What test equipment for DIY repair?

[Skywave]

Originally Posted by AlanBeckett

I spent many a long year beating into the Production Test people that they weren't trying to redesign the piece of equipment that they were testing. It should work. . . . etc.
Alan

I understand the essential point that you are making, Alan, but I think an expansion and extension of it is required.

Having spent many a year as a Test Engineer, it has always been an understood (if not explicitly stated) responsibility of a Test Enginer not to merely 'find the bit that doesn't work' but also to explain why it failed and, ideally, to make suggestions to the Design Team of what - on a purely technical basis - should be done to reduce or eliminate such a failure.
And thereby hangs a classic problem. The Test Engineer's outline task is as defined above. But the Design Engineer has many other factors to consider, apart from total reliability: this, I appreciate, you are only too aware of. Consequently, there will always be compromises that have to be made by the Design Team - and sometimes these will be to the frustration and puzzlement of the guys in the Test Department. This, in turn, will sometimes trigger the more technically-capable Test Engineers to contemplate subtle, unauthorised design changes. Such activity by these guys may well be professionally mis-guided - but their intentions are well-meaning.

However, when I have a sample of vintage kit on the bench undergoing repair / renovation, a different scenario ensues: I am no longer in a 'current production, mass-produced, new item' environment. So I can wear my Post-Design hat and implement whatever changes I consider worthwhile, subject, of course, to the restraints of any impact on re-sale value and originality. But the item belongs to me, not my employer, giving me many more degrees of freedom.

Two different situations; two different sets of criteria.

Al.

[ppppenguin]

I've started a new thread on design vs test etc to save hijacking the OP's question.
https://www.vintage-radio.net/forum/...ad.php?t=64220

Perhaps the mods could move a few posts across.

[stuarth]

So back to the original subject of test equipment for DIY repair, then obviously a multimeter is very high on the list.

For most repairs it doesn't have to be too high spec, but it does need to be reliable (so a well made meter) and not give misleading results (so some digital meters and even some electronic analogue meters which are affected by, for example, AC signals superimposed on DC may not be ideal). Personally, I prefer a good analogue meter, it gives a good "feel" for what's going on in a circuit (put it on a high range and prod around the circuit) and it's side effects (mostly loading on high impedance nodes) are predictable (eg an AVO 8 on the 250V range puts a 5 meg load on the circuit. It's doesn't inject noise into the circuit, it doesn't depend an a battery, it just puts a load on the circuit. And it shows slowly moving voltages as slowly moving voltages, not as a bunch of apparently random numbers). Analogue meters are much more expensive to make (consider the last listed price of an AVO 8 compared to a good digital such as a Fluke), but are relatively cheap second hand, and will probably outlive the digital (50 year old Fluke anyone?). AVOs are well regarded, but my (OTT) meter collection includes many European and American meters which are very good, and invariably better than any far eastern "Tokyo wonder" meters.

But before moving on to more exotic equipment, I'd address power supplies. The use of isolation transformers, Variacs, lamp limiters, connector blocks, etc is discussed in many posts here, and a decent mains supply arrangement is invaluable for safe valve equipment servicing. For transistor equipment, a variable voltage low voltage power supply is also invaluable, especially for sets where the battery holders won't hold batteries when the case is dismantled. I first came across this problem in the 1960s with sets such as the Philips Popmaster, which used 4 AA cells (I think they were HP7s then) held in place by the 2 halves of the case - no way to contain them with the case opened. Back then, I used a "winner" grid bias battery for powering transistor radios under test, but you had to be careful that the fault wasn't being made worse by a failing battery. I find a metered mains supply invaluable, mine is a home made unit, 0 - 30V set by a 10-turn pot with a (calibrated) turns counting dial and variable current limit (recently useful to set a maximum current to 30mA when supplying heater power for a Pye P131MBQ under test, Dx96 range valves with series heaters.)

I know power supplies are not very exciting, but in practice they are very very useful, and should be given a reasonably high priority on your "shopping list" (or build your own).

For fixing audio and radio equipment, a 'scope is not used all that often, but a signal tracer (eg the audio stages of a known good radio) is. TV repairs can make use of a 'scope, anything from a fairly basic single channel instrument for a quick check around the timebases to a more exotic instrument with dual delayed timebases to examine colour bursts or front and back porches on video waveforms. For digital circuit, hmmm...for boards full of 74nnxx devices you may need a really good scope, especially if the design was poor and you have clock skew or timing problems, but for modern consumer "digital" devices, the digital bit is all in a proprietary IC, and I'm not sure what you'd use a scope for. Having said all that, a 'scope is definitely a "nice to have" thing, when my Tektronix dies, I'd go for a reasonably modern (= compact) instrument with 2 DC coupled channels and maybe 25MHz bandwidth. For my money, a 'scope is more of a design and development tool than a tool for fixing equipment known to have previously worked.

Stuart

[Station X]

Originally Posted by evingar

Indeed. A lot of early digital broadcast kit had timing done with strings of monostables. Additionally they contained a relative sea of badly designed PPL's, clocked to other badly designed PPLs, hardly a recipe for a stable and jitter free outcome !

What are PPL's? Did you mean PLLs (Phase Locked Loops)?

[evingar]

Originally Posted by Station X

What are PPL's? Did you mean PLLs (Phase Locked Loops)?

Indeed

[TuningIndicator]

Originally Posted by stuarth

.................and will probably outlive the digital (50 year old Fluke anyone?)...............

Well, not far off; I've got one somewhere that's 34, I've no doubt it will last another 16; short of being run-over by a lorry.......