Silicon ChipHindsight is a wonderful thing - September 1999 SILICON CHIP
  1. Outer Front Cover
  2. Contents
  3. Publisher's Letter: Email us with your ideas for editorial content
  4. Feature: Automatic Addressing on TCP/IP Networks by Greg Swain & Bob Dyball
  5. Feature: BreezeNet: Wireless Networking Without The Hassles by Greg Swain
  6. Project: Autonomouse The Robot by John Clarke
  7. Serviceman's Log: Hindsight is a wonderful thing by The TV Serviceman
  8. Project: Voice Direct Speech Recognition Module by Ross Tester
  9. Feature: Internet Access - Reduced Prices by SILICON CHIP
  10. Order Form
  11. Vintage Radio: Vintage hifi stereo AM radio by Rodney Champness
  12. Project: Digital Electrolytic Capacitance Meter by Eugene W. Vahle Jr.
  13. Project: An XYZ Table With Stepper Motor Control; Pt.5 by Rick Walters
  14. Product Showcase
  15. Book Store
  16. Back Issues
  17. Project: A Peltier-Powered Can Cooler by Ross Tester
  18. Notes & Errata: Burglar alarm extensions / Audio-Video Transmitter / Daytime Lights for Cars / Line Dancer Robot
  19. Market Centre
  20. Advertising Index
  21. Outer Back Cover

This is only a preview of the September 1999 issue of Silicon Chip.

You can view 34 of the 96 pages in the full issue, including the advertisments.

For full access, purchase the issue for $10.00 or subscribe for access to the latest issues.

Items relevant to "Autonomouse The Robot":
  • Autonomouse The Robot PCBs patterns (PDF download) [08409991-3] (PCB Pattern, Free)
Articles in this series:
  • Autonomouse The Robot (September 1999)
  • Autonomouse The Robot (September 1999)
  • Autonomouse The Robot; Pt.2 (October 1999)
  • Autonomouse The Robot; Pt.2 (October 1999)
Items relevant to "Voice Direct Speech Recognition Module":
  • Voice Direct Speech Recognition PCB (PDF download) [07109991] (PCB Pattern, Free)
Items relevant to "Digital Electrolytic Capacitance Meter":
  • Digital Electrolytic Capacitance Meter PCB (PDF download) [04109991] (PCB Pattern, Free)
  • Digital Electrolytic Capacitance Meter panel artwork (PDF download) (Free)
Items relevant to "An XYZ Table With Stepper Motor Control; Pt.5":
  • DOS software and sample files for the XYZ Table with Stepper Motor Control (Free)
  • XYZ Table PCB patterns (PDF download) [07208991-2, 08409993] (Free)
  • XYZ Table panel artwork (PDF download) (Free)
Articles in this series:
  • An X-Y Table With Stepper Motor Control; Pt.1 (May 1999)
  • An X-Y Table With Stepper Motor Control; Pt.1 (May 1999)
  • An X-Y Table With Stepper Motor Control; Pt.2 (June 1999)
  • An X-Y Table With Stepper Motor Control; Pt.2 (June 1999)
  • An X-Y Table With Stepper Motor Control; Pt.3 (July 1999)
  • An X-Y Table With Stepper Motor Control; Pt.3 (July 1999)
  • An XYZ Table With Stepper Motor Control; Pt.4 (August 1999)
  • An XYZ Table With Stepper Motor Control; Pt.4 (August 1999)
  • An XYZ Table With Stepper Motor Control; Pt.5 (September 1999)
  • An XYZ Table With Stepper Motor Control; Pt.5 (September 1999)
  • An XYZ Table With Stepper Motor Control; Pt.6 (October 1999)
  • An XYZ Table With Stepper Motor Control; Pt.6 (October 1999)

Purchase a printed copy of this issue for $10.00.

