February8, 1967

Dear
Enclosed is a short Progress Repoft of our wor
k
to date pursuant to-Very truly yours,

AMk
ol

Agri.
I January to 31 January 1967
PROGRIESS REPOR'R
r. PREZOELECTRIC CRYSTALS
Several PZT 5-B series connected ceramic birnorphs
have been
obtained from Clevite. A test rig was built to a-Uow e
lectrical and
mechanical measurements on these cer

amic crystals.

The test rig

allo%vs a thin rectalngular crystal to be mounted canti
lever fashion;
i.e., clamped at one end with the rest of the element s
uspended in air.
A stop was machined to allow the free far end of the cr
ystal to be dep.-essed only .020". Foz the 1. 7" by . 7" crystal whi
ch has been
slubjected to test this amount of deformation re-)resen
ts a strain of
about 2 x 10 -4 which is 209,lo of the e' ec@ted fiactu
re strain and sufficieat
XP
to produce a peak voltage transient of 13 volts. The
experimentation with
this crystal has bten c2oncerned vn"th estabushing
the transient or rin"
characteristics. Electrical voltage output is monitore
d on an oscilloscope.
and the crystal is set in (decaying) oscillation.by dep
ressing the end to
the stc-p with a piece of plastic and then slidin,c the
plastic off the end of
the crystal. Oscillation at 150 cps occurs with an app
roximately ecponential- amplitude Zecay at a tir@e constant of 83 mill
iseconds (with a 10
meg electrical load). This ti@ae constant can be inte
rpreted as equivalent
to saying that the system has a Q of 39 with am electri
cal open circuit. As
the resistance connected to crystal is decreased the Q
is, lowered and the
2
tromechanical model

oscillation dies out faster. A lumped const,--nt elec

of the crystal was used to de%,-clop expressions for to

tal system Q and

total electrical power delivered to the load resistor a
s functions of e.\-ternal
load resistance. The optimurn resistance was thus ca
lctilated and when a
resist-or of this value is connected to the test crysta
l, Q and voltage decreas3 to values within the e.\porimental error
517,)) of the calculated
values. According to our calculations the marimum
efficiency of enerery

-2 conversion is 50@'a with a pure resistive load of

the optimum value and
the quantity of power for a 2017a strain is 20 j2 J
oule per square inch cf
crystal area for each transient mechanical excitati
on. Since the actual
load presented by an electronic packa2ge will not b
e a pure resistance
the actual conversion efficiency will be.less than
501*. Our previously
assumed value of 207* efficiency still appears a re
asonable one.
In the remaintng contract time work %vM conti
nue to establish the
desi-n requirements of a piezoelectric ener,7,y con
verter.
11.

ENIG SOURCE RESISTANCE

Our consultants have determi@ed that the most

promising electrode
materials a.@-e pla.tinum, gold, stainless steel,
and tantalum. 3 mil
diameter wires of all four metali@.,have been order
ed. The stainless
0 PT I 00,'o rRi wires hav
e been rec
2

steel and platinum (90T

eived

already
Teflon coated. When the other wires'arrive they wi

ll be Teflon coated
by a local compa@y. Ch@r animal experiinent co
nsultant is expected to
begln work immediately using the materials on hand
with laboratory rats
as the experimental 3stibjects.