Troubleshooting.
Appendix A
Mechanical Disassembly.

Taking something apart may seem so simple as to be not worth mentioning even in an appendix. In actual practice mechanical disassembly can be quite complex.

If you look at the back panel or bottom panel of an instrument you may see quite a few screws. Some of them hold the panel to the rest of the case while others hold internal components to the panel.

Screws which hold the panel to the rest of the case are often sheet metal, self tapping or machine screws in tapped holes. The screws which hold such things as transformers and circuit-boards may have nuts on the inside end.

It is easy to tell the difference between a screw in a tapped hole from one which is "nutted". When you loosen a screw in a tapped hole it will back away from the panel. When you loosen a screw which is "nutted" it will not back away from the panel and will act as if it is stripped out. Do not continue to loosen such a screw; try to tighten it back.

Most (but not all) manufacturers are aware that their equipment will have to be repaired one day and make it possible to get at the innards for repairs. (The worst case in the "but not all" category that the writer has seen was a piece of equipment in which a bracket had been spot-welded to the inside of the case after the chassis had been installed. The bracket prevented removal of the chassis from the case.)

As mentioned earlier in this book there is a joke among technical people which goes "When all else fails, read the instruction manual." In practice, the manual should be consulted first, not last. A first-rate manual will tell you what screws to remove and even give drawings showing how to disassemble the instrument.

Never be ashamed to read the instruction manual; it can save you a lot of time, frustration and money. Unless you are working on equipment as a hobby, time is money.

Appendix B.
Desoldering.

Reams of pages have been filled with articles about the proper techniques for removing components from printed circuit boards. The basic problem is to remove the solder so that the component lead can be removed from the hole.

There are many vacuum devices on the market for the purpose of removing solder. These range in price from two dollars to five hundred dollars. The cheap ones don't work very well; I don't know about the expensive ones.

The best and most economical method of removing solder is a product called Soder-WickR. The product consists of copper braid which has been coated with a substance which prevents oxidation and causes molten solder to flow around and along the small wires in the braid.

The writer has read magazine articles in which it was said that the braid removed from shielded cable could be used in place of Soder-WickR. Statements like that were written by people who have never tried it. The chemical coating is absolutely necessary; the braid will not "wick" the solder without it.

A recent development in electronics is the surface mount board. Special devices have been developed to make repairs to these boards. One of these is a hot air blower in which the air is actually hot enough to melt solder. You don't want to get your hand in front of one of these. Someone on the email list recently suggested that one of these devices is good for removing ICs with lots of pins from a board. I can see how this is true. However a stream of hot air may be a rather blunt instrument for cleaning up a solder bridge in the middle of a dense board. For such a job I'll bet on solder wick every time.

The proper use of Soder-WickR is a learned skill and the best way to learn it is to do it. A few pointers will be given here in the hope that they will help.

Use a soldering iron with a lot of heat capacity. Notice that I did not say a hot iron. The iron should not be any hotter than is normally used on PC boards but it should be a bit more massive than the needle tips usually used.

The wick must be hot. It goes without saying that the solder must be molten, but the wick must be above the melting temperature of solder. Hold the iron tip to the joint to be desoldered until the solder melts and then lift the iron tip and trap the wick between the tip and the joint. You will see the copper-colored wick turn silver-colored as the solder is drawn into it. Move the solder wick around a bit to wipe up the solder. Remember you are wiping up a liquid with an absorbent material. The principle is the same as cleaning up spilled water with a paper towel. The only difference is that the liquid and "paper towel" are at a temperature of 700 degrees F.

If at first you don't succeed, melt some new solder on to the joint (resolder it). That may seem to be counter-productive but solder which has been held liquid for a while "gets tired". New solder will refresh it and allow the wick to soak it up. This technique is especially helpful with plated-through holes.

Appendix C.
Component Substitution.

Substitution of a component which is not the exact replacement is usually not recommended. In spite of the best laid plans of mice and people, emergencies do arise. A substitution may be necessary to get the production line going again, the station back on the air, the lab instrument working by 1:10 PM.

Resistors.

Resistors have three important parameters--resistance, percentage tolerance and wattage rating.

Obviously the resistance must be right, but not always. There are some cases in which a resistance value which is close will get you through the weekend until you can get out on Monday to get the right one.

