This exercise arose following the purchase on eBay of a Ferrograph Test Set RTS2 in approximately 2008. It was described as being in good general condition except that the meter movement was broken. The unit was purchased with the expectation that a second hand replacement meter movement would be relatively easy to acquire. In the event, that was not to be the case.
The original meter movement was manufactured by Sifam in the UK as part of its Clarity Focus range- model 42AF. It is relatively large having front face dimensions of 120mm x 97mm and has the unusual feature of the pointer zero adjuster being on the rear of the movement rather than the usual position on the front. This enables the meter to be adjusted when mounted inside the front panel of the Test Set. The meter movement has a full scale deflection (F.S.D.) of 100uA and an accuracy of +/-1%. Unfortunately this range of Sifam meters has been obsolete for a number of years.
After years of scouring the internet and searching at radio rallies there was no sign of a suitable meter movement. Recently however, a sifam 42AF meter was offered on eBay but it had a 1mA FSD rather than 100uA. This meter was purchased and the challenge was then to adapt it, if possible.
Physically the new meter was identical to the original. The scale from the original meter could easily be transplanted. The zero adjuster was appropriately on the rear of the meter. The problem was to increase the meter’s sensitivity by a factor of 10. It is simple enough to reduce the sensitivity of a current meter by shunting it with a suitable resistance, but increasing its sensitivity would require some additional electronics.
After a little thought the solution began to become apparent.
- Replace the 100uA meter in the Test Set with a resistor of equivalent value to the original meter’s coil resistance.
- Design a high impedance volt meter around the new 1mA meter movement such that it reads FSD when measuring the voltage across this resistance when passing 100uA. This should result in a precise linear function with no disturbance to the operation of the Test Set.
- Improvise a power supply for the voltmeter so that it is powered from the Test Set and requires no batteries.
- Examination of the old meter case and the Sifam data sheet on the internet revealed that the original meter movement had a resistance of 1020 ohms. Therefore at FSD, when passing 100uA the voltage which needed to produce FSD on the new meter would be 102mV.
- A little memory dredging and a search through old issues of Practical Wireless came up with an elegant and simple design for a high impedance voltmeter by John Thornton Lawrence, GW3JGA, published in 1986. His design was stripped down and adapted for the purpose in hand.
- According to the Ferrograph service manual, the meter coil is not directly connected to ground which means that the power supply to the voltmeter would also need to be isolated from ground.
Fortunately the Test Set mains transformer has an independent secondary winding supplying 12v AC to a festoon lamp mounted above the meter movement to provide scale illumination. The power is supplied to the bulb via a single core screened cable with the screen connected to one end of the winding but also to the chassis.
The solution here was to replace this wiring with twin screened lead carrying the 12V, with the screen grounded but the transformer winding floating in relation to ground. A 12V AC supply for the voltmeter was derived from a connection to each end of the illumination bulb holder.
The final circuit is shown below.
The 12V AC from the transformer is full-wave rectified by diodes D1-4. C2 provides a degree of smoothing and IC3, a 78L09 regulator provides a stabilised 9v supply. C3 and C4 decouple the input and output of the regulator and are intended to prevent instability. R6 and R7, as a potential divider across the 9V, provide a 4.5V reference for IC2. IC2 will strive to keep both inverting and non inverting inputs at the same voltage and thereby provide a 4.5V rail.
We therefore have effectively a +4.5V – 0 – -4.5V supply for IC1.
R1 and R2 in series are equal in value to the original meter winding resistance (1020 ohms). R3 and C1 were included in GW3JGA’s design as a low pass filter to keep a.c. and noise from reaching IC1 and it was felt that their retention would do no harm. The potential developed across R1+R2 is applied to the non inverting input of IC1.
IC1 has a mosfet input stage with an extremely high input impedance of 1.5 Tera- ohms and therefore drawing an infinitesimal input current. Again IC1 strives to keep its inverting and non inverting inputs (pins 2 and 3) at the same potential. With the 1mA meter in the feedback loop and when indicating FSD for a voltage of 102mV at pins 2 and 3, resistors R4 and R5 need to total 102 ohms (102mV divided by 1mA). VR1 allows for input offset correction. It is adjusted give a zero indication with the input shorted to earth.
A breadboard version of the circuit was built and checked for accuracy against a 100uA meter connected so as to indicate current passing through resistors R1 and R2. The result gave precise tracking of the two meters to within a needle width.
Two M3 bolts provide terminals for connection of the original wiring from the Test Set. The power connection is via flying leads from the bulb holder to pins on the top edge of the circuit board.
The solution works perfectly, resulting in a fully functional Ferrograph Test Set.
G4KQK Nov 2013
Ferrograph RTS2 Service Manual