Fullscreen HTML5Med Res (??MB)HTML4

Vintage EQUIPMENT AVO Valve Testers and Valve Characteristic Meters By Ian Batty The ultimate evolution of the AVO Valve Characteristic Meter – the MkIV. “I checked it on the AVO.” For decades, AVO valve testers were the standard for testing valves (their multimeters were also extremely popular). This article explains the differences between the various AVO meters and describes how they work. Warning: Electrocution Hazard All AVO valve testers apply AC voltages with peak values ~1.57 times the indicated voltage on the voltage selectors. From the MkI onwards, they can apply AC voltages with peak values exceeding 600V. Even the initial Valve Tester can apply peak voltages close to 400V. Exercise care with all AVO Valve Testers. Never touch any exposed contacts on valve socket panels. Be careful when measuring voltages. 88  Silicon Chip Australia's electronics magazine siliconchip.com.au A VO was typically used to refer to the AVO Valve Characteristic Meter (VCM), based on a design first made by the Automatic Coil Winder & Electrical Company in the late 1930s. This company would become the famous AVO, best known for (and named for) its most prolific product, the Amp-Volt-Ohm meter. With its initial patent lodged in 1922 by Donald MacAdie, the AVOmeter would become the sub-standard meter of choice, with the final one made in 2008 (Photo 1). Note that I wrote sub-standard and not substandard; in measurement circles, a sub-standard is an instrument second only to the physical examples stored at the National Standards Laboratory. But there wasn’t just one “AVO”. The initial release was the 1936 Valve Tester, registered as British Patent 480,752: “An Improved Method and Apparatus for Testing Radio Valves”. Lodged by Sydney Rutherford Wilkins on August 26, 1936, the patent describes the AVO Valve Tester circuit and gives the design principles described below. Notably, there is only one non-­linear component, the rectifier in the SET ZERO circuit, which applies pulsating DC to the meter circuit in opposition to the valve’s pulsating anode current. It’s the balancing of these two currents that allows the meter to settle to zero in readiness for the gm measurement. It’s a remarkably elegant design, so let’s look into how the problem of valve testing was definitively solved. Valve testing basics Simple valve testers heat the filament or cathode and measure the emission between the filament/cathode and the anode (in a diode) or the first grid (in all other valves). You can use an ordinary ohm-meter for this job. You would need a list of various valves types and their expected resistance readings, and such charts were the manufacturer’s specified anode current. 2. Shift the grid voltage up and down by half a volt each way and observe the swing of the anode current. Using a 6V6 with 250V on the anode and screen, reducing the bias voltage of -12.5V to -12V and increasing it to -13V should give a total anode current swing of 4.1mA, confirming a gm of 4.1mA/V or 4.1mS. But that would demand up to three adjustable, regulated supplies, and the Valve Tester hails from the 1930s. Regulated supplies of the day were bulky and prohibitively expensive. Imagine designing and building two indepenPhoto 1: an AVOmeter (amp/volt/ dent 0~400V, 100mA supplies before ohmmeter) Mk8, the multimeter. the invention of the 6L6 beam tetrode. supplied with some multimeters, such Knowing that they could design and as Hansen’s FN/SU models. build mains transformers that would This is emission testing, useful deliver well-regulated AC supplies, for sorting out dud valves and mak- the engineers at ACWEC decided to let ing like-for-like comparisons. How- the valve under test do the rectifying. ever, emission testing does not test With the valve performing rectificathe entire valve’s performance under tion, the anode current is pulsating DC. typical applied voltages, doesn’t test The indicating meter would simply be at the valve’s full rated voltage or (for calibrated to respond to the pulsating power valves) typical operating cur- DC and give a reading equivalent to a rents, and doesn’t check for inter-­ steady direct current. electrode shorts or leakages such as The applied anode (and screen) heater-­cathode leakage. supplies would effectively be half The emission tester also fails to test sinewaves since the valve would not a key valve characteristic: its mutual conduct during negative half-cycles. conductance (gm), now commonly A simple implementation would called transconductance. This is the see the indicating meter settle to, say, ratio of anode current change to grid 45mA for a 6V6, rising to 49.1mA voltage change. It was initially mea- when the grid voltage is made 1V more sured in microamps (of anode cur- positive. That would work, but you’d rent) per volt (of grid voltage), with have to observe, accurately, only about the unit of the micromho (“mho” is a 10% change in the meter reading. ohm backwards). Fig.1 shows the problem. It looks It is now measured using the SI unit like the standing current is about of microsiemens (µS). It’s a form of 45mA, and the on-test current is about conductance (G = I ÷ V) because it’s 49mA, so the valve’s gm is maybe about the inverse of resistance (R = V ÷ I). 