#1
|
|||
|
|||
1941 GE J620 Ye old ballast tube
This unit is the first series string radio I have entered and though I get the concept, it is new territory for me. The filaments are fed AC right off the line through what was a BL42D ballast tube. Someone apparently long ago tried to repair this as the tube was gutted and some ancient power resistors were in place under the chassis. The socket just became a terminal. It was botched pretty badly. I deