[simpits-tech] MarvMan and Craigo Fixed Stick Update

[Marv De Beque]

Yeah, just did a quick test of one of the coils now and it looks like it
works.  I need to build the circuit so I can properly test it with installed
in the transducer base.

All I had was a square wave generator to excite it, so the output was a set
of spikes.  Still, as I moved the core inside and out it changed amplitude.

Next weekend I should have it wired up!  Well, might get both axis wired up,
too.  ;-) 

Marv

On 12/20/02 7:07 PM, "Craig Rochester" <johncraigroch@msn.com> wrote:

> Marv,
> 
> Holy *&^%!  Winding your own LVDTs!!!  Now that's impressive!
> 
> The variable displacement for pitch-up and pitch-down sounds great!
> 
> Thanks for the update.
> 
> Craig R.
> 
> 
> 
> -----Original Message-----
> From:    simpits-tech-admin@simpits.org
[mailto:simpits-tech-admin@simpits.org]
> On Behalf Of Marv De Beque
> Sent:    Friday, December 20, 2002 5:45 PM
> To:    simpits-tech@simpits.org
> Subject:    Re: [simpits-tech] MarvMan and Craigo Fixed Stick Update
> 
> Well I have not been very impressed with the results of the load cells and
> the one version of strain gauge that I have.  I have ruled out the load cell
> as it does not have acceptable accuracy and repeatability in the null
> position (hands off the grip should return to center position).
> 
> I have ordered a different strain gauge set to test which should be in by
> mid January.  This set is a little more work to install, but supposed to be
> much more accurate than past experiments I have tried.
> 
> Over the next week I am going to test some LVDT sensors.  Actually, I am
> going to roll my own for this test.  I spent a little time this morning with
> our companies bobbin winder in the lab.  Last night I cut two bobbins out of
> nylon and spun them with about two hundred windings of 36 gauge transformer
> wire on each coil (there are three coils per sensor).
> 
> I have ordered some special interface or signal conditioning ICs that should
> arrive some time after Christmas.  I'll bread board a circuit and run the
> same series of tests on the LVDT system as the other load cells.  By then I
> should have the other strain gauge parts in to test that system.
> 
> So far I can say that the strain gauge system holds promise in that I think
> I can get it to work satisfactorily, but it will be more labor intensive to
> build than LVDT.  However, LVDT sensors are expensive.  My single piece
> price on those is $92 each!  Now you see why I am rolling my own for testing
> purposes.  The interface circuit is also a little more complex.
> 
> I completed a mechanical design enhancement that better replicates the feel
> and movement of the grip relative to the transducer base.  As you may know,
> the real side stick has different amounts of physical movement of the grip
> depending on direction it is moved.  Pushing forward has virtually no
> movement at all, but pull back and it can move 1/4"!
> 
> I have been able to replicate that movement precisely.  What's more is it is
> adjustable!  However, you need to take the base apart to get at the set
> screws that control the amount of flex.  I plan to set it to factory specs
> when it leaves my place if I use the design.  Unfortunately, it requires
> more parts and machining to do it, but we will see.
> 
> I must say that this project has taken a lot more engineering time than I
> had first budgeted.  The good news is that I have been slowly refining the
> design so it will be "right" when it is done and offer significant amount of
> realism.
> 
> The downside is probably going to manifest itself as price creep.  The worst
> case scenario is the LVDT sensor system.  That could cost an additional $150
> to $200 depending on actual costs of components.  Obviously, that is my last
> choice for a system if I can't get satisfactory results from the strain
> gauges.
> 
> Marv
> 
> 
> 
> _______________________________________________
> Simpits-tech mailing list
> Simpits-tech@simpits.org
> http://www.simpits.org/mailman/listinfo/simpits-tech
> To unsubscribe, please see the instructions at the bottom of the above page.
> Thanks!