Engineer‎ > ‎Electronics‎ > ‎

Electronics - Electronics

Created; 05/02/2015, Changed; 27/04/2018, 28/04/2018
Previous page; Mechanical Design for good RF

Compensation - Ensure that the circuit is compensated properly and is therefore stable;

Below is an example of a super regenerative radio receiver (type of TRF Tuned Radio Frequency receiver). I made something like this in the 1970's as a boy and it worked well with reasonable sensitivity. But I did not fit the reaction capacitor (C2) in fear of there being too much positive feed back and the circuit becoming a radio transmitter. This is an example of a circuit that is designed to be almost unstable, not only is their positive feedback in the RF path via C2, but there is also positive gain feedback in the audio signal path from the detector fed into the transistor base. Diodes and transistor would be low junction voltage drop germanium types such as diode 0A91, and Transistor OC44. The radio is therefore quite selective with this arrangement. The circuit below is from memory and may be inaccurate, the original circuit can be found at; http://www.mds975.co.uk/Content/trfradios02.html

Particularly of note is that although this design uses few components but it is not a simple design. Take care when developing with a minimal component count, elegant design because invariably there will be complex interaction, be difficult to fault find and the modelling will be complex. Therefore trade off elegance of design with simplicity of function many blocks to suit your ability and the requirement. On the other hand there can also be benefits in minimal component usage and dealing with an issue early in the signal path when it is small. This circuit really is not complicated but I have use it to demonstrate the point in the same way that I did for the camera and it's lens on the first page.


Oscillator & stability

Circuit oscillation or instability occurs when there is a phase shift of 360', and the loop gain is at least one. The circuit below is a phase-shift oscillator but such circuits don't always oscillate - I tried and failed with a similar design as a boy. The timing capacitors and resistors are transposed for a BJT shown in the second circuit below (because the feedback is current for a BJT). In first circuit which is voltage feedback (rather current) based you can sort of see one capacitor charging then the next and then the last capacitor charging. Phase shift oscillators are sometimes formed inadvertently see Amplifier below.

If you analysis the circuit as a DC circuit there is negative feedback (180' phase shift). When you include the capacitors and resistors, they can produce up to a 90' phase shift each, this is an important point. The circuit oscillates at a frequency that occurs when the sum of each phase shifting component, which includes propagation delay within the transistor, and the gain is greater than one. We can stabilise this circuit, that is prevent it from oscillating, by increasing or reduce in the second case the value of one capacitor, say C1, to the point were it's phase shift (reactants) predominates, and there can no longer be as much as  180' phase shift and a gain greater than one. The phase shift would be nearer to 90' when C1 for example predominates.
Phase shift oscillator (LT Spice AL-0024-03B); This circuit above is a variation of the standard circuit below.
 I find this circuit easier to see how it works M1 could be a BJT + series input resistor
 but I tried that and there is insufficient loop gain for the circuit to maintain oscillation.


Phase shift oscillator (LT Spice AL-0024-02B); This is the standard circuit.
 The feedback is current rather than voltage as in the circuit above - But like the AC
current and voltage on a previous page I find it more difficult to visualise how it works. 
The BJT has just enough loop gain for the circuit to maintain oscillation.

Operational Amplifier    

The circuit below (AL-0001-01A) is an Inverting amplifier has input and output filtering, formed of R6, C2, and R3, C3. Also R1 and C1 form an active filter. R2 and R3 have been chosen to be high enough to avoid amplifier instability.

The operational amplifier is unconditionally stable, that is it is stable with R1 = 0, maximum feedback factor, which is where the amplifier gain would be zero. This circuit may become unstable though with R3 = 0 so that C3 is directly connected to the amplifier output, in this case the propagation delay is increased but the loop gain would not have reduced enough for the loop gain to remain below one. If the value of C3 is reduced below 1nF this particular amplifier would become stable again. See individual data sheets for the operational amplifier or any other amplifying component to assess stability.

It is also likely that this amplifier would become stable (R3 = 0 as before) with C3 increased to 100uF or more as the loop gain would again be reduced at low frequency. This is because the operational-amplifiers behave like transconductance amplifiers in that there gain is reduced when driving a large load. The other condition where the amplifier becomes unstable is where R2 = 0, it is important that capacitive loading be kept away from an amplifiers summing junction (the negative input).


The consequence of adding capacitance to the output or input of an op-amp will be that the amplifier will amplify transients so the output overshoots, causing the output to be less accurate.

Alternatively adding large value capacitors is done with voltage regulators, but you may need a small series resistance to prevent the capacitor Q from being too high, and causing the amplifier section to become unstable.

If their are varying currents flow through the ground or high EM fields nearby, Adding capacitors to inputs and outputs is not the proper solution. Increase the distance, sort the varying currents, and add electrostatic (aluminium/copper/gold for HF) or magnetic (soft iron for Low Frequency) screening, are good strategies.The best approach to avoiding varying ground currents injecting noise is to use differential pair paths for small signals particularly.

