This is only a preview of the March 2010 issue of Silicon Chip. You can view 30 of the 104 pages in the full issue, including the advertisments. For full access, purchase the issue for $10.00 or subscribe for access to the latest issues. Articles in this series:
Items relevant to "A Solar-Powered Intruder Alarm For Sheds & Boats":
Items relevant to "Digital Audio Signal Generator, Pt.1":
Items relevant to "A Low Capacitance Adaptor for DMMs":
Purchase a printed copy of this issue for $10.00. |
A Low Capacitance
Adaptor for DMMs
This neat little adaptor allows a
standard digital multimeter to
measure low values of capacitance
– from less than one picofarad to
over 10nF. It will allow you to
measure tiny capacitors or stray
capacitances in switches,
connectors and wiring.
By JIM ROWE
A
lthough some modern digital multimeters do
provide capacitance measuring ranges, these
are generally not particularly useful when it
comes to measuring low value capacitors or the stray
capacitance associated with connectors, switches and
other components.
For most of these small capacitance measurements
you normally need to use a dedicated low-value
capacitance meter and these can be a bit pricey.
The Adaptor is easy to build, with all of the
components mounted on a small PC board. The
board fits into a box which is small enough to
be used as a dedicated ‘low capacitance probe’
for the DMM, making it well suited for measuring stray capacitances. Just about any modern
DMM is suitable for the Capacitance Adaptor,
provided it has an input resistance
of 10M or 20M.
How it works
Essentially the
Adaptor works as
a capacitance-toDC-voltage converter, as shown
in Fig.1.
First we generate a square wave ‘clock’
signal with a frequency
of between 110kHz and
1.1kHz (depending on the
measuring range) using a simple relaxation oscillator based on
capacitor C1, resistor R1, trimpot VR1
70 Silicon Chip
siliconchip.com.au
R2
SQUARE WAVE
OSCILLATOR
VR1
R1
SCHMITT
BUFFER
EXCLUSIVE-OR GATE TRUTH TABLE
VC1
(NULL
STRAY C)
EX-OR
GATE
BUFFER
INTEGRATOR
R4
1.000
DC VOLTS
R3 (=R2)
T1
C1
T2
Cx
(CAPACITOR
UNDER TEST)
and a Schmitt trigger inverter. This square wave signal is
then passed though a Schmitt buffer stage to ‘square it up’
and produce a waveform with very fast rise and fall times.
The output from the Schmitt buffer is then split two ways
and passed through identical resistors R2 and R3. Then
they are fed to the two inputs of an exclusive-OR (XOR)
gate. The signal which passes through R2 has a small trimmer capacitor VC1 connected from the ‘output end’ of R2
to ground, while the signal which passes through R3 has
the capacitance which is to be measured connected from
the output end of R3 to ground (ie, between terminals T1
and T2).
So each signal is fed to the inputs of the XOR gate via
an RC delay circuit. The combination of these two RC
delay circuits and the XOR gate form a simple ‘time delay
comparator’.
Remember that when both inputs of a XOR gate are at
the same logic level (either high or low), its output is low.
And whenever the two inputs are at different logic levels,
its output switches high. This is summarised in the truth
table associated with Fig.1.
Now consider the situation where there is no discrete
C2
+
–
DMM
(SET TO
DC V)
INPUT A
INPUT B
L
L
L
L
H
H
H
L
H
H
H
L
OUTPUT
Fig.1: it’s essentially a capacitanceto-DC-voltage converter, as this block
diagram shows. The truth table for
the exclusive-OR gate is shown above.
capacitor connected between the test terminals, so there
will only be a small ‘stray’ capacitance between them. As
a result, there will only be a very short delay in the signal
passing through R3 to the lower input of the XOR gate.
If trimmer VC1 is set to provide the same low capacitance
for the signal passing through R2, the two signals applied
to the inputs of the XOR gate will be delayed by the same
amount of time, and so will arrive at the gate inputs ‘in
sync’ – rising and falling at exactly the same times.
In this situation the output of the XOR gate will remain
low at all times, because both inputs of the gate are always
high or low, both switching together between the two levels.
