Lately, something has been going sick inside my RACAL RA1772. Something is a bit fishy in the state of Denmark.
OBSERVATIONS FROM SCOTLAND 28 May 2008 GM1SXX
My RACAL RA1772 calls in sick.
Lately, something has been going sick inside my RACAL RA1772.
Something is a bit fishy in the state of Denmark.
This radio is my pride and joy and I love it like a brother. I NEED to fix it!
I first noted the problem about a month ago when tuning between stations on
Medium wave. I quite often use the radio in the shack to listen to medium wave
commercial stations while dong other stuff. The problem first manifested itself
as a faint buzz (like mains leakage) when tuning off a strong carrier. On tune,
the buzz disappeared except on the strongest stations. My first thought was the
power supply. The RA1772 uses a number of regulated supply lines, all of which
can be monitored via the front panel meter. All were OK voltage wise. Since the
receiver has an IF out at 1.6Mhz, I monitored that using my ICOM PCR-1000
receiver and at the 1.6Mhz IF, the buzz was still present. All the signs still
pointed to power supply problems, they still do, but perhaps in the front end
circuitry.
Such a complex receiver as the RA1772 calls for a somewhat systematic approach
to faultfinding. I have a block diagram of the RX but no worthwhile schematics.
The few I have are bad copies of copies and cost me a pile of money for a load
of old rubbish from a well-known UK supplier who will remain nameless to protect
the guilty. He knows who he is!
.
The convenient IF out socket allowed me to get quickly to about the half-way
point within the receiver circuitry. I shifted my attention to the first mixer
stage. A case of divide and conquer, or as they say in the computer business...
'the successive approximation method' (divide and conquer) for quickly finding the solution.
I've tracked the source of the hum to the first local oscillator unit. This
tunes on the high side, tracking the signal at 36-64Mhz and feeds the first
mixer. On first listen, it sounded fine on my little ICOM RX but tuning away
from the carrier, I got a strong buzz. So the problem IS in the front end area
and afflicts the first mixer and everything else after it.
Photo thieved from Andrew Holme's website.
RACAL's unusual quad fet commutating mixer.
Please visit Andrew's site at http://www.holmea.demon.co.uk/Radio.htm
This is not your usual little Mickey-Mouse level7 mixer found inside many quality radios. This is what to expect inside a premier quality RACAL eceiver.
The first mixer is a high level type. None of
your little SBL-x's or the like. This is a 'quad FET commutating mixer' fed by a
powerful signal from the Local Oscillator. A FET wheatstone bridge for big
signals if you will.
The LO signal is amplified to a pretty high level by two transistors that run
hot enough to take the skin off your fingers (yes they did!). It then goes to
the commutating powerfet mixer to be mixed with the incoming signal from the RF
amplifier /preselector unit. The mixer uses four powerfets in a bridge
configuration. The resulting signals are then fed to the first crystal filter. I
could copy the buzz on the output so I switched my interest to the Local
Oscillator input to the mixer and sure enough... the buzz was there too.
I'm on to it, and it's definitely fun to work
on gear of this calibre. It's as well built as anything I've ever seen. The next
lid to be ripped off will be the HF LOOP unit... part of the synth, and the
obvious source of the hum. Although I don't have the schematics for this radio,
it's proving quite easy to work on because its so modular. Incidentally, the
first mixer module is actually bigger than the ICOM PCR-1000 that I'm using to
do most of the testing with! The first local oscillator delivers more RF than
some PA's I've worked with. Level 7 it most certainly is NOT!
These radios are only intimidating if you treat them as
single entities. Really, they are just a collection of fairly simple circuits.
That's the approach and mentality you need to develop when fault-finding.... and
the only one that works. This is not work, its fun.
The best advice I can give to boatanchor radios is 'get to know your radio'. By
that I mean, have a look and a listen to what the signals at all the stages
sound or look like (on a healthy radio) on a scope or another radio used as an
IF. That way, if you keep decent notes, you KNOW what to expect at each stage
and what sort of signal level to expect. Oh, and although I own a spectrum
Analyser and sweep generator plus two signal generators, I have NOT used any of
them in this attempt at repairing my pride and joy. The only test gear I've used
to date are a cheap digital multi-meter and an ICOM PCR-1000 radio. The DMM
allows static voltage measurements to be taken while the PCR-1000 radio is being
used simply as an upmarket RF 'sniffer' to sample the signals on the various
stages.
Test equipment is nice to have and cuts the time needed to find faults, but a
second radio, preferably a wideband one, a multi-meter and should you be well
off, a cheap DFM can do almost everything you might need to keep these old
radios working. I'm deliberately making this project difficult for myself so
that it might be of use to others.
To monitor the various lines for hum, I used a simple LM386 amplifier feeding a
set of headphones with an old 'recycled' waxed paper blocking capacitor on
the input.
Today, I called on 'Mr RACAL.. Henk PA8PDP for some assistance. Henk very kindly sent me some RACAL manuals for the RA1772. These were of far higher quality than the load of rubbish I paid quite a lot of money for from a very well-known UK company.
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
After coming home today, I launched back into trying to repair the RA1772.
The buzz I found on the -7V line feeding the HF Loop board should have given me a strong clue, but being tired and a bit harassed I missed it and spent more time monitoring the -7V line to the board. Eventually I decided to do the sensible thing and revisit the power supply. That -7V regulated line I hadn't noticed before turned out to be the source of the buzz. I have to eat humble pie about my earlier comments about the front end being at fault. My only excuse is my lack of sleep over the last week or so.
