Soldering surface mount ICs

I was interested in the Serviceman’s technique for removing and replacing a surface-mount IC using a "flow soldering" approach (February 2005, page 50). Would it be possible for someone to spell out exactly how you do this?

There are a few SILICON CHIP projects which interest me but which I have so far not had the courage to attempt after reading the recommended way of soldering an SM IC.

The first problem I foresee is holding the IC in place while making the first soldered joint. Could this be done with a very temporary, quite weak, adhesive such as BluTak? What about then following this by "flow soldering" as described, followed by desoldering braid? (J. N., via email).

First, we should mention that the Serviceman uses this "flow soldering" technique because he lacks the correct tools for the job, which would be prohibitively expensive if rarely used.

The technique will work on all fine-pitch surface-mount chips that have rows of pins on only two sides. The idea is actually quite simple, as follows:

(1). Short all the pins on one side together by bridging them with solder. This is done by applying a lot more solder than is normally necessary to each pin.

(2). Heat the IC to soften the glue that is used during assembly to hold it in place. Heat should also be applied to the opposite side of the PC board, under the IC. A hair dryer might be suitable for the job.

(3). Slip a fine-bladed instrument under one side of the IC and apply light upwards pressure.

(4). Heat the solder bridge formed in step (1), which should simultaneously melt the solder on all pins on that side.

(5). Lever the IC up to lift the pins away from their pads, just clear of the solder.

The second side can be removed in a similar manner. A desoldering tool or fine desoldering braid can then be used to remove excess solder.

Note that two disadvantages to this method are the possibility of damage to the PC board due to lifted pads as well as pin malformation of the IC itself.

We don’t describe many projects employing SMD ICs because they can be very difficult to hand solder. In fact, most manufacturers do not recommend hand soldering for fine-pitched devices.

We described how to solder a 0.5mm-pitch device in the "High Efficiency Power Supply For 1W Luxeon Star LEDs" project in the January 2004 edition. As you point out, one of the more difficult parts of the job is locating and holding the IC precisely over its pads. Unfortunately, short of using commercial rework gear, we know of no easy way this can be achieved.

Improving A 240VAC Inverter's Output I have a question or two for an expert in electricity. Recently, I have been powering my 240VAC 0.9A 50Hz freezer from a modified sinewave inverter (12V DC input, 1500W continuous). I also have an 8μF 400VAC capacitor from an old water pump (240VAC 2.5A 50Hz). Would there be any advantage in making a junction box with the above capacitor connected to the Active (as it was with the water pump) to: (1) Reduce the surge maximum on the inverter, therefore making 1200W available for other applications during fridge start up? (2) Improve the quality of the modified square wave supply by smoothing out the signal? I don’t have an oscilloscope! (A. M., via email). We would strongly recommend against connecting any capacitor across the output of your inverter. It will increase the harmonic currents from the inverter and will actually increase the total load current. It will also increase the initial surge current, if it is installed as part of the freezer. It may smooth the waveform somewhat but there will be no benefit since your freezer motor naturally blocks the harmonics of the waveform and draws current mainly at the fundamental frequency.

Marine TV antenna

I have a question which relates to life on-board a yacht. How can I automatically keep the TV antenna pointing at the transmitter while the boat yaws about from the influence of winds and tides? I guess if cost is no problem, that a kind of auto-pilot arrangement with fluxgate compass, feedback unit and motor would do it? Or maybe some kind of gyro? Any ideas? (P. R., via email).

We are not sure how you would solve that problem. You would first have to find the correct heading for the antenna (could be difficult, if you don’t know where the transmitter is) and then maintain it. A servo circuit controlled by an electronic compass would do the job but we have not published anything in this area.

Ozone generator wanted

I am trying to find an ozone generator kit, to produce ozone. I understand that SILICON CHIP or EA did one a long time ago. (K. R., via email).

