Learn How to Series - Valves (Tubes)

your tuned to 'boatanchors
downunder' radio
on 3402 Megs... A collaborative effort from collective sources resulting in 3.4GB
of data (2,000+ pages)ONLINE

How To Series: About - Valves (Tubes) a.k.a bottles, jars

Valves - There is a widespread public misconception that valves are extinct and unprocurable. I suspect that this myth was fostered by some sections of the electronics retailing industry as part of the planned obsolescence philosophy and a ready excuse for avoiding uneconomic servicing work. Receiving valve production has long ceased, but vast inventories have survived and there will be no critical shortage of post 1936 types in the foreseeable future. Some earlier varieties are scarce. Pre 1930 American battery valves are in demand, but there are firms still selling them. Most difficult to find locally are the pre 1935 English and European types. They were never imported into New Zealand in any large numbers and wartime destruction and depletion left few stocks in Europe. Even so, it is possible, with a bit of perseverance, to locate many spares. American and most of the post war European types are readily available from the Antique Electronic Supplies whose address is at the end of the series.

As these servicing notes have an emphasis on post 1935 receivers, valve supplies concerned are not likely to be a major problem. A most common type encountered in this group is likely to be the octal series. This is fortunate, because the octals were produced internationally, and large quantities of war surplus stocks were released at bargain prices. Later, the all glass miniatures were, with the exception of the Rimlocks, also produced internationally and supplies are still good. Similarly Loktals are not a real problem. One unexpected development has been the resurgence of valve technology in the HiFi and musical industries, whose enthusiasts insist that the only sounds suitable for their golden ears are generated by big valves and transmitted along gold plated cables made from oxygen free copper. The outcome has been continuing manufacture of major audio amplifier valves in “Eastern Bloc” countries.
The first problem is ascertaining whether or not a valve is useable. Although common only with battery valves, the easiest fault to locate is an open filament. Absence of any light and a quick test with an ohmmeter will give the answer. A related fault can be puzzling. The filament has continuity and the envelope gets warm after a few minutes, but there is no glow from the filament. The problem is that some air has leaked into the valve.

Most common and readily understood is wearing out or more correctly, loss of emission. To judge just at what stage a valve becomes unsatisfactory requires experience and an understanding of what a valve is doing in its particular application. To give an extreme example-, some- receivers used a general-purpose triode as a detector by earthing the anode and using the grid as a diode. Even with a very strong signal, the cathode was only supplying a few microamperes, well within the capabilities of the most worn out valve. Conversely he same type of valve may have been used as the heterodyne oscillator in a shortwave receiver, where a low emission specimen would not work at all well. Even brand new valves vary considerably from their published performance figures, some parameters having a tolerance of plus or minus 20%. Thus the mutual conductance of a new 6K7 may be as low as1.15 ma/v or as high as 1.74 ma/v and yet the still be within specification.

Valves can suffer from many faults. Open filaments are common in directly heated types and leave no doubt as to the valve's worth. Less obvious is leakage- or a short circuit between elements. Sometimes these can be picked up with a resistance check between pins, but unfortunately, these faults may not occur until the valve is hot. Particularly serious is short-circuiting between heater and cathode in some indirectly heated rectifiers, such as 84, 6X4, 6X5, and the EZ-- series. It was common practice in cheaper receivers to use a common, earthed filament winding on the transformer. A contact between the rectifier heater and cathode will, in this system, result in the full-rectified H. T. being short-circuited to ground. This can destroy a power transformer in a few moments, so be very alert to this problem and switch off promptly if the rectifier shows any signs of distress.

Few beginners are likely to own a valve tester, but this is not necessarily a major problem, as probably more valves have been incorrectly diagnosed as faulty by inadequate interpretation of valve tester readings than for any other reason. Many testers check little more than cathode emission and for short-circuited elements. With their BAD-?-GOOD scales, they may have impressed the customer, but did not necessarily tell the full story. Still, don’t pass up the opportunity of acquiring one if you get the chance, these checkers do provide some useful information. The major testers such as the Weston, Taylor, Precision and AVO, which test mutual conductance as well as emission characteristics are really necessary for a comprehensive assessment, but they do require a good knowledge of valve parameters to produce the best results. Valve testers, and test instruments generally are like the doctor’s stethoscope. By itself it won’t tell you anything that makes sense, but requires processing between the operator’s ears.
It is most important that if you have the chance of buying a valve tester, make sure that its manual of test figures is available.

