Fullscreen HTML5Med Res (??MB)HTML4

Not long ago, Queensland endured some of the worst flooding on record. During this time I was offered the opportunity to recover some radio receivers that would otherwise have ended up as landfill. SALVAGING ELECTRONICS AFTER THE FLOOD T he equipment concerned was ex-Army communication receivers, Rohde and Schwarz ESMC 0.5-650MHz, about fifteen years old. The receivers were stored in a shed that that went to the roof in floodwater and remained so for a day or two. The owner had picked them up at auction and was in the process of selling them on a popular internet auction site. Electronics gear, especially the older stuff, is like artwork – its beauty is in the eye of the beholder. Where some would have turfed these straight on the scrap heap, the magpie nature in me couldn’t resist a challenge. Rohde and Schwarz equipment is well built and has a good name, so I picked up six of the better looking units to take home and repair. The rest were consigned to the pile of ruined memories on the footpath outside, like the contents of so many other flood-affected houses in our area. If you are going to try this sort of flood recovery of electronic equipment you need to pick your targets. Some things, like speakers and gear 22  Silicon Chip with non-removable batteries can be a lost cause. As with most flood events, the effects on equipment change with many different factors. Floodwater might be clean, like rainwater runoff or a broken pipe; the flood event may only last a few minutes or even seconds. In this flood, the water was brown, silt-laden (and who knows what else) and lasted a day or two, the water backing up a small creek that led to the local river. The longer gear is submerged in the flood water, the more it will accumulate silt and muck. The last local flood of this calibre was in 1974 and most places were flooded for a week or more. After that length of time immersed in flood water, equipment recovery gets very difficult. Each situation requires its own assessment as to the worthiness of an attempted recovery and special consideration needs to be given to high voltage and mains-powered gear. Never try to power up electrical by Robert Googe equipment that has come in contact with water until it has been fully checked by a suitably experienced person. And remember your personal hygiene: flood water can contain all sorts of contaminants, from industrial chemicals to raw sewage. If you are at all suspicious about any contaminants, use gloves and wash your hands! Even a face mask might not go astray. Having never really attempted flood recovery before, it was time to sit down and have a think about how to go about this. More speed, less haste! Time is the enemy but I had other commitments for a few days and could not begin the rescue process straight away. So the first thing I did was take freely available oxygen out of the equation – the less oxygen available, the slower the corrosion process. How would I do that? Well, counterintuitive as it may seem, I drowned the equipment/gear again. I figured that if I could cover the receivers with clean and (almost) ionfree water from our rainwater tank, I siliconchip.com.au Just as they were about to be auctioned off on the internet, the once-in-a-lifetime flood decided to wreak its havoc (for the second time this lifetime!). Some of these Rohde and Schwartz receivers were consigned to the scrap heap . . . but I couldn’t resist having a go at trying to clean some of them up and get them going! could achieve several things; give the gear/equipment a good flush of clean water, removing some of the silt and reducing the oxygen availability that would have been higher in air. Initially I was stumped as to how I was going to achieve this, as these units aren’t exactly on the small side, about as big as a size 13 shoe box (yes, I should know!). Something large enough to completely submerge them, obviously water tight . . . Then it hit me, a wheelie bin! So the contents of our recycling bin went onto the ground, the receivers were stacked up inside and the bin was filled with water from the tank. Check for batteries! It was just as I finished filling the bin, I realised there was something I needed to do. The first thing I should have done was to check for batteries and disable or, preferably, remove them. Batteries in water accelerate corrosion by electrolysis, so removing them is essential. That’s why so many “I-xxx” devices made by that fruit company (as Forrest Gump called siliconchip.com.au it) do not survive immersion – the (non-removable) battery voltage kills circuitry by corroding, or by the voltage being in places it shouldn’t thanks to moisture. Quickly pulling apart one of the units, I found there was a rechargeable Nicad backup battery on the processor board. I did not have time to remove each battery but it was a simple move of a jumper to take it out of circuit. This was done for each unit. The other enemy is bi-metallic corrosion. This is where two dissimilar metals can set up an ion flow due to their