4mS (49mA − 45mA). So, the question then is – how to We’d prefer a direct indication: a gm measure it? In principle, the steps are: of zero means the meter does not move 1. Apply the correct grid bias, at all, a gm of 4.1 gives a meter indiscreen voltage (tetrodes/pentodes) cation of 4.1, and so on, as shown in and anode voltage, and trim to get Fig.2. This requires two supplies: the Fig.1: the difference in reading you would expect applying a 1V signal or step to the grid of a 6V6 with the specified bias of -12.5V. It’s hard to read this with any precision. siliconchip.com.au Fig.2: by increasing the meter’s sensitivity and offsetting the reading so that it’s at zero with the specified bias of -12.5V, it becomes easier to read the difference in current accurately. Australia's electronics magazine August 2022  89 Read the instruction manual before operating an AVO meter This article includes basic lists of steps for using each type of AVO meter. This is mainly to give you an idea of how they work. I recommended that you read the full instructions before using any of the valve testers. Note that in each case, the recommendation is to set the switches with the power off or, where available, with the FUNCTION set to CHECK. Doing it this way prevents accidental short circuits and valve damage. selected anode voltage & the backing-­ off supply, which adds to it. This is depicted in the simplified circuit of Fig.3. Although it’s not shown, S1’s SET position is applying a sinewave causing an effective -0.5V grid bias. The anode current flows through the mA/V pot, which acts as a variable shunt, controlling the meter’s full-scale sensitivity. Let’s say we have selected an anode voltage of 250V, and the backing-off control is adjusted for minimum effect. The valve will draw a current of Ia, so there will be some voltage drop across the mA/V pot. The meter will deflect, with the indication depending on the shunting effect of the mA/V pot’s setting. Let’s say the valve draws 45mA. Adding current from the backing-off supply will raise the voltage at the anode end of the mA/V pot, reducing the total current through the mA/V pot. If the backing-off supply is adjusted to give enough current to raise the anode back to 250V, there will be no voltage drop across the mA/V pot, and the meter needle will fall to zero. Now, applying the test bias to the valve will increase the anode current, but the backing-off supply is still set to 45mA and cannot entirely cancel the new anode current. The difference between the new anode current and the backing-off current will be shown directly on the meter scale, as in Fig.2. The “SET M. A./V.” (referred to as “SET mA/V” for future references) can be adjusted to the expected gm value; in our example, 4.1mS. This control is a continuously-­variable current shunt across the meter movement, so this setting gives the meter itself a full-scale deflection of 4.1mA. After doing that, the key switch is set to the mA/V position. This inverts the sinewave voltage on the grid, replacing the effective -0.5V with +0.5V. This step will push the total anode current to about 49.1mA. But, as there is a counteracting current of 45mA from the backing-off supply, the meter will indicate 4.1mA (49.1mA − 45mA). 90  Silicon Chip And that is the sensitivity we set using the mA/V control, so the meter will show 100%. Alternately, setting the “SET mA/V” control to the “mA/V” position gives 10mA full-scale. In this case, our 6V6 will deflect the pointer to the 4.1 mark. This confirms the previous measurement, but it also allows a direct reading for any valve without having to look up a table of specs and adjust the “SET mA/V” accordingly. So that’s the principle used in the AVO Valve Tester. A description of how the follow-on Valve Characteristic Meter operates will come later. The AVO Valve Tester The AVO Valve Tester (Photo 2) used a case similar to their existing multimeters, with an extension board carrying the selector switches and valve sockets. It could test valves with anode voltages ranging from 30-250V and screen voltages from 60-250V. The test range was either a direct reading of 0~10mA/V or by setting a dial for the specified gm and reading the valve’s merit (“goodness”) from the scale. Heater/filament voltages matched common valves of the day, with selections of 2, 2.5, 4, 5, 6, 7.5, 10, 13, 16, 20, 26, 30, 35 and 40 volts provided. The test panel added a ÷7 switch so that, for example, 1.4V valves could be tested with a selected supply of 10V, reduced to 1.4V by actuating the ÷7 switch. The Tester also offered a heater-­cathode insulation test. The instrument’s accuracy depended on the mains voltage, with an internal selector panel allowing settings of 200V AC to 250V AC in 10V steps. The Valve Tester set an instrumentation standard that saw “the AVO’s” widespread use in civilian and military contexts. I recall using a CT160 at the Williamtown Air Force Base near Newcastle in the