We warrant each of our products to be free
from defects in material and workmanship. Our
obligation under this warranty is to repair or
replace any instrument or part thereof which,
within a year after shipment, proves defective
upon examination. We will pay domestic
surface freight costs.
To exercise this warranty, call your local
field representative or the Cleveland factory,
DDD 216-248-0400. You will be given assist-
ance and shipping instructions.
REPAIRS AND RECALIBRATION
Keithley Instruments maintains a complete repair service and standards laboratory in Cleveland, and has an authorized field repair facility
in Los Angeles and in all countries outside the
United States having Keithley field representatives.
To insure prompt repair or recalibration service, please contact your local field representa-
tive or the plant directly before returning the
instrument.
Estimates for repairs, normal recalibrations,
and calibrations traceable to the National Bu-
reau of Standards are available upon request.
Page 3
MODEL 515 MEGOHM BRIDGE
CONTENTS
TABLE OF CONTENTS
Section
INTRODUCTION . . . . . . . . . I-l
I.
SPECIFICATIONS . . . . . . .
II.
III. OPERATION. . . . . . . . . . . III-1
Outline of Procedure . . . 111-1
A.
Description of Controls
B.
and Terminals . . . . . , III-2
Operation Steps. . . . . . . III-3
C.
Standardization. . . . . . . III-4
D.
Connecting the Unknown
E.
Resistor. . . . . : III-4
F. External Bridge Voltage
Supply. . , , . . . . . , III-8
Voltage Across Unknown and
G.
Standard Resistors. , , , . III-8
Accuracy Considerations. . . III-9
H.
1. Null Detector
Sensitivity . . . . . . III-9
Null Detector Zero
2.
Drift , . . . . . . . . III-10
Resolution of the
3.
Readout . . . . . ; . . III-10
Accuracy of the Read-
4.
out and Standard
Resistors . . . . . . . III-10
5. Temperature and Voltage
Coefficients in the
Bridge Resistors. . . . 111-12
Leakage Resistance
6.
Across the Unknown. . . III-12
Errors Caused by Guard
7.
to Ground Resistance. . 111-12
Page
II-1
Section
8. Time Constants - Slow
Response . . . . . III-13
9. Transients Caused by
Push-to-Read Switch. III-14
Verification of the
10.
Accuracy . . . . . III-14
CIRCUIT DESCRIPTION . . . . .
IV.
A. power Supply. . . . .
B. Null Detector . . . . .
C. Bridge Circuitry. .
Overvoltage Protection. .
D.
MAINTENANCE . . .
V.
REPLACEABLE PARTS . . . . .
6.
6-l. Replaceable Parts List.
6-2. How to Order Parts. .
Model 515 Replaceable
Parts List . . . .
Model 515 Schematic
Diagram 14522D . . .
Green Calibration
and Repair Form . . . .
* Change Notice
*Yellow Change Notice sheet is included
only for instrument modifications affect-
ing the Instruction Manual.
Last page
Page
IV-1
IV-1
IV-1
IV-3
IV-4
V-l
6-l
6-l
6-l
6-2
6-10
6-11
0568R
i
Page 4
Page 5
SECTION I -
INTROlXJClTON
The Model 515 Megohm Bridge
resistors from 105 ohme to 10
1.0%.
It is complete, with an electrometer null detector, shielded
if5a Wheatstone Bridge for measuring
ohms with accuracies from 0.05% to
enclosure for the unknown, and a bridge voltage supply.
A unique system of switches is provided to allow corrections to be
made for the slow changes in resistance of the standard high megohm
resistors.
This enables all values of resistance to be read with rated
accuracy, directly from the bridge dials.
Bridge voltages in one volt steps up to 10 volts are available from
the internal supply.
With external supplies, voltages as high as
1,000 volts csn be used.
A
connector is provided so that unknown resistors can be measured
outside the instrument as well as in the built-in shielded enclosure.
Page 6
Page 7
SECTION II - sPFXXF1cATI0NS
RANGE:
ACCURACY: As tabulated below, if bridge is operated so that volt-
age *cross stendard resistor does not exceed 10 volts.
RangeOhms
105 to 10-f
10
;I
10
109 to 1010
1010 to loll
loll to loI2 0.25 1t0lOo0 1
12 to lo13
El3 to 1014
1014
For less than minimum voltage, accuracy decreases in proportion
to the ratio of applied voltage to minimum stated voltage.
INKTT:
insulated triexial cable.
GROUNDING:
lO5 to l&5 ohms with a six-dial in-line readout.
Accuracy
0.05%
to 108 0.05 1tolocn
to 109
to lo15
Built-in compartment or Remote Test Chamber with teflon-
One terminal of unknown is at ground potential.
0.10
o.l.5
0.2
0.30 1 to 1000
0.5
1.0 10 to 1000
Possible Bridge
Voltage
1tolooOV
1 to loo0
1toloco
1 to 1000
1to 1000
Min. Volt. for
Rated Accuracy
5v
50
3
2
1
1;
100
yy4DmmR
volt per diw in decade ranges. Reading is non-linear past l/3 of
full scale for ease in balancing.
ZERO CREYX Normally closed zero-check button shorts out null de-
tector input except when depressed.
BRIWEFQTERTI&z
selectable from the front panel. External:
or 241Power Supply, from zero to 100 volts on any resistance reading,
from 100 to 1000 volts provided the readout dial is at least in the
x10 position. Bridge interlocks, and the inherent overload protection
of the Keithley power supplies, prevent damage if readout dial is inadvertently placed in the wrong position. Since other types of power
supplies do not provide the correct overload protection, only the
Keithley Model 240 or 241 is recormnended.
POWER:
TIJBE AND TRANSIS!t.OR COMPLE!MER?
6-21il381.
ACCESSORIES AVAILABLE:
rubber feet.
60" triaxial cable and bridge connector; Model 5153 60" trisxial. cable
with bridge connector on one end.
amperes and sensitivity ranges of lvolt perdiv. to 1 mill-i-
100-130 or 200-260 volts, 50-60 cps. 10 watts.