But when we connect an unknown capacitor (Cx) between
terminals T1 and T2 the signal passing through R3 will be
delayed more than the signal passing through R2.
So now the lower gate input will switch high and low a
short time after the upper input and as a result, the logic
levels of the two gate inputs will be different for short periods of time following each H-L or L-H transition of the
square wave signal.
The output of the XOR gate will switch high during these
transition delays, generating a series of positive-going puls-
Here’s a view inside the
open low capacitance adaptor,
looking towards the unknown
capacitor terminals. The jacks on the
right-hand end connect via banana leads to the digital
multimeter – although elsewhere in this article we
give a possible “plug-in” alternative which saves you
using leads at all.
siliconchip.com.au
March 2010 71
es with their width
as those for lower
directly proporvalues because of
Specifications
tional to the extra
the increasing curThree measuring ranges –
delay time caused
vature of the R4-C2
Range A: 0.1pF = 1mV, [gives a range from below 0.3pF to above 100pF.]
by the unknown Range B: 1pF = 1mV, [gives a range from below 1pF to above 1000pF (1nF)]
charging/dischargcapacitor Cx.
ing exponential.
Range C: 10pF = 1mV, [gives a range from below 10pF to above 10.0nF.]
In fact the width
Accuracy:
Within approximately 2% of nominal full scale reading,
Circuit details
of the pulses will
(assuming you can calibrate ranges using capacitors of known value).
be directly proThe full circuit
Power:
9V alkaline or lithium battery.
portional to the
of
the Capacitance
Current drain: less than 5mA.
value of unknown
Adaptor is shown
capacitor Cx, bein Fig.2. Schmitt
cause we deliberately limit the delay time to a relatively inverter IC1a operates as the square wave clock oscillator.
small proportion of the half-wave period of the square The only difference from Fig.1 is that switches S1b and
wave ‘clock’ signal.
S1c allow three different C1/VR1 combinations to be used,
The rest of the circuit is used as a simple integrator, to for oscillation at three different frequencies, to provide the
convert the positive-going pulses into a DC voltage. We three measurement ranges.
feed the pulses through a non-inverting buffer, to ensure
The remaining inverters in IC1 (a 74HC14 device) are
the pulses are all of constant peak-to-peak amplitude and used to form the non-inverting Schmitt buffer following
then through the integrator formed by series resistor R4 the oscillator. IC1b squares up the signal initially and then
and shunt capacitor C2.
drives IC1c-f in parallel to re-invert the signal and square
The average DC voltage developed across C2 is directly it up even further.
proportional to the width of the pulses and it is this DC
The paralleled outputs of the clock buffer drive the upvoltage that is measured by the DMM.
per and lower arms of the ‘time delay comparator’. Here
Although we are only using a simple RC combination the two 10k 1% resistors correspond to R2 and R3 in
to perform this integration, the relationship between the Fig.1. However, the signals from the two delay circuits R2/
pulse width and the output DC voltage is reasonably linear VC1 and R3/Cx now pass through another pair of Schmitt
because we have deliberately limited the integration to the inverters, IC2c & IC2a, which are part of a second 74HC14.
initial 20% of the exponential RC charging and dischargThis has been done to ‘square up’ both signals, to ensure
ing curve.
that the width of the output pulses from IC3a maintain
That’s why the nominal full-scale reading on each of our their linear relationship to the value of the capacitor becapacitance ranges is only 1.000V, even though all of the ing measured.
Adaptor circuitry operates from a 5V supply rail.