mid-1960s, and in Darwin. The photo opposite (Photo 3) shows the interior. From top to bottom, the major components are the high-­voltage transformer, meter, function keyswitch and low-voltage transformer. The dualgang Set Zero (backing off) pot can just be seen at lower left. The socket panel’s connector is at top right. For all its brilliance, the Valve Tester had a serious drawback: it tested with 0V of standing bias. This meant that the anode current under test might not be that recommended by the valve manufacturer. This matters, as transconductance is anode-current dependent. It’s low for low anode currents, and increases Fig.3: a greatly simplified circuit for the original AVO Valve Tester. The twogang potentiometer at upper left is used to zero the meter before starting the test, while the pot below the meter adjusts its sensitivity so that FSD (full scale deflection) can be set to the expected reading. A good valve will then provide FSD, while a weak valve will give a somewhat lower reading. Australia's electronics magazine siliconchip.com.au as anode current increases to the permitted maximum. Let’s consider the 6AU6. With zero grid bias, it draws around 17mA to give a gm around 5.5mS. But it’s often used in audio amplifiers at anode currents as low as 300µA. What is its gm at such a low current? The Valve Tester cannot apply variable bias (and we’d need around -4.5V to get such a low anode current), so it’s impossible to find out. The Valve Tester also swings the grid positive, with possible grid emission effects giving false readings. To explain the remaining features of Fig.3, diode D1 rectifies the backing-off supply to balance the anode current indication back to zero for testing. In the SET position, S1 applies an AC voltage to the grid. Setting S1 to the TEST position reverses the polarity of the grid signal, causing the anode current to rise, and allowing the meter to indicate the change in anode current as a transconductance reading. Diode D2 ensures that the screen cannot go negative during the valve’s non-conducting cycle. Allowing this could disrupt the instrument’s measurement accuracy. This diode is not included in all diagrams; I have included it in case you find a Valve Tester with it fitted. So, while the Valve Tester gave reliable indications for valves (mostly triodes) that specified low (essentially zero) grid bias voltages, it could not be relied on for those that required a negative grid bias for normal operation. That’s pretty much everything with an oxide-coated cathode. Also, one had to trust that the calibration was accurate. Valves are specified for a range of filament/heater voltages, and it was luck whether the Valve Tester actually applied the correct voltage on any one particular test. While manufacturers allow as much as ±10% variation of heater voltage, deviations from the nominal voltage affect results. On test, an ECC82/12AX7 returned gm values of 2.05mS and 1.5mS for heater voltages of 6.9V (+10%) and 5.7V (-10%), with a reading of 1.8mS at the specified 6.3V. That’s a variation of +14%/ -17% over the recommended operating range. Basics of operation 1. With the power off, consult the AVO data book and set the roller switches to the required positions. Set the filament/heater voltages. Be careful with 1.4V valves; you need the ÷7 setting on the socket panel with the 10V setting on the Tester. 2. Set the mA/V control to the value shown in the data book to get an indication of relative functionality, or to 10 to get an actual transconductance reading. 3. Push the key switch to the mA/V position and read off the meter indication. The Valve Characteristic Meter (VCM) The Valve Characteristic Meter was a significant rework of the design. First, it was unitised and made more ergonomic. The meter and controls were mounted on a sloping front panel, making operation and observation much easier. The socket panel was located on the top surface, removing the previous connecting lead, plug and socket. Sockets that had been recently invented were included. The socket panel was protected against debris intrusion by a flip-up cover. Second, the mains voltage selector was brought out to the front panel, with an indication on the test meter. Third, they added a variable grid bias control. Operators could set up all of the valve manufacturer’s specified parameters. Fourth, the VCM incorporated a short-circuit relay which appears to have been included in some issues of the Valve Tester. This needed to operate at any anode/screen current selection. To achieve this, the relay’s core held enough residual magnetism to stay latched in with no current flow. In regular operation, the anode/ screen current is pulsating DC due to the rectifying action of the valve under test. The resulting uni-directional magnetisation added to the residual magnetism, holding the relay in. But a short circuit would draw current on both half-cycles of the internal Photo 2 (above): the original AVO Valve Tester. The part on the right was an expanded version of their AVOmeter ‘multimeter’ (a term that hadn’t been coined yet), while the part on the left