Regardless of what tester, if any is used, the ultimate assessment is whether or not a valve works properly in a receiver that is known to be performing well. This presupposes of course that the test receiver uses the same type of valve, or its equivalent. Tune the receiver to a weak station, and compare the performance with the original valve. If the sensitivity and hiss are similar, there is a fair chance that the valve is OK In the case of converter valves, check that there are no dead spots, indicating low oscillator performance. Confirm at high volume that output valves can deliver full output without distortion. Tapping the valve with the handle of a small screwdriver should check intermittent mechanical faults'. There should be no crashing or banging noises, indicating loose welds. Note that directly heated filament valves are likely to be microphonic. This is normal. Push the valve around to in its socket to check for intermittent operation. This may result from base pin soldering problems and will be covered later. Be alert though for. Faulty sockets in receivers creating confusing contact problems of their own.
Rectifier condition can best be checked under operating conditions by comparing the HT voltage with the service data or t hat produced by a known good valve. Sometimes the two sections of a rectifier will become unbalanced. This is not of any great consequence if the HT is satisfactory, but, with one side more efficient it is possible for the ripple to have a 50Hz component which may not be filtered as well as the normal 100Hz hum. If the voltage is reasonable and hum level satisfactory, keep it in service.

VALVES (tubes) continued

Apart from bent pins, which can be carefully straightened, the glass based valves have few physical problems that can be corrected. However, the older valves are prone to a number of faults in soldering and cementing which are repairable. Today, there is a large range of adhesives for all purposes, including those that adhere strongly to glass. Considering that in the heyday of valve manufacture there was nothing of this nature available, remarkable results were achieved with glues inherited from lamp making, some as simple as a mixture of resin and plaster of Paris for fastening bases and grid caps. Now, 70 or more years later, these cements are losing their grip and due care should be taken. The most difficult time for a valve is during extraction from a socket, so ease bases gently out of sockets. One way is to use a screwdriver as a lever under the base, and grid clips should be loosened in a similar way.

If a grid cap is loose, but the wire is intact, simply unsolder the top of the cap and after carefully straightening the wire, take the cap right off the valve. With a. knife, scrape off any adhesive adhering to the glass. Make sure that the wire is well tinned. Now clean out the cap and glue it back on to the valve with a two pack Epoxy resin, superior in my opinion to instant "Power Glue" which is not gap filling and relies on intimate contact between mating surfaces. Wait a day until the glue has hardened before resoldering the lead. Excess adhesive can be removed with a sharp knife a couple of hours after gluing.

Inserting some adhesive between the glass and the base can often refasten loose bases. In this case, if the original cement has not crumbled, power glue may be satisfactory. The most serious results of base loosening occur with valves with metallized coatings such as the gold and red Philips series and grey zinc. At the junction of the base and the glass is a fine earthing wire, which also loosens when the cement bond is broken, resulting in instability and noises. The most practical repair is to first re-glue the base, and then, after cleaning some of the paint off the earthing wire, tightly and evenly wind on about a dozen turns of 5 amp. fuse wire immediately above the base. Carefully solder the earthing lead to the fuse wire and smear some epoxy resin over the winding. The result is likely to be a bit unattractive, but can be disguised by wrapping a narrow strip of tape or paper around the top of the base.
One of the best features of the all glass valves was the elimination of soldering to the base pins and grid caps. In the older separate base construction, only the tips of the lead out wires and the base pins are soldered. The bond is easily broken when the base loosens, and creates an intermittent contact, often triggered by pushing the valve around in its socket. This is the cause of a particularly annoying fault in which the heater will start to cool, causing a the receiver to fade over a period of 5-10 seconds, and even happens in valves with well cemented bases. Resoldering is sometimes successful, but not always, because often, the wires are black with oxides and probing around in an attempt to clean the surfaces is unsuccessful. Then the only remedy is to remove the base and clean the leads properly. This may seem to be a daunting process, but with a bit of patience, it is really quite straight forward. It is not a bad idea to practice on a discarded valve first!

