Like almost all such ‘designs’, they are published without supporting measurements or simulations.
The transformer is intended to be used with a load such that the input impedance Zin is approximately 50+j0Ω, Gin=0.02S.
Analysis of a simple model of the transformer with a load such that input impedance is 50+j0Ω gives insight into likely core losses.
Let us calculate the magnetising admittance of the 2t primary at 7MHz. The core is a stack of 3 x FT240-52 ferrite toroids.
Gcore is the real part of Y, 0.00104S.
If Yin of the loaded transformer is 0.02S, we can calculate the core efficiency as 1-Gcore/Gin=1-0.00104/0.02=94.8%, core loss is 0.23dB.
Whilst this might be smaller than many similar designs, it must be considered in the context of the 500W CW rating.
The average power of A1 Morse code modulation is about 44%, so the average power of a 500W CW transmitter is about 220W, and 5% dissipation is 11W average which is probably within the capability of the cores, especially with ventilation as shown.
Losses in the matching transformer (in this case 25W of the 500W CW input) are only part of the total system loss, and overall system efficiency will be lower than estimated here for the transformer alone.
]]>
A LED power meter that I had ordered finally arrived (slow boat from China syndrome).
Above, the upper rail contains a RCD, the power meter which displays Volts, Amps, and kW, or pf, hours, and kWh, a DIN mount terminal block for mains, and a 40A SSR on a heatsink. A clip on CT can be used for oscilloscope observation of mains current.
The lower rail has a DIN mount terminal block for signals, a switched mode 12V PSU and piggybacked in the black ABS enclosure is a DC-DC converter to supply 24V for the 4-20mA devices, a DIN rail mount Pt100 4-20mA converter, and a hcctl prototype (with flying lead for bootloader programming of EEPROM calibration constants and firmware).
Above is a generic PID controller zip tied to a DIN rail bracket made for SSRs.
Above is a DIN mount relay, I have these in a range of DC and AC voltages to test various project designs.
Above is a small opto isolated 5V relay board with inverting and non-inverting logic level drive. The relay PCB is mounted on a small piece of acrylic sheet which is in turn fixed to a couple of DIN rail PCB clips.
Above is a generic PCB vice drilled and screwed to a DIN rail bracket made for SSRs.
Above, one of several small switched mode mains power supplies in different voltages, and fixed to a DIN rail bracket made for SSRs.
A warning: don’t mess with this stuff unless you are competent, don’t take safety for granted, don’t allow access to persons who might not have the needed competency.
]]>Over recent years to 2002, the number of issued amateur licences was declining, the trend was about 2.8% pa decline over the five years to 2002.
This has concerned some people, who took the view that the decline was a harbinger of the impending demise of Amateur Radio.
Figure 1 shows the recent history of issued Australian Amateur licences.
The licence numbers shown in the graphs at Oct 2005 are aggregated on their equivalent type under the changed Amateur LCD which came into effect shortly afterwards. Note also that the figures include clubs licences, beacons, and repeaters. Figures post April 2007 are not broken down as the breakdown has not been published by the WIA or any other organisation.
Clearly the growth rates of almost 6% pa attained up to April 2007 have not been maintained since then. Growth in all amateur licences June 2007 to June 2008 according to the ACMA 2008 Annual Report was much lower at 1.8%pa. It is interesting that of the 858 Certificates of Proficiency issued, only 12% of the certificates were Advanced, a strong indicator of reduction of Advanced licences from 85% of the licence base at October 2005. It is also notable that although 858 certificates were issued, and at least the 588 Foundation certificates represented potentially new licencees, licence numbers grew by only 269 for the same year. The underlying attrition continues, and it is likely to be mainly in the Advanced grade as amateur radio is dumbed down to a populous social hobby and Foundation Licencees reshape ham radio.
The WIA published a breakdown of licences at 1 April 2007. Note that it is difficult to reconcile the WIA’s published figures with the ACMA’s annual reports.
On the WIA’s published figures, the number of Advanced equivalent licences have declined at a rate of 1.33% pa from 04/10/05 to 01/04/07. This includes club licences which are usually pretty stable, so the number of individually held Advanced licenses is likely to have fallen by even more than 1.33% pa. It is interesting to note that in that category are the then Limited and Intermediate licencees who obtained increased band access with the recent licence reforms.