SERVICEMAN'S LOG Hindsight is a wonderful thing When it comes to servicing equipment, it’s sometimes all too easy to accidentally create an additional fault – one that often doesn’t appear until after the job has gone back to the customer. Such was the case with two of my stories this month. It is a chastening fact that despite doing everything one can possibly do to get something absolutely 100% right, it lets you down at the critical moment. A friend’s wife, Patricia, had 30  Silicon Chip picked up what looked like a bargain in a garage sale. It was a white Sharp R-2A55 Carousel microwave oven with touch controls and a cooking sensor – all for only $70. Unfortunate- ly – and inevitably – there turned out to be a catch . . . it didn’t work! I was called in to check it. When I removed the covers it was all too obvious that this wasn’t the bargain of the century. The whole thing was a mass of panicking cockroaches. Fortunately, I was able to get the whole thing outside before too many had fled into the workshop. There then followed an unpleasant half hour of spraying insecticide and cleaning up the damage their excreta had caused. The fuse was blown and the protection diode and capacitor were short circuit but the death knell for the viability of this oven was the touch switch pad (key unit) which didn’t work eith­er. I discovered this only after I had to purchased the service manual ($35.50) and established which of two connectors needed to be joined to reset/clear the flashing display. My reasoning was that if I could operate the oven without the touch pad, I could eliminate the microprocessor as the cause of the trouble. But the switch pad was the stumbling block; it would cost $65 and what with all the other items, including a missing plate and display window, the estimated cost (trade) came to a shade over $200. On this basis, the project had to be abandoned, which left Pat without an oven. Luckily, I had an old but otherwise immaculate Toshiba ER-562ETA oven, surplus to requirements. This somewhat more modest oven hadn’t been used for years but after checking it out, it proved to be in good working order. And so I offered it to Pat with a 12-month warranty for $70 (at least that would partly cover me for the expenses involved in working on the other one). Her husband picked it up and said he could install it and I thought no more about it – at least not until he casually men­tioned at a social gathering a week later that when he got it home, it didn’t work. Of course, I was rather embarrassed about this, espe­cially as this was a trigger for the other members of our group to voice their shonky secondhand technician stories. There was nothing for it but to eat humble-pie until I had it back in the workshop to investigate what had gone wrong. What’s cooking As I said earlier, this was a very simple oven consisting of a heat/defrost switch, a rotary timer and a start button. And sure enough, everything was dead. The cook light didn’t illu­ minate when the start button was depressed and neither did a fluorescent tube (with the end connectors removed) or a glass of water indicate any microwave activity. I removed the cover and examined it. Everything looked in order. The fuse was OK, the oven light came on and, with it disconnected and the capacitor discharged, the diode measured OK. Interestingly, the start switch is not a simple switch; it is a spring loaded plastic lever connected to a lower door interlock relay and at first, I couldn’t quite understand how the magnetron circuit is switched on. The fundamental circuit of a microwave oven is very simple. It consists of a 240V power transformer which has a filament winding and high voltage secondaries which feed the magnetron. The cathode is connected to the high voltage via a capacitor and an asymmetric rectifier diode. The whole lot is switched fully on or off via the timer, by applying 240V to the primary. From this, I not only found out how the oven was switched on but in the process, why it had failed. The lower interlock lever on the oven door depresses the relay to the “on” position, when the start lever allows it to do so. Or that was what was supposed to happen. However, this oven hadn’t been used for a very long time and the lower Items Covered This Month •  Toshiba ER-562ETA Microwave Oven •  Akai CT2105A TV Set •  Mitsubishi HS-M54(A) VCR •  Panasonic NV-HD100 VCR •  Akai CT-2868 TV Set lever, which is spring-loaded, wasn’t returning to its normal position. And it wasn’t doing so because its lubricat­ing grease had hardened. A small squirt of CRC 2-26 on its axle and a little working back and forth restored its function imme­diately. I returned the oven and demonstrated it to Pat that same afternoon. However, I don’t think that this is the last of the innuendo I will cop from my “mates”! Blue Akai Jim Makim is a 68-year old retiree and a nicer man it would be hard