[Collins] 516F-2 Query

[Dr. Gerald N. Johnson]

The 516F-2 is already has choke input filters. Choke input filters 
smooth the current peaks through the rectifiers and are essential for 
rectifier tube longevity. Unfortunately rapidly changing loads can cause 
inductive kicks to stress the rectifier inverse voltage. Tubes have 
reserve, diodes, especially early silicon diodes didn't. But the silicon 
diodes can handle a much higher peak to average current ratio that the 
5x4 family of tubes.

A choke input filter has a minimum load current to prevent the capacitor 
voltage rising towards the peak AC input voltage, and a capacitor input 
filter has a maximum load current to hold the voltage near that peak 
value. Both can give adequate voltage regulation but the bleeder to set 
that minimum current for the choke input filter sometimes represents 
serious power up to 15 or 25% of the maximum load. One limit is that as 
one increases the filter capacitor to load ratio in the capacitor input 
filter to improve voltage regulation, the conduction time of the 
rectifier gets shorter, and hence the peak current gets larger. A prime 
cause of poor power factor in power supply loads, much harmonic current 
from the power line. That comes from the rectifier only conducting for a 
short time at the line voltage peaks. The leakage inductance of the 
transformer winding works somewhat to limit the current peaks.

Choke input filters with solid state rectifiers in the 208U-3 and its 
variation used as the driver for the 821A-1 included selenium transient 
absorbers in parallel with the input filter chokes. In those days, 
rectifier diodes didn't have controlled avalanche in the reverse 
direction so the rectifier stacks included resistors to divide the 
reverse voltage evenly and capacitors to divide transient reverse 
voltages evenly. Sometimes I have found that the resistors and capacitor 
were not up to the task and in a supply of mine a resistor failure from 
too much voltage took out the rest of the diodes in the series string. 
Today the 1N4000 family has controlled avalanche and a series string 
distributes reverse voltages better without voltage limited capacitors 
and resistors.

There was a classic article in the Proceedings of the IRE about 1948 
about power supply design that has a fine chart illustrating the voltage 
regulation of power supplies taking into account the current limiting of 
the transformer and the use of choke and capacitor input filters. 
Similar sets of curves have been in many ARRL Handbooks, the first one 
to appear in the pile this morning is from 1988, and its figure 13 of 
chapter 6. I have a scan of the full wave portion of it in my on-line 
collection at http://www.geraldj.networkiowa.com/papers/Capfilterps.jpg

73, Jerry, K0CQ, Technical Adviser to the Collins Radio Association


On 1/22/2012 9:16 AM, Jim Whartenby wrote:
> I have read this topic many times on various lists with most replies agreeing
> with what is found below.
>
> If you look at the problem from the standpoint of "heat is the enemy of
> electronic equipment" then anything you do to reduce the heat load will extend
> the life of the equipment.  If you use either resistors or voltage regulators to
> lower B+, the heat load will not change.  The power transformer will still
> provide the same power to the equipment.  For the most part, the power
> transformer is unobtainium so anything done to reduce the load on the power
> transformer is a good thing.
>
> The only method I can think of is to go to a choke input filter with relatively
> low values of filter capacitors and put up with a little hum in the speaker.
>   Using a choke input filter should drop B+ significantly compared to
> a capacitor input filter.  The only issue with this approach is finding the room
> for the new filter inductor.  If the equipment already has a choke then
> eliminate the first filter capacitor.  Perhaps leave a .05 uF ceramic capacitor
> in place to eliminate any RF or noise from getting into the B+ from the AC line.
>
> Converting the B+ rectifier from vacuum to solid state will eliminate the
> voltage drop of the rectifier and raise the B+ by 30 or so volts.  This, as you
> noted, frees up the filament winding on the transformer to buck the primary
> voltage.  This will also drop the filament and B+ voltages to the other tubes in
> the radio.  As long the resulting drop is less then the 10% heater voltage bogie
> number, all should be OK.
>
> One other thing to check is that the power transformer just gets warm to the
> touch after all is said and done.  There have been issues where the chassis is
> coupled to the power transformer and causes the chassis to act as a shorted
> transformer secondary.  Spacing the transformer a bit above the chassis helps to
> decouple the chassis and reduce the effect.  The easiest way to check for this
> problem is to remove all tubes and power up the set.  The transformer should be
> only lukewarm after a few hours of operation.  If it is hotter then this, try
> spacing the power transformer above the chassis with appropriate washers.
>   Hopefully there is enough slack in the transformer leads to accommodate the
> higher mounting height.
>
> I hope this helps,
> Jim
>
>