Electrometer with a grid current of less than 5 x
Internal: From zero to 10 volts in one volt steps
With Keithley Model 2443
i-5886, 2-6418, 1-0~2; 1-2~1535,
Model 5151 End Frames with mounting hardware,
Model 5154 Cabinet; Model 5152 Remote Test Chamber with
515
II - 1
Page 8
DIMENSIONB:
Model
5154 Cabinet, 21" w x 25" h x 16&f a.
Model 515 bfegolpn Bridge, 19" w x 14" h x I@" d.
NE2 WEIGR!J!z Model 515
inet, 52 lbs.
Megolm
Bridge, 24 lbs. Model. 5154
Cap-
515
II - 2
Page 9
SECTION III - OPFRATION
OUTLINE OF PROCEDURE, (taken from instructions fastened to the
A.
inside of the door of the Shielded Measuring Compartment).
1. Connect power cord to 115 volts, 50/60 cps unless specified
on rear for 230 volt.
To change line voltage see Section V -
Maintenance.
Turn on power; set MILLIVOI/IS PER DIVISION switch to 1000;
2.
release HJSH 'IO READ switch. Set meter to zero with FINE ZERO.
If necessary use COARSE ZERO. Increase sensitivity and rebal-
ace. Drift which msy be apparent at maxirann sensitivity will
become negligible after a short warm-up.
STANDARM- Set FUNC'IION switch to STANDARDIZE: MULT-
3.
1PLIERdialtom;
and RESISTAEtX, OHMS dials to 10.000. Bring
to exact null with ORMS dials; at the same time increase the null
detector sensitivity to ma%-. Release HJSH 9 READ button
and set FUNCTION switch to CALIBRATE. Adjust l0 CALIBRATE
potentiometer to give a null when T II 9?3 FOND switch is oper-
ated.
Next set exponent dial t&IO and repeat step 3. Do
the same in sequence UJP thrn I.0 . This completes the bridge
standardization.
4. 0PEwJ!I0N:
Place resistor to be measured in compartment.
Locate the ground clip to suit the resistor length, and close
compartment.
Select bridge voltage.
or external voltages up to 100 volts may be used
precautions.
Above
Internal voltages from 1 to 10 volts
with no
special
I.00 volts the xl0 or xUl0 dial must not be
set at zero. With the Kelthley 240 or 241 Power Sypp3y, the
interlock circuit will prevent damage.
Set null detector sensitivity to minimum and operate PUSH M
READ button.
If miU. detector deflects to left the readout
dials (RESISTAECE, OBMS) are set below value of resistance.
Increase the indicated mISTAWCE until a null is obtained.
If null detector deflects to right reduce the indicated RE-
SISTANCE.
If no deflection is observed increase null detector
sensitivity. Final balance should be made tith enough sensltivity to give required accuracy.
For external operation, attach special cable only to input con-
nector and set FUEtX!ION switch to
door
interlock
is now inoperative observe care with high bridge
EiilWNU OPEW. Since the
voltages.
ACCURACY: As tabulated below, if bridge is operated so
5.
that voltage acrose stendard resistor &es not exceed 10 volts.
515
III -1
Page 10
Rang% ohms
Accuracy Possible Bridge
Voltage
Min. Volt. for
Rated Accuracy
I.05
to lo;
I.07
: $9
108
I.09 to 1010
lo~tolo~
lo~tolLG
lo12 to I.013
lo13 to lo*
lo14 to I.015
For less then mlnimm voltage, acouracy decreases in proportion
to the ratio of applied voltage to
B. DEWRIPl!IONOFCONTROU3AND~AIS:
BRIR3EVOUCS:
to the bridge In 1 volt steps up to I.0 volts and &so Is used
to
energize
position.
ON: Toggle switch is the main power switch. Preseme of power
is indicated by the i-ted meter dial.
MIUXVOIfE3 PER DIVJSIOIk Rotary switch provides decade step6
of nuU detector sensitivity.
0.36 1 to 1000 ;
0.5 ltolcoo
1.0 lotoKxxl
This rotary switch adjusts the voltage applied
the external supply circuit when it is in the EK!l!
1tolooov
lo to loo0
1to lGQ0
1 to loo0 2
1
to loo0
1tOlOOO
lE
minirmpm stated voltage.
5v
50
3
1
FINE ZERO: Ten-turn control18 used for sett%ng the null detector to zero.
COARBEZERO:
,&thin the range of the FINE ZERO controls. It mey be swltchee8
with a screwdriver from the fxont panel.
NUT& INDICAMR: Three-inch meter, Incorporating a non-linear
;mvement for easy bridge balancing,
'RTSH 9X) READz Push-button switch mxnNJy shorts the miJ.l de-
tector @put. It may be locked In the open position.
'FUNCTION Four position rotary stitch provides the necessary
circuit arrangements for calibration of the standards, and also
employed when the unknown resistor is outslde the instr7.mvd.
'RESIS'JANCE, OHMS: These seven di.als include five decade step
xw-itches and one rheostat which forn the variable arm
bridge.
the unknown resistance is read directly from these dials.
Below these dials is the shielded test chamber. This contains
the external input connector and 81s calibration controls in
addition to the guarded test terminal. The external input connector is a teflon insulated triexial receptacle
Eleven position rotary switch sets the meter zem
of
the
!W seventh dial is a nailtiplier switch. At b&lance,
(Gremw
5632~).
515
III - 2
Page 11
EUSE: A fuse extractor post Is located on the rear of the in-
strmnent . For ILL5 volt operation use a 3 AG, t amp. fuse; for
230 volts use a 3 b3, l/8 mqp.
POWER CORD: The three wire cord with the NlNA approved three-
prong plug provides a ground connection for the cabinet. An
adapter to allow operation from twn prong outlets is provided.
ACCESSORY OUTLET:
vided on the rear for operation of an external power supply.
It is wired directly to the powr cord and is not controlled
by the bridge power stitch.
EXTEBNAL INHTl!z UHF receptacle on the rear of the instrument,
used to connect an external power sqpply when bridge voltages
above 10 volts are desired.
OPERATION STEPS
C.