Start by working the base a few degrees each way and it will soon loosen up. Then melt the solder out of the pins and carefully pull the base straight off so as to retain the lead configuration. The wires will be seen to emerge from the press at the top of the stem and a bit of quiet study will show that they follow a sequence with the filament leads at the middle of the press, and finishing with the anode at one side. Draw a diagram of the leads and their pins. Carefully clean the bottom of each lead with fine sandpaper and coat it with solder. It would be nice to be able to resolder the leads directly to the pins, but invariably they first need extending and again, 5 amp fuse wire is suitable. Wrap the fuse wire around the leads so that they will enter their pins without catching and solder them. Now make sure that there will be no contacts between leads. Sometimes manufacturers will have used sleeving if there was a chance of a short circuit and if necessary you can do the same. Thread the leads back into their pins and after applying adhesive to the inside of the base, carefully position it back on to the bulb. Resolder the pin tips, trim off the surplus wire and the job is done.

This technique can be used on odd occasions when a valve base is damaged, but the internals are thought to be OK A typical case is when a rectifier base arcs across and carbonizes a track. Putting on a new base will often save an otherwise unserviceable valve. There may also be times when an original type is hard to find, but a later equivalent is common. A typical example is the 6D6 and the 6U7G. Modifying a valve with a new base and grid cap is far less drastic than the alternative of attacking a chassis to change to a different type of socket.

Defective soldering to a grid cap is common and can cause frustrating symptoms similar to those caused by a faulty capacitor. The receiver seems a bit insensitive and a sudden noise impulse such as from an appliance switching will cause the sensitivity to return to normal. Unsoldering the cap will often reveal a completely un-tinned and black lead that should be scraped clean and resoldered
Frequently, a grid lead will break off flush with the surface of the glass and at first sight, the valve appears to be a write off. With a bit of care, it is possible in about 90% of cases to resurrect the valve. Most grid lead seals have a fair depth of glass, and can spare up to a half a millimeter. With a sharp, very fine triangular file make three or four cuts in the glass at the tip of the seal. On the fourth cut, the glass tip is likely to break off, leaving a short piece of wire exposed, hopefully with the internal vacuum intact. Now take a piece of the ubiquitous fuse wire and make a small loop in the middle. Lasso the projecting tip of grid wire with the loop, and carefully apply some solder. Twist the two tails of fuse wire together to become the new grid lead.

VALVES (tubes) continued

Fortunately, valves are available for the majority of receivers, but there are occasions when an exact replacement takes some locating. At this stage the question of equivalents versus alternatives arises. For example, several Columbus models introduced in 1940 were supplied with Marconi/Osram DH63 detector/1st audio valves. The DH63 is identical to a 6Q7G, and only a perfectionist would be concerned about using the more readily available latter type as a replacement. However, it is quite likely that in fact a lower gain alternative Philips EBC33 will have long since been installed. This was a common practice, due to restrictions of US dollar imports in the 1950’s. Similarly it is common to find EF39 and ECH35 as substitutes for the American pentodes and converters. There is little practical difference in performances.

It was only during the last decade or so of valve development that numbering and interchangeability became internationally universal. Even then, Europe and America retained their own traditional naming systems, but they did at least make direct equivalents. Thus Europeans made the American 6BA6 and called it the EF93, while the Americans copied the EL84 and called it the 6BQ5, but most importantly, both names appeared on the packets.

Prior to 1950, only the American Radio Manufacturers Association valves had anything like worldwide acceptance. The Philips system was the nearest that Europe had to a standard. Many of the smaller makers had their own naming systems, and even experienced technicians can get into difficulties when trying to find replacements for some of the lesser known British or European valves. Lists of equivalents are published but they are not always complete or absolutely reliable. These lists were often compiled by valve manufacturers who naturally encouraged the use of their own product, even if it involved a socket change. Old timers will remember the series of BRIMAR advertisements about substituting valves. The catch phrase was "Be Wise -Brimarise". Actually, socket changing should be a last desperate resort. Better to modify a substitute valve with a base of the required type, or make up a plug in adaptor from a valve socket and base. The exact type of valve may turn up later. Obviously, it would be impossible to cover every type of valve likely to be encountered, but I will deal with some of the main groups and some of, which may have particular problems.

