SCHEMATRONICS

The RF sensing alarm would also have to be self-contained, which means without any hookups to my radios. After a bit of reading and thinking, I came up with a solution that meets all the initial objectives. Here's circuit in detail.



The circuit shown above may look scary for some of you, but it is not. It can be broken down into four stages. Let's look at them one at a time. The first stage acts as a RF sensor circuit. It is made of U1C, one of the four operational amplifiers of a LM324 chip, and its associated input circuitry. U1C is used as a voltage comparator. Note that the two U1C inputs (plus and minus) have similar DC circuits connected to them. The plus input has R7, R8 and D3, and the minus input has R6, R10 and D5. In these two circuits, D3 and D4 are partly biased (about 200 mV of forward voltage) in order to better exploit the variation of voltage versus current that the diode produces. This translates into increased RF sensitivity.

More circuit description in detail

The schematic diagram for the simulator. Closing the switch engages the simulator. Turning the knob clockwise simulates a lean condition, turns the LED off, and the car should start running rich to compensate. The big "V" is a digital voltmeter(not shown in the pictures). Using a smaller value for C1, perhaps 4.7 uF, will make the circuit oscillate faster and might be more like a real oxygen sensor(a new sensor switches more often than an old one).





The adapter cable. Note the connector recycled from an old oxygen sensor. Hard to see under the black tape: 100K resistor.





More about Oxygen Sensor Simulator

The ESR meter uses 8 operational ainplifiers. An op-amp is an idealized basic amplifier with two inputs. The non-inverting input (+) has an in-phase relationship with the op-amp output, and the inverting input (-) an out-of-phase relationship. Op-amps are usually used with negative feedback and reach a stable operating condition when their two inputs are equal in voltage.


Op-amps IA & 1B form a regenerative 100 kHz oscillatnr circuit. Capacitor C1 is the basic tiining capacitor and RI is selected to set frequency. Diodes D2 & D3 clip the bottom and top of the output waveform so that the output level and frequency are resistant to battery voltage changes.

The oscillator output of op-amp 1B drives 10-ohm source resistor R8F. The test-capacitor, thru the test leads, couples this 100 kllz signal to 10-ohm load resistor R9F. The amount of voltage developed here is indicative of the capacitors ESR value. (The 10-ohm resistors determine the basic iieter scaling.)

Capacitor C3 blocks any DC voltage present on the test-capacitor. Diodes D4 & D5 protect the ESR Meter from any initial charging current to C3. Resistor R7 discharges C3 after test.

A DC operating bias of 0.55 V is established by diode D1 for the oscillator stage and for all subsequent stages, which are DCcoupled and operated class A. DC bias from D1 and ESR signal from R9F are combined at the input of op-amp 1D. Both voltages are amplified by 1D, 1C, & 2A. Each of these three stages has an amplification factor of about 2.8 due to the ratio of output-voltage to feedback~voltage at the (-) input, which is determined -by feedback resistors R13F & R14F, etc.

Op-amp 2D is configured as a peak-to-peak detector. when the in-corning AC signal goes more positive than the normal bias level of about 0.77 Volt, the output of 2D also goes positive. But it must go positive enough to overcome the voltage drop across diode D6 before a fully equalizing positive voltage can be fed back to the -(-) input thru R20 to stabilize the op-amp.

-Capacitor C4 is charged to the peak value of the AC signal and accurately represents the peak of the incoming AC signal. The voltage drop across the diode becomes almost inconsequential due to the feedback process, and the circuit works down to a few mV.

A similar action occurs during the negative peak, using D7 & C5.

Resistor R21 provides a constant minimum amount of negative feed--back around op-amp 2D. The negative feedback increases the op-amp bandwidth which, most importantly, keeps the amplifier input-to-output phase-shift low enough for proper circuit operation.

The two outputs from the peak-to-peak detector are connected to two high-input-impedance unity-gain DC amplifiers, which drive the 1 mA meter movement differentially.