Connect power cord to ll5 volt, g/f50 cps, unless specified
1.
on resx for 230 volts. !Co change line voltage see Section V,
Maintenance.
2.
the null detector is normeSLy shorted corresponding to sn open
galvanonder key).
Set null detector to l.CW milklvnlts per cllvislon and un-
lock PUSH 'JB READ snitch so null detector input is shorted (note
!Cum power on and allow 30 seconds for warm-up.
3.
should indicate zero.
if necessary. If the detector cannot be set to zero, use the
coarse zero control.
A three-terminal convenience outlet Is pro-
Increase mill sensitivity and re-zero
The meter
4. Standardize the bridge if necessary (See D - Stsdardizaz, following).
Insert the component to be measured in its test fixture.
5.
Set the function switch to OPRRATE when using the self-contained
shielded measuring compartment, or set it to EWERNAL OPFJWCE
when the unknown is located in the Model 51.52 Ren&e Test Chamber
or in another external sample holder. See E. Connecting to the
Unknown Resistor, which follows for detailed instructions for
connecting the unknown.
6. Set the BRnaE MU16 to the desired value. For external
bridge supplies, see F. External. Bridge Voltage Supply following.
Operate the FWH M RENI button and b&Lance for null with
7.
the resistsnce dials.
(See H.Aa?uracy) to give the desired accuracy at final belsxxe.
The resistance of the comnent is then read. directly from the
resistence dlaJ.8.
Use x M to x O.OOldi
Increase the null detector sensitivity
515
III - 3
Page 12
STAEDARDIZ4TION
D.
Wire-wound resistors have the greatest accuracy and keep their calibrations over long periods of time. Values greater then about onemegohm, however, are too large and too expens
Carbon film resistors provide values up to
e to be widely used.
ei
10
ohms and higher with
reasonable SUCC~QS and this type resistor is used in the Model 515.
But then value of these resistors changes wlth time, sometimes one
or two percent per year.
The Keithley Model 515 Megohm Bridge has been designed so that fre-
quent compensations can be made for variations of its high-megohm
standard resistors. This process is called Standardization and is
carried out as given below. Section IV - Circuit Description dis-
cusses the circuitry involved.
The bridge should be restandardized following a chenge in tempera-
ture of greater than about lOoF, and at least once each week, to
compensate for the errors introduced in the carbon standards by temperature and time. For the utmost accuracy possible from the bridge,
it can be standardized. daily, hourly, or immediately before a crit-
ical measurement.
To Standardize the Bridge:
1. Set the Multi 8 lier
OfJim dials) to 10
2.
Set TXINCTION switch to STANDARDIZE
Set NCLtL DETECTOR to 1000 mv per division.
3.
4.
Operate IUSH ltl READ switch and balance the bridge as in
normal operation.
.
The reading will be close to 10.00. The
(the farthest right of the RESISTANCE,
final balance should be made with maxinnun null sensitivity.
Release BUSH To READ switch and set FUNCTION switch to
5.
CWBRATE.
6., Operate PUSH To READ switch and re-balance the bridge with
the 10 CALIBRQE potentiometer located in the Shielded Measuring Compartment.
Turn FUNCTION switch back to
7.
STANDARDIZE.
8. Turn multiplier to I.07 and repeat steps 4 thru 7. Do this
for each successive mult
is ot used since the 10
I.?
multiplier position.
lo
lier thru 1011. The 1Ou position
f?i
ohm standard is calibrated in the
E. COIVNKTING TREUEKWOWWR.E3IS'lrOR
1.~ Using Internal Test Chsmber
515
III -
4
Page 13
Fig. 1 Shielded Measuring Compartment,
With Unknown Installed
The bottom section of the bridge contains the shielded compartment for holding the unknown resistor, and is accessible when
the hinged door has been opened. The compartment has been de-
signed for greatest user convenience.
Its being shielded elim-
bates troublesome pickup, and the unit construction eliminates
the necessity for having cables running from the unknown to
the bridge, with their associated flexure noise.
The measuring compartment will accept resistors up to about
eight inches long.
banana jacks.
Connections to the bridge are made through
A convenient clip to use with the banana jack
is the readily available Grayhlll Test Clip #2,1; it has a bansna plug on the bottom snd spring clips on the top for holdin@;
the resistor heads;
three are supplied with each bridge. These
clips are illustrated in Fig. 1, holding a typical high-megohm
resistor.
A number of ground jacks have been provided so that the ground
clip can readily be placed for conveniently holding the unknown
resistor, irrespective of its length.
In measuring high
resistances,
the many precautions necessary
in electrometer techniques must be borne in mind; most important
are the need for dryness and cleanliness so that leakage resistance paths from the HI terminal to ground will not affect the
accuracy of measurement, and mounting the resistor so that its
body does not touch conductors or other insulators setting up
undesired or inadvertent leakage paths.
53.5
III - 5
Page 14
2. Unknown Resistance Ecternal to ths Bridge.
The Moe1 51.52 Remote Test Chamber shown in Fig. 2 Is used for
testing irmilation or making other external shielded measurements.: This test chamber is equipped with an integral &I-Inch
teflontinsulated trisxial cable fitted with a conuector for
attaching to the mating cqnnector in the Shielded Measuring
Compar&nentlnthebr.Mge. The chamber and connecting cable
are rated for continous operation at temperatures as Wgh as
125w.
Flg. 2.Model 5152 Remote Test Chamber
The eLectrical connections are made throngh banana jacks in
the oh&her.
are furnished to facilttwte installing unkuowns tith axisJ~lea&s.
The banana jacks of course, can be used with any other connectors
or resistor holders.
To use:the Model 5152 Remote Test Chamber, fasten its cable
connector into the mating connector located in the Measuring
Compartment in the Bridge, and connect the unknown resistor
betweeq the RI and GROUND banana jacks in the Test Chamber (us-
ing the Cks+yhil.l test clips if possibae).
The third bsnsz.% jack in the Renwbe Test Chamber Is QUARD; it
Is cdn@ected through the inner shield braid of ths triaxial
cable to the guard connection in the Bridge.
