[R-390] Ballast Tubes (no, not the dead horse!)

[Flowertime01 at wmconnect.com]

Paul,

If an operator walked into the shop and said, my receiver just died, odds 
were someone had pulled his antenna patch or the ballast tube died. We ask do 
you have cal tones or not? On all bands or just some?

What I want to know is - is this a common failure mode? If so, why?

The element in the ballast tube is more like a light bulb than a vacuum 
tube. Yes, they do fail more often than any tube. Every six months the tubes 
were checked in a tube checker and then measured for noise in the receiver. 
Weak and noisy tubes then got replaced long before they failed to the point of 
no signals. A ballast tube was run in the receiver until it burned open. So 
ballast tubes always failed in use. Thus the perception that they were 
problems. But they do have a shorter life expectancy than any of the tubes. This 
has been known since they were designed into the receivers to start with.

Then I measured the voltages, 25 Vac in, 15 Vac out. Correct tubes in BFO 
and PTO. So at the correct current, they are not dropping the specified 
voltage. No problem, I put in the 42R resistor.

The R390 had all 25VAC circuits. The R390 had an alternate power supply and 
could be powered from a nominal 24 Volts military vehicle power system. The 
objective of the 6082 regulators was to filter all the generator crud and 
whine off the B+ in these applications. 

The R390/A gave up on the 24 Volt source and went to just 50 - 60 Hertz 120 
/ 240 volt power sources.

The R390/A keep the 25 Volt filament because they keep the 26Z5's rectifier 
tubes. They also keep the two 5749's (VFO and BFO) in series with the 
ballast tube to stabilize the two oscillators. The goal was not rock solid 
stability. The goal was to keep the signal "readable" copyable during power line 
sags, lighting strikes and other power source problems. 

You should watch a couple Caterpillar diesel power plants get swapped on 
line at a Field Station every eight hours. We calculated that we used more 
power in the barracks with the lights and little room air conditioners that we 
did in the Field Stations operations with the receivers big air conditioners 
and stuff.

The ballast smoothes the filament voltage through the fast flashes. The B+ 
is regulated. This provides a BFO and VFO that would stay on a CW signal so 
you could type with both fingers and not need to keep one hand on the knob 
and poke it out at 25 WPM with just one finger. 

So the VFO tube is on the end of the line with one end at ground. The BFO 
is next. then the ballast is on the top of the string. You will measure 6.3 
volts on one side of the BFO tube filament and 12.6 volts on the other side 
of the filament. The 5749 tube filament current is .3 Amp. It is as such 
printed in the tube manual.  On one side of the ballast we expect to see the 
12.6 volts to ground and the other side of the ballast is what ever the 
transformer winding is producing for you at the time of measurement.

The Ballast has a characteristic property of droping .3 amps at 12.6 volts. 
As the voltage drop goes up and down a few volts the tube still wants to 
conduct .3 Amps. That's the design and it works good enough. The time constant 
is long and that's good enough to get the job done.

We use a resistor today only because it is less expensive than a ballast 
tube. We can use a resistor only because the power source in most radio shacks 
today is so much more stable than the range of source voltage and variation 
the receiver was designed and expected to operate in. 

Simple calculations suggest we use a 42 Ohm 5 Watt resistor. I have found 
that 50 Ohms works OK and 10 watts just provides a little more surface area 
to radiate heat. Running a resistor at half its rated dissipation is expected 
to increase its life expectancy.

Hope this helps

Roger Ruszkowski 33C4H   1968 - 1975</HTML>