Although this squaring up is only necessary for the lower
In fact, you can use the Capacitance Adaptor to measure (Cx) signal, because of its longer delay and hence greater
capacitors with a value of more than the nominal full scale ‘rounding’, we also pass the upper (VC1) signal through an
value on each range but the readings won’t be as accurate identical inverter to ensure that it is inverted in the same
D1
1N4004
A
100nF
2
1
4
3
IC1b
10k
VR3
4
5
14
IC1c-f
8
9
11
VR2
VR1
IC1:
74HC14
VR1-VR3: 5k x 25T
6
10
12
13
7
2
10nF
10 F
1nF
100nF
3
2
1
S1c
RANGE
FUNCTION
1
(POWER OFF)
100pF (0.1pF/mV)
1nF (1pF/mV)
2
3
4
2009
GND
47 F
IC1a
3
SC
+5V
OUT
S1: RANGE
/POWER
S1b
100nF
IN
3
1
3
REG1 78L05
2
4
9V
BATTERY
4
1
S1a
K
4
IC2a-f
10k 1%
VC1
3-10pF
NULL
STRAYS
5
6
9
8
10
11
IC3:
74HC86
1
IC3a
3
2
IC3b 14
4
5
10
10k 1%
Cx
(CAP
UNDER
TEST)
14
100nF
+
–
1
2
13
12
9
12
13
7
6
IC3c
1k
8
IC3d
11
10 F
7
IC2: 74HC14
–
78L05
10nF (10pF/mV)
DMM CAPACITANCE ADAPTOR
+
OUT
TO
DMM
GND
IN4004
A
K
IN
OUT
Fig. 2: the complete circuit diagram. The three active switch positions give a range of about 0.3pF to 10nF.
72 Silicon Chip
siliconchip.com.au
way as the lower signal. Thus both signals have the same
nominal phase and both signals have the same propagation
delay, ie, via IC2a & IC2c.
IC3a is the XOR gate of the time delay comparator, while
the remaining three gates in IC3, a 74HC86 device, are used
as a non-inverting buffer to drive the RC integrator.
Here the 1k resistor corresponds to R4 in Fig.1, while
the 10F tantalum capacitor across the output jacks corresponds to C2. Gates IC3b-d are used simply as non-inverting
buffers by tying the second input of each to ground logic low.
90 x 50.5mm and coded 04103101. This fits snugly inside
a plastic instrument box measuring 120 x 60 x 30mm.
The only components which are not mounted directly on
the PC board are the binding posts and the output ‘banana’
jack sockets (or banana jacks themselves) for connection to
the DMM. The former mount on one end of the box while
the latter mount on the other end.
In each case the posts and jacks connect to PC board
pins. Note that the binding posts and jacks are both spaced
apart by the standard 19mm (3/4”), to make them compatible with double-plug connectors etc.
Before you begin fitting the components to the PC board,
it’s a good idea to open up the box and check that the board
will slip inside the lower half (the half with the countersunk
holes for the final assembly screws). You may need to file
off a small amount from all four sides of the board so that
it will slip down to rest on the support pillars moulded in
the inside of the box.
You may also need to file small shallow rounded recesses
in the two ends to clear the larger pillars around the box
assembly screw holes. It’s much easier to do this before any
components have been mounted on the board.
Begin board assembly by fitting the three wire links,
followed by the six PC pins: two each for the input terminals and output jack connections and two for the battery
clip lead connections (just below the positions for D1 and
REG1, at lower centre).
Next, fit the three 14-pin IC sockets for the three ICs,
noting that the socket for IC1 should have its notched end
to the right while those for IC2 and IC3 are to the left, as
on the overlay diagram of Fig.3. Then fit the four fixed resistors, followed by the three 5k 25-turn trimpots. Make
Power supply
Power is supplied by a 9V alkaline or lithium battery,
with diode D1 used to prevent any possibility of reversepolarity damage. Switch S1 acts as a combined power and
range switch, with S1a is used to switch off the Adaptor in
the fourth (fully anticlockwise) position.
The Adaptor circuit needs to run from a regulated DC
supply rail, so that the measurements don’t vary as the battery voltage droops with age. Regulator REG1 is therefore
used to provide a regulated +5V supply rail, provided the
battery voltage remains above 7.5V.
Since the current drain of the circuit is below 5mA, we
are able to use a 78L05 regulator (TO-92 package) for REG1.
The 47F, 10F and 100nF capacitors are used to filter any
noise and switching transients which may appear on the
+5V supply line.