The circuit below is as above but modified for a capacitive input device. It is a zero bias voltage, current amplifier. R2 has little bearing on the design equation but largely removes the diode capacitance from the summing junction in order to make the circuit very stable; (AL-0004-01a)

Be careful of symbolic circuits (circuit diagram show to illustrate but does not have all the detail) such as those showing photo diodes connected directly to an operational amplifier's summing junction (Negative input). Photo diodes do have capacitance and a series resistor, R2 that I have added, would have little effect on the design equation but reduces capacitance directly connected to the amplifier. The resistor provides an AC summing junction at the -input with the feedback capacitor (which you usually must have to meet amplifier stability criteria), and the DC summing junction at the photo diode.

In this low frequency amplifier the biggest contribute to noise would be the feed back resistor R1, therefore chose a low ppm/'c type this would most likely happen to be a 0.1% tolerance part. Some of the high performance photo cells made by Hamamatsu (better photo diode than others in 2005) and Centronic currently are better (at the time of writing 2014) of say 10mm2 contribute lower noise than the resistor in this low noise zero bias voltage configuration. In the past a PIN diode, although not designed for this configuration, may have performed better than a PN diode with zero bias and there are other more modern low capacitance diodes with 100pF rather than 1-2nF at zero bias. Speak to the maker and ask them to advise you rather than entirely rely on data-sheet comparison.

 

This link is to the stability graph for a low speed, low power op-amp TSV71x ST, operates at 15uA and 150KHz  

https://community.st.com/servlet/JiveServlet/showImage/2-143959-13207/RISO_TSV711.png

https://community.st.com/thread/35976-tsv71124-op-amp-output-resistor-for-stabilty

Micropower op-amps that some operate at about 1uA and have a bandwidth of about 10KHz can become unstable with output capacitance of only 30pF in the case of Microchip MCP603x. That would occur with a very short PCB track length. Microchip suggested a solution is based on a similar graph to the one above using a 1K resistor in series with a 1uF capacitor network。 See figure 4-3 of the Microchip MCP603x data-sheet;

http://ww1.microchip.com/downloads/en/devicedoc/22041b.pdf

ST's TSU10x  and TSU111 micropower op-amps can drive 90pF or 100pF without a series resistor but similar the PCB track would have to be kept quite short.

Faster (15MHz) and higher operating current FET or MOSFET input op-amps are usually the best all round choice having both low noise, accuracy and stability.

MOSFET input op-amps have a very low input current but this increases with temperature. Since the early 1970's probably the best over a wide temperature range have been the LM308 and its successor LM11C types these had a compensation pin so that you could add a capacitor to increase the amplifier's stability but reduce the op-amps bandwidth. The modern equivalent do not have the a compensation pins brought out so you can not over compensate these amplifiers. This is probably a good thing because you could introduce errors because the amplifier could appear to work with a high level of noise that would be rectified and cause an offset error.

The link below is to a TI, LMP2234 it's input is bipolar transistor which imposes limitations but the input offset current is the lowest of any I have discussed. 

http://www.ti.com/product/LMP2234?keyMatch=LMP2234B&tisearch=Search-EN-Everything

Analog Devices MOSFET input 10uA op-amp type ADA4505-4 quoted stability with RL=10K up to 1nF but they do not provide a stability curve. The op-amp is similar to the first op-amp and operates at 10uA.


Radio - Intermediate Frequency Amplifier (I.F. stage)

This is interesting lesson from my first year at collage - C4 looks like it is providing positive feedback, but due to the propagation delay through the germanium transistor (storage time) of this VHF radio I.F. stage operates at a frequency of 10.7MHz, the feedback is actually negative feedback. The transistor itself has capacitance and miller capacitance between the Collector and Base would become positive feedback not negative feedback expected, is cancelled out by the addition of C4.

Useful links;

Component manufactures websites.

http://www.freescale.com/ Very good micro-controllers such as HCS12. Also specialised complex ICs. 

http://www.st.com/stonline/ ST microelectronics Switch mode power supply ICs (which they invented) including voltage mode, Stepper motor drivers.

http://www.nxp.com/ Philips - Switch mode power supply ICs including voltage mode types.

http://www.linear.com/ Linear Technology Current mode switch mode power supplies, and design spreadsheets. See LT1072 (obsolete)

http://www.diodes.com/ Zetex design spreadsheets. A few unique discrete transistors developed from Ferranti E line series.

http://www.coilcraft.com/ very good power density wound components and very good design spreadsheets.

http://www.we-online.com/ Wurth provide good advice on EMC issues.

https://www.fairchildsemi.com/ A good range of basic parts - transistors, logic, power supply for example.

EPCOS make common mode chokes that don't have split magnetic but the bobbin is split put together on the ferrite and then the coil wire is wound by spinning on the bobbin on the ferrite arm. This should minimise EM field leakage Wurth, Coil Craft and Vishey have some very high current surface mounded Inductors in which the ferrite material is molded around the conductor therefore similarly minimising the EM field, copper losses and size.