to find – intelligent and articulate. In his younger days, mainly during the valve era, Jim was an electronics engineer and so could understand a lot about his set and its symptoms. His Akai CT2105A decided to go all blue and so Jim asked me to call. He would have delivered it himself but he was due to have an operation shortly and so decided against it. It was obvious that the voltage on the blue cathode of the CRT was low, causing it to go blue, but was it due to a heater cathode short or to a problem on the video output? The easiest way to check this was to remove the blue drive transistor (C505, 2SC2482). If there was no blue with no transistor in (which was what happened) the problem was unlikely to be heater cathode. Next, I swapped this transistor with the one from the red amplifier (C503) but no change. I then disconnected the drive from the main board to the blue output transistor on the CRT socket and no blue was displayed. I now felt sure that the prob­lem was not on the CRT socket but on the motherboard, mainly the drive from IC301 AN5601K. Rather than take the whole set back to the workshop, I removed the motherboard and CRT board and took them with me. I then checked all the components associated with the blue amplifi­er. There was a lot of silicone rubber compound on the board which may have corroded them but they all measured OK. The only thing that I couldn’t check was IC301 on the moth­erboard, which I now felt was the most likely culprit. I ordered a new one and made another visit when it was ready. I plugged it all back together, fully confident that I had fixed the problem but no such luck. I was mortified (again!) to find that the fault was still there. There was September 1999  31 Serviceman’s Log – continued Fig.1: the CRT board circuit for the Akai CT-2105A colour TV set. The blue drive transistor (Q505) is at extreme right, while the red transistor, Q503, is at extreme left. The blue signal comes in on pin 3 of connector CN301. Resistor R513 is circled in red. nothing for it but to return to the workshop with the whole set. Fortunately, another identical model set had just come in for a different problem, so I removed the CRT board and swapped it over. This fixed the fault, restoring the blue function completely. So what had I overlooked? Well, ridiculously enough, I have had this fault before, on early Samsung models. But because of the way I had tackled the problem, I had forgotten the cause. The component that was faulty was R513, a 12kΩ 2W resistor feeding C505’s collector from the high-voltage rail, which had gone very high. So when the blue transistor was turned on, the collector voltage dropped too low and turned the blue gun hard on. If I had checked the base voltage of the blue output tran­sistor after swapping it with the red output transistor, I would have found it to be correct. However, this wouldn’t have picked up the fault in the collector circuit. Jim was most understanding of my faux pas. The leftover part I have recently taken on a young casual assistant, Tom. He is doing 32  Silicon Chip a technical course and is seeking as much practical experience as he can get. He is getting on very well and although he has been mainly involved with audio repairs, he also dabbles with video equipment. In my experience, audio technicians often make better VCR technicians than those with exclusive backgrounds in television. They tend to have a better understanding of electromechanical interfacing. I was out doing a service call when Mrs Thomas brought in the family Mitsubishi HS-M54(A) VCR with the tape jammed inside. It was fairly urgent as Mrs Thomas needed to return the tape to the rental store as soon as possible to avoid late fees. Tom obliged by removing the tape for her but Mrs Thomas brought the VCR back a day or two later complaining that it wouldn’t rewind. This time the job fell to me, as Tom was at college that day. There was a note in Tom’s writing taped on top of the VCR, saying that it wouldn’t rewind, together with a piece of white plastic that had been found inside the machine. I removed the covers and examined the video deck which is a Mitsubishi Fo chassis Series 3. There appeared to be nothing untoward and I could see that the main pinch roller assembly had already been replaced. I made sure that the shaft was well lubri­cated and that all the other functions were working. I did notice, however, that the tape cassette was intermit­tently catching on the lefthand side as it went in and down but I didn’t put much store on that. The deck could fast-forward nor­ mally, so why wouldn’t it rewind? The idler looked fine and I confirmed that it could rewind with a dummy test cassette in the machine. The supply reel had lots of torque, the brakes were off and the take-up reel was free. Yet when a proper tape was inserted, it wouldn’t rewind at all. There was no friction worth writing