Construction
As you can see from the photos and the PC board overlay
diagram of Fig.3, virtually all of the components used in
the Adaptor are mounted on a small PC board, measuring
CAPACITANCE
MEASURING
BINDING POSTS
2x
100nF
IC3
74HC14
100nF
Cx+
+
5k
VR1 VR2 VR3
REG1
D1
5k
100nF
5k
IC1 74HC14
+
DMM TEST
LEAD
JACKS
+
1nF
10k
BOX
END
PANEL
OUT+
10 F
+
4004
VC1
3-10pF
BOX
END
PANEL
9002 ©
19021140
9V BATTERY
47 F
–
OUT–
78L05
9V
+
10k
10 F
–
10k
Cx–
S1 RANGE
10nF
ZERO
NULL
-
+
74HC86
1k
IC2
E C NATI CAPA C
RETE M
R OTPADA
S M M D R OF
Fig.3 (above): life-size
component overlay diagram,
with posts and jacks, plus a
slightly enlarged photograph of
the same thing. The only thing
not shown here is a small cable
tie which should be used to
secure the battery snap leads to
the PC pins – flexing of the leads
when the battery is changed is
a sure-fire recipe for them to
break off at the solder joints.
siliconchip.com.au
March 2010 73
Connecting to your DMM:
another
approach
While this project
was being prepared for
publication, it occurred
to us that there was
another, perhaps even
more logical way to
connect the adaptor to
a DMM – particularly if
you would like a more
“hands free” operation.
This takes into account the fact that the
overwhelming majority
of DMMs which use 4mm sockets (and we would have to say
ALL pro-quality units) have a standard 19mm spacing between
those sockets.
Therefore, we reasoned that it would be quite sensible to
replace the banana jack sockets on the “output” end with
banana jacks – thus allowing the unit to be plugged directly
into the DMM.
At the expense of some flexibility, this would mean that there
would be no need to make up a set of Adaptor-to-DMM leads.
Try as we might, we could not easily find a set of these already
made up. You can get banana to probe, banana to alligator clip,
banana to multiple adaptors, even banana to blade fuse fittings
(for automotive use) but banana to banana? Nada. Zilch. Nyet!
So the only alternative would have been to buy some figure-8
red and black lead (believe it or not, also getting hard to find in
lightweight, flexible type!), two pairs of red and black banana
plugs and solder them onto the lead.
The alternative approach, as shown above and below, is to
fit a pair of red and black banana plugs through the end of the
case. We used a scrap of PC board, cut and shaped the same
as the end panels, with a strip of copper removed down the
middle. Drilled appropriately, this gave us a handy “platform”
to which we soldered the two banana plugs (inside) without
their plastic shrouds. The plugs were then soldered back to
their respective PC pins using short lengths of tinned copper
wire (eg, resistor/capacitor lead offcuts).
Presto – a plug-in adaptor. And if you want to use it off the
DMM? Simply use a banana-to-alligator clip lead set.
sure you place the latter with their screwdriver-adjustment
screw heads at lower left.
Now add the fixed capacitors, taking care to place the
polarised 47F and 10F caps with the correct orientation,
as shown in the overlay diagram. Then fit the mini trimcap
(VC1) in position, with its ‘flat’ end to the left as shown.
Rotary switch S1 is fitted next, after cutting its spindle
to about 10mm long and filing off any cutting burrs with a
small file. The switch mounts on the board with its moulded
locating spigot at approximately the ‘7:30’ position, viewed
from above and with the board orientated as shown in the
overlay diagram (ie, with IC1 at lower left).
S1 is a “universal” type of switch offering a number of
switch positions so after it is installed, it needs to be set
for the four positions we require.
Remove the nut and lockwasher from its threaded bush
and then lift up the stopwasher as well. Then turn the
spindle anticlockwise by hand as far as it will go and refit
the stopwasher with its ‘stop pin’ passing down through the
hole between the digits ‘4’ and ‘5’ moulded into the switch
body. Then replace the lockwasher and the nut, threading
the latter down until it’s holding down both washers firmly.
You should now find that if you try turning the spindle
by hand, it will have a total of four positions – no more
and no less.
Don’t be caught out by the old trap of thinking you only
have three positions because it only clicks three times.
Remember it clicks to three more positions from its end
position.
Then you can fit REG1 and D1 to the board, noting their
correct polarity. Plug IC1-3 into their respective sockets
and your board assembly will be complete. You put it aside
while you drill the various holes which need to be cut in
Another way of measuring “C” – using a small length of
4mm brazing rod with a point and slot, (shown below) you
can fashion a “probe” to get into tight spots.