Guarding is used exbenslvely in the Bridge to reduce the e&or6
causedfby spurions Le&age cwrents. WaxIing should also be
en@.oyed in the construction of test electrodes fitted to the
RemoteITest Chamber, in order to obtain the greatest accuracy
from We bridge measurement,
The hw.yhilJ. #2-1Test Clips as shown in Fig. 2
515
III - 6
Page 15
The guard conductors are driven from the galvanometer junction
of the low tiedance standard arms of the bridge; a total re-
sistance less than 10" ohms from guard to gmund till stit
the standards sufficiently to create errors great enough to
impair the rated accuracy of the bridge. Great care has been
taken in the construction of the bridge to
GROUND resistance substantial& higher than 10
should be taken by the user to maintain that high level.
Fig. 3 is a simplified schematic diagram showin@; the electrical
connections of the standard and readout resistors, the unknown,
the null detector, and the guarding. A more extensive discus-
sion of the circuit operation and guarding will be found in
Section IV Circuit Description.
kee the GUARD to
Yl
ohms, and cexe
Fig. 3 Model 515 Megohm Bridge, Simplified Schematic Diagram
In cases where measurements with the unknown external to the
bridge are necessary and the Model 5152 Remote Test Chambe?
is not suitable, the user can make his own holding fixture
and connect it to the bridge.
Teflon insulded trisxial cable should be used for the con-
nection. The central conductor is the High Impedance conductor;
the inner shield braid is the Guard, and is driven from the
low impedance arms of the bridge; and the outer braid is GND,
to provide shielding. Amphenol 21-529 is a suitable cable.
515
III - 7
Page 16
The connector should also be teflon insulated. Gremar 7991
is satisfactory. Fig. 4 shows the connector and cable.
MTERWAL BRIIGEVOLTAGESUPPLY
F.
Bridge voltages higher than the 10 volts available from the internal
4 are desired when measuring resistances greater than about
;;r;
ohms, or in studying the voltage coefficient of a resistor.
A UliF connector labelled EXTERN& LLNFVT is mounted on the rear of
the bridge cabinet for ready connection of a high voltage source.
The shell of the connector is at ground potential, and this grounds
one terminal of the external bridge supp4. The central conductor
is the high - voltage lead. The bridge is insulated so that the
external bridge voltage can be as high as 1000 volts.
Either the Keithley Model 241 or the 240 Regulated High Voltage Sup-
ply makes a very satisfactory source for external bridge voltage.
The over-current protection on each is an important feature in pre-
xenting damage to the bridge resistors or to the unknown.
connecting the supply to the UHF receptacle on the rear panel. Do
not app4 more than 100 volts unless the x ICC or the x 10 dial is
in a position other than "O",
the bridge resistors. With the recommended Model 240 or 241 Regulated
Voltage Supp4 the over current protection will prevent damage in
the eventthis precaution is not observed.
for too much current will flow through
In making voltage coefficient measurements, it should be kept in
mind that the voltage applied to a Wheatstone Bridge is greater than
the voltage appearing across the unknown resistance being measured.
The relationship between the bridge voltage and the voltage across
the unknown is given in Section 0, below.
The shielded measuringcompartment in the bridge has a safety switch
which is operated when the door is closed. This switch operates
a relay whiah applies the voltage from the external bridge supply
to the bridge circuit. With the door open, the voltage is renmved,
so that the unknown can be changed without possible harm to the operator.
When the unknown is located outside the bridge, and the FUNCTION
switch set to MTERNAL
the circuit< Unless the external bridge voltage supply is turned
off or disconnected, voltages dangerous to the operator may be present
at the unknown terminals.
source is to switch the BRIDGE VOLTAGE from EXT to zero.
G. MLTAGEAC!RGSSURKD3WNARDSTANBARB
In many cases, particular4 in measuring coefficients of resistors.
it is important to know the voltege across the unknown. In measuring
OPERATE, this safety interlock is removed fmm
A convenient means of disconnecting the
515
III - 8
Page 17
the leakage resistance of capacitors,
the applied voltage must be
known to avoid breakdown. Also, for rated accuracy, the voltage
across the standard resistor must not exceed ten volts.
If the bridge voltage is E, the unknown resistance X, and the standard resistance S, then the voltage across the unknown is:
and the voltage across the standard is:
The bridge voltage is read from the BRImE VOLTS dial or from the
external bridge voltage supply.
The standard resistance is the value
indicated by the m&tiplier dial.
H. ACCUHACYCONSIDEHATIONB
The accuracy of measurement of an unknown resistor in a Wheatstone
Bridge depends primarily on the accuracy and stability of the other
three arms in the bridge, upon the resolution of the variable arm,
and upon the ability of the null detector to respond to the small
incremental changes in the variable arm. There are also numerous
secondary effects.
These will aU be discussed below.
1. Null Detector Sensitivity.
To be able to detect a desired fractional deviation of the unknown, corresponding to the wanted percent accuracy of the measurement, the required null detector sensitivity is given by the
approximate expression*:
e is the null detector signal in volts
x is the incremental pa-t of the unknown resistance
E is the Bridge Potential in volts
S is the Standard Resistance, in ohms
X is the Unknown Resistance, in ohms
For resolutions of O.l$ in the unknown,
x = 0.00l.x
If X and S are approximately equal,
and
the Bridge Potential
is 10 volts,
o.oolx = 0.0015,
OY63R
e =
0.0025
v0its
(2.5 millivolts)
*See Electrical Measurements by F. K. Harris. John Wiley & Sons, N. Y. 19.52
515
III - y
Page 18
In the case when X is approximately 10 S,
e = .W8 volts
(o .8 miuvolts)
The maximum sensitivity of the null detector in the Model 515
Megohm Bridge is one millivolt per meter dial division, and
is thus sufficient for the rated accuracy of the bridge. Care
should be taken, however, to be certain that the detector sensitivity and the bridge potential are great enough end the resistance of X end S are sufficiently close to each other to
obtain the expected accuracy of measurement.
A check on the sensitivity of the system may be made by unbalanc-
ing the bridge readout dials a given percentage and observing
the null detector deflection.
