[MRCG] 800Hz AC Power for Navy Radios

[Scott Johnson]

Mark-
Most of the noise comes from magnetostriction in the stator.  I have tried
to quiet those things dowN, but to no avail.
Even the 700 VA solid state inverter I modded made a lot of noise
(laminations in the transformer), so will the transformers in the rig. (The
TBW was very noisy)
It's pretty clear why 400 Hz was chosen over 800 by the end of the war!
The big problem is that the lamination alloy that has lower loss at elevated
frequencies also exhibits more magnetostrictive effect, to the point 
That it becomes a pretty effective acoustic transducer.  


Scott W7SVJ
-----Original Message-----
From: mrcg-bounces at mailman.qth.net [mailto:mrcg-bounces at mailman.qth.net] On
Behalf Of Mark J. Blair
Sent: Sunday, May 26, 2013 2:30 PM
To: West Coast Military Radio Collectors Group
Subject: Re: [MRCG] 800Hz AC Power for Navy Radios


On May 23, 2013, at 12:42 , Jay Coward <jcoward5452 at aol.com> wrote:
> I'm not too familiar with s/s inverters but I would think there would be a
reference oscillator/frequency divider/ PLL that gets you accurate 60  cy.
How does the 60 cy of the inverter compare to the line  60 cy? I would
presume the inverter manufacturer would attempt an accurate output
frequency. Also, the output filter would not be a good filter if it passes
harmonics of 60 cy so you may need to look into that.

I had some time to explore the circuit some more yesterday. The 60 Hz comes
from a 30.72 kHz crystal oscillator which is divided by 512 to make a 60 Hz
square wave, then filtered to a sine wave with an R-C + op-amp filter. I
should be able to convert it to 800 Hz by replacing the crystal with a 25.6
kHz one, and changing the R-C filtering components.

I swept out the AC output filter yesterday, too. Th 3dB corner is just above
1 kHz. Loss at 800 Hz is around 1.2 dB, which translates to way too much
power dissipated in the filter at maximum load. So, I'll need to move the
corner frequency up. It looks like it would work as-is up to around 200-250
Hz. I think the easiest way to change it might be by removing some turns
from the big toroidal inductor in the output filter. The inductor is glooped
to the PCB with RTV, but I think I should be able to cut turns out of the
middle of the coil without dismounting it from the board, then splice the
wire back together.

The output filter doesn't need to cut out 60 Hz harmonics in this design; it
just needs to cut out the PWM switching frequency (up around 25 kHz or so)
and its harmonics. Thus, its corner frequency can be well above the AC
output frequency. The 60 Hz harmonics are filtered out at the point where
the low-voltage 60 Hz sine wave reference is generated.

I still haven't tested the unit for bad EMI radiation under full load.


How available are the original 800-1 rotary inverters? Looking at a picture
of one, it seems like it should be possible to quiet one down a lot with a
redesigned end housing at the blower end. The existing design looks just
like a mechanical siren.


-- 
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/

______________________________________________________________
MRCG mailing list
Home: http://mailman.qth.net/mailman/listinfo/mrcg
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:MRCG at mailman.qth.net

Website: http://www.mrcgwest.org

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html