74 Silicon Chip
siliconchip.com.au
the top, bottom and end panels of the box.
Preparing the box
Two holes need to be drilled in each of the end panels and
five holes in the top of the box. You will also need to cut
away a small amount from the sides of the assembly screw
surround pillars on both the top and bottom of the box,
to provide clearance for the ‘rear ends’ of the capacitance
measuring binding posts and DMM test lead jacks, when
the box is assembled. This cutting away can be done with
a small milling cutter in a high speed rotary tool or done
manually with a sharp hobby knife (careful!).
Both pairs of holes in the end panels need to have a diameter to suit the binding posts and banana jacks you are
using. They are located on the centre line of their panel
but 9.5mm away from the centre-line in each case - so the
binding posts and jacks both end up spaced apart by the
standard figure of 19mm (3/4”).
The five holes in the top of the box can be located quite
accurately using a photocopy of the front panel artwork (or
a printout from siliconchip.com.au) as a template, because
you’ll see that this includes a dashed outer rectangle to
show the outline of the box itself. The central hole for the
power/range switch is 10mm in diameter, while the other
four holes are 3.5mm in diameter, which allow adjustment
of the zero null trimcap and calibration trimpots when
fully assembled.
The exact location and amount of material which must
be removed to clear the binding posts and banana jacks will
depend very much on the actual posts and jacks that you
use. You can see from the internal photos where material
needed to be cut away for the posts and jacks used in the
prototype.
(By the way, the binding posts used were the PT-0453 &
PT-0454 from Jaycar, while the banana jacks were the PS0406 & PS-0408 – also from Jaycar. Other posts and jacks
may need the removal of either less or more material but
you should be able to fit in most types that are currently
available.)
The last step in preparing the box is to make another
photocopy or printout of the front panel artwork on either
an adhesive-backed label sheet with a piece of clear selfadhesive film over the top or, for really long life and best
protection, plain paper laminated in a plastic sleeve. The
label is then cut out and applied to the front of the upper
half of the box, lining up the holes of course.
First assembly steps
The first step in assembling the Adaptor is to mount the
binding posts and banana jacks on their respective end
panels, tightening their mounting nuts to make sure they
won’t be able to rotate and work loose. Note that in the
case of the banana jacks, you also need to mount them with
their solder tags orientated vertically downwards so that
after the nuts are tightened, the tags can be bent up by 90°.
This is to allow the holes in the tags to be shortly slipped
down over the terminal pins in the PC board.
Next lower the PC board assembly into the lower half of
the box, and fix it in place using four very small self-tapping
screws (no longer than about 5mm). Then you should be
able to lower the end panel with the output jacks down
into the slot at that end of the case, with the tags on the
rear of the jacks passing down over the terminal pins of
siliconchip.com.au
Parts List –
DMM Low Capacitance Adaptor
1 PC board, code 04103101, 90 x 50.5mm
1 Utility box, 120 x 60 x 30mm
(eg Jaycar HB6032, Altronics H0216)
1 3 pole 4 position rotary switch (S1)
(eg Altronics S-3024, Jaycar SR-1214)
1 Instrument knob, 16mm diameter
1 Binding post, red
1 Binding post, black
1 Banana jack socket, red
1 Banana jack socket, black
1 9V alkaline or lithium battery
1 9V battery snap lead
3 14-pin DIL IC sockets
6 1mm diameter PC board terminal pins
1 small cable tie
4 Small self tapping screws, max 5mm long
Semiconductors
2 74HC14 hex Schmitt inverter (IC1,IC2)
1 74HC86 quad XOR gate (IC3)
1 78L05 low power +5V regulator (REG1)
1 1N4004 1A diode (D1)
Capacitors
1 47F 16V PC electrolytic
1 10F 16V PC electrolytic
1 10F 25V TAG tantalum
3 100nF multilayer monolithic ceramic
1 100nF MKT metallised polyester
1 10nF MKT metallised polyester
1 1nF MKT metallised polyester
1 3-10pF mini trimcap (VC1)
3 known value reference capacitors (see text)
Resistors (0.25W 1% unless specified)
3 10k
1 1k
3 5k 25T cermet trimpots (VR1,VR2,VR3)
the board. When the panel is down as far as it will go, you
can solder the jack tags to the terminal pins to make the
connections permanent.