Null Detector Zero Drift.
2.
Vacuum tube electrometers drift about one to two millivolts
per hour, and this rate can be expected in the Null Detector.
Obviously, a false balance is indicated if the meter points
to zero, indicating balance, when in reality thereare several
millivolts at the input.
This error is easily eliminated by adjusting the null detector
to zero,whlle the PUSH 7.0 READ button is released, then depres-
sing the button end balancing the bridge.
Resolution of the Readout.
3.
Using only the readout dials x10 through x.001, full rotation
of the x.001 dial is O.l$ of the total setting. The dial can
be easily read to one-twentieth of its full rotation, giving
a readout resolution of 0.005%. This is ten times the best
accuracy specified for any range.
When using all the dials, the readout resolution is very much
greater than the maximum accuracy.
4. Accuracy of the Readout Resistors,the Standard Resistors,
end the Standard Calibration Controls.
location of each of them in the Wheatstone Bridge Circuit).
The accuracy of the resistors on the switches controlled by
each Readout Dial (RFSISTANCE, OHMS) is:
(See Fig. 3 for the
515
III
- 10
Page 19
Xl00
0.5%
O%$
x.01 X.001
0.1%
1.0%
ma~i.smm accuracy with the bridge is obtained when using the
dials xlOthroughx.001.
!l!his is because the most accurate
readout resistors are used, and also because the unknown re-
sistance and the standard resistance are sufficiently close
that the null detector has enough sensitivity with bridge Wit-
eges less than ten volts (see Section III R. 1).
!Che xl00 dial has only 0.5% resistors associated with it
because of the extremely high cost of more
value resistors, and because resistors above 10 ohms
act
te high
Y
are not very stable high accuracy measurements are not
warranted.
The resistors on the x.01 end x.001 dials are less accurate
because they are not followed by enough diels to give high
resolution, and their accuracies are great enough for rated
accuracy when using the x1.0 to x.001 dials.
With the MCLTIPLIER dial in either the 105 or 106 position
all three aTm6 in the bridge itself are wirewound resistors
accurate to .02f&, permitting the unknown to be measured to an
accuracy of .05$.
With the MUTEIPL.lRR in the lo7 position, after the Standardization process, the bridge accuracy is that of the previous re,nge
(.05$) plus the error introduced by standardizing, which is
conservatively set at .05$.
FoUowing this pattern, the accuracy of the bridge at each suc-
cessive step of the lrmltiplier dial is the accuracy of the pre-
vious step plus the .05$ standardizing error. It is in this
fashion that the accuracies in the specifications up thru 1012
ohms were derived.
FromlO~tto l.014
to warrant the 0.5% rating, and above I.0
ohms, enough secondaryl~ffects are present
ohms, the xl00 dial
is used, adding enough further error to bring the overall accuracy
rating to l.O$.
The standard resistors used are as follows:
Multiplier
q
3
30
I&
ld2
Resistor !Pype
Wire Wound
0 (1
Dezosit,ed Carbon
Sealed,?-Meg
II
I! II
II 11
Accuracy
0.0s
0.0s
1.0 %
1.0 5
2.0 $3
2.0 $
515
III - XL
Page 20
The Standard Calibration Controls arm is either a wirewound
resistor accurate to 0.02$, or deposited carbon resistors in
series with trinnaing potentiometers.
5. Temperature and Voltage Coefficient of Bridge Resistors.
The wirewqind resistors employed are free from voltage coefficient.
They use one of the lowest temperature coefficient of resistance
alloys available, changing 20 parts per million per 'C, or O.OG~$,/~C.
They are measured at room temperature, 25oc, end for greatest
accuracy, the bridge should be used near this temperature.
The depositedcarbon and Hi-Meg resistors have substantially
higher temperatuqe coefficients of resistance than the wire-
wound resistors.
But if the bridge is allowed to come to its
working temperature and standardized, it will have its rated
accuracy unless the temperature changes. In this case it should
be restandardized.
Deposited carbon and Hi-Meg resistors also exhibit voltage coefficientof resistance.
The'Hi-Meg resistors used in the bridge
are spiralled and have about one tenth the voltage coefficient
of standard Hi-Meg units.
Nevertheless, the voltage across
these resistors should not exceed 10 volts for ms.xitmsn accur-
acy . Seem Section G. Voltage Across Unknown and Standard Resisters .
6. Leakage Resistence Across the Unknown.
6
lOlo ohms shunting one megohm (I.0
ohms) produces a change of
0.015; and 1015 ohms shunting 1012 ohms produces a 0.1% change.
With high resistance resistors end high accuracies leakage resistance is an important consideration.
The termipals of the Model 515 Megohm Bridge have been carefully
made with teflon insulation, and guarding has been employed
extensiveily .
The major concern of the operator in using the
bridge is to keep the insulation clean and dry. The user, how-
ever, should be greatly concerned with the bobbin and housing
or casing of his unlrnown resistor and with any specielly built
holdllng fixture. Paper base bakelite which has been handled
and allowed to remain in a humid atmosphere has a surprisingly
low resistance.
Glass envelopes which have been handled and
have finger oil and salt paths between fused-in wire conductors,
or simple water vapor paths, also csn have a surprisingly low
resistance. E&reme care is necessary to avoid unsuspected
errors or: instabilities in measuring high resistances.
Errors Caused by Guard to Ground Resistance.
7.
Guarding, as described in Fig. 3 Section III E, is used exten-
sively in.the construction of the
Model
515 Megohm Bridge to
515
III - l2
Page 21
reduce errors caused by undesired leakege currents. The Guard
conductors are driven from the low impedance side of the Null
Detector.
on the Readout Mel switches and the Standard Calibration con-
trols.
Resistance from Guerd to Ground shunts the resistors
The Readout Mal resistors of xl0 to x.OClmay be as
hlghas10me@ms
shunted with 5 x 10
In the buildl
shunting error was establish;d for the lowest Guard to Grouud
resistence, and the user should da nothing to lower it.
The Guard &onductors are mstly inside the instrument cabinet
but Gusrd is exposed in the connector on the external unknown.
