Aden wrote:

I just uploaded details of the supplier, and then saw this and another
message from Mr Tracy Crook (may I call you Tracy?). A couple of

points:-

The transformer used is rather critical in terms of air gap (reluctance)

for

efficiency, and the 1uF capacitors must be of the type specified for
longevity. The ICs and IRF822 FETs are from International rectifier and

are

a bit
difficult to source here in Oz. A data sheet for the IC is available off

the

net.
The FETs are not critical, but must be of the same type with integral

clamp

diodes
- Siemens (Philips) BUZ42 work fine, MTP3055s are available everywhere
(Motorola).

I also note Tracy's concerns about the validity of using a substitute

for

the Mazda ignition
system. I know nothing about the stock Mazda ignition, but from what has
been written,
I infer that it is non CDI. If this is the case, then there are distinct
advantages to be had in
using CDI - particularly for lean burn conditions. I agree with Tracy
regarding the selection
of quality coils, and point out that even though I did not find any

problems

with a new
standard coil (the oil filled, metal cased variety) commonly available
(although many times
during testing, the coil's internal spark-gap arced over), I have only

built

two of these units
with these coils.

However, my concerns lie elsewhere. i.e.: Firstly, this is a kit; and

not a

thoroughly tested
production item.  As such it requires thorough testing in its intended
environment under load
conditions as it will be a critical item in the airplane. Secondly, this

is

not a kit suitable for
someone who has not soldered before unless constructed under experienced
guidance.
Whilst the article gives good clear directions on assembly, one bad

solder

joint (dry or
peaked say) around 300VDC will likely cause failure in time. Now having

said

that, I do
think that with meticulous attention to detail, and given the level of
redundancy planned (one
unit and coil per plug) I  believe it to be a practical construction
proposition.

Its your call.

Kind regards to all.
Aden

Good synopsis of CDI circuit requirements Aden, and please do call me Tracy.
Not sure how the "Mr."  Tracy got started :-)  The information on
transformer reluctance is especially handy for anyone planning on
manufacturing these things.  I do think this design is the best I have seen
and it would be my choice if going with CDI.

The reason I am not sure it would be an improvement over the stock 86 - 91
Mazda 13B ignition is the results I saw when doing my lean burn tests.   The
engine did not go into lean stumble until long after the best BSFC point had
already passed.  The nature of the lean stumble is also a telling point.
When the ignition is marginal under lean conditions, the engine will indeed
"stumble".  In my case, the engine didn't do this but just suddenly stopped
as if the engine had been switched off when the mixture passed the minimum
fuel-air ratio needed to sustain combustion.  Now, it is possible that an
even better ignition system could have kept it running at even leaner
mixtures but it would seem to be a moot point since the power had already
fallen off by around 40% and BSFC was worse than at richer mixtures.  Could
the BSFC have been improved in the range where my engine had already passed
the optimum?  Possible, but I wouldn't have guessed so because the engine
was still running dead smooth but I'd love to be wrong.  I hope some of you
guys will try it and let us know.

Tracy Crook
rws@altavista.net
www.rotaryaviation.com

 
The AirCraft Rotary Engine NewsLetter.              Powered by Linux.

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