With the rear panel folded down (it carries the PSU board and pass transistors), I poked around with a plastic trimming tool until I found one of the bridge rectifiers with very dry-looking solder joint. I re-soldered the connection and the fault has not returned.
The RA1772 power supply is a linear type using multiple supplies fed from individual transformer taps. As such, it runs rather hot in normal use and the wiring tends to dry out and the insulation become brittle. The pass transistor on the rear apron and the various resistors and rectifiers on the PM370 PSU board generate quite a bit of heat in normal operation.
My particular radio had previous PSU problems which I traced to damaged insulation on wiring and solder joints to the large bridge rectifiers becoming 'cooked' and unreliable. This is no big deal and can be fixed. I just took a note of each wire in turn... where it connected to and then cut out the affected section of wire and replaced it with a short piece of stranded hookup wire with a soldered splice and a bit of sleeving over each joint.
The PSU reservoir capacitors are the originals and two of them are showing their age with dried-out electrolyte crusted around the pressure relief vents on each. I'll probably replace all of the large electrolytic capacitors since their capacity will have been reduced significantly due to heat and natural ageing.
There is not a lot of space available within the cabinet of the RA1772 but I suspect that a small sized computer type 12V DC brushless fan jammed into what little space is available would be a good thing to help stir up the airflow around the PM370 board and keep the bridge rectifiers in particular a bit cooler.
That's something I'll look at in the near future. Meanwhile, I'm delighted to have my favourite radio working properly again. Tomorrow, I'll be shopping for some 10,000uF and 4700u Electrolytics.
===================================================================================
CATASTROPHE!
The +ve connection to the smoothing capacitor ICI feeding the -7V DC circuit came adrift! Loud buzz followed by ... err, umm, deadly silence!
Could it get any worse? I looked at the connections to all of the large electrolytics and some of them looked poor to me. I re-made them and reconnected the +ve line that had come adrift. Now I had no -7V DC line ..... AGAIN!
How to fix it. I phoned around looking for a replacement metal canned UA723 but the prices I was being quoted were downright silly.. like £15 plus postage and VAT!
To check whether the receiver had been damaged by the -7V line going bad, I traced all the wires to the -7V line (violet ones) and lifted them from the PSU board. I then used a variable positive supply 'floated' 'upside down' as a variable negative supply. I fired it yp and checked the current... about 150ma. I then switched on the receiver and the current went to 500ma but the radio seemed dead. I was rather disappointed with this. I did other things for an hour and went back to the radio. That's when I discovered a very fine black wire connected to the -7V line on the PSU board that I had missed earlier. It seemed to read infinity on a resistance check to ground, so I connected it anyway and tried again.
VOILA! A working RA1772, but with some hum on signals.
 |
|
The RA1772 in pieces, again... and that BAD PSU board... but it's ALIVE! |
Poking around the PSU board with my fingers while the radio was on, I burned myself on the IC ML1. Regulator ML1 was pulling down the -7V line.... hence the mains hum.
This IC is the in the round metal package near the top centre of the PCB above. It is the ua723 for the -7V line. Since it was no longer in use anyway, I switched off and clipped it out. Trying the radio again afterwards revealed no trace of mains hum. The entire PSU is a conventional linear type using a number of ua723's and external pass transistors (2N3055's) .
I don't expect any more problems with the -7V regulator after I've replaced it with an LM377 negative variable regulator IC and a few components. I'll build the regulator board into the space vacated by the 2N3055 pass transistor on the original -7V line that I'll be ripping out later.
Oh Joy, miracles do happen!
 |
| A veritable mess of wires, PSU's and other stuff, but I don't care! I KNOW I can fix this easily with a DIY negative variable IC in a simple circuit. I'm happy again.! |
=====================================================================================
11th March2007
Well, the idea of using a little regulator board went out of the window. There's nowhere it can fit! Despite being a large radio, the RA1772 has little spare space inside. Instead, I built the regulator straight into the space vacated by the 2N3055 pass transistor. I used 'copper island' pads cyano glued to the rear of the heatsink and placed the few components used on that. A nearby grounding lug was hi-jacked as the regulator ground and new wires were run to the incoming power pin on the PSU board and the output wire connected to the bundle of wires (-7V lines) that go into the main wiring loom. Rather than use a variable resistor to set the voltage, I made a resistor ladder to set the correct voltage with 220R in one arm and 547R in the other. My 547R resistor was a 560R in parallel with a 27K. This sets the output voltage from the regulator at -7.140.V .. 140millivolts on the high side. Too lazy to go and try to find a fitting kit (mica washer, plastic stepped insulating spare and nut and bolt) for the regulator, I cut the old TO3 mica in two and enlarged the fixing hole in the regulator to suit the plastic spacer. I carefully filed flat the area around the hole where the metal was a little ragged and fitted the new LM337 regulator. It fits neatly in the space vacated by the 2N3055 and my radio is now back in service.
A BIG thank you to Henk PA8PDP http://www.pdp-11.nl/racal/index.html for the help given. Thanks are also due to Bertrand Velle of RXcontrol http://rxcontrol.free.fr/
73 Al.
GM1SXX
Back to OBSERVATIONS