atomic structure; water in this case can promote this process. Thankfully, this did not present a problem in this case, probably because I was using tank (and therefore quite pure) water. Documenation Documentation is another issue – the more the better. Unfortunately for specialised gear such as this (and even more so as it is ex-military), there’s very little (read: no!) information freely available on the Internet. Thank- Even after rinsing in fresh (tank) water, there was significant evidence of the flood inside. This pic isn’t of corrosion or dust due to age, it’s good old Brisbane River mud. Further inside, I found quite a bit of the Brisbane River itself! June 2011  23 Once again, I was surprised at the lack of “damage” when I opened the lid after the big rinse. But delving down revealed a liberal coating of dirt and, in some cases, captured areas of water. All of these required thorough cleaning/drying and testing before there was even a thought of reapplying power. fully the original owner could give me an electronic copy of the operation manual which did have a short section on troubleshooting. This included a few steps on narrowing down any issues and outlined supply voltages etc. A logical order Two days later I finally had the time to take the next step and strip one of these units. After some thought, I decided on the following procedure: • Strip each board, removing covers, RF shielding, etc. • Wash each part in a tub of water with a toothbrush/paintbrush/rag. • Rinse in rainwater. • Spray liberally with methylated spirits (metho). • Gently blow excess fluid off with compressed air. • Suitably label the dismantled components. • Leave in the hot Queensland sun to dry (we were actually getting some by that stage!) The methylated spirits act as a dewatering agent, actively excluding moisture from hard to get at areas. I did think about using liberal amounts of WD40 or such but as I wasn’t sure of its effect on RF gear, it was only used on some of the connectors. I also avoided the use of a high pressure washer – this can do more harm than good by destroying fine parts and pushing contaminants into inaccessible places. Gentle water flow is best. The tools I used (apart from screwdrivers, pliers etc) included a digital camera and notebook to aid reconstruction, an electrical safety tester and multimeter. Last in, first out: the receiver at the top of the pile was duly removed from the water onto our outdoor table. I started to remove the panels and was surprised at how little mud and residue was inside. The basic construction consists of a chassis/motherboard, front and rear panels, transformer, DC-DC power supply – these units can run from both AC (100-240VAC/50-440Hz) and DC (10-32VDC) – and ten shielded boards that slot into the motherboard with multiple coaxial cable connections between boards and to the rear panel. Take copious notes and pics! For complex equipment it is impor- A little daunting, perhaps – but if approached in a logical order, disassembly, cleaning and reassembly should achieve the desired result. It’s important to handle the cleaning gently, especially when it comes to things like coils and trimmers. 24  Silicon Chip tant to spend the time taking notes, photographs and maybe a little reverse engineering for diagrams. This will aid the reconstruction. A block diagram or rough circuit will help you understand the operation when it comes time to power on and you have the inevitable faults. Back to the job at hand: each of the boards had a shield front and back, with some having more push-on tin shielding inside. It turned out that this level of shielding is ideal for water containment! This emphasises the effort and importance of stripping equipment to its lowest possible level. RF/IF boards required careful attention to remove dirt and contaminants, without the movement of numerous air wound coils. Removing the brown mud residue was important; it seemed to dry well, with little electrical conduction properties. But leaving this is fraught with danger: the problem with silt-like residue is that it readily absorbs moisture in high humidity. This could easily be a problem in the future with it becoming conductive and/or corrosive. The boards themselves seemed to have a protective coating – again this helped in the restoration. Any socketed chips were carefully removed and cleaned, as were their sockets. While most modern components are impervious to methylated spirits, one problem with using it was that it attacked the dyes on the coaxial cable labels. It did not remove them completely but it is worth keeping this in mind when you spray it around – as the commercial cleaners say, test on a small area first. The other point to note is that methylated spirits is extremely flammable and due care must be taken when using it, especially from a spray-bottle. The power supply The DC-DC switch-mode power supply PCB was removed from its covers and given the same treatment as the boards. The mains transformer was more difficult. It was a toroidal type which on first look