Finn wrote:
If I understood it right, Tracy's explanation is that the variable
venturi
formed by the slider does not generate the same high pressure drop
that "normal" carbs do.
Bob Moore wrote:
I think that there is an unfortunately poor choice of words in the above
paragraph.
I too use a "slider" fuel metering device on my 1/2 VW engine. Mine is a
POSA
device, hard to call it a "carb" because it has no float bowl or "venturi".
It does
make use of a "variable inlet", but it is not a "venturi" and therefore
there is no
pressure drop as there would be in a venturi and as a result, no air
temperature
drop leading to ice formation and no throttle plate structure for the ice to
form on.
Yes, the induction pipe downstream from the POSA does get cold from the
vaporization of the fuel, but by this time, it is too close to the hot head
to form ice,
and it certainly isn't going to form on the intake valve! :-)
Can anybody point me to an authoritative source (other than Tracy)
that explains why constant velocity carbs are not subject to icing?
I'm certainly no authority on engines or carbs, but I've never had an icing
problem. Constant velocity means no pressure drop, no pressure drop means
no air temperature drop and no moisture condensed out, therefore, no ice.
Bob Moore
Jeff Spitzer wrote:
If there was a section on non-experts on carbs in the dictionary it would
have my picture next to it. That said, I believe that "constant velocity"
carbs have a pressure drop and velocity increase that is constant for all
airflows, hence the name. If there were no pressure drop and no venturi
then there would be no way to meter the fuel. My basis for the statement
is the Bing carbs on the Rotax 912/914. They are constant velocity carbs
that DO have a pressure drop.
Jeff Spitzer
BTW Jeff got these to work up at 57,000 feet. I think. No small
accomplishment. Unless of course he switched to EFI :-)
If I knew that for sure Jeff would have to kill me :-)
Paul Lamar
The Aircraft Rotary Engine Newsletter. Powered by Linux.
http://home.earthlink.net/~rotaryeng/ http://www.linux.org