Digital Test

Manufacturer: RACAL
Model: SA.520
Year: 1963
Schematic: RE
Max. Rate: 300 KHz
Display: 1-of-10 incand.
Digits: 4
Logic: discrete transistor Ge-PNP

A simple, low-speed model from the British radio manufacturer. While only 4 digits are displayed it is possible to get 6 digits of precision by taking a reading on each of the 3 provided gate periods.

The stack of PCB modules from the top ..

.. and from the bottom.

Servicing these units is very awkward as the stack of boards are soldered into a cube.

Side view showing the power supply.

There are 6 different module types used in these counters, 5 are shown here. Clockwise from top-left: Decade Counter, Control, Timebase, Input, Decade Divider. There are two types of decade counter, a high-speed version and a low-speed version. Only one is shown here as they appear identical, differing only in component values.

- Unit Log -

Serial Number: N0653
Year of Manufacture: 1963 (capacitor dated May 63).
Date of Receipt: 2008 Dec 18
State upon Receipt: Clean, very good condition. Count is incorrect on "*1" setting. MSD "0" bulb burned out.

Date: 2008 Dec 19
Procedure: MSD "0" bulb replaced. (Several replacement bulbs provided in unit.) Loose solder joint on lead of LSD "1" bulb resoldered.

Date: 2008 Dec 21
Symptoms: Count is incorrect on "*1" setting.
Analysis: Gate period is only about half a second, it should be 1S. 2nd decade divider is counting modulo 6 rather than 10. 2nd stage flip-flop is not toggling but does pass a pulse through 1 for 1.
Solution: V3 of 2nd divider has leaky CE junction. Replaced with 2N1377.

Date: 2008 Dec 21
Procedure: Reassembled. Fully functional.

- Unit Log -

Serial Number: N0385
Year of Manufacture: 196?.
Date of Receipt: 2012 Nov (customer repair)
State upon Receipt: Clean, good cosmetic condition. Sample cycle and reset is operative but count sequence is not correct. Input board had previously seen extensive repair, appears that input was smoked. DV2 board also had been extensively repaired.

Date: 2012 Nov
Symptoms: 8 lights with 6, 9 lights with 7 on LSD decade.
Solution: DC1-VT24 turning on during 6 & 7. Diode and resistor of gate seem OK. Presumed leaky BC junction in VT24. GET536 replaced with 2N4037 (Si).

Date: 2012 Nov
Symptoms: Count is inconsistent across samples.
Solution: DC2-VT3 leaky, GET536 replaced with 2N526 (Ge).

Date: 2012 Nov
Symptoms: Count is good on *1 range but inconsistent on *100 and *10 ranges.
Solution: DV1-VT1 leaky, GET536 replaced with 2N526 (Ge). DV2-VT8 leaky, GET536 replaced with 2N3645 (Si).

Date: 2012 Nov
Procedure: Input board cleaned up, some resistors replaced. DV2 board cleaned up, numerous components replaced. PCB stack reassembled with new bus wires, tabs slotted so boards can be removed.

Date: 2012 Nov
Symptoms: Does not work in TEST mode, but does with external input.
Solution: Traced to level sensitivity. PCB trace to 15K resistor on INP-VT2 base had been cut at an earlier time. Patched. Also replaced INP-VT1 2G1303 with 2N5023 (Si), and INP-VT2 2G1303 with 2N5023 (Si).

  .. | RACAL SA.520 | ..
Early Digital Test Equipment