No magazine that we know of has ever produced an ozone generator. We assume you are referring to a negative ion generator. These had the undesirable side-effect of producing ozone which is why EA and SILICON CHIP never produced one. ETI magazine did produce a negative ion generator in the April 1981 issue, however if you wanted it specifically to produce ozone, say for a sterilisation application, it would not be very effective.

To produce reasonable amounts of ozone in a corona discharge, you would need much higher voltages than the 3kV generated by a negative ion generator.

We can supply the negative ion generator article if you want it. Photocopies of SILICON CHIP, "Electronics Australia" and "ETI" articles are available from Silicon Chip Publications for $8.80 each including GST and P&P (Australia) or $A10.00 each including airmail P&P (outside Australia). It is the same price for a back issue if we have it in stock.

Lost RFID tags require system reprogramming

I want to build two RFID units (SILICON CHIP, June 2004) for our shop to activate and deactivate our alarm system and as an automatic door lock for our rear door. As we share the premises with a Subway store, there would be around 15 tags in circulation.

The question is, if a tag was lost (very likely to happen) is there a way to eliminate that tag’s number from the RFID module? From what I can see, you need the tag to delete it from the module. (P. D., Marshall, Vic).

The only way to remove a lost tag is to erase the entire memory and reprogram the tags again. You may need a more comprehensive system for your application. Commercial units with keypads and/or computer connections that allow editing of tag codes are now available. You’ll find some examples at: www.microgram.com.au

Magnetic field strength meter

I am trying to find a kit that was available about eight years ago. It was a digital EMR meter kit which discriminated between 50Hz and 60Hz and measures in nanoteslas, microteslas and milliteslas. This instrument could also pick up harmonics and it factors in all radiation, not just primary frequencies. Can you help me? (P. B., via email).

The project was probably the Magnetic Field Strength Meter published in October 1991. No kit was produced but the parts should all still be available.

Concern about hot components in SC480 I recently constructed two of your SC480 amplifier modules (January & February 2003) and they sound great! However, after only perhaps half a minute of operation, Q4 and Q5 get very warm. For peace of mind, I have wedged a small heatsink between the two and affixed them using thermal compound (to the heatsink) and hot glue (to the PC board). Nevertheless, the transistors still remain hot at around 55°C (with the heatsink). I realise that this is within the operating limits of the transistor but such a temperature rise (in excess of 30°C) is much higher than I expected. Q6 also seems to get quite warm at around 60°C but according to your schematic, there is around 38V across it. In addition, the 1W resistors which are raised around 3-5mm off the PC board on the power supply board get very warm as well (around 60-65°C). I am not powering anything from this ±15V power supply. Is this temperature to be expected? The output transistors do NOT get hot. I know that most of these temperatures are (well) within devices’ operating limits, so perhaps the thought of the lot "blowing up" and the financial distress involved enforces this need for reconfirmation. I hope that you can tell me if these temperatures are to be expected. (Y. P., via email). Q4, Q5 & Q6 have to dissipate about 0.33W and will therefore get quite warm in normal operation. As you point out, this is within the operating parameters of the transistors and therefore no heatsinking is required. The 2.2kΩ 1W resistors on the power supply board each dissipate about 0.28W. Again, these resistors will get quite warm. Check that the output voltages are correct (±15V), as overheating can be caused by accidental reversal of the zener diodes. Assuming that all of the DC voltages around the circuit approximate those shown on the circuit diagram, you can be reasonably confident that all is well and your investment is not about to disappear in a puff of smoke!

PortaPal plagued with hum

I have recently built the PortaPal portable PA system described in the February & March 2003 issues but have problems with it. With a good battery and mains plugpack connected, I get loud hum and "motor-boating". If the plugpack is disconnected, the noise goes but of course, there is no earth and the (unbalanced) microphone produces bad hum.

With the plugpack connected but with low battery volts (11V), I get no noise so the problem is voltage-dependent. Has anyone else experienced problems with this circuit? Are there any notes/errata I should implement? (J. W., Gladesville, NSW).

