The subject for this month's article is the care and feeding of old tubes. Let's say you've managed to acquire a really great old radio or two, or perhaps you've picked up some old tubes at a swap-meet. In any case, it's a good bet some of these tubes are going to have problems (which may be the reason the old radio was put on the shelf and forgotten - until you came along!!). However, it's also a good bet, with a little patience and work, some of these problems can be overcome.
Although many things can go wrong with an electron tube, the most typical failure is an open filament or the filament's electron emission has weakened beyond usefulness. In the early days the filament was used as the actual electron source, whereas most modern tubes use the cathode element for the electron source and the filament is used as a "heater" - a common filament synonym. Of course an open filament means you've just became the proud owner of a new high-tech fishing float. However, with certain older tubes there is hope for a "weak" filament.
Fairly early in the tube design process it was discovered that the oxides of certain rare-earth elements were efficient producers of electrons when heated. Unfortunately these compounds are neither good conductors of electricity nor particulary strong physically. Consequently they were used as coatings on Tungsten filaments. This coating practice is also used on modern cathode equipped tubes. However, it turns out there were a few tube types that used filaments that had the electron producing material, Thorium, actually imbedded in the Tungsten as well as being used as a coating. Impregnated Tungsten was known as Thorated Tungsten. It's this type of tube filament that has a chance to be rejuvenated by a technique known as "Flashing". Fortunately, this group of tubes include the ubiquitous 01A.
The flashing process consists of two steps. First there is a relatively high "flash" voltage application to the filament, then a lower "aging" voltage is applied - each for a specific voltage and time duration depending on the tube type. Fig. 1 is a chart showing the voltage and duration values for each tube capable of using this process. Be VERY careful to note the tube types that CANNOT use the flashing process and, in fact, may be damaged by it's use. You might be interested to know this technique was worked out many years ago in the labs of RCA.
Before you get "flash happy", you're going to want to make sure the tube you're about to flash REALLY needs it. Flashing a good tube will materially reduce It's remaining useful life. So, run your suspect tube through it's paces on a tube tester first. In this case we're only interested in filament emission values. Figure 2 is emission test data to provide minimum guidelines for determining if a tube has impaired filament emissions. Although you could use this information in lieu of access to a tube tester, note that Fig. 2 is unfortunately not as complete as Fig. 1. If you use the Fig. 2 data in a simple hook-up, remember to tie the grid to the plate first as if the tube were a diode.
O.k., now it's time to flash! It's unimportant if the flash voltage is AC or DC. Perhaps the easiest source of an adjustable voltage would be your tube tester. Just make sure that if you do use a tube tester DO NOT press any test buttons, as flashing must be done without any plate current flowing. Other voltage sources might be a Variac (a continuously adjustable transformer), or a variable DC power supply. In any case, your voltage supply should be capable of 2 - 3 times the normal filament current draw for the tube being flashed.
Using FIG. 1, simply apply the flashing voltage, then apply the aging voltage. The object of this process is to drive whatever Thorium is left in the Tungsten to the surface as a replenishing oxide. Be aware that the actual results you get will be dependent on the individual tube and it's condition prior to flashing.
Loose Tube Wires:
Another problem with many old tube types is that a lead wire becomes disconnected from it's base pin. The most probable reason for this is poor or incomplete solder tinning of a lead wire and/or it's pin. After time, corrosion, arcing or thermal stress can cause the lead wire to separate Sometimes the relatively heavy current draw of the filament will fracture a poor solder connection and, as it turns out, this is the most common failure mode.
A tube that tests as totally dead, or there's no control when the test grid bias is changed, is the clue for a loose wire. However, an open filament is really the most likely cause of a dead tube. You could start with a conservative repair approach by placing the tube in a clamp of some sort, with the.pins pointed downward, and try to re-sweat (re-solder) the wires to the pins - starting with and testing the filament pins first. The problem here is that the wire or pin may be corroded, the wire is too short or the wire is actually broken. If you still suspect a loose wire you'll need to remove the bakelite base of the tube in order to inspect and repair it.
First you will need to disconnect all the wires from the pins by heating each pin with your soldering iron and sucking or wicking out the solder with an appropiate tool. Next, the tube base is removed by softening the base cement with Iso-propyl alcohol. Brush the alcohol liberally around the base of the tube, held in an upright position, keeping the cement wet until gentle twisting and pulling will allow the base to pull away from the tube.
Now you can make a final check on the filament with an ohmmeter. If you've found the problem was a loose wire then begin reassembly by making sure each wire is tinned and straight. If a wire is broken or too short then solder on an extension. Make the splice close to the tube to allow for as much distance as possible between the solder splice and the re-soldered wire/pin joint. Also, do your best to see that the inside tips of the pins are as clean and tinnable as possible.
You can use the old cement, while it's still soft and tacky, to re-attach the base after you've made the repair. After you've made sure that the old cement is very soft and in position to "squish" against the tube glass, slide the base back onto the wires and press the base back into place against the tube. Be careful not to disturb the glued base as you begin to re-solder the wires. When re-soldering the wires, try to get some of the solder to tin around the outside bottom of the pins - just enough to "catch" the pin were you know it's clean, but not so much that it would obstruct the pin from entering the tube socket.
One of our members told me that there was a tool available, at one time, that would neatly cut out a small section of the bottom of a tube pin so that you could more efficiently make a repair.
I've had some luck with tubes that have had the grid cap wire break off by "nipping" away a little of the glass around the stub of wire left at the top of the tube - just enough to solder on an extension wire, but not so much that the glass fractures. Make this extension wire flexible and have it be long enough to solder into the grid cap and still have room to work on the stub. After soldering you can use the old cap cement as above, or can use a little epoxy to replace the cap.
An old timer, that one of our members knew, use to re-align the element stacks that had gotten out of line inside some older tubes by carefully nesting the tube within the grasp of his hand and hitting the heel of his hand against his thigh (ouch!).
You might have run into hum problems that any tube can have, but were particularly troublesome in quite a few 6xx tubes of the 1930's (6F6, 6D6, etc.). This is usually caused by inter-element high resistance shorts - most often between the filament and cathode. This type of short is usually at it's worst when the tube is cold, and can be measured with an ohmmeter. Using a tube base diagram, measure from each element to all the others. Typical shorts, that cause hum, measure from around 500K to about 2M. If you find one don't throw the tube away just yet - particularly if you don't have a replacement. Sometimes the short gets "better" after the tube warms up, and/or it may work suitably in another radio.
Maybe this is old news, but a good way to identify a tube, whose number has faded away, is to fog it with your breath as you would when cleaning your eye glasses. Then, hold the tube up to the light to catch the faint image in the moisture. The glass needs to be as clean as possible for this technique to work well, BUT you don't want to use anything but a very gentle DRY wiping to clean the surface. One collector I know puts his tubes in the freezer as method of invoking the moisture. Once you've found the tube type mark It on the base with a scribe.
I hope this information helps you to save some of those good 'ole "hollow state" devices.