So whilst we have gained 1186 Foundation licences and 85 Standard licences, we have lost 235 Advanced licences.
Figure 2 expands the licence statistics for the last two observations, and it shows the 4.08%pa increase in overall numbers, and the 2.36%pa decrease in Advanced Licences.
The information on Australia’s main amateur radio training site Radio & Electronics School suggests that students should expect to be able to complete the Advanced course in about 66 hours in addition to the regulations and practical components which suggests about 80 hours of training activity in total for the Advanced licence. By contrast, the training investment in the Foundation licence is more like 6 hours.
When the increases and decreases are weighted by the relevant skill level or training investment, we have taken a step backwards.
That should be a concern to all amateurs, and to the WIA.
But apparently the WIA sees this as an unqualified success. A story Australia reverses amateur radio’s downward trend – Amateur Numbers continue to grow on Southgate ARC and attributed to the WIA calls this out as a success “For those who have not yet heard, the great news is out. We have finally reversed the trend of our declining amateur population and we are now in positive growth mode!! “.
The graphs above have been updated for new data from the latest ACMA Annual Report.
The ACMA’s annual report reveals that in FY 2016/17, there was a increase in total amateur licences was just 15 licences or 0.1%. The more interesting question is the dissection of the total into the numbers for each grade and the trend.
Fig 3 shows the total issued licences over the years since the introduction of the licence reform which introduced the Foundation licence and gave significant additional privileges to all existing licencees except the ‘full’ licencees.
Fitting a polynomial curve to the data gives some insight into where it might be heading. My forecast in 2012 based on that model was that licence numbers would peak between 15,000 and 16,000 around 2014 and to then resume the decline that was apparent in the 1990s.
That has indeed happened, licences peaked at 15,760 in 2012, and the decline is not only evident, though this year there was a small increase.
Whilst the number of licenced amateurs is not a valid indicator of the health of the hobby, this trend will probably cause hand wringing in some quarters, especially the WIA whose economic viability is linked to subscription numbers and we can expect another wave of dumbing down to try to prop up membership.
Data prior to 2001 was from IARU statistics published on their website. Data from 2001 to 2004 is from the ACA Annual Report 2004, Oct 2005, Aug 2006 data is from the WIA website, April 2007 data is from the WIA AGM reports.
Data after April 2007 is from ACMA Annual Reports.
Owen Duffy 05 November 2017
]]>Is a ham transmitter conjugate matched to its load?
The answer speaks to the relevance of Walt Maxwell’s Conjugate Mirror proposition to ham stations.
If an antenna system is carefully adjusted (eg using an ATU) so that the transmitter sees a load with extremely low VSWR(50), the load is almost exactly 50+j0Ω.
It is conjugate matched to the source is the source is well represented by a Thevenin source with equivalent source impedance equal to 50+j0Ω.
Although ham transmitters are commonly designed and specified to deliver output power to a nominal 50Ω load, that does not necessarily mean that they are well represented by a Thevenin source with equivalent source impedance equal to 50+j0Ω. Indeed it would be most unusual for a HF transmitter to do so, but may be the case for microwave transmitters that use a suitably terminated circulator or isolator at their output. (Hams have a tendency to appropriate well known terms to different meanings, and isolator above does not mean a common mode choke as often meant by hams.)
It is important to understand that source impedance and load impedance are not referring to the same thing, they are two distinct entities, and ham transmitters are typically specified to suit a nominal load impedance, but most unlikely to specify the source impedance.
So, even if you have an antenna system with extremely low loss presently an extremely low VSWR, it is unlikely Walt Maxwell’s Conjugate Mirror utopia applies because of source mismatch.
Should we have expected this outcome?
Let us solve a very similar problem analytically where measurement errors do not contribute to the outcome.
Taking the load impedance to be the same 10.1+j0.2Ω, and calculating for a T match similar to the MFJ-949E (assuming L=26µH, QL=200, and ideal capacitors) we can find a near perfect match.
The capacitors are 177.2 and 92.93pF for the match.
Now turning the network around by swapping the capacitors and changing the load to 50+j0Ω.
Above, the impedance looking back towards the 50Ω load is 17.28-j0.71Ω, which is quite close to the value obtained by measurement, 18.0-j0.8Ω (which is dependent on the actual Q of the ATU elements).
So, in answer to the question Should we have expected this outcome?
, the answer is yes, it is not surprising and quite similar to what we might expect from a network of this type.