Wreme care must be taken at all times, with the instrument
cover on or'off, to maintain the cleanliness and dryness of
the Guard to Ground insulators.
The Guard conductor is also exposed in the 5152 Remote Test
Chamber.
T&M&% 5 5 G
8. Time Constants - Slow Responses
If sny resistor-holding electrode is connected to
ohms to ground, for meximum accuracy.
of ~~~nt~~~$o~~<C~~f~~.
T
12
uard, it too, must have a resistance greater
Ten picofarads and ICE
the wiring capacitances in the bridge end null detector input
combined with an unknown of about 1012 ohms produces a time
constant of several seconds.
The time constant is apparent in the length of time required
for the null detector meter to reach its final position after
an ed.justmellt has been made in the bridge.
For maxin~~~ measuring accuracy, the bridge null mnst be care-
fully determined, and readings taken only after the null detector
meter pointer has stopped moving.
The bridge has been carefully designed to keep the stray capacitances as low as possible, so that measurements can be made as
rapidly as possible. In measuring resistors greater than lOl2
ohms, the standard resistor is never greater than 1012 ohms,
thus the time constant is never longer then several seconds.
Measuring the leakage resistance of capacitors with the Model
515 Megohm Bridge can be a very tedious process, for with good
capacitors with very little leakage, the time constants with
the bridge impedances c&n be as long as several days. It is
recommended that this sort of measurement be done by charging
the capacitor to a known voltage and measuring its voltage at
known times later with a Keithley electrometer voltmeter.
ohms have a time constent of I.0 seconds;
515
III
- 13
Page 22
Transients Caused by Push-to-Read Stitch
9.
Whenever two conductors have been m&zing contact and are.separated, a~ charge appears on the contictors. In the Model 515
Megobm Bridge, this charge transfer is apparent in the null
detector meter when the Push-to-Read swltch is operated, removing the short circuit across the detector input. It
is mxt~noticeable when using the
EP
ohm multiplier and measur-
10
ing unknowns of 10” ohm or greater.
The Push-to-Read switch has been very carefully designed and
constrnqted to minimd.ze charge transfer, but a few nKU.Llivol.ts
are often induced in the bridge circuit by its operation.
!Phis is not harmful,
but it is necessary to wait each time the
switch Is operated, for several time constants during which
the voltages come to their steady-state value, and the null-
detector meter pointer stops drifting.
10. Verification of Accuracy
In chec
105,
ng the ccuracy of prototype bridges, resistors of
F ?
10 , ani 10 ohms were coqpared between the Model 515,
and. a Leeds and Northrup Guarded Wheatstone Bridge, Catalog
4232-B. These measurements were thus traceable to the National
Bureau gf Standards, e& were verified with various resistor
manufackurers. Agreement was within O.Ol$, which is weld. tithin
Model 515 specifications.
Resisttlllce values between 107 and. I.010 ohms were s-ted
a deltzwye
Transformation;
AIEE !!kansactlons Paper
58-556
by
gives the details.
Resistance values up to 10U
ohms were also measured carefulJy
by meas~ing the discbarge times of accurately known capacitors.
Above I.+ ohms, stray capacitances Introduced too much
Both the
well within the specified accuracy of the
ohm range was checked by measuring ten 10
,,,,cogy$2hm Y
delta-wye and capacitor
series. Agreement was obtained within
to 10 5 ohs, accuracy is assured by the care-
discharge measurements were
de1 515. !Che 10”
3Y
ohm resistors and
error.
ful. meaqurement of the resistors In the xl00 and xl0 switches
in the peadout arm of the bridge.
Ten resistors connected in parallel, measured accurately, and
then cowected in series is another method of obtaining high
value resistors to great accuracies, end was used extensively
in developing the bridge in production tests. This method is
described in “ElectrlcalMeasurements” by Harris. See Section
H, part 1.
515
III - 14
Page 23
In the manufacture of each Model
the bridge resistors is measured and found to be within its
rated limits before the bridge is assembled. After assembly,
each step on each decade is checked with a precision decade
box as the unknown resistor.
veloped standard resistors are used to check each step of
the multiplier switch. The internal applied bridge voltage
supply and nul.l detector sensitivity and zero drift are also
checked.
515
Megohm Bridge, each of
Following this, Keithley de-
515
III
-
15
Page 24
Page 25
CIRCUIT DESCRIPl'ION - SECTION IV
The circuit of the Model
components:
the Dower supply, the electrometer null-detector and
515
Megohm Bridge consists of three major
the Wheatstone Bridge.
A. FowEFl SUPPLY
The power supply consists of a transformer, rectifiers and filters,
a transistor regutitor supplying I.2 volts dc, and a transistor converter which supplies voltages Isolated from ground to the null-
detector. A portion of the I2 volts dc Is used to polarize the
bridge.
A detailed description of the power supply is as follows: Drawing
14522-D at the rear of the manual is the complete schematic. Tl
is the power transformer operating from the power line. The primary
may be connected for either IlO or 220 volts. One secondary is centertapped and, with diodes r@ and DlO, supplies 18 volts at 250 ma. The
other secondary and half-wave rectifier Dll supplies 20 volts at 10
ma. The 18 volts is applied to the collector of $1 and is dropped
to I.2 volts through the action of Ql and the associated regulator
circuitry.
Q4 and
Q5
form a difference amplifier which compares a portion of
the I.2 volt output with the reference voltage derived from zener
diode Dl4.
Q3
and Q2 are used as amplifier and driver for output
transistor Ql. D13 is supplied from the 20 volt supply referred
to above.
This supply is "boot-strapped" on the I2 volt regulated
supply to furnish a regulated return point for R2l2, the load resistor
for
Q3.
The circuitry is so designed that any change in load cur-
rent, or line voltage is compensated perfectly within the operating
range.
consisting of transistors
The regulated I2 volts supplies the transistor inverter,
Q6
and Q7 and transformer T2, and also
the ten volt bridge polarizing potential.
