josok a crit :
>
> Gilles,
>
> With all respect to your students work, it's always a good idea to check if
> the
numbers add up. From the top of my head, with (low quality) thyristor junction
loss of 0.6 V at 20 A makes for 12 watts to dissipate, not 80 as "actually
measured". Add 10 watts for the rest of the components and we are still far
---From the "actual measured" figure. Common sense also tells me that 80 watts
dissipation
would make this item scalding hot in minutes. 20 W also is more in
line with the Rotax idea of ambient temperature. Where is the mistake?
>
>
Josok and all,
When I was working on our project some years ago, there was a lot of
hangar tales about the Rotax regulator and wiring. The only hard fact I
could rely upon was that several of my buddies, and some Europa builders
experienced failures, and that the Rotax 914 was electrically dependent.
This is why I started some investigations concerning the unit and
designed some redundancy into the wiring.
Among other things we dicovered :
- The efficiency of the unit is not great, and its thermal
characteristics are poor
- Cooling will definitely improve matters.
- The unit wont start to life unless the c wire senses some voltage, so
there are failure modes where the regulator will stall and stop
supplying current to the ship, rendering the Rotax circuit suggestion
unsuitable for the 914.
- Asking more than 12 amps continuous from the unit seems not reasonable.
The numbers have been actually measured, in a very well known and
reputable electricity lab (LEG and IUT 1 in Grenoble) with professional
equipment.
The numbers you give seem remarkably low, so maybe the unit you measured
is a different model ? Rotax may have modified his regulator, after all...
Of course, everyone is invited to check the numbers and facts, and I
would be glad to publish the results of other experiments to complement
the article.
Best regards,
--
Gilles
http://contrails.free.fr
|