My new spectrum analyser
which covers up to 1.5GHz (click picture to see more)
Note the invisible leads...
Note again the invisible leads...
Click picture to read its spec.
Here's my old AVO valve
tester which is virtually indispensable unless one has loads
of valves and plenty of time to swap them around. It was used
by Electronics Lincs in North Thoresby to test TV valves and
it's still got its original handbook.
The handbook is vital as it
provides the settings for the tester. The array of switches and
knobs needs to be carefully adjusted for each valve being tested.
Under the lid is a set of valve bases and the pins on these are
connected to roller switches which are turned to a number 0-9.
These switches set the valve electrodes correctly, after which
the various voltages are set up, the heater or filament, anode,
screen etc. Depending on the type of valve other switches and
knobs are adjusted so that emission and gain can be checked.
Weirdly, the equipment uses
AC not DC to power the valve being tested.
Signal Generator TF2008
This is a Marconi TF2008,
one of the last analogue professional products of this type from
that company. It uses transistors (not valves) of course and
can do most things. One really needs a counter to check exact
frequency output but it has a built-in device to carry out spot
checks if one isn't available. The first counter I had was cheap
and cheerful and had one significant drawback..it leaked RF so
that the superb attenuator ccould't be increased beyond background
level of the leak. To get down to microvolts required the counter
to be unplugged.
The generator has one really
unusual feature. The pointer traverses to the right going from
zero frequency to the top of the first band then, when the next
band is selected, it tunes "backwards" from right to
left, next tunes "forwards" from left to right and
so on in 12 bands to over 500MHz. This gives you essentially
a continuous tuning scale of about 12 feet in length.
I bought it in a "non-working
but easily repairable" condition years ago for a lot of
money. I think mine had been connected to a transceiver which
had inadvertently been put into transmit mode, damaging the output
of the generator. I find it nicer to use than a digital equipment
but a bit fiddly when setting to exact frequencies. This is probably
because I use a digital counter with too many digits.
Signal Generator TF1370
This was my first audio
signal generator, a Marconi TF1370 which is authentic enough
to use valves but suffers from the drawback of having to warm
Oscilloscope Type 454
I can't abide an oscilloscope
which takes hours to warm up and stabilize so I ditched the various
valved models I had and invested in this Tektronix for which
my bid of £50 was accepted, being surplus to Plessey's
requirements. It's transistors warm up immediately and one can
make measurements within half a minute of switching on.
I no longer use this because
I bought a brand new scope (GDS-1102U)
when the supplier listed them at reduced price for a short period
in September 2013. I paid £276.94 inc VAT and delivery.
Pricing of products such as
this is sometimes very odd. For example, in Dec 2014, Farnell
lists the GDS1102U for £568 whilst sister company CPC has
them at £383. Maplin has them at £419.99, whilst
Amazon lists four at £908.50, £509.99, £761.10
and £481.60. Ebay has them at £438.70, £678.29,
£436.38 and £627.41. A US company has them at $488
which is £312 and in South Africa they are £418.
What on earth is going on. Can
anyone offer an explanation?
The new model has some really
useful features which were undreamed of when Tektronix made the
model 454. I can press a button and immediately see a locked
stable picture of the input just the right size for the screen
and another press I can see the RMS value of a sinusoidal input.
Output Transformer Tester
This is my Line Output Transformer
The key part of all TV sets is the transformer
which develops the high voltage to drive the cathode ray tube.
The LOPT, as it is known, is also pressed into service to provide
subsidiary voltages for many other parts of a TV's circuitry.
Energy that would ordinarily just go to waste as heat is harnessed
to provide things like power for the CRT's heater, voltages to
drive the focus and screen grisd electrodes, and several low
voltage power sources. The LOPT is very highly stressed and a
typical fault is a breakdown within the transformer's windings.
This is usually not an easy fault to diagnose, at least it's
one that could be mistaken for an entirely different fault. The
simple way to find out if a LOPT is faulty is to swap it for
a new one. Unfortunately as there are hundreds of different types
it would be an expensive proposition to carry stocks of all of
them. One of the largest manufacturers of transformers came up
with this little tester which can accurately diagnose most LOPT
faults by emulating the circuitry connected to it in a way that
can determine short circuits between windings, shorted turns
in windings and faulty rectifier diodes in the EHT circuit. Since
I started using it I must have saved many hundreds of pounds
in transformers I never needed to use.
It's now many years since I used this.
This is a government surplus
variac which is essentially an auto transformer which one connects
across the mains supply to provide an adjustable source of power
for testing faulty TV sets and the like.
I've had this one in the workshop
for ages on loan but recently swapped it with its owner for a
By monitoring input current,
usually across an open fuse holder one can gently increase the
mains voltage and check nothing nasty is afoot.
Be warned though with TV sets
the degaussing posistor presents a low impedance across the imains
input until after a second it gets hot and effectively disappears.
