[simpits-tech] Need a power supply...

[Matt Bailey]

On Fri, 4 Apr 2014 08:47:04 -0700 (PDT)
geneb <geneb at deltasoft.com> wrote:

> I need to get my hands on a 115v 400HZ, single phase power supply for
> the ADI in the F-15.  Anyone here have one for sale?
> 
> Matt, what system are you using for driving synchros right now?  The
> ADI has four synchro inputs on it.  I'm wondering if it wouldn't be
> easier to get four synchro transmitters(?) and drive them with
> stepper motors...

Not sure which Matt you are addressing, but I have been playing with
synchros, so I will chime in. :)

First, synchros don't seem to care what frequency you run them at.
400hz was selected (as I'm sure you know) because of various harmonics
and interference concerns on real aircraft systems. I have a very
coarse method right now to READ synchro/resolver position (which you'll
have to do if you drive the synchros electronically), and they respond
just about as well at frequencies from the tens of hertz all the way up
to thousands. I can read position, but I need an improved circuit that
allows waveform analysis in order to know what quadrant the
resolver/synchro is in, and to make precise measurements near the end
of each quadrant. (I am working on one, with help from an Arduino guy
who made a metal detector).

Voltage, at least for resolvers, also does not seem to be a huge
factor. My little test circuit started out switching something close to
the resolver's labeled voltage (it was incredibly low, like 2.3 volts
or something), but I can also run the Arduino's 5v logic through it no
problem. Same for resolvers with HIGHER rated voltages (a 26v AC
resolver responds just fine to 5v logic). 

In fact, resolvers don't even seem to care about the SHAPE of the wave
you are sending them. The coarse method I use now is actually driven by
a SQUARE wave produced ENTIRELY by an Arduino...! As I recall, the sine
wave produced by a function generator yielded *slightly* more precise
readings, but like I said, I still need a different circuit design
anyway to yield sufficiently precise results.

Now, most synchros are LABELED at a particular voltage and frequency,
but these are just the DESIGN parameters. The synchro or resolver may
or may not work OK in other parameters. I have not tried driving a
synchro with oddball voltages/frequencies yet, I suspect there may be a
danger of sending too MUCH voltage. Too little would probably just not
move the synchro, or move it very slowly.

If you have a synchro or two lying around extra (that are labeled
115vac), try powering them up with 115/60 house voltage! Of course, you
really need a control transmitter synchro to position the first
synchro, to see if it's actually working.

If you are really set on 400hz, I see 1-phase 400hz inverters on ebay
from time to time...obviously a static inverter is desirable, but the
old rotary inverters (like what's in my Sabreliner) come a little
cheaper, like under $200. You could get a cheap 750VA "instrument
inverter"...my Sabreliner has one of these, and I was able to run it
with a 27.2v 54A DC power supply.

As far as manually positioning control transmitter synchros, this is
definitely an option. It will be as smooth and precise as how you move
the CX. I've seen it done only with servos so far. Good solution, but
it does limit you to <360*. Instead of dealing with steppers (which can
"stutter" and look not-smooth, and require some sort of
position-sensing additions anyway), you might want to consider just a
plain old DC motor geared to the CX, and a Hall sensor geared or
connected directly (1:1) to the CX. Obviously, the Hall would be setup
for angular sensing as opposed to linear. I have such a sensor, it's on
the back burner...but it does allow position sensing over 360*,
according to its documentation. I sorta set it aside once I got
seriously into reading and driving synchros/resolvers. I will probably
end up using it for some of my non-synchro real instruments that have
motors in them.

> I've also gotten the SFS wire out DONE!  Check it out here: 
> http://www.f15sim.com/?p=315

NICE stuff, it hardly looks like a reproduction part! I really wish I
had some milling equipment here, or at least a good 3D printer...I
could make my old sim some nice bezels and controls. :)

-- 
Matt Bailey

Keeping It Real:
Sabreliner 60 Flight Simulator
Serial # 306-61 - N1JX
http://sabrelinersim.com