houses all the valve sockets plus some extra controls. Photo 3 (right): the inside of the AVO Valve Tester is busy but there are very few actual components. Most of it is (very neat) wiring! The meter movement is right in the middle, while the transformers are at the top (HV) and bottom (LV). siliconchip.com.au Australia's electronics magazine August 2022  91 Fig.4: the Valve Characteristic Meter (based on the MkIII/IV VCM) is a refinement of the original concept that added a great deal of flexibility. Its main advantages are the ability to test the valve over a wide range of bias voltages and a built-in overload/short circuit protection relay that ends the test if too much current flows. Fig.5: the final evolution of the AVO Valve Tester, the VCM163, included a solid-state sinewave generator and amplifier/rectifier to provide even more accurate results. 92  Silicon Chip Australia's electronics magazine alternating voltage supply. The relay might hold in on the first half-cycle (depending on polarity) but would be thrown out as the opposite-­polarity half-cycle began. Once thrown out, it was reset by pushing the RESET button on the control panel. Fig.4 shows the basics of the VCM circuit. The backing-off/zero circuit has been modified: it now applies the opposing current directly to the meter, but with the same effect. Notice that the meter now reads the voltage drop across the fixed 200W resistor (R36). You can regard the meter as a sensitive, multi-range voltmeter calibrated in transconductance when testing. The overload relay (RLYA) senses anode and screen currents in separate windings. As described, the alternating current resulting from a short circuit will throw the relay out, demanding that the operator reset it manually. As with the Valve Tester, diode D2 ensures that the screen never has negative voltage applied. The bias supply is in two parts. In SET mode, the operator uses potentiometer VR5 to apply the specified grid voltage. Switching to TEST mode makes the grid voltage 1V more positive. This causes the anode current to increase above the balanced value when the backing-off was set. That extra current will be read as the valve’s transconductance. Along with this, the design rework provided for anode current measurement. The name “Valve Characteristic Meter” is a clue. This rework allows the operator to record the anode current for any combination of control grid bias, screen voltage and anode voltage. It was possible to plot the entire set of grid-anode characteristics for any valve that would fit the extensive set of sockets. In effect, the VCM offered a complete test bench for any valve, of any kind, for any test conditions. Operators could also identify weak valves, which would work fine at low anode currents, but lacked the emission to deliver full performance at full current. Matching valves to each other (important for high-performance push-pull operation) was also made much easier. A manufacturer aiming to operate a particular output valve from a lower-­ than-specified high tension (HT) supply could easily measure that valve’s characteristics and could refine a siliconchip.com.au design to suit. The venerable 6V6, for example, can give up to 4.5W of output. But a small mantel set can get by with just one or two watts to the speaker. Could an ‘economy’ set do this using a 6V6 with just 150V HT? Sure, and the VCM could confirm that. Basics of operation 1. With power off or the function setting in the CHECK(C) position, consult the AVO data book and set the roller switches to the required positions. If the VCM was off, switch on in the CHECK(C) position and adjust the SET~ control for the correct mains indication. 2. Set filament/heater voltages. 3. Set grid, screen & anode voltages. 4. Set the METER SELECTOR (MkI-II) or METER SWITCH (MKIII-IV) to 100(mA). 5. Switch to C/H.ins to warm the valve up before testing. 6. Switch to TEST and read the anode current. Set the METER SELECTOR/SWITCH to a lower range if needed. 7. Set the SET mA/V control to the expected gm value and set the METER SELECTOR to mA/V. 8. Adjust the SET ZERO (MkI-II) or BACKING OFF (MkIII-IV, COARSE and FINE) to bring the meter to 0. 9. Press the mA/V button or switch to mA/V and read the valve’s merit from the coloured scale. 10. To get the actual gm value, repeat the above, but with the SET mA/V control at 10. Press the mA/V button and read off the valve’s actual gm value, treating the calibrations as a 0~10mS scale. The CT160 The ‘clamshell’ CT160 used the same basic electronic design. While it did not offer laboratory testing capability, it became the standard ‘quick, accurate and ready’ instrument used in many workshops and service centres. The CT160 only operates as a gm tester; it does not give anode current readings. The electrode voltage settings (grid, screen and anode) work as for the MkI-IV and the VCM163. But the anode current settings take the place of the backing-off controls in all previous models. A simplified version of its circuit diagram is shown in Fig.6. The CT160’s meter is fixed at 700μA FSD. Perhaps confusingly, the 1mA/V mark, at around 