Source

Juste because it's damn small does not mean that you have to settle down for second best when it comes to performance. The repeatability is around 0.01 pH and the accuracy, while depending on how well you will calibrate it, is around 0.02 pH. The main characteristics are:

• very small footprint (11cm2)
• very lightweight
• pluggable module for easy replacement
• requires only an external transformer and a display unit to work
• slope/offset settings
• repeatability 0.01 pH
• accuracy 0.02 pH
• low power
• low-cost single-sided PCB
• total unit price (including case and display unit): less than 100 euros.
  

The circuit input is pin 15 of K1. The probe signal enters IC1 via an RC circuit designed to allow only relatively slow signal variations (and avoid getting parasite HF signals). IC1 is a CMOS op-amp and thus has a very high impedance. The gain of IC1 is adjusted with the potentiometer R14. C2 is there for the amplifier stability. The R5/R11 circuit is the adjustment of the amplifier offset which is necessary for a high-precision application like this (see calibration below).

Once the signal has been amplified it enters an offset circuit built around IC2. IC2 is a more classic TL081 op-amp commonly found in audio devices, among others. The offset is defined by two potentiometers R12 and R13. The first one is on the PCB and the second one on the front panel. This improvement on the original design (single pot) allows the range swept by R13 to be symmetric, albeit smaller than without R12. It can be skipped if you wish (those small SMD trimmers can be damn expensive...). The circuit is designed to provide an average offset of 2V.

After the offset circuit the signal passes through a voltage divider before reaching the display unit. The divider roughly changes the signal range to something that is acceptable for the display. The real setting will be done on the display itself which contains a multiturn trimmer to precisely adjust its input gain.

The voltages for the signal evolve in the circuit as follows:

• Before IC1: -0.414/+0.414V (this might depend on the electrode used and its age, hence the gain/offset control)
• After IC1: -2/+2V
• After IC2: 0-4V
• After the voltage divider: 0-140mV (roughly)
• After the on-display trimmer: 0-140mV
• On the display: 0.00 - 14.00 pH (the display measures mV but the decimal point is placed accordingly to show a 0-14pH range)

As you can see the electrode voltage is symmetric and must undergo a linear transformation to fit the 0-14 pH range. This is all very classic stuff... Note that even if the supply rails are at +/-5V the circuit can cope with a 0-4V signal because the output swing is almost equal to the rails (no 0.7V drop, more around 0.3V IIRC).

A little remark concerning the integrated power supply circuit: it is a very small circuit that supplies a maximum of 50mA. Be careful of you want to add a power LED or something like that as it might be too much for the circuit. Check the total power used by the circuit before adding extras.

PCB
The PCB is very small and you are advised to build it with through-hole mounting components if you're not familiar with SMDs. That means start the PCB design from scratch. I personally think that it looks much cooler with a small footprint... No other special remarks concerning the PCB, except that the PCBs that were manufactured were slightly different (see the photos below). This is actually also true for the schematic. No functional difference, but I changed from Protel to Eagle for designing the circuit so I had to reenter the schematic and PCB manually. Hence some differences in layout but this is not a big deal.


Component list
This is the list of components used in this circuit. I only mention the display and probe at this time as the other components are generic. Maybe more info will follow in the future.

A little link to the display unit used in this project. I chose this one because it has a nice 'pH' unit that can be activated on the display.

Another link to the probe used with this circuit (IIRC). Most probes should work but I only tested the circuit with this one.
Construction

Random remarks: start with the smallest components, go slow, don't forget to set all the solder bridges correctly on the display unit (what you want is a 0-200mV range, an appropriately set dot and 'pH' shown as the unit). Check your cables,... before powering up.

The cabling diagram of K1 is:
* 1: AC 1
* 2: GND
* 3: +5v OUT (to display)
* 4: SIGNAL OUT (to display)
* 5: R13 / 2
* 6: R14 / 1
* 7: -5V OUT
* 8: R14 / 3
* 9: AC 2
* 10: GND (from transformer)
* 11: GND (to display)
* 12: GND (to BNC input)
* 13: R13 / 1
* 14: R14 / 2
* 15: INPUT

Source: ©Damien Douxchamps