Chem. 407
electronics experiment
edited: October 18, 2000... document chem407_elect.htm
A few comments on selected portions of the electronics experiment
1. Gain of a simple inverting amplifier
- the voltage gain is defined as Vout / Vin
- the design rules say gain = - Rf/Rin
- In practice there are a couple of errors that affect the results
- the resistors we use are +/- 1% so they should not cause errors in excess of 2%
- the amplifiers often have a small offset error of ,say, 0.3 millivolts
- that is, with zero volts in, the device behave like 0.3 mV are present
- at large gain, this can end up a significant error
- the error is significant when you are measuring small signals
- likewise, the voltmeter or the oscilloscope often has a small zero offset
- that is, it doesn't quite start at 0.000 volts
- again, it's a minor error in a 1 volt signal, significant for a 1 mV signal
- Both of these errors represent a constant contribution to the circuit
- so the best working definition of gain is
- gain = (change in) Vout / (change in) Vin
- that is, the slope of a Vout vs. Vin graph.
Noise and Filtering
The thermocouple system involves a relatively low voltage signal
of a few millivolts. It also involves a relatively long loop of
wires that can act as an antenna, picking up small signals of other
types. For example, it is easy to pick up 60Hz and 120 Hz signals
from the power lines in the room. There are also some other signals
generated when electrical wires are flexed or when electrical contacts
are pulled or bumped. These undesired extra signals are considered
electrical noise. In a typical situation, this noise may be in the
neighborhood of a few hundred microvolts.
If the thermocouple signal were large, this noise would be negligible.
However, the thermocouple response is in the neighborhood of
500 microvolts per degree C, so the noise represents an uncertainty
of 1-2 degrees C.
Fortunately, the interesting thermocouple signal differs from the
noise in a substantial manner. The thermocouple signal is steady,
changing only slowly over a few seconds or even minutes.
The noise changes every few milliseconds. What we need is a
frequency sensitive circuit that allows the DC and very low
frequency signals to pass but stops or reduces signals of higher
frequency. Technically this is called a Low Pass Filter.
We can implement such a filter by adding a capacitor across the
feedback resistor of our amplifier. A capacitor acts like a
low resistance element for high frequency signals.
- This means that the ratio of Rf/Rin is small and the amplifier
has very little gain.
- Typically the gain is less than one so the signal
is actually reduced by the amplifier
A capacitor behaves differently with a DC signal
- The capacitor momentarily passes a current, but then it is charged
and no further current passes
- Therefore, the amplifier behavior is fixed by the feedback
resistor value
- If this is large, then the circuit will amplify the DC signal
the net result...
- the AC components decrease or vanish
- the DC component (temperature signal) dominates the
amplifier output
- the signal to noise ratio is greatly improved
- small temperature changes are no longer lost in the noisy response.
the pH meter circuit
This is an example that shows the advantage of taking a step by step
approach to design and assembly.
Basically there are four rather distinct features or stages
of our pH meter and all must work properly
- the electrode and the pH samples (buffers)
- that is, we need a signal from the electrode
- the voltage follower circuit
- the choice of gain so we get 100 mV per 1.0 pH unit
- a voltage offset so we get the meter to read the actual pH
If any of these components isn't working properly
(or connected together properly) the device won't work as intended.
There are too many steps to hope we won't have an error.
Instead, let's try to build this circuit in a series of short steps, each with a built in test.
1. Let's test the electrode.
- Connect the pH electrode into a suitable plug.
- Plug this into the digital volt meter
- Set the meter to Voltage, 200 mV full scale, DC.
- Place the electrode in one pH buffer and note the voltage
reading.
- Remove and rinse the electrode, transferring it to another
buffer
- You should see a definite change in voltage as the pH
of the buffer changes
- Technically we expect 0.059V per pH change
- E.g., going from a pH 4.01 to a pH 6.98 buffer
should produce about 0.18 volt change
- In practice you'll only see about 20-30% of that value,
but it should be obvious and reproducible.
If you don't see a change you either have the meter set wrong, a malfunctioning electrode, bad connections or bad buffers. Find out which before proceeding.
2. Check the voltage follower circuit and the electrode response
- move the electrode connector to the voltage follower circuit
- this must be set up on the FET op amp unit
- connect the output to the digital meter
- again, note the reading in two different buffers
- in this case, the change should be within 10-20%
of 0.059 x (delta)pH
If this isn't working, find out what's wrong before doing part 3.
3. Setting up an amplifier with the desired gain
- The notes suggest the use of a 59K input resistor and
a 100K feedback resistor
- send the output of the voltage follower into this circuit
- send the output to the digital voltmeter again
- move the electrode between buffers
- you should now see 100 mV change for every 1 pH change in the buffers
4. Adding a suitable voltage offset so the meter reads pH directly
- set up the additional input resistor
- connect it to the DC millivolt source
- start with the mV source set to zero
- decide if you need to increase or decrease the output voltage.
- add voltage until the output reading matches the
pH of the buffer
- e.g.., reads 401 mV in a pH 4.01 buffer
return to home page