The other end panel (with the binding posts) is then
fitted in much the same way, except that in this case there
are no solder tags at the rear of the posts. Instead you may
need to bend over the terminal pins on the PC board so
that they clear the rear spigots of the binding posts and
are alongside them, ready for soldering. Then when this
panel is down as far as it will go, the binding posts can be
soldered to the board pins.
The next step is to cut the battery snap lead wires fairly
short -- about 20mm from the snap sleeve - then strip off
about 5mm of insulation from the end of each wire, tin them
and solder them to the PC board pins just below REG1 and
D1. The positive (red) wire goes to the pin immediately
below D1, as you can see from the overlay diagram and
pics. Ideally, these wires should be secured to the PC board
pins with a very small cable tie.
After checking that everything looks correct, connect
the battery to the battery snap and your Low Capacitance
March 2010 75
Adaptor should be just about ready for its initial
set-up. All that remains is to fit the operating
knob to the spindle of switch S1 temporarily, to
make things easier during the set-up operation.
Initial set-up & calibration
NULL
B 1mV = 1pF
–
Fig.4 same-size front panel artwork
which can also be used as a template
for drilling the five holes required.
This can also be downloaded from
siliconchip.com.au
76 Silicon Chip
OUTPUT TO
DMM (DCV)
UNKNOWN
CAPACITANCE
Select the DMM that you are going to use with
the Adaptor and make up a connecting lead
to connect the output of the Adaptor to its DC
voltage inputs. In most cases the lead will need
standard banana plugs at each end.
Then connect the Adaptor and DMM together
using this lead and turn on the DMM, switching
it to a fairly low DC voltage range, eg the range
Arguably the wrong way to
with a full-scale reading of 1.999V or 1999mV.
measure a small capacitor –
Turn S1 to the first position (‘Range A’) for the
there is too much lead on it
present. You should find that the DMM will give
so stray capacitance could
distort the reading. However,
a relatively low reading - less than 10-15mV.
This reading is due to the fact that the ‘stray’ we got away with it in this case
– as you can see, the capacitor
capacitance of the Adaptor’s input binding posts
is labelled “6” (6pF) and the
is not as yet being nulled by trimpot VC1. So
DMM is reading 6.08pF
the next step is to use a small plastic or ceramic
alignment tool to adjust VC1 very carefully to
get a minimum or ‘null’ in the DMM’s reading. You should actual value in picofarads. For example, if the capacitor
be able to bring the reading down to below 1mV.
has a known value of 1.013nF or 1013pF, adjust VR2 until
If you are able to achieve this null, your Adaptor is very the DMM reading is 1.013V.
likely to be working correctly and the next step is to caliFinally repeat the process again for Range C, this time
brate each of the three ranges.
using the 10nF reference capacitor and trimpot VR1 to
For the three calibration steps you’re going to need three make the adjustment. The correct setting for this range is
polystyrene, polyester or silvered mica capacitors whose where the DMM reading in millivolts corresponds to the
values are accurately known, because the accuracy of your capacitor’s actual value in tens of picofarads. For example
Adaptor will depend upon them. The three capacitors if the capacitor has a value of 9.998nF, the DMM reading
should have values close to 100pF, 1nF and 10nF respec- should be 999.8mV or 0.9998V.
tively, because these are the nominal full-scale readings of
That’s all there is to it. Once you have calibrated each
the Adaptor’s three ranges.
range in this way, you can turn off the Adaptor using S1,
They needn’t have these exact values but ideally you remove the knob from its spindle and then fit the top of the
should know their actual values, measured using a cali- box carefully - making sure you don’t catch the battery snap
brated digital capacitance meter or LCR meter.
wires under the side. Then turn the complete box over and
Once you have these three known-value or ‘reference’ fit the four countersunk head screws used to fasten the top
capacitors the calibration of your Adaptor is relatively to the bottom. After this all that should remain is to refit
straightforward.
the knob to the spindle of S1.