about from the opposite reel, nor was there any along the tape path, with the tape wrapped around the head drum. So the problem only occurred when a real tape was used but it worked fine with the dummy. It had to be something to do with friction between the cassette and the driving wheel hub. Why is it catching? By this time, I was becoming infuriated with the cassette catching on the lefthand side as it went down and so I decided to take a closer look. The only way to find out what was causing this was to remove the ejector and run it in and down by hand, so that’s what I did. But I could find no reason at all why the ejector should be sticking. There was no friction that I could feel and so, ex­asperated by all this, I decided to fit the ejector back in the deck. This isn’t nearly as easy as taking it out and requires a bit of a deft hand and experience to install it. Basically, one has to angle it down slightly at the front, to engage two tongues with locating lugs on each side, and then push it down and forward at the same time, using a long thin screwdriver to locate the loading gear into its well. If all goes well, the ejector will line up with the screw mounting holes at the rear. The first time I did this, I used a lot of force to line up the holes and when I tried loading a cassette it was still catching and not rewinding as before. It still worked OK with the dummy tape cassette, though. By now, I was beginning to get an idea as to what was caus­ing this, so I took an old tape, unscrewed the case, removed the two spools inside and reassembled it. When I loaded this, it was easy to see what was happening. The cassette went in easily but when I looked at it from the top, I could see that the spool wasn’t in the centre of the hole in the cassette housing. Instead, it was displaced towards the rear but it could turn on engaging rewind. I removed the doctored cassette and removed the ejector again. This time, I reinstalled the ejector more carefully, without using any force, ensuring in the process that all four anchor points were lined up as well as the loading gears – as per the service manual. Now when I put the tape cassette in, it lined up exactly in the centre of the hole. And that solved the problem. When I tested it with a real cassette, I found that it would now go in and rewind properly. It wasn’t until Tom came back the next day that the mystery was completely solved; only then did I learn that he had removed the cassette housing to remove the faulty tape. I tested the machine thoroughly to make sure it wasn’t responsible for damaging the hire tape but it performed fault­lessly through many operations, with different tapes. Finally, one might ask where did the piece of white plastic come from? Unfortunately, I haven’t a clue – I couldn’t find any gears broken or chipped and can only conclude that it came from the hire tape. The Panasonic VCR And now for a couple of less traumatic episodes. I thought Mrs Laruso’s VCR was going to be a doddle. She described her Panasonic NV-HD100 VCR as having an intermittent problem in which the mechanism would stick between various play functions. The common cause of this problem is the loading motor coupling. This coupling splits, then slips on the shaft on which it is normally a press fit. It was one of those busy days and my mind was preoccupied with other things, so I worked on it in a sort of automaton mode. I had done a few of these so it wasn’t difficult. It involved completely removing and replacing the deck and loading motor assembly but I soon had the job completed. The machine performed like a bought one and I handed it back to a suitably grateful Mrs Laruso after demonstrating it performing its tricks. I thought nothing more of it for a couple of weeks until Mrs Laruso brought it back in. This time, she was complaining that the eject mechanism was going back and forth and turning the VCR off without fully accepting the tape. She could insert the tapes but they were spat out again almost straight away. This was different. In order to accept the tape, the cas­sette pushes a lever across the right­ hand photocoupler and the mech­ a nism senses that it is down when both end sensors see light from the centre LED source. By maintaining an even pressure with my hand, I could push a cassette to the bottom of the ejector assembly and get the deck to load the tape around the head. All play functions worked but I noticed when the tape came to the end that the machine did not stop immediately, implying that the end sensor wasn’t working. When I rewound it to the beginning of the tape, it also didn’t stop immediately. As it was unlikely that both end sensors were faulty, my suspicions turned towards the centre LED. With the tape in fast forward, I could cause it to stop and rewind by shining a torch onto the left end sensor and, similar­ly, stop it by shining the torch on the right