-spark system.  The II is only a single spark
unit.  Also, to be fair, some loss is probably due to the fact that Klaus
wants those of us running 8:7:1 compression or higher to back the timing off
by 5 degrees, so instead of running at a static 25 degrees BDC as before, I?
m supposed to be at 20.  I get more like 5% increase in fuel economy.  (It?s
a function of altitude too - the higher you go, the better your economy
improvement will be.  We generally cruise at 10,000, though, and I?d think I
?d already be at max timing up there.)

This brings up another ?flaw? in the instructions - there?s no information
about how you should fly with the new ignition system.  That is, should you
manage your mixture any differently?  Lean for takeoff or no?  Favor
different power settings over others (if C.S. equipped)?  I don?t know.
>From what I?ve read lately, it seems you lean as normal once at altitude.
You?ll notice that you can lean a little further than before, due to the
advanced timing.  Just how far you can go will be a function of your fuel
distribution system - someone with balanced GAMI fuel injection can probably
lean a lot further than we can with a carburetor.

I don?t have any way of knowing if I?m getting the advertised 4% HP
increase - again, that figure is quoted for the multi-spark system.  I do
climb better at altitude, but don?t seem to go any faster once level.  I
suspect those with a fixed pitch prop will see more benefit in the speed
department, since the extra power helps turn the prop faster, which in turn
helps to make more power.

An interesting downside of advanced timing is that it?s even harder now to
descend in the Glasair.  When you pull the throttle back to 15 inches, the
engine makes more power than it used to (or isn?t braking like it used to)
because the timing advances so much, forcing even shallower descent
profiles.  We?ve had to start our descents sooner than before, or pull the
throttle about 3 inches lower than before for the same descent which cools
the engine more than I like.

Noise

I knew this would be a problem going into this.  The spark plug noise at low
frequencies is pretty fierce.  Spiral wound wires work well at suppressing
high frequency noise, but no so at low frequencies.  Since we fly a plastic
airplane (no metal shielding) and still use ADF, LORAN and Stormscope, this
is an important issue for us.  We flew a trip from Phoenix to Florida just
after the system was installed.  While the LORAN lost lock a few times (it
usually never looses lock for that trip), the ADF rarely locked on any
beacon for the entire trip.  Even listening to Rush Limbaugh on the AM band
was difficult.  The Stormscope didn?t paint anything noise wise, but it?s
difficult to know if it?s performance is degraded by all that noise.

Since then, I?ve made a few changes that, while not eliminating the spark
plug noise, have made the noise manageable.  I changed from the automotive
plugs supplied over to a Champion model that has a ?inductive-resistor?
inside instead of just a resistor.  I bought some braid from Spruce and
installed it over the MSD spark plug wires.  (Another flaw in the manual -
the Plasma I on-line manual describes how to do the shielding.  The Plasma
II manual does not.)  I also put braided shield on the coax cables that come
from the CDI box inside the aircraft and go to the ignition coils,
essentially making them tri-ax cable.  (Despite the fine looking design, if
you take a portable AM radio, and use it to ?sniff? around the ignition
system, the CDI box really sings.  Kinda like a strobe power pack on
steroids - which I guess it is!)  I made some 5 sided copper boxes to
partially shield the coils themselves and grounded those as appropriate.

All this has helped a lot.  I can now lock up on NDB?s in the 200 KHz range
at about 25 NM.  (I used to be able to lock up at 50, without noise!)  AM
radio reception on the ADF is much better.  LORAN S/N is higher, but haven?t
done a trip to Florida yet to see whether it looses lock or not.  I?m hoping
that the Stormscope is okay, but it listens down in the 50 KHz range, so it?
s going to suffer the most.  There?s no practical easy way to completely
shield the system as much as I?d like, so as a last resort, I may move the
CDI box forward of the firewall (which is not recommended) to get all the
wiring further away from the radios, and take my chances with longevity of
the box in a hot environment.

Beep Beep!

We have toggle switches for our ignition system in the Glasair, which are
nice, but unlike a keyed ignition system, it?s easy to forget to shut off
the mags after engine shutdown.  (Especially when the Approach Plate holder
is blocking your view of the switches.)  I used to have a Sonalert beeper
mounted on the firewall for my old inductive ignition system that would beep
when the ignition was hot, because it was really easy to kill yourself by
moving the prop - either way - with that system.  (You couldn?t hear the
beeping in flight over the engine noise.)

Klaus claims you don?t need a beeper with his system, because his system
only sparks when the prop turns faster than a certain set speed.  That is,
if you turn the prop very slowly, as by hand when pushing the airplane into
your hangar, it won?t spark.

That?s true, but if you pull a spark plug wire to watch his system as you
move the prop, I think you?ll be surprised how easily you can get it to
spark.  Especially if you move the prop back and forth through TDC, where
the Hall Effect triggers.  Considering how easily the engine comes to life
once hot, I don?t see any downside to having an aural warning system on this
ignition.

Biggest Bang for the Buck

For almost twice as much money, you can buy a Plasma I multi-spark ignition
system, with all sorts of fancy features I didn?t need.  As I wasn?t sure
how bad the noise problem was going to be, I didn?t want to invest too much
money in my first Plasma system.  (In retrospect, the noise would probably
be worse with a multi-spark system.)  I don?t know how much performance
difference I?d notice with multi-spark over single-spark (if you?re racing,
you might notice), but I figured 1) even with the single spark system, I?m
doing better than with a magneto and 2) for the extra money, I could buy two
Plasma II?s and probably equal the performance of just one Plasma I.  So I
think I got the biggest bang for the buck.  I don?t know how well Jeff Rose?
s Electroair system performs compared to Klaus? (I wonder if Klaus shouldn?t
tailor his curves for various engines, as apparently Jeff allows for with
his two pot RPM curve setting?), but now you have a review of the ?other?
system.

Mike Palmer <><


The AirCraft Rotary Engine NewsLetter.              Powered by Linux.

http://home.earthlink.net/~rotaryeng/            http://www.linux.org