With the Adaptor still switched on and set to Range A,
Just before you declare your Adaptor ready for use,
first connect the 100pF capacitor to the Adaptor’s binding though, it’s a good idea to check the setting of null trimcap
posts using the shortest possible lead lengths. Then adjust VC1, because the stray capacitance associated with the intrimpot VR3 until the DMM reading in tens of millivolts put binding posts does tend to change very slightly when
corresponds to the capacitor’s actual value in tenths of a the box is fully assembled.
picofarad (pF). For example, if your
capacitor has a known value of 101.5pF,
adjust VR3 until the DMM reading becomes 1015mV or 1.015V.
Once this is done you repeat this process on Range B, this time using the 1nF
reference capacitor and trimpot VR2
+
+
OFF
to make the adjustment. VR2 should
ZERO
be adjusted until the DMM reading in
A 1mV = 0.1pF
millivolts corresponds to the capacitor’s
–
C 1mV = 10pF
C
B
A
CALIBRATE
siliconchip.com.au
So switch the Adaptor on again, in Range A but with
nothing connected to the binding posts and if necessary
adjust VC1 using the alignment tool (passing down through
the ZERO NULL hole in the front panel) to see if you can
improve the null reading on the DMM.
Using the Adaptor
Putting the Adaptor to use is also quite straightforward.
Basically it’s just a matter of hooking it up to your DMM,
setting the DMM to the 0-2V DC range and then turning on
the Adaptor to the appropriate range and connecting the
capacitor to be measured to its binding posts. Then you
read the voltage on the DMM and convert this to find the
capacitance, using the legends printed on the Adaptor’s
front panel.
But there are a few things to bear in mind if you want to
achieve the best measurement accuracy. For example when
you are measuring really low value capacitors in particular
(ie, below 100pF), try to connect them to the binding posts
with the shortest possible lead length. This is because any
excess lead length will add extra stray capacitance, as well
as a tiny amount of lead inductance. Both of these will degrade reading accuracy, because measurements on Range
A are done at a frequency of about 110kHz.
If you can’t connect a capacitor directly to the binding
posts with minimum lead lengths, an alternative is to make
up a pair of short but stiff (ie, heavy gauge) test leads, each
with a banana plug at one end and a small crocodile clip at
the other. The leads should then be plugged into the binding posts, and zero null trimcap VC1 then adjusted with
an alignment tool (on Range A) to null out the additional
stray capacitance.
Then you can connect the capacitor to the test lead clips
and measure its capacitance as before.
You can follow a similar procedure to use the Adaptor
as a handheld ‘probe’ to measure stray capacitance, as opposed to measuring the value of discrete capacitors. Here
it’s a good idea to make up a small ‘probe tip’ out of a 30mm
length of 4mm (5/32”) diameter brass rod (eg, brazing rod),
with a fairly sharp point ground or filed at one end and
the other end slit down the centre with a fine hacksaw for
about 8-10mm. The slit end can then be expanded slightly
with a small screwdriver, so that it will just slip inside the
socket on the front of the Adaptor’s positive (red) binding
post and stay in position. You also need to make up a short
but stiff test lead for the ‘earth return’, with a spade lug at
one end (to be clamped under the negative binding post)
and a small crocodile clip on the other end to connect to
the reference metalwork for the stray capacitance to be
measured. The probe tip and earth return lead I made up
are visible in one of the photos.
Here again you need to null out the additional stray
capacitance associated with the added probe tip and earth
return lead, before making the actual measurement. But
this is again easy to do: simply fit the probe tip and earth
return lead, turn on the Adaptor to Range A and adjust VC1
with an alignment tool for the deepest null in the DMM
reading. Then you can proceed to make your measurements
of stray capacitance.
Get the idea? It’s quite in order to use test leads and/or
measuring jig attachments to connect whatever capacitance
you want to measure to the Adaptor’s binding posts, providing you null out the added stray capacitance using VC1 (on
Range A) before making the actual measurements.
SC
Custom Battery Packs,
Power Electronics & Chargers
For more information, contact
SIOMAR BATTERY ENGINEERING
Phone (08) 9302 5444 or email mark<at>siomar.com
www.batterybook.com
siliconchip.com.au
March 2010 77
|