We published Notes & Errata in the April 2003 issue to the effect that the 10μF at the adjust terminal on the REG1 (charger circuit) should not be used in the PortaPal. This is because it can cause the problems you are experiencing.

Also check the stability of the amplifier. It may have a faulty decoupling capacitor.

High cost can make TV sets unrepairable

Is it normal for TV sets which require a new yoke to be unrepairable? My NEC FS 6331 has a faulty yoke and I was told to throw it out and purchase a new TV as new yokes are not available without a new picture tube.

If this is the case, could you please offer an opinion as to a brand and model of TV you consider reliable and well made, etc. (P. B., via email).

We ran your question past our Serviceman writer for his response and he replied, as follows: often, the deflection yoke is only sold as an integral custom-built part of the picture tube. In this case, this set uses a Thomson picture tube A63ADT10X505 which has the yoke (I believe!) cemented on to it (part no 33026904). I guess if you want to spend about $1000 on a 10+ year old TV, it can be repaired.

I cannot guess which brand of TV currently on sale is the most reliable – only time can tell.

Charging deep cycle batteries on a boat

On my yacht, there are two deep-cycle batteries that are charged by the diesel alternator. We stay away for periods of up to a week and when at anchor, the internal system runs on one battery with the other held in reserve so that we are able to start the diesel.

Looking at the circuitry in the Deep Cycle Charger (November & December 2004), it would seem that if two chargers were built with input from the rectified output of the alternator, the advantages of the designed charging cycle could be obtained. I assume that the display could be switched between the two controllers.

I would be pleased to have your comments. (J. R., via email).

You can use the one alternator to drive both chargers. The display could be switched between chargers by switching the data and control lines. It would require a 6-pole 2-way rotary switch. It might be easier to have two displays to avoid the switching.

Help needed with Speedo Corrector

I have constructed a Speedo Corrector kit from your book "Performance Electronics for Cars" and cannot get it to work.

Several experts have followed the decision diagram to no avail. We have installed LK1 at X, gone to procedure 2 and installed R2 and LK3 at Y. No output signal to the speedo results. The speedo and sender are from a 1991 Toyota Corolla Levin AE101 import. What is wrong? (J. O., via email).

First, make sure that the speedo-meter still works without the corrector. Then, when the corrector is connected, check for voltage at the signal input and output. In your case, with LK1 at X and LK3 at Y and R2 installed, the voltage at the signal input should be around 6-7V (multimeter set to DC volts) and the same voltage should be at the output.

If this is not the case, check through the procedure again. Make sure the correct resistor (R1 or R2) is installed and that the setting for LK3 is correct.

Solid-State Transformers No Good For DC I have purchased a 240/12V solid-state transformer for use with a 50W halogen lamp. I want to rectify the output to give DC. I have used a 6A bridge rectifier but it gets extremely hot and drops about 5V across it. Is there some reason why the output of these power supplies cannot be rectified? (G. H., via email). These so-called solid-state transformers are, in effect, high-frequency switchmode power supplies with an AC output. They are set to deliver 12V RMS to the halogen lamp but their peak output is a lot higher than that so you can’t use them to produce a given value of DC. Nor will ordinary rectifiers work with them, as you have found, as the switching frequency is very high at around 40kHz. Nor can they drive a standard rectifier/capacitor combination. The accompanying waveforms show the operation of a typical 50W solid state Halogen transformer. In waveform A, the scope sweep speed is set to 4ms/div to show the 100Hz modulation of a high-frequency waveform. This has an RMS value of 11.9V – ideal for a halogen lamp but useless for anything else. In waveform B, the scope sweep speed is set to 10ms/div so we can see the high-frequency waveform without the modulation. In this case it is running at about 38kHz. Note the peak-to-peak voltage of 39.4V. So the unit is quite useless if you want to use it to produce a given value of DC.

How to motorise Santa

We decorate our house at Christmas and have a motorised Santa that pops in and out of a chimney. This unit is home-made and is powered by a 12V DC windscreen wiper motor.