Walt Maxwell’s Conjugate Mirror (Maxwell 2001 24.5) which imbues a magic system wide conjugate match with certain benefits, a utopia, which does not apply to systems that include any loss, it does not apply to real world systems. Maxwell does not state that limitation of his proposition.
Is a ham transmitter conjugate matched to its load? Watch for a follow up post.
One of the most useful (and sometimes astonishing) principles in radio is the Conjugate Match theorem. In the simplest terms, what this says is that the maximum power will be transferred between a source (like a transmitter) and a load (like an antenna), when the source impedance is the COMPLEX CONJUGATE of the load impedance (or vice versa).
Here’s a neat little experiment to prove the conjugate match theorem. You need four basic ingredients: an antenna analyzer like the MFJ259 (or an actual impedance bridge, if you know how to use one). A good low loss antenna tuner. A good 50 ohm resistor. And a good 200 ohm resistor. And some appropriate connecting hardware, namely some short bits of coax.Step 1) connect the 50 ohm resistor to the OUTPUT of the antenna tuner. Connect the antenna analyzer to the INPUT of the antenna tuner.
Step 2) Adjust the antenna tuner to get precisely 50 ohms, zero reactance on the antenna analyzer. This step simply confirms everything is working.
Step 3) Replace the 50 ohm resistor with the 200 ohm resistor. Readjust the antenna tuner to get 50 ohms, zero reactance on the antenna analyzer. Do not disturb the antenna tuner adjustments after this point.
Step 4) Remove the 200 ohm resistor and insert the antenna analyzer in its place (at the OUTPUT of the antenna tuner).
Step 5) Insert the 50 ohm resistor at the INPUT of the antenna tuner.
Step 6) Take a careful reading of the antenna analyzer. (What do you think it will say?)
10 points for anyone who will correctly explain why this works.
Jacobi published his maximum power transfer theorem in 1840. It states that maximum power is transferred from a (Thevenin) source to a load when the load resistance is equal to the (Thevenin equivalent) source resistance.
It was later adapted to apply to AC circuits with sinusoidal excitation, maximum power is transferred from a (Thevenin) source to a load when the load impedance is the complex conjugate of the (Thevenin equivalent) source impedance.
(Maxwell 2001 24.5) states
To expand on this definition, conjugate match means that if in one direction from a junction the impedance has the dimensions R + jX, then in the opposite direction the impedance will have the dimensions R − jX. Further paraphrasing of the theorem, when a conjugate match is accomplished at any of the junctions in the system, any reactance appearing at any junction is canceled by an equal and opposite reactance, which also includes any reactance appearing in the load, such as a non-resonant antenna. This reactance cancellation results in a net system reactance of zero, establishing resonance in the entire system. In this resonant condition the source delivers its maximum available power to the load. …(1)
Note that it states that if a conjugate match is established an any junction, then a conjugate match occurs in any (all) other junctions, simultaneously a conjugate match exists everywhere.
KL7AJ’s experiment never actually directly measures power transfer, much less that it is maximised.
What it does do is create a conjugate match for a 50+j0Ω source at the input looking from transmitter port to output load on antenna port in Step 2, then attempt to show in Step 6 that when the instrument looks back from the antenna port to the matched load at the input, that measured impedance is the complex conjugate of the original output load. This is actually testing Walt Maxwell’s Conjugate Mirror proposition.
I am going to perform a very similar experiment which will test the stated “Conjugate Mirror Proposition”. The experiment is at 1900kHz to minimise the effect of adapters and load resistor imperfections and uses the ubiquitous MFJ-949E ATU. Measurements are made with a Rigexpert AA-600 which demonstrates good accuracy on a known 50Ω load.
Above, the nominal 10Ω reads 10.1+j0.2Ω, this is the critcal value that will be tested later.
2. Adjust the ATU for a perfect match for 50Ω using the AA-600 connected to its input (TRANSMITTER) and the 10.1+j0.2Ω on the output (COAX1).
Above, ATU adjusted for near perfect match (better than the inherent accuracy of the instrument) with the 10.1+j0.2Ω output load.
3. Without changing the ATU controls, connect a 50+j0Ω load to the input port (TRANSMITTER) and measure the impedance looking into the output port using the AA-600 (COAX1).
Under the Conjugate Mirror Proposition, the measured impedance should be the complex conjugate of the original load resistor, 10.1-j0.2Ω.