Transformer T2 and transistors Q6 and Q7 form a dc inverter operat-
ing at approximately 200 cps. The feedback winding to the bases
of the transistors provides the oscillator drive. The primary in-
ductance of the transformer determines the frequency of oscillation.
T2 Is especially well insulated to provide the necessary insulation
of guard to ground in the bridge circuit. Since the l2 volt source
is exceedingly well regulated, all the secondary voltages are unaffected by line voltage variation. Therefore, the operation of the
null-detector is not affected by line voltage variations.
NULL,Dhl'FZ!tDIC)R:
B.
Fig. 4 is a simplified circuit diagram of the null detector. The
component designations are the same as used In the complete schematic.
The power supply potentials are represented by batteries for simplicity.
515
Iv -1
Page 26
DRIVEN
GUARD
RI50
LO
GUARD
FIG. 4
The filaments of Pl, V2 and V3 are in series and supplied with 3.6
volts. This voltage also is used to bias the electrometer tube,
Vl. V2 is an amplifier and V3 is the output tube. The output clrcuit is somewhat unique and operates as follows: The meter will
have zero deflection only if there is no potential difference between the plus terminal of B3 and minus terminal of FE. This will
occur only if the drop across Rl@+ is equal to the potential of B3.
This zero output condition is set by adjusting the screen voltage
of Vlwith the fine zero control, ~164, and the coarse zero control,
An input signal will cause the plate current of V3 to change,
89.
and. the drop across Rlk.4 will increase or decrease, depending on the
polarity of the signal.
A current will then flow thru the meter
and one of the range resistors on SlO, the sensitivity switch. Since
the low impedance side of the input is the minus meter terminal the
potential drop across the range resistor will alter the potential
of the filament circuit in such a way that the grid-filement potential
of Vl will remain nearly constant.
The filament is the "Driven Guard"
since its instantaneous ac potential is nearly equal to the ac poten-
tial of the input signal. All the guarded points of the bridge em-
cept Sll are returned to this point.
This acts as a driven shield
and a considerable increase in response speed is realized in certain
cases.
J
51.9
IV - 2
Page 27
The simplified bridge circuit is shown in Fig. 5. One ratio arm,
A. is the six read-out dials.
I.cO K ohms for each multiplier rage.
105 the standard resistor S and the ratio arm B are both 105 ohm
.O
f~ wirewound units.
2
10
ohm ,025 wire wound and B is 105 ohm .02$. With the multiplier
at 107 or higher the standard resistor is a carbon film unit of limited
accuracy and B is a 95 K fixed resistor in series with a 10 K rheo-
stat. By the standardizing procedure described below B is set to an
appropriate value to compensate for error in S and the bridge will
be direct reading.
The first step in standardizing is done with the multiplier at 106
and the FUNCTION switch at STANDARDIZE.
as shown in Fig. 6.
standard resistor for the 107 range is in its place.
bridge with the read-out
ard as compared to the 10 standard.
accurate wirewound unit the value in cated for the 107 resistor
is quite accurate. In Fig. 3 the 10
be 5% high and the read-out dials will indicate 10.500.
Leaving the read-out dials set the FUNCTION switch is next set to
CALIBRATE. As shown in the simplified schematic of Fig. 7 the
standard and unknown resistors are replaced with a network of 1O:l
ratio, *ccurate to
a sociated with the 10
6f
B
a null will be obtained when B is 1.05 X 10 ohms.
When the multiplier is set at 106,
The unknown X has been disconnected and the
gi
.Ol
Ratio arm B is now B7, the adjustable leg
q* d"r"d.
ohm stan
The other ratio arm, B, is nominally
With the multiplier set at
S Is a
The basic circuit is then
Balsa
eJ.s will give the value60f the 10
Since the 10 standard is an
"i'
ohm resistor is assumed to
Rebalanc
ng the bridge with
f
ng the
3
stand-
This completes the standarization of the 107 ohm standard resistor.
After this, when using this resistor as a standard its 5% error is
exactly corrected for by B7 being 5% high.
515
IV - 3
Page 28
To standardize the 10' ohm standard the multiplier is set at 107,
the FUNCTION switch is returned to STANDARLWE, and the bridge is
again balanoed withgthe read-out dials. The indicated read-ou
the value of the 10 ohm resistor compared to the corrected 10 ohm
is
ft
standard. Next the RJETION yitch is set to CALIBRAT'g and the bridge
is rebalanced with ratio arm B .
Af$3r folio* this procedure in sequence for multiplIer positions
10 thru l0 the bridge is completely standardized and will be direct
reading on *Xi ranges.
OVEBVOLTAGE PF0TEcTION
D.
The function of V4, 8 gas regulator tube, is to prevent damage to
the readout resistors from excessive bridge voltage. It is connected
thru auxiliary contacts on the xl00 snd xl0 dials across the bridge
voltage.
Thus if both dials are at zero the circuit is complete
and, if snore than 133 volts is applied to the bridge, the tube will
conduct current.
With a Keithley Model 240 or 241 Voltage Supply
this current W-XL be enough to trip the overload relay and no damage
will be done.
If an unprotected source is used it is possible that
V4 would be ruined and then possibly the readout resistors would
overheat from excessive current.
IV - 4
Page 29
SDTJIONV - MAIRTRNARCE
Very few maintenance problems will arise from ordinary use of the
bridge. The components used have adequate safety margins and, since
the total power consumption is only 10 watts, very little tempera-
ture rise will occur even with continuous operation.
If it becomes apparent that the bridge is not working properly the
first step is to check the voltage on the printed circuit boards.
Remove the six screws on each side and the two on top and remove
the cover.
null detector board as shown on Drawing 14363-c.
If the difficulty is determined to be in the null detector proper
the following procedure will be helpful in isolating the cause:
Short circuit the feedback by jumpering the feedback resistor in
use.
The sensitivity w-ill. now be about 500 microvolts per division and
it will be rather difficult to keep the meter on-scale with the ZERO
control. However, if it is possible to swing the tube voltages thru
the values indicated on 14363-c the stage is working satisfactorily.
Start with Vland workthru to V3.
located check the tube itself and then the associated components.