sensor. I removed the deck once more and checked the LED sender. It was OK and activated a remote control infrared tester. And by very carefully suspending the deck upside down, while still connected to the rest of the VCR, I measured the voltage across the LED – there was nothing. With an incredible amount of patience and a lot of time, I traced (without a circuit) the power source to the LED through the connectors and onto the main PC board, until finally it came to an unmarked surface mounted resistor alongside a screw near the front – and then the whole mystery fell into place. When I repaired the original fault, I had replaced the original Panaso­nic screws, including one which has a head with a flange on it. This screw did not belong there; the diameter of the flange was oversized for this location. And I had carelessly fitted it so that it hit this resistor and cracked it, thus breaking the voltage supply for the LED. Fortunately, I could read 211 on one part of the resistor, which I interpreted as 210Ω but I measured it to be 200Ω. I fitted two 100Ω resistors and the correct screw to get me back to where I’d been two weeks earlier. The moral of the story is to place the screws into contain­ers, so that one reassembles them in the exact reverse order to which they were removed. Anyway, I wasn’t about to admit to Mrs Laruso that my carelessness had wiped out all the profit I had made out of the original repair – and then some. One just has to put it down to experience. The Akai TV set Mrs Clyde’s Akai CT-2868 AT TV set was still under warranty and because it was 72cm set, it required a service call. The problem was that it had a vertical hold that wouldn’t lock – intermittently! As there is no external vertical hold control, I was fairly sure it was just faulty reception. If there is serious ghosting and it is spaced at the appro­priate distance apart, it can confuse the set so that September 1999  33 Looking for an old valve? Serviceman’s Log – continued or a new valve? BUYING - SELLING - TRADING Australasia's biggest selection Also valve audio & guitar amp. books SSAE DL size for CATALOGUE ELECTRONIC VALVE & TUBE COMPANY PO Box 381 Chadstone Centre VIC 3148 Tel: (03) 9571 1160 Fax: (03) 9505 6209 Mob: 0411 856 171 email: evatco<at>mira.net AUDIO TRANSFORMERS Manufactured in Australia Comprehensive data available Harbuch Electronics Pty Ltd 9/40 Leighton Pl. HORNSBY 2077 Ph (02) 9476-5854 Fx (02) 9476-3231 it doesn’t know which vertical sync pulses to lock onto. Some earlier sets had additional modules that could be fitted, to help the set decide which pulses to select. Of course, when I arrived her set was going fine and I confirmed that the reception was ghost free. It was 34  Silicon Chip only then that she told me that it took up to a couple of hours before it misbehaved. Why didn’t she switch it on much earlier? I told her there was nothing I could do then and there but I would call back later after my next job a few streets away. In the meantime, I asked her to leave the set running. Fortunately, it was faulty when I returned but the only clue I had was that it was probably a temperature sensitive component. But it was not a job I could do in the home – it would have to come back to the workshop. When the set was finally plonked on the bench, I tackled it with an array of hairdryers and freezers. Gradually I managed to isolate the problem to several components near IC302, especially two tantalum capacitors C331 (pin 34) and C333 (pin 33), but replacing these made no difference. I also replaced electrolytic capacitors C340 and C341 on the 12V feed (pin 29) to the IC but there was still no difference. The preset vertical hold (VR303), zener ZD301, capacitor C332 and diode D306 were also all ul­timately cleared of guilt. I was beginning to feel unhappy about R351 (180kΩ), as it was very sensitive to heat and cold – but that could have been due to the frost and moisture condensing on the outside. First, I measured it in circuit with a Philips PM2505 multimeter and it read high. I then measured it again and it checked OK. By now I was thoroughly suspicious, so I unsoldered one leg and remeasured it more carefully just in case diode D308 was affecting it. When I put my prods on it, it measured exactly 180kΩ but when I re­peated the measurement a short time later, it read 250kΩ. How could this be? I repeated the procedure several times using crocodile clips (so that my fingers didn’t affect the measurement) and amazingly it measured 180kΩ one way and 250kΩ the other way. I won’t attempt to explain this phenomenon but enough was enough. I fitted a real resistor – one that measured 180kΩ both ways – and that fixed the fault. I can only surmise the resistor had gone high – for whatever reason – and the heat from the iron had made it good again but I leave all the speculation to you, and yes, I was SC sober when it happened.