Above, measured input impedance is 18.0-j0.8Ω.
It is quite different to the value predicted by the Conjugate Mirror Proposition, 10.1-j0.2Ω.
The experiment does not give the outcome that disciples of Walt Maxwell’s Conjugate Mirror Proposition would expect, it does not support the proposition and therefore questions its validity.
Should we have expected this outcome? Watch for a follow up post.
My own article Exploiting your antenna analyser #12 is but one of many.
Measurements consistently hint that the defect is that the characteristic impedance is typically somewhere between 30 and 40Ω.
Above is a dimensioned drawing from Amphenol (https://www.amphenolrf.com/connectors/uhf.html).
Let’s focus on the right hand part, and in particular let us try to estimate the characteristic impedance (Zo) of the transmission line section formed by the outer surface of the centre female pin and the inner surface of the body with some solid dielectric occupying most of that space. This is the longest unavoidable transmission line section contributed by a connector pair with good (short circumferential) coaxial cable terminations on both connectors.
If we assume that the dielectric has a permittivity of 2.0, and dimensions scaled from the diagram, we can calculate Zo to be 30Ω.
Parameters | |
Conductivity | 5.800e+7 S/m |
Rel permeability | 1.000 |
Inner diameter | 0.00539 m |
Outer diameter | 0.01098 m |
Velocity factor | 0.707 |
Loss tangent | 0.000e+0 |
Frequency | 300.000 MHz |
Length | 0.010 m |
Results | |
Zo | 30.19-j0.02 Ω |
Velocity Factor | 0.7070 |
Length | 5.10 °, 0.014 λ, 0.0100 m, 4.718e+1 ps |
Line Loss (matched) | 5.73e-4 dB |
So, even with the best transition from cable to male pin, and female pin to cable there is an unavoidable 10mm of 30Ω line.
The Insertion VSWR of 10mm of Zo=30Ω, vf=0.707 line at the bottom of UHF (300MHz) is 1.10 , and 1.36 at the top end of UHF (1000MHz).
Observations that UHF series connectors have non-ideal Insertion VSWR, and that measurements hint that they act like short sections of transmission line with Zo between 30 and 40Ω are partly explained by calculation of the expected Zo based on a dimensioned drawing of the female connector.
]]>They were regarded as the epitome of the art… but it was not a very well understood art.
Lets analyse the common implementation as a Ruthroff 4:1 voltage balun in a 50:200Ω scenario.
In this implementation, Amidon’s instructions show 16 bifilar turns on a T200-2 core.
A very simple model is to consider the device as an ideal transformer with a shunt magnetising impedance equal to the impedance of the 16t winding that appears across the 50Ω terminals. This has its greatest effect at low frequencies and although it is specified from 2-30MHz, lets analyse it at 3.5MHz.
The powdered iron core has very low loss at 3.5MHz, sufficiently so that we can ignore the imaginary component of µr for this analysis and take µr to be 10+j0.
Above is a calculation of the magnetising impedance and admittance under those assumptions. The magnetising admittance (0.00-j0.0134S) appears in shunt with the transformed load admittance (0.02S) so we can simply add them to find the admittance seen by the transmitter (0.02-j0.0134S).
We can now calculate the VSWR that the transmitter sees.
Above, the transmitter sees a VSWR of 1.93, so the balun has an Insertion VSWR of 1.93.
That doesn’t make it unusable, it just means it significantly transforms the load impedance by shunting it with an inductive reactance of around 75Ω.
The thing that does make it less suitable to most wire antenna applications is that it is a voltage balun, and voltage baluns deliver current balance ONLY on symmetric loads, and backyard wire antennas are rarely very symmetric due to the effects of nearby structures, other conductors, soil variation etc.
Although Guanella’s design is older than Ruthroff and better suited to use with asymmetric loads, Ruthroff’s balun captured the ham application. For example, most ATUs with an integral balun use a Ruthroff 4:1 balun.
(Reisert 1978) described a Guanella balun which although it did not give details of common mode impedance, would have relatively low common mode impedance due to the core choice and turns used. The metal box and feed throughs also detract from good performance, but his application was a fairly symmetric Yagi so not very demanding. Reisert was an early if not the first published design based on Rule 500.
In the last decade or so, there has been wider recognition of the need for higher Zcm baluns to be effective in reducing common mode current, even in apparently innocuous applications.