74% of FSD, is a DC siliconchip.com.au Fig.6: a simplified circuit diagram of the CT160. It doesn't provide all the features of its predecessors (eg, it lacks anode current readings), but it is still a useful instrument and was widely used. equivalent of 520μA. With the SET mA/V control at 1mA/V, the applied grid voltage decrement is 0.52V. Using the formula ∆Ia = ∆vg x gm, a valve with a gm of 1mS will give an anode current increment of 520μA, resulting in a scale indication of 1.0. So, while a 0.52V decrement would give a 1.0 indication for a valve with a gm of 1mS, applying the 0.52V decrement to a valve with any higher mutual conductance would overswing the meter. The SET mA/V control does, indeed, give a 0.52V decrement on its 1mA/V position, but it gives proportionately less for each higher dialled-in gm value: 260mV for gm = 2mS, 130mV for gm = 5mS and so on. I was, again, awed by the elegance of this instrument’s design. As with the previous VCMs, the CT160 is calibrated with simple DC values, so this preceding complexity is hidden from the operator. Basics of operation 1. With the power off or the function Australia's electronics magazine setting in the SET~ position, consult the AVO data book and set the roller switches to the required positions. There are plug selectors and a switch beneath the transparent lid just below the meter. Be aware that these are at mains potential. Adjust for the correct mains indication. 2. Set filament/heater voltages. 3. Set grid, screen & anode voltages. 4. Set the anode current’s coarse switch and fine potentiometer controls to the specified values. 5. Switch to C/H to warm the valve up before testing. 6. Rotate the mA/V control to the Cal position and set the function switch to TEST. 7. Be ready to adjust the anode current, as the meter may swing wildly back past 0, or forward past full scale. I find it easier to adjust the grid voltage when the meter overswings – it has the same authority as the two anode current controls combined, but it’s a single control and is easier to manage. Once the meter gives a safe indication, August 2022  93 trim the grid voltage and anode current controls. Aim to get the specified anode current, even if the grid voltage is not close to the specified value. Anode current has the most effect on gm, so the correct setting of anode current has priority. Be aware that a very low grid voltage implies a valve with poor emission. The VCM163 Finally, the VCM163 introduced a solid-state measurement design (Fig.5). This revolutionary instrument uses a transistor oscillator to generate a sinewave signal that is applied to the grid of the valve under test. This high-frequency signal modulates the half-wave 50Hz applied to the grid. The VCM163 uses a high-pass filter in the anode circuit to pick off the amplified high-frequency modulation from the anode current. This signal is further amplified and rectified to drive the transconductance meter. Since the transconductance is measured by the amplification of a high-frequency signal, AVO removed the entire backing-off section. This allowed continuous measurement of anode current by a dedicated meter. No longer did operators need to set anode current, back off, measure transconductance and then remove the backing-­off setting to check that the anode current had not drifted. Half-wave rectification is now done by silicon diodes, removing the possibility that high-voltage negative half-cycles applied to valve electrodes will affect the instrument’s accuracy. The VCM163 retains the fundamental AVO principle: mains transformers can deliver sufficient regulation to permit accurate valve testing without the need for regulated DC supplies. Setting the valve up as a signal amplifier gave the highest accuracy. It also took the gm meter out of the valve’s current path, meaning that overloads caused by incorrect settings, or shorts, would not pass damaging amounts of current through the meter’s delicate moving-coil winding. Basics of operation Set the CIRCUIT SELECTOR to CHECK(C) and LEAKAGE to ~. Check that the meter settles to the calibration mark. If the front-panel SET~ control won’t adjust, remove power, open the voltage selector panel on the left side and adjust the coarse mains tapping. 2. With power off or the function setting in the CHECK(C) position, consult the AVO data book and set the roller switches to the required positions. 3. Set filament/heater voltages. 4. Set grid, screen & anode voltages. 5. Set the anode current and mA/V controls to the expected values. 