One of the least understood components found in radios is the common electrolytic capacitor. These capacitors work on a slightly different principle than a simple paper capacitor.
When certain metals, such as aluminum, are immersed in an electrolyte (conductive solution) and current is applied, oxygen from the electrolyte oxidizes the surface of the metal forming a layer of oxide. This oxide layer is usually highly insulating and will build up as the current flows until it totally insulates the metal, thus blocking further current flow. The insulating oxide layer is extremely thin, about 1/10 the thickness of a sheet of paper. In spite of this thin nature, the capacitor can withstand voltages up into the hundreds of volts. In the most primitive capacitors (up until the mid 1940’s) all electrolytic capacitors were “wet” electrolyte types. These capacitors had an aluminum anode that was folded up into a round can and insulated from the can with a cardboard or Bakelite insulator sleeve. The outer can was the negative capacitor plate and the folded anode was the positive. The can was filled about 90% full with the electrolyte solution, which was a saturated mix of borax and ethylene glycol.
These capacitors had relatively low capacitance per unit volume for an electrolytic capacitor, the maximum being about 20mfd for a reasonably tall capacitor. The very thick oxide coating of these capacitors was not subject to serious time related dissolving, as were the later capacitor types. Any leakage from electrolyte breakdown was very quickly “healed” the next time power was applied due to the abundance of electrolyte present.
The bane of the capacitor was the slow leakage of the electrolyte, which caused the capacitor to dry out and eventually lose all of its capacity since the electrolyte was effectively the negative plate of the capacitor. As the electrolyte level dropped, so did the capacitance!
The next evolution of the electrolytic was called the “dry” electrolytic. These capacitors were identical in function to the older wet electrolytic but physically they were totally different. Instead of using the outer can as one plate of the capacitor, a second aluminum plate was used, there now being two aluminum plates (made of thin aluminum foil), which were separated by a carefully made paper separator layer. This separator layer was soaked in the electrolyte, which was the same borax mix as in the wet type capacitor. The foils were rolled up into a tight roll and put in the housing, which served only as a container. (The housing was usually connected electrically to the negative terminal for manufacturing convenience) Thus, a substantially greater surface area could be crammed into a given size can, resulting in a significant reduction in capacitor size. The can was sealed with a rubber disk on the end to prevent (hopefully) the loss of what little electrolyte was present.
This design was instantly popular, the capacitors being substantially smaller for a given voltage and capacitance rating. There was a serious problem with these capacitors. As they aged, the electrolyte did tend to migrate out of the case causing a gradual decrease in the liquid contained in the capacitor. The ethylene glycol is essential in the mix to prevent dissolving of the oxide layer in the capacitor! So as the capacitor aged and dried out, the remaining borax attacked the electrolyte causing it to become weaker, sometimes to the point of being more of a resistor than a dielectric. Some people try to “re-form” these capacitors by applying voltage in a controlled way to plate more dielectric onto the aluminum but this is futile because the liquid concentration is so low. The capacitor will immediately begin degrading again as soon as the power is removed, and will again revert to no more than a resistor that can load down the circuits until something gives. Thus, “re-forming” is an exercise in futility. The liquid content of the electrolyte would have to be replenished also to have this be a viable technique. The old capacitor must be replaced to be reliable.
The next evolution of the electrolytic capacitor was the changing of the electrolyte from the borax solution to what is called a polymer dielectric. Certain organic compounds have a chemical structure that allows them to conduct current. This material is coated in a very thin layer onto the aluminum and current is applied. The oxygen in the polymer forms the oxide layer for the capacitor dielectric when current is first applied at the factory. Since there is no liquid involved, the capacitors do not “dry out” or degrade like the old formula capacitors did. This totally eliminated the severe leakage that developed over time with the old technology. Virtually all capacitors made after 1980 use the polymer electrolyte so no re-forming is necessary. Another benefit of these capacitors is the lifespan of the capacitor is typically 40,000 hours at 25C. (Compared to less than 5 years with the old formula capacitors)
A major disadvantage of the polymer electrolyte capacitor is it has NO tolerance to reverse voltage. Any more than a volt or two applied in reverse will cause the polymer to break down the oxide layer re-adsorbing the oxygen, thus causing the dielectric layer to disappear. The oxide layer will get thinner and thinner until the capacitor will short out the next time full voltage is applied.
A second disadvantage is tolerance to over voltage. The old technology electrolytic capacitor had a “working voltage” and a “surge voltage” rating. The working voltage is the safe voltage that can be continuously applied to the capacitor for long periods of time. The surge voltage is the maximum voltage that can be applied for a short period, like ten seconds or so.
The modern polymer electrolytics have no surge voltage rating! The voltage marked on the case is the maximum that should ever be applied to the capacitor. Thus, you must make allowances in the design to account for surge voltages. This was not necessary using the old capacitors. This is a major difference in the rating of the capacitors made after 1980 and must be carefully considered when replacing capacitors in antique radios.
Typical old technology capacitors were 450 volts working voltage, 525 volts surge voltage. Thus you could use these as the filters of a 400-0-400 transformer power supply and the surge voltage rating would take care of the turn on surge. Not so with the modern capacitors! These are rated at 450 volts absolute! There is no allowance for surge! While these capacitors may take the surge for a few dozen times, the capacitor will fail early if the surge voltage is above voltage marked on the case.
The Chinese “blue” 500 volt capacitors Bill sometimes sells are ideal for these sets that use the 400 volt secondary.
For sets that use even higher transformer voltages, series connected equal value capacitors with equalizing resistors are necessary.