seemed to be sealed. But on close inspection I could see internal droplets of condensation on the clear wrapping. I just had to let it bake in the hot sun and hope for the best. Other devices that could trap water, such as trimpots, switches and connecsiliconchip.com.au Even if I do say so myself, the clean-up and restoration job has come up a treat. The receivers probably look a darnedsight better than when they were taken out of service. Of course, there’s been a little bit of “elbow grease” between then and now! tors could only be dried with the rest of the board and “hope for the best”. Sunbaking in Queensland So there we had it. What seemed like a thousand screws later the whole unit was reduced to its component parts, cleaned and placed on a tarp in the sun, baking. After a full day of sun, the parts were covered, left overnight, then around mid-morning the following day, the unit was re-assembled while each part was still hot from the sun. The power transformer had lost its droplets of moisture, the power switch was no longer weeping when pressed – not a sign of moisture anywhere. Fingers (and everything else!) crossed . . . By this stage I had done two units and it had taken most of two days – you have to be keen! The next step I took was to unplug the rear fan from the power supply and connect it to my bench supply. Each of the fans was connected to an independent supply and run. Both were fine, no unusual noise, they sounded just like, well, electric fans. I decided to let the fans run for another two days, drawing air through the units and hopefully removing any further moisture. Then came the electrical safety tests. As with any piece of electrical equipment, mains insulation needs to be intact (greater than 1M at a test voltage of 500V). Using a multimeter may give the same resistance reading but only with an applied voltage of a few volts. It is important to test with an electrical safety tester for this reason – the insulation needs to stand up to 500V without breaking down. Obviously, low voltage equipment doesn’t have this requirement. To do a safety test, make sure the power switch is on and all AC mains siliconchip.com.au circuits are connected and then simply short active and neutral of the mains plug and measuring between there and ground. Unit 1 was fine but unit 2 was measuring only a few thousand ohms! This turned out to be the heatshrink covering over the power switch connections trapping dirt and moisture. Removing the heatshrink and a spray of WD40 fixed that. Applying power came next. All boards were popped from their sockets, so the DC supply was only connected to the motherboard. Initially I tested the unit from DC using a 12V car battery. The DC supply burst into life and all voltages were correct – so far, so good. Next the AC – I connected it to the mains and powered up – again fine. Plugging in all the boards, reconnecting the coaxial cables and the backup battery went without a hitch as well. Soon there was static coming from the speaker. It’s alive! This receiver is controlled via software through either a GPIB or RS-232 port. Using a terminal program, I was able to initiate the self test routines. Everything passed except for the Synthesiser 2 board, which got a fail. After pulling the board out again I closely inspected what I initially assumed to be some sort of tuned, solid aluminium RF block. In fact it turned out to be a heat-transfer mechanism from a hidden SOIC socketed chip to the shield cover. Removing the block and chip, then giving it and the socket a good clean fixed this. Unit number one was now fully operational, I could tune into the local ABC AM radio station down at 612kHz all the way up to the 468MHz emergency services. As far as the operational specifications, such as selectivity etc, are concerned – I do not know as I don’t have the facilities to test this. All I know is that it works and works well, even when compared to my little Yaesu scanner. These results really show how well built these units are. Their military equipment application says they have to be – and they are made to some of the most demanding and robust construction standards. I don’t think I’d have the same success rate with such a complicated piece of electronic equipment out of a consumer factory in China (not that it might stop me trying!). Honestly, there were only three areas of iron rust stains – two screws (the rest were stainless steel), the captive screw blocks on the rear “D” connectors and the end of one feed line cable. There was some white residue on PCBs, probably an oxide of tin or lead from the solder. So if you do come across flooded gear and you think it’s worth having a go – do it! Just remember the safety of you and others when you attempt this. For 240V applications, always make sure the equipment passes electrical safety tests before plugging it in. SC And here’s the “icing on the cake” – the label might say “fully functional” but it certainly wasn’t when removed from the floodwater. Here it is passing the software test routines. Yay! June 2011  25