I made this decoration some four years ago and it operated fine. It was run on a motor-bike battery that had some dead cells. However, this battery was totally "gone" by the next year and upon trying a car battery or even a good bike battery, the whole thing runs too fast. Can you suggest a circuit that will slow the motor’s speed while still leaving enough power for it to turn and lift the decoration? (P. M., Tuggerah, NSW).

Have a look at the 12/24V speed control from the June 1997 issue. It should do the job.

NAND Schmitt gates explained

I wish to know what the four lines in middle of a NAND gate symbol mean. When I did my amateur course, we were only informed of the four types of gates. So what is the significance of the NAND gate symbols featured on page 69 (Circuit Notebook) of the December 2004 issue? (G. A., via email).

The symbol you refer inside gates and inverters indicates that the device is also a Schmitt trigger. In fact, the symbol indicates the square hysteresis curve of the input signal.

Different crystal for Big Digit Clock

Can a 4.194304MHz crystal be substituted for the 4MHz crystal in the Big Digit Clock featured in the March 2001 issue? The way I read and look at it, in theory when the adjustment cycle is on zero, timekeeping with the 4.19MHz crystal would be nearly spot on. Not knowing the code inside the chip, I thought I would ask. (B. H., Creswick, Vic).

If you use a 4.19 MHz crystal, the time-keeping will be much too fast. Admittedly, the binary set crystal frequencies are easier to divide down to 1Hz using discrete logic but when using a microprocessor, the division can be done using standard crystals and dividing by non-binary numbers will obtain an exact 1Hz signal.

The adjustment available in the clock is to allow for the crystal inaccuracy (typically 50 parts per million).

Auto-start on deep cycle charger

With reference to the Deep Cycle Battery Charger (November & December 2004), I would appreciate clarification of the auto-start function. Were I to build the charger, I would find this a useful facility as I would normally want to charge a deep-cycle battery but would occasionally want also to charge a starter battery (Specific #1) which I understand I can do with the link removed.

However, does the PIC controller remember the last charging cycle selected or revert to "normal" parameters as distinct from Specific #1 on a future charging cycle?

I would guess this comes down to whether the control out pin gets its data from EEPROM or RAM. (J. A., via email).

The charger remembers the last settings and uses these during the auto-start feature. If you decide to change them, just press the set switch when power is first applied to the charger and change settings. The settings are in EEPROM and so are stored until changed.

Notes & Errata Pool Lap Counter, March 2005: on the circuit diagram (Fig.1), the seven outputs from the 4511B (IC3) are shown connected to the LED array in reverse order. For example, pin 13 of IC3 should go to pin 7 (segment "a") of the LEDs, pin 12 of IC3 to pin 6 (segment "b") of the LEDs, etc. Also, the decimal point (pin 8) of DS1 is shown connected to ground via a 1.5kW resistor. Instead, this resistor should go to the +12V rail. The overlay diagram (Fig.2) and PC board (Fig.4) are correct.

WARNING! SILICON CHIP magazine regularly describes projects which employ a mains power supply or produce high voltage. All such projects should be considered dangerous or even lethal if not used safely. Readers are warned that high voltage wiring should be carried out according to the instructions in the articles. When working on these projects use extreme care to ensure that you do not accidentally come into contact with mains AC voltages or high voltage DC. If you are not confident about working with projects employing mains voltages or other high voltages, you are advised not to attempt work on them. Silicon Chip Publications Pty Ltd disclaims any liability for damages should anyone be killed or injured while working on a project or circuit described in any issue of SILICON CHIP magazine. Devices or circuits described in SILICON CHIP may be covered by patents. SILICON CHIP disclaims any liability for the infringement of such patents by the manufacturing or selling of any such equipment. SILICON CHIP also disclaims any liability for projects which are used in such a way as to infringe relevant government regulations and by-laws. Advertisers are warned that they are responsible for the content of all advertisements and that they must conform to the Trade Practices Act 1974 or as subsequently amended and to any governmental regulations which are applicable.