So the little multi band radio arrived.  And as suspected, noise across a broad spectrum.

With the system running only one coil pair I could pick up noise frequency bands as low as on the 19M band at 15.07 as loudest.  Next loudest (and louder) on the 22M band was at 13.22.  By the 25M band all noise frequencies were about the same "volume" as the noise tuned in while cycling through that band.  On the 31M band I started to get the "tuned" lamp on the radio to light around 9.32.  The "tuned" lamp then would turn on at every resonate frequency on 41M and 49M.  I can also back several feet away from the coil and keep the tuned lamp lit while needing to be right on top of the coil to even hear the noise when starting at 19M.

Nothing on FM.

But AM and LW tune in nicely.

Let me know if anyone wants any specific testing.  For me this just confirms that the virtual coil shorting this circuit provides is tossing out RF everywhere.  This caused me a lot of problems on my DC drive circuit lines.  Those problems were mitigated by keeping all input lines as short as possible.

M.

Quote from: gyulasun on July 29, 2011, 06:09:32 PM
Xeno,  you could use two reed switches in series connection to shorten the resultant ON time, of course their position should be found out by watching the scope.  Their ON times can be shifted with respect to each other, from the overlap position towards a non-overlapped one, while finding the smallest ON time just before the non-overlapped ON windows.

Gyula, thanks a lot for the tip. That sounds like a good idea.
I wonder if i could interconnect 2 Hall sensors in a way to achieve the same.
I will try to do that until i set-up the Arduino for peak detection somehow
because i wanna keep things simple. Multiple unshorts per peak seem also
to be important with this, ringing the radiant-bell hard hehe.

Quote from: xenomorphlabs on July 29, 2011, 06:16:55 PM
...
I wonder if i could interconnect 2 Hall sensors in a way to achieve the same.
...

I think you can by connecting the second Hall's negative supply pin to the output of the first Hall and the load resistor goes between the second Hall's pin output and supply positive as usual. First Hall's supply pin also goes to positive as usual.

Making two consequtive shorts could basically be done by paralleling two reeds (or Halls) instead of series connection. (of course to complicate things, you can use two series devices in the paralled setup... lol

I was able to tune up coil pair #2 to almost the exact Vout levels as coil pair #1 by adjusting the distance of the top coil to rotor.  I scoped them and found the top coil to be slightly lower Vout than the lower.

But when I add both coil pairs in parallel to the Dump Cap, the Pout/Pin ratio does not increase much at all.  I would have predicted it would have nearly doubled!  Pin did not increase much at all!  RPM also didn't drop that much!

So I have been noodling about this.  And here is a theory:  Can the RF noise being introduced by coil pair #1 be negatively affecting the output of coil pair #2 and vise versa?

If so, how did RomeroUK's device work?  Do the parallel, twisted single conductors in Litz wire have an inherent noise cancelling characteristic?

M.

Quote from: mondrasek on July 29, 2011, 07:40:09 PM
I was able to tune up coil pair #2 to almost the exact Vout levels as coil pair #1 by adjusting the distance of the top coil to rotor.  I scoped them and found the top coil to be slightly lower Vout than the lower.

But when I add both coil pairs in parallel to the Dump Cap, the Pout/Pin ratio does not increase much at all.  I would have predicted it would have nearly doubled!  Pin did not increase much at all!  RPM also didn't drop that much!

So I have been noodling about this.  And here is a theory:  Can the RF noise being introduced by coil pair #1 be negatively affecting the output of coil pair #2 and vise versa?

If so, how did RomeroUK's device work?  Do the parallel, twisted single conductors in Litz wire have an inherent noise cancelling characteristic?

M.

Have you tuned the first coil pair with bias magnets like Romero suggested before adding more coil pairs?