If it is necessary to replace the electrometer tube avoid touching
the glass near the lead wires.
These resistors are mounted on S-10, MILLIVWIS PER DIVISION.
Check the voltages on the main power supply board and
Once the defective stage is
Once the bridge is operating a good check on the accuracy may be
amde by placing the multiplier at 105 and the IUNCTION switch at
STARDARDIZR. A null balance should be obtained with the readout
dials at IO.000 f.O5$. If the reading is not within these limits
it is recommended that the unit be returned to the factory for repairs.
The overvoltage tube, Vk, will ordinarily not carry any current and
should last indefinitely. However, if the instrument has been used
with a supply lacking adequate overload protection it is possible
the tube may be damaged. If this has happened it is likely that
other components have been damaged also. Check the resistors in
arm A and B of the bridge (RIJJ. thro RI.23 and Rl71 thrn Rlgl).
230 volt operation:
To change the power line voltage to 230 volt remove the two jumpers
at the bottom of the main power supply board and connect a single
jumper as shown in the circuit schematic, 14522-D. Be sure the line
cord is removed from the power when doing this. Replace the 4 ampere
yrC: fuse with a l/8 ampere unit.
515
V-l
Page 30
Page 31
MODEL 515 MEGOHM BRIDGE
REPLACEABLE PARTS
SECTION 6. REPLACEABLE PARTS
6-1.
Model 515.
manufacturer, the manufacturer's part number and the Keithley Part Number.
address of the manufacturers listed in the "Mfg. Code" column are in Table 2.
6-2.
Part Number, the circuit designation and a description of the part.
and those parts coded for Keithley manufacture (80164) must be ordered through Keithley
Instruments, Inc. or its representative.
Parts List, completely describe the part,
ment, Keithley Instruments, Inc.
REPLACEABLE PARTS LIST. The Replaceable Parts List describes the components of the
The List gives the circuit designation, the part description, a suggested
TIE name and
HOW TO ORDER PARTS.
For parts orders,
a.
b. Order parts through your nearest Keithley representative or the Sales Service Depart-
sw
CerD
Camp
3Cb
EMC
ETB
E
:Cb
<
LOC.
ampere
Ceramic, Disc
Composition
Deposited Carbon
Electrolytic, Metal Case
Electrolytic, tubular
farad
Glass enclosed carbon
kilo (103)
Location
include the instrument's model and serial number, the Keithley
All structual parts
In ordering a part not listed in the Replaceable
its function and its location.
M
2,.
0
PMC
Poly
I-r
v volt
w watt
ww
WWVar
mega (106)
milli
Manufacturer
ohm
Paper, Metal Cased‘
Polystyrene
(10e3)
micro (10-6)
Wirewound
Wirewound Variable
1065R
TABLE 1.
Abbreviations and Symbols.
6-l
Page 32
REPLACRABLE PARTS
(Refer to Schematic Diagram 14522D for circuit designations)
ABLE 2. Code List of Suggested Manufacturers. (Based on Federal Supply Code for
Manufacturers, Cataloging Handbook H4-1.)
6-8
0668R
Page 39
v3
v2
6418/
j’
6418~
P
i !
Page 40
Page 41
Page 42
Page 43
.REPAIR AND CALIBRATION FORM
For
repair or calibration, please fill out this form
Sales Service Department
Keithley Instruments, Inc.
2877.5 Aurora Road
Cleveland, Ohio 44139
user’s Name
company
Division
Data
1.
Reason for Return
0 Repair and Recalibration 0 Report of Calibration Certified
0 Recalibration only (No report, except Traceable to N.B.S.
as specified in item 4 on reverse)* 0 Calibration Report
*If repairs are necessary to meet speci-
fications,
the calibration.
3.
To help repair the instrument, briefly describe the problem:
1s the problem -Constant
4.
Under what conditions does the problem occur:
Control setting e) Line voltage
=)
b) Approx. Temperature OF
b) Approx.
c) Approx. Temperature variation * OF
c) Approx. Temperature
d) Approx. Humidity (high, medium, low)
d) Approx. Humidity (hrgn, n
they will be in addition to
Telephone
Address
City
Model No.
2. Calibration Report Desired
(for details, sea reverse side of this form)
and return it with your instrument to:
RDo not write in this space.
state
Serial No.
a Certificate of Compliance
0 Node
1 Intermittant
f) Other (such as line transients,
line variations, etc.)
Ext.
Zip
Please draw a block diagram of the system using the Keithley.
5.
nent data which can help in the repair. Include charts or other data if available.
Signal Source
Source Impedance
Readout Device:
aRecorder
0 Oscilloscope
- Other
a None
Lengths & Types of Connecting Cables
What repairs or modifications have been made on this instrument which are not on file
6.
with the Keithley Repair Department?
Please enclose any other pertinent data and charI.? which you feel might help the
7.
Repair and Calibration Department
Signature Title
1267
List any other perti-
Page 44
Listed and defined below are the four types of calibrations and their associated
report formats which are presently available at Keithley Instruments.
They fall into
the following categories: .
Report of Calibration Certified Traceable to the National Bureau of
1.
Standards
Calibration Report
2.
3.~ Certificate of Compliance
Recalibration
4.
ALL calibration and certification performed by Keithley Instruments is in accord
with MIL-C-45662A.
Prices shown below are in addition to repair charges for any work necessary to place
a customer’s unit into first class condition prior to the calibration.
I
Page 45
U.S.A. and Canada sales offices
european services
Keithley has a manufacturing facility in Germany and a warehouse
and office in Switzerland to provide prompt delivery in the European
area. The facilities. staffed by Keithley field engineering personnel.
give on-the-spot assistance to your Keithley representative so that
he may better serve you. A list of the worldwide Keithley rspresentatives is given on the back.
K
EITHLEY INSTRUMENTS; INC.
28775 AURORA ROAD TELEPHONE: (216) 248-0400
TELEX: 98.5469
CLEVELAND,;OHIO 44139
Page 46
international sales offices
NETHERLANDS
PAKISTAN
REPUBLIC OF SOUTH AFRICA
Protea Physical 8. Nuclear