Hi All ,

Can anyone give any input to the findings I have discovered?

Example at hand , 1956 AMI jukebox amp , still works and sounds great , but it has all its original caps through out.

Figured I would give it a resto and replace them , but they all have no signs of leakage , wear and tear , all have good ESR and all in tollerance to their capacitance.

The new caps multican as the originals, have higher ESR and not as close to their capacitance as the old ones .

So why should I replace them is the arguement , if it ain't broke leave it alone , or will the new caps work better anyway.

Got me stuffed any suggestions, also all my other tube amps prior to 1963 are the same , did they make the caps better 56 years ago ? Yet younger amps of 1963 vintage have bad caps.

Regards
Vic.

did they make the caps better 56 years ago

In my experience think the answer to that is 'yes' and 'no', depending on brand.

I have a 30-valve 1959 Hammond organ which I'm restoring. So far I've been amazed at how good the original caps are, including the high voltage electros which, on the basis of age alone, I don't trust and plan to replace. Back in the day Hammond were known to use quality components in their organs, but I wonder what the capacitor manufacturers would have stated as the expected life of their products? I think we can guess they would certainly not say 50+ years!

The organ in question came from a family that used it fairly frequently until recently so I guess the caps have been kept "maintained". Electrolytics tend to dry out if they are not used.

By way of example, I've had electros explode on me in a 1970s Teac tape recorder and a 1970s Trio oscilloscope (both solid state circuitry) when they were powered up after a long period of non-use. (I have since learned to power such gear up slowly with a Variac after looking for obvious signs of component distress.)

The composition of electrolytes has changed over the years, as has the physical size of the electros. I think the older ones perhaps vent more easily and perhaps are less prone to explode?

And the wax caps (Ducon, etc) that you find in Aussie radios have a poor reputation for longevity with many restorers simply replacing them as a matter of course.

If you looks at the specs for today's electros you'll see many with a stated life of only 1,000 or 2,000 hours. In the USA you can find electros rated for 8,000 hours but they are comparatively expensive. I guess components are made to suit the throwaway nature of gear these days.

Regular use would have extended the life of your amp.

The old capacitors are large, whereas today they are tiny and highly stressed, crammed in next to hot components.

There is a tale on the internet about faulty capacitors which involved industrial espionage. Apparently they knew someone was trying to steal their formula so only a fake formula was kept on their computer system.

Here is an article you might like to read:-

http://en.wikipedia.org/wiki/Capacitor_plague.

If you are using the amp yourself, I would not replace them.

^ Very interesting Wiki article on "the capacitor plague".

Interesting article,

Having learnt to check and repair the old fashion way has done me well in the past but I wanted more this time around.

I my stable of test equipment I have various types of tube testers, and now have a high voltage capacitor testers. I decided to check my theory on some old caps that looked ok and some that had visual signs of leakage.

The old non leaky caps tested ok up to their voltage rating and the leaky caps only just failed the test ,but at their actual operating voltage where still good. So yes unless they are shorted prematurely because of manufacturing faults these large capacitors still work.

So what I found was unless you test at high voltage capacitors show up ok on all other test equipment.

I noticed some members don't like tube testers, well I have tubes that may work in an amp or radio but have shorts or leakage in them, unless you replace all tubes with new how do you know what u are doing. Plus I have found tubes that show faults on tube testers really don't work 100% in real use.

Now I never used tube testers before , but now find how can anyone that repairs electronics do without proper test equipment. See in my teens tube stuff was relatively new, so resistors and caps were the major fault, then along with tubes. Now days these old resistors and caps are way off their values, so i'm taking the guess work out of the equation.

Quite amazing the theorys and results that have come out since I have started using high voltage test equipment.

These are my personal views and findings that have evolved over the years and that I thought I would share them with members.

Regards
Vic.

I noticed some members don't like tube testers, well I have tubes that may work in an amp or radio but have shorts or leakage in them, unless you replace all tubes with new how do you know what u are doing. Plus I have found tubes that show faults on tube testers really don't work 100% in real use.

Valve testers have their uses but they are not the be all and end all indicator of a valve's operating condition.

Back in the heyday, simple DIY shorts and emission testers were positioned in repair shops so that customers could bring in and test their own valves as being simply 'Good' or 'Bad'. A 'bad' indication didn't necessarily mean that the valve was at the end of its working life, but it sure made the task of selling replacement valves a whole lot easier.

Even the most sophisticated valve tester can only approximate the operating conditions of the plethora of valves it accommodates. The old saying goes that the best test of a valve is the actual circuit it was designed to be used in, hence swapping new for existing in the hope that the fault will disappear.