Easiest range to deal with is the pre-octal American 6.3 volt series. With a few exceptions, each type was unique, and without a direct replacement or European equivalent. There are two well known interchangeable pairs though, the 77 = 6C6 and the 78 = 6D6, which, although physically different have only minor electrical differences. The main thing to watch for is that the 6C6 and 6D6 require better shielding than the others. In the majority of cases, types 41 and 42 can be interchanged, but the 42 has nearly double the heater current of the 41. Frequently, a 37 triode can be replaced by a type 76.
The American 2.5 volt A.C. heated valves can be divided into two categories. The early group comprised the 27, 224 or 24A, 35 or 51, and the 45, 46 and 47. Philips produced the early 2.5 volt series using their own numbering system, but these valves are rare:- F203 = 45. F209 = 27. F215 is similar to 56. F242 = 24A.

A type 56 will often be found substituting for a 27. There is no direct equivalent for the 35/51. A pair of type 45 output triodes might be replaced by a pair of 2A3's BUT an extra 2 amperes would be required by the, filaments -too much for many transformers. Several of the later 2.5 volt series had 6.3 volt equivalents and this resulted 'in a servicing ploy whereby a 2.5/6.3 volt transformer was connected to the sockets involved, and the more readily available
6.3 volt valves substituted. For example, it is common to find a 42 in place of a 2A5. Be on the look out for this trick because the sockets may have their original labeling, and plugging in a 2.5 volt valve could well result in a burnt out heater.

Very collectable radios from the 1930's were made in the U.S.A. by Majestic, and in Canada by Rogers. These were unique in North America in that they used valves equivalent to the RMA standard, but which were zinc spray shielded. Some had an extra base pin to earth the shielding, but many had standard bases with the screen connected to the cathode or were earthed via a chassis mounted contact finger. Replacements can be hard to find, and the standard versions may have to be used. Other manufacturers who made replacements, overcame the shielding problem by supplying the valves fitted with "Goat" shields, made of two pieces of tinplate contoured to fit like a suit of armour and held in position with a wire circlip. Goat shields were quite common and can still be found.

The Octal series was so prolific that I doubt if any collector would have difficulty in finding correct replacements. British Marconi Osram had their own numbering system. A few types were peculiar to M/O, but most were identical to standard American valves. Some European manufacturers, including Philips, made alternative octals. Their characteristics were different from the American series, although in many cases they could be used as plug in replacements. These valves were used extensively in N.Z. made receivers. Philips unwittingly created problems for collectors. During the 1950's, as already mentioned their Red Octal Series valves frequently replaced the RMA series, but they were of course, shielded. As a result many sets are now minus some valve shields!

Loctal valves were made in Europe as well as the U.S.A. and apart from the ECH21 and EBL21, most of the common types have equivalents in both systems. Replacements can generally found without trouble. Rimlocks were only made by European manufacturers and may be the first of the post war valves to become scarce. The real problems occur with the pre 1940 English and European valves. Most of the large number of makers had individual numbering systems, and some types with different bases had the same names. As an example, here is a list of some equivalents for a common type of valve, a 4, volt heater double diode / medium mu triode:-
Mullard...TDD4, Philips...ABCl, Triotron..DT436, Ever-Ready ….A23A, Marconi/Osram HD4 or DH42, Ekco. DT41, Tungsram..DDT4, Hivac.. AC/DDT, Mazda ..AC/HLDD. These are all interchangeable!
Clearly. finding replacement valves of this period entails a fair bit of research and luck. I strongly recommend obtaining lists of equivalents and above all, valve characteristic tables. Learn what the various valve parameters mean, because you can then work out your own substitutes.




2022 Copyrights \| oldradios.co.nz