As the control is turned up it is customary to pause while the
posistor gets to operating temperature bedfore looking for meaningful
This particulatr variac has
plenty of power handling capacity unlike my previous model which
I bought from Maplins which blew up after only a couple of uses.
I've now fitted this device
into a case.
This is another variac.
It was being chucked out because the owner didn't know what it
was. Inside the home made wooden case, disguised as metal, is
a variac with a horizontal frame rather than the vertical type
This model is wired in a different
way to the one above. It can be adjusted to provide an exact
output voltage over a limited range.
Power Supply type SRS153.2
Clearly designed for testing
valve equipment because you can see the output for standard valve
heaters (6.3v). The PSU provides outputs not unlike the Type
234A used for powering the R1132 and R1392 receivers. I recall
looking inside after I bought it and found a series pass circuit
using power valves to provide a variable output voltage.
It uses five valves, a pair of 5B/254M
(CV428) beam tetrodes for output voltage control and three diode
rectifiers, type EY84. As I recently (2019) needed to repair
this equipment I checked the circuit details. The two 5B254M
valves are connected in a series pass circuit with their anodes
supplied by two HT rectifiers in a full-wave swinging choke arrangement
with the HT transformer centre tapped to ground. The third diode
rectifier cathode is fed from the same HT transformer winding
to produce a high voltage negative supply. This is filtered through
a pair of capacitors and a 470Kohm resistor to supply a bias
voltage to the grids of the 5B254Ms. The HT output voltage is
determined by the setting of the slider of a potentiometer wired
across the HT and the bias voltage. All nice and simple and extremely
rugged and straightforward when compared with its modern solid-state
When mine failed, due to connection
across something modern that happened to fail short-circuit,
the excess current through the 5B254Ms resulted in failure of
screen grid resistors. One burnt up and the other went high in
value. These appeared to have once been 47ohm resistors. I fitted
two new resistors plus two new anode "stopper" resistors
which had risen in value from their marked 100 ohms. The 1Kohm
grid resistors were both a bit high at 1.5Kohm but as they're
not critical I left them. There are a couple of cathode resistors
which are high stability 10ohm ceramic bodied components in perfect
condition. The failed anode and screen resistors were very low
wattage and perhaps chosen as such to act as fuses in the event
of an overload?
a new power supply I'm building
Here's some pictures of the
inside of the Solartron SRS153.2
This is the best example of the circuit
diagram I could find, but as the circuitry is so simple you can
easily follow it.
It's often useful to have
two independent HT power supplies and I was pleased to receive
this second Solartron (above) from Steve Kaplan. This model looks
slightly newer than my first example, having a different meter.
He also presented me with this oscilloscope.
Frequency Counter type PM6611
The Philips PM6611 has a single input
connector which accepts up to 80MHz with a sensitivity of a nominal
10mV dropping off at higher frequencies. Input impedance is 1Mohm
and some switchable signal processing which helps it accept either
sine or square wave inputs and improved performance below 100KHz.
In practice this example handles inputs at over 100MHz.
Frequency Counter type 1991
The Racal 1991 has two input sockets,
0-100MHz (Input A) and 0-160MHz (Input B) with sensitivities
of 25mV rising to 50mV towards the upper frequencies. Input impedance
is selectable at 1Mohm or 50 ohms.
Counter type 1953A
This model of the Fluke 1953A
has three input sockets designed to accept 0-25MHz (Channel A),
0-125MHz (Channel B) and 0-512MHz (Channel C which includes a
prescaler). Generally the sensitivity is 30mV, falling off towards
the maximum frequency ranges. The input impedances are aound
1Mohm rather than typical 50 ohms.
Transient Recorder type DL901
I have no information on this
equipment which I bought about 10 years ago
Audio Signal Generator type FG3
The FG3 is a very hand instrument
providing extremely low frequency signals up to a couple of hundred
KHz. It'll need an external attenuator if the variable control
is found to be too inaccurate. It also has a swept output.
Frequency Counter type 9915
The Racal 9915 has a better sensitivity
(down to 10mV) than the earlier 1991model above and has two input
sockets tailored for up to 60MHz (Input B) and 40MHz to 520MHz
(Input A). Note that Input A is rated at 50 ohms whilst Input
B is 1Mohm.
Signal Generator type 2407
Power Meter type 431C
Packard VHF Oscillator type 3200B
Like some of my receivers, the
R206 and DST100 plus the W2508 Wavemeter, this VHF Oscillator
has a turret tuner carrying coils for its different wavebands.
The Oscillator has a pair of 6DZ4 valves operating as a multi-vibrator.
A design weakness is the rather small knob which not only operates
the turret, but also flips the tuning scale to match the selected
Power Unit Type
This is the AC power supply
for the R1392 and R1132 receivers although it's fine for powering
lots of wartime stuff.
Mine is missing its plug-in
moving iron meter, but works OK without it. Under the light grey
fuse panel lid there's a setting for changing the HT output between
high and low.