ConMed Hyfrecator 2000 Service Manual page 43

2000
1. Signal Tracing
This method involves injecting a signal at or near the input end of a suspected signal chain and looking for
the expected effect farther down the chain. The choice of which chain to investigate is based on the
technician's hypothesis of which part of the system could fail in such a way to explain all of the known
symptoms. This method is effective only when you understand approximately how each stage in the chain
affects its input signal.
One may start looking at the output of the first stage, and then on to the next, and so on until the output
signal fails to respond as expected. One may conclude that the fault is in the last stage checked. The stage so
discovered may be a single replaceable component, such as an IC or transformer, or an entire subassembly.
A variation on this scheme, called "divide and conquer", is to start looking for the expected response at the
farthest output. If the response there is OK, then the problem with the system is not likely related to the
signal chain first suspected, and another theory can be formulated. This can save time if one is not confident
in the initial hypothesis. If the fault does appear at the end of the chain, then look midway down the chain
to discover in which half it is located, then go to the middle of that half.
Another variation is to monitor the same point in the system while injecting test signals at different points
upstream in the chain. This method can be more difficult, because some signal stages may not be amenable
to signal injection, and it may be difficult to find a source for a proper signal.
Signal tracing is most effective where signal chains have only a few inputs and outputs, such as simple
amplifiers and power supplies. Digital systems having many inputs or complex single-chip processors
may not yield well to this method, unless one separates the signal chain into parts, each ending at a processor
pin.
2. Board-Swapping
This method is probably the least time-consuming in getting a system back into service, but it also can be the
most costly in terms of replacement costs if the defective board is simply replaced rather than being repaired.
It also requires a local stock of spare boards or a spare system which can be "cannibalized".
Board-swapping does pose the risk, in that the problem may reappear later, unless the defect in the replaced
board is confirmed by component level troubleshooting, or duplication of symptoms when installed in
another properly functioning system. All too often, the process of removing and replacing a board may jar
an intermittent component elsewhere in the system, or restores a marginal connector's continuity.
3. Shot-Gunning
This method is the component-level variation of board-swapping, where individual replaceable components
are replaced one at at time until the symptoms disappear. In some cases, a suspect component may be
left in the circuit and temporarily shunted with a known-good part; this method works well for open-circuit
faults on simple two-lead components such as capacitors and resistors.
Shot-gunning can be effective where the number of suspect components is small, and signal tracing fails to
isolate the problem further. However, it can also be both time-consuming and costly if the replaced
component count becomes large. Further, repetitive board removal and component desoldering and
installation may create failures where none existed before. Service-induced failures will void the factory
warranty, and may require replacement of an entire subassembly.
37