One example is in switching circuits. It is impossible to make any rules except one--if it works it's all right.

Percent tolerance can be substituted on a permanent basis if care is taken. Obviously a 1% resistor can substitute for a 5% resistor of the same value.

If you need to substitute a 5% resistor for a 1% resistor it can be done. You need access to a very accurate resistance measuring instrument. You also need a large number of resistors of the value which is needed. (These assumptions are quite reasonable for an industrial or university laboratory.) Simply measure the resistors until you find one which is within 1% of the stated value and put it in the circuit.

A lower wattage resistor cannot be substituted for a higher wattage one, at least for no more than a few tens of seconds. A higher wattage resistor can substitute for a lower wattage one if it will fit. Higher wattage resistors are physically larger and may not fit into the space provided for a lower wattage (smaller) resistor.

Capacitors.

Capacitors have other parameters than capacitance, percent tolerance and voltage rating. These are dissipation factor and temperature coefficient of capacitance.

The rules for the resistance and percent tolerance of resistors applies to the capacitance and percent tolerance of capacitors--if it works, do it.

A lower voltage capacitor cannot be substituted for a higher voltage one but you can substitute a higher voltage for a lower voltage one if it will fit.

The dissipation factor is never given as part of the specifications of a capacitor but it is measured by capacitance bridges. Some circuits are very critical with regard to dissipation factor while other circuits don't care. Different types of capacitors have different dissipation factors. This is the reason why there are so many different types of capacitors. For permanent replacement, you should always replace a tantalum capacitor with another tantalum, a ceramic with another ceramic, a mica with another mica and so forth. On a temporary basis if it works, do it.

Replacement of one type of capacitor with another may not have any immediately observable effect on circuit operation. If you replace a tantalum electrolytic capacitor with an aluminum electrolytic capacitor the circuit may work just fine when the capacitor is new. A few days or weeks later the circuit may begin to operate erratically.

There is one "never never" in substituting capacitors. Never, never replace a non-electrolytic capacitor with an electrolytic capacitor.

The temperature coefficient is often stated in code on ceramic capacitors. The code takes the form of a letter and a number. P for positive and N for negative. The number gives the parts per million per degree C change in capacitance. For example a P1500 is a capacitor with a temperature coefficient of + 1500 PPM / °C (parts per million per degree C). A capacitor designated N750 has a temperature coefficient of - 750 PPM / °C. A capacitor with a zero temperature coefficient is designated NPO.

A capacitor in a timing circuit or a frequency determining circuit should always be replaced with one with the same designation as the original. Failure to do so will result in unstable and unreliable operation of the equipment.

Inductors.

It seems as though there are as many different inductors as there are grains of sand on a beach. Substitution of inductors is usually not even possible, let alone practical.

You can't measure the inductance of a defective inductor. If the service manual gives a good description of the inductor you may be able to find something which will work on a temporary basis.

The usual case requires that you get the exact replacement part from the equipment manufacturer. Transformers

Everything said for inductors applies to transformers.

Transistors.

It has been said that 90% of the transistor types will work in 90% of the circuits. If you get the polarity right, that statement is true. So why are there over 10,000 transistor types? Who knows?

It is easy to tell the polarity of a transistor. Study of the schematic diagram will reveal the maximum voltage which can be applied to the transistor at any time. The physical size of the original transistor will give a good clue as to the power dissipation requirements. If the replacement transistor meets the requirements of polarity, voltage and power odds are almost unity that it will work.

Integrated Circuits.

ICs are quite a different story from transistors. Not only must the circuit function be the same but the pins must be in the right places. An MC 1310 and an LM 1800 have the same function--an FM stereo demodulator. One cannot be substituted for the other because the pin connections are different.

The pin connections of op amps now have a de facto standard and some substitution is possible. An FET input op amp can be substituted for a bipolar, a low noise for a standard, a high voltage for a standard, a wide-band for a standard, but not vice versa.

There is a whole new family of CMOS ICs which are specifically designed to substitute for TTL ICs. These can obviously be plugged right in. The 74CXX series are pin compatible but are slower speed and may not work in all circuits. For guaranteed operation you must use a series which is fully TTL compatible.

Back to Fun with Transistors.
Back to Fun with Tubes.