3-lead LVDT?

[James Arthur]

Is there such a thing as a 3-lead LVDT?

I picked up a Starrett model 712-2 gaging head at
a swapmeet, without the electronics.

This ancient gadget measures linear displacement very
accurately and repeatably over a small range of travel.

Googling suggests it's an LVDT, but the connections seem
strange--it only seems to have three active leads.
Does that make any sense?


The connector looks like this:

3
. - .
. O .
/ \
2 . O O . 4
. O .
. 6 .
\ O O / 5
1 . .
._n_.



And the continuity measurements are:

1. ------- ---
^
2. n/c 43 ohms
V
3. ------- ---
^
4. n/c 43 ohms
V
5. ------- ---

6. shield, 2 ohms to LVDT casing


I opened the connector to confirm 2 and 4 are definitely
not connected, the cable only has 3 wires + braided
shield, the shield does connect to the 712-2 case,
and nowhere else.

Inside, a plunger moves a ferrite-looking rod
in and out of a bobbin, presumably the LVDT coils,
but I dare not peek--the tiny windings must be
micro-fine.

Are there LVDTs like this needing only three wires,
or are they possibly measuring just the differential
inductance to get displacement?

TIA,
James Arthur

 

[James Arthur]

At work we use something they call a DCDT, a DC version of an LVDT.
Evidently the electronics is built in the unit, I'm wondering if that's what
you have.

Perhaps it's like one of the GCD-SE gage heads?
http://onlinecatalog.digikey.com/WebProject.asp?BookCode=dik08flx&SectionIndex=0&PageIndex=2503#

RogerN

Yep, that's the physical look of the thing. This eBay auction has
a picture of a later sensor + electronics:

http://cgi.ebay.com/STARRETT-LAST-W...IDGE-HEAD_W0QQitemZ250363586264QQcmdZViewItem

I don't think mine's a DC gage--the cavity inside the case is too tiny,
and I read that Starrett's manual for the electronics is dated 1966,
long before SMT. That would've been sweet though.

Thanks for the info. This is for fun, for learning about LVDTs.
They're a lot cooler than I thought.

James Arthur

 

[[email protected]]

They're probably measuring differential inductance.  Put 2500Hz or so
across pins 1 and 5, and see what pin 3 does as you move the plunger in
and out.

--

Good advice, but it would be a nasty way to wire up an LVDT - you end
up with the excitation current flowing through one of your signal
leads. It's nominally 90 degrees out of phase with the voltage you
want to measure, but it wouldn't help if you were trying to measure
small displacements.

 

[James Arthur]

They're probably measuring differential inductance. Put 2500Hz or so
across pins 1 and 5, and see what pin 3 does as you move the plunger in
and out.

Thanks for the thoughts Tim. I suspect you're right, and I'm off
to my lab to figure it out. I'll report back.

This sure is a neat gadget...

Cheers,
James Arthur

 

[James Arthur]

They're probably measuring differential inductance. Put 2500Hz or so
across pins 1 and 5, and see what pin 3 does as you move the plunger in
and out.

Hmmm. It's still a little unclear what all's inside there,
but I measure 1.5nF from pin 3 to pin 6. It looks like
pins 1 and 5 take the drive, with output a.c. coupled on
pin 3.

Doing that, anyway, with a 1v @ 5kHz sine gives a sinewave on
pin 3 that shifts first in amplitude, then phase, then
amplitude again as the ferrite core plunges into and through
the winding bobbin.

I'm doing this manually. The ferrite needs gluing back onto
the plunger mechanism, so I can't tell which range of motion
they're using, amplitude-varying, or phase. I think phase,
but either mode should be a snap to demodulate.


Cheers & thanks all,
James Arthur