Really awesome TPS...

[Donald Whisnant]

Timothy Coste wrote:

>   I'm not sure if these are exactly what you're talking about, but
>from a C&H Sales Company (surplus and misc sales) catalog...
>
>   Bearing Module #blah This unit contains...the following:
>   one resolver #CSHblah, one resolver #CVHblah, one differential,
>   one motor tachometer...and many high precision and anti-backlash
>   gears, precision bearings, standoffs, etc. Dim.... Price: $15 US.
>
>   Obviously (I think) a mil. spec. fire control or such unit.  They
>also have a Range Module with similar stuff.  If these are what you
>speak of, and if they are durable enough to most likely still be
>in good shape, it may be a way to try your idea on the cheap. 
> 
>   Your post jogged a little something loose in my brain (as if I
>can spare any little things being jarred from my brain) and I
>remembered reading about these things in this catalog and wondering
>what the heck they were (and might be good for).  In the true
>spirit of diy, I really get into surplus catalogs.  
>
>   Just FIY...
>
>   Tim Coste
>   tlcoste at mtu.edu
>

:)  .. This sounds similar to what I was talking about.  I presently don't
have a surplus catalog from them (and yes I get into surplus cats too!) so
I can't look it up and see...  ...  Resolvers are really very simple devices
with the price reflecting the accuracy with which they were wound.  ...
Basically they are two windings (well, actually three windings) and
are set up kinda like a motor.  The "primary" winding is excited by the
driver/receiver circuit to a known frequency AND phasing.  As the device's
shaft rotates, the coupling between the "primary" and "secondary" vary by
the sine/cosine of the angle of incidence.  The driver/receiver circuit 
decodes this and outputs a voltage proportional to angular movement.

Another neat device to play with is a Synchro (basically a resolver with only
one output set -- instead of sine and cosine).  You can actually hook two
synchros together and the movement performed on one will be replicated on
the other (i.e. synchro movement)...  This is very neat to play with in things
like model planes, etc...  

Also (in addition to the resolver) you will need the driver/receiver circuit.
Analog Devices makes several of these -- some even with digital outputs to
hook directly to a cpu...  However, most of the Analog Devices chips are
expensive (especially the mil spec ones).  I think they are in the 
Data Converter Reference Manuals (there are 2 of them) and the Military/
Aerospace Reference Manual.

Someone on this forum (I don't remember who) mentioned an LVDT (Linear
Variable Differential Transformer) --  Those are basically a resolver that
is stretched out over a linear distance (actually they work by voltage
coupling coming out of one winding pair and going into another) ...  Those
would work equally as well EXCEPT you have to convert radial motion into
linear motion -- where as the resolver is already radial... ... 

As far as experimenting with these things, we have all 3 of these (resolvers,
synchros, and lvdts) where I work .... I've just never tried putting them
on as a TPS...  ...  For sources, since the accuracy of a tps doesn't need to
be that of a missile guidance system, surplus markets would be good -- but
possibly have slightly flawed windings ...  For those of you seeking missile
quality units, check the Thomas Register -- I know of a pretty good place in
Charleston, SC that makes them for mil applications...

Donald Whisnant
dewhisna at ix.netcom.com