Nonetheless, if you have a batch of valves of unknown condition, and don't have a radio or TV chassis to test them in, then a valve tester is your next best choice in order to try to sort the outright duds from the maybes.

The non polarised paper caps always tended to leak. I have an RC bridge here that I have to repair however, it's instruction book actually gives an ohmic value for them in a particular position at which point if they go below it, they should not be used.

The book is burried in the move but I sent Kevin Chant a copy in case mine got lost. ( "Honor" made for Lafayette )

In this day & age most of the paper caps are past their use by date and I rarely test them. These leaking caps as coupling caps can seriously impinge on bias, by bleeding positive onto the control grid. The new one's like Metallised Polyester, don't leak.

The electrolytic will fail if it does not leak, as they tend to need this leakage to maintain the oxide coating within, that is the insulator.

The general formula for that current (in micro amps) is 0.01 times capacity in mfd by applied voltage. The modern ones are less suceptible to loosing polarity, when not used, (form) than the old ones, but over time they will.

Not all modern electrolytics are of good quality and I have had 450V types fail after 2 years.

I havea variety of ways of testing caps, however the insulation tester is best on non polarised caps.

You do not use such a device on electrolytics. Valve & Circuit testers like the Paton VCT actually measure the current through the cap.

Marc

I think the main reason I don't have a valve tester is because there's not many available at a reasonable price. Those that get sold on Ebay usually sell for inflated prices, driven by last-minute bidding frenzies.

With that in mind, I've always been content with trying valves in radios that are known to work. A valve tester will come one day but it should be remembered that test equipment is just as vulnerable to faults and failure as a radio.

I think the main reason I don't have a valve tester is because there's not many available at a reasonable price. Those that get sold on Ebay usually sell for inflated prices, driven by last-minute bidding frenzies.

In many quarters they have become a collector's item, snapped up by people who want to display them rather than use them ... like radios.

Many of the Australian models that I see offered are either incomplete or in pretty poor condition. The transformers in valve testers are purpose built and almost impossible to replace -- other than by salvage. Another problem is lack of documentation.

Another problem is lack of documentation.

I was giving that some thought when I read the updates to this threat this morning. With a circuit diagram and some suitable parts, the meter with the correct markings, the various valve sockets and rotary switches, etc it would be easy enough to build new machines. Some key cutting joints are agents for businesses that engrave formica and it'd be the best way of making front panels and the cabinets could be of plywood.

It is am ambitious exercise I reckon but not impossible.

It is am ambitious exercise I reckon but not impossible.

It would depend on the functional spec and hence complexity and cost. If your intention is to test a limited range of valves, with a limited range of voltages for plates and heaters, etc, then it would be possible.

Then you need to decide if it's to be a solid state or valve unit itself.

I don't think RT&H ever published a DIY valve tester, but I guess one of the many American electronics magazines would have covered it at some time, back when.

If I was to ever try such a project I'd make it as authentic as possible though because I don't have a perfectly imtimate knowledge of valve circuitry I admit I would only do it if I could emmulate a commercial circuit.

I should add that the difference between a comparatively simple shorts and emissions tester and one that can measure transconductance ups the ante, too.

It is not impossible to build a valve tester in this day and age and the transformer issue is easily overcome.

One of the members of our Radio Club has recently built one based on an EA circuit I think.

There are intigrated circuits and other modules that can provide a wide range of voltages, that can be pre-set and the unit does not have to be a basic emmission tester. The Radiotron handbook actually shows a method of feeding signal into a Valve that will allow you to study the signal.

I have used an LR8 Voltage regulator in several applications this will take up to 450V input and can be used to set voltages from around 5V to 400V. Of course you can shunt it if you need more current from it.

You can do similar with the filaments & heaters as the majority that are not DC polarised, won't be too fussed by being fed DC.

I did a few years ago get two identical Krieslers from the same family. One was a daily driver the other had, typical of them, an open OP transformer.

I had to put new Filters in the daily driver as the hum said they were not far off going. It also had an intermittent short.

The other one I left them in as I checked them as a matter of interest & they were OK. The set with the replaced OP transformer, actually came back this year after a resistor failed. The filter caps were still holding. I did not leave paper caps in either set.

Marc

the transformer issue is easily overcome.

I mentioned transformers in the context of repairing or restoring vintage valve testers which provide varying voltages via numerous multi-tapped secondaries in transformers which were purpose-wound for the manufacturer.

If you start from scratch today then of course you can take advantage of today's technology. As I said, you need to decide if you want to go the solid state or valve route.

One of the members of our Radio Club has recently built one based on an EA circuit I think.

I wasn't able to find reference in RT&H or EA to a DIY valve tester. Out of interest, I'd like to know which magazine and issue it is from.