6. Switch to C/H to warm the valve 1. up before you start testing it. 7. Switch to TEST and read off the anode current from the left-hand meter. Read the transconductance value from the right-hand meter. Model identification The Valve Tester is immediately identifiable by its two-part construction. Valve Characteristic Meters can be identified as follows: ] MkI: Grey aluminium exterior case, unitised design, flip-top lid over valve sockets, side carry handles, sits flat on the bench. ] MkII (Photo 4): Similar to the MkI with added front handles, standup runners raising the instrument off the bench and a valve data book tray underneath. ] MkIII (Photo 5): Revised design with ‘roll-over’ handles, panels over the frame, black front panel, large dials for grid voltage (left) and transconductance (right). ] MkIV (see lead photo): Revised design with combined grid voltage variable/range switch and transconductance variable/range switch. ] CT160 (Photo 6): clamshell design, transconductance only. ] VCM163 (Photo 7): has two meters. Special handling Never tap any meter on the glass. Be aware that the original Valve ► Photo 4: the AVO VCM MkII looks similar to the MkI, also having a flip-top lid with extra handles fitted to the front. Photo 5 (above): the AVO VCM MkIII has roll-over handles. Its grid voltage and gm controls are on the front panel, while the MkIV has them behind protective windows. Source: Rodney Champness 94  Silicon Chip Australia's electronics magazine siliconchip.com.au Tester meter movement is not enclosed, as the interior photo shows. Opening the back of the Valve Tester exposes the meter movement, making ‘clean room’ maintenance essential. The instrument is well-constructed but my example had a two-wire power lead. I did notice that slight ‘tingle’ that you get (due to mains leakage) when I ran my fingers over the front panel. I recommend the fitting of a three-core power lead to provide Earthing. You would need to make connections to the metal frames of the two power transformers. If you decide to take on an AVO to repair, get all the info you can first. All VCMs are compact, and the MkIV is tight to the point of inaccessibility/ invisibility for some components. Further reading The available circuit drawings are often difficult to interpret. I welcome discussion and corrections regarding my simplified illustrations. I have not found a single, easily-­ comprehensible circuit for any AVO. An example is the calibration circuit – the critical first area to examine when repairing or calibrating. I found the original AVO documentation hard to understand, mixing operating instructions with technical descriptions. If you’re a newcomer to the AVO, consider getting help from an experienced owner. You can find the detailed manufacturer’s instructions online, so I have not attempted to make this article comprehensive. You can find out a lot more Differences between voltage readings and applied voltages AVO valve testers rely on the tested valve’s self-rectification, so the applied voltages and currents are not the same as those selected on the controls, or indicated on the meter. On their DC ranges, meters commonly display average values, so they indicate 0.637 of a half-sinewave’s peak value, rather than the correct RMS factor of 0.707 for AC. The conversion factor from average to RMS is (0.707 ÷ 0.637) = 1.11, so with a selected anode voltage of 400V – the DC-equivalent mean – the instrument applies 444V RMS to the valve anode. While you won’t usually measure it, this is a peak value of some 630V. AVO’s meter is calibrated to deflect to twice the valve’s anode current. The grid voltage is even stranger. Selecting -10V bias on the Grid Voltage setting measures as -5.2V on an average-reading meter. This is a bit confusing, but you only need to consider it if you’re testing or calibrating an AVO valve tester. In the main part of the article, I treat all currents and voltages as DC values, unless the AC values are critical to description or calibration. Just to reiterate, the controls and the meter are calibrated for the equivalent DC values. by reading those instructions. See the links to just some of the many valuable references at the end of this article. Next month In the follow-up article next month, I’ll describe three AVO Valve Testers/ VCMs that I was given to test (plus my own CT160) and some of the problems that I encountered. In some (but not all) cases I was able to fix the problems and get them working properly again. Useful links Martin Forsberg’s excellent entries on the UK Vintage Radio Repair and Restoration Discussion Forum, in collaboration with Euan MacKenzie and permissions from Yutaka Matsuzaka: siliconchip.au/link/abeh (be aware Photo 6: the CT160 is the only AVO Valve Tester in a clamshell case. While it’s a later design, it only offers direct measurement of gm. siliconchip.com.au these texts are copyrighted). For the MkIV, see Guido Pedrali Noy’s thorough reconstruction of the user manual at: siliconchip.au/link/ abe5 Frank Philipse’s extensive list of resources for the MkII/III/IV, CT160 and the VCM163: https://frank. pocnet.net/instruments/AVO/ Extensive discussions for AVO products at: siliconchip.au/link/abei A must read (!) article on the VCM163 at: www.schmid-mainz.de/ Radio-Bygones_140.pdf Even more information on the AVO MkIV, including meter replacement: siliconchip.au/link/abe8 For information on servicing and repairs, see pages 3-10 of the PDF at: SC siliconchip.au/link/abeg Photo 7: the VCM163 is the only one with two meters! They show DC anode current and transconductance. Source: Jerry Aldrich, UK Vintage Radio Repair Forum Australia's electronics magazine August 2022  95