However, people still implement baluns like the Amidon AB_200_10 balun and feed they are effective, though I cannot see any reports that give measurement data, either Zcm or Icm.
A superficial analysis is that the feedback to the grid from the anode via the anode to grid capacitance (Cag) is in phase with the anode voltage, which because of inversion in the valve means it is negative feed back. How can it cause self oscillation?
Well it does, so the superficial analysis is probably inadequate.
Text books tend to gloss over the detail of how it works. Understanding is not helped by some folk lore, try to differentiate between what you know that is truly fact and other ‘knowledge’.
Where does the additional 180° phase shift necessary for self oscillation come from?
Lets take a fairly high level approximation of what happens in a simple circuit, one with identical parallel tuned circuits in the anode and grid circuits, and a very very small equivalent Cag.
Some important concepts:
Here are the steps around the loop considering only phase:
The total phase lead from anode current to grid voltage is 180+45+90+45=0° which satisfies one of the criteria for oscillation. If the magnitude of the total loop gain is greater than unity, then the circuit will self oscillate at this frequency.
So how does it start?
If the stage is biased so that some anode current flows, it contains noise components, and if the feedback circuit has loop gain greater than unity and phase=0°, the noise currents will be amplified greatly and quickly lead to self oscillation.
Note that in this example, the necessary phase shifts mean the circuit oscillates just a little below the self resonant frequency of the two identical tuned circuits.
The circuit does not depend on identical tuned circuits, and just one could be variable to adjust frequency of oscillation over a small range. Though the explanation used a very small ideal Cag, again it does not depend on that though the frequency of oscillation is sensitive to the phase relationship between Iag and Va. Obviously phase shift may differ at each stage with these small variations, it is the frequency where the loop phase shift is 0° that oscillation will occur if loop gain is sufficient.
I have used an example with LC parallel tuned circuits, but one or both could be tuned cavities (for UHF and above).
Was it a good oscillator in its day?
Probably not for a host of reasons, and probably why its use died out pretty quickly. The last transmitter that I had that used a TPTG oscillator was a modified aircraft transponder on the 23cm band (it was originally a TPTG oscillator and needed more feedback to run at lower power in continuous mode)… that was in the late 1960s.
]]>The article describes a test panel to fill that need.
The panel is constructed on a piece of 3mm aluminium sheet, drilled and tapped to take two sections of 35mm DIN rail for flexible mounting of accessories.
Above is a pic of the test panel in use to test the generic heating / cooling controller (hcctl), a flexible bang-bang controller based on an ATTiny25.
The top row of the panel is essentially the single phase 230VAC components. From left to right a 10A/30mA RCD, a 12V 3A switched mode DC power supply (with a 24V DC converter tied to it for this test which includes a 4-20mA loop), a terminal block for 230VAC, and a 40A 480V SSR. The panel is cut into an ordinary 3m 10A extension cord, input from the plug end and the socket end controlled by the SSR. The switching device can be changed to a 230VAC relay, 12VDC relay, 5VDC relay to suit the module under test. Likewise the power supply can be easily swapped.
The lower DIN rail is for the LV components, in this case a terminal block (spare), Pt100/4-20mA converter and the hcctl module being tested.
If need be (eg for oscilloscope probe attachment to the 230VAC side), the whole thing can be run from an isolating transformer and a neutral-ground tie installed on the terminal strip.
Above is a wider shot of the test where a pot with 1l of water is being heated by a portable stove, temperature sensed with a Pt100 probe and the stove controlled by the test panel output. The device at lower left is a data logger monitoring temperature with an independent thermistor probe.
Above is the logged temperature for the test run. The controller SV is 40° and differential 5°. (The EEPROM SV is actually 44° to compensate for measured 3.8° error @ 40° in the Pt100 sensor which as a Class A probe should have error less than 0.23° – cheap Chinese junk, and residual error in the 4-20mA converter, the sense resistor, and the ADC reference in the microcontroller.) There is a largish overshoot on the first cycle and just a little on the subsequent cycles, typical of a bang-bang control loop.
Other accessories include a 15A/1V clip on current transfromer, a DIN rail mounted generic PCB clamp, and as mentioned, an isolating transformer.
A warning though: don’t mess with this stuff unless you are competent, don’t take safety for granted, don’t allow access to persons who might not have the needed competency.
]]>