The attached document describes shunt regulators - they would be most
likely to fail with an overvoltage result (failing shunt).
The Ducati is not a shunt regulator - just a diode bridge with
controlled time-outs for regulation - the most likely failing mode is
undervoltage (open bridge branch).
I believe Schicke and SH are the same, with possibly more robust
implementation and better cooling.
The possible advantage of replacing thyristors with MOSFETS is
potentially lower heat because of lower voltage drop. But unless you do
something clever you need 2 MOSFETs for each thyristor and an
interesting way to switch them.
Regards,
Jan de Jong
On 1/21/2014 7:06 PM, nigel_graham@m-tecque.co.uk wrote:
>
> I have followed this discussion with great interest and with an
> increasing sense of Dj vu.
> When the issue of regulator/rectifier failures occur withing the Rotax
> community, the standard fix seems to be to simply replace like with
> like (at elevated Rotax pricing) and fly on until the next unit fails.
>
> Up until about five years ago, nearly all modern Japanese motorcycles
> used the same technology as that found on the Rotax 91x series
> engines, permanent-magnet generators controlled by SCR
> rectifier/regulators. It may come as some surprise that for an
> industry renowned for precision and reliability, nearly all of these
> major motorcycle manufactures have at some time, been plagued by
> problems with their SCR based rectifier/regulators failing and burning
> out alternators and wiring, cooking batteries and in extreme cases,
> squirting unregulated AC into the wiring loom and blowing up ECUs. It
> was just such a failure that "sparked" my interest :-(
>
> Known as "Shunt Regulators" the SCR technology runs extremely hot and
> requires more cooling air than modern styling and space allows.
> The solution was a move towards MOSFET controlled regulators and the
> numerous web-based one-make discussion groups were full of information
> and advice on the reasons for failure, the benefits of the change and
> practical advice on how to convert.
>
> It was by trawling these groups that I was able to build up a good
> understanding of how these R/Rs worked, what was causing the problem
> and what the potential solution might be. I compiled what I considered
> to be the best informed snippets of information into the attached
> document.
> The first half gives an overview of a typical design of an SCR
> Rectifier/Regulator for use on a permanent-magnet generating system
> and the second half discussed the relative merrits of SCR versus MOSFET.
>
> It would be interesting to know which technology the the after-market
> Schicker and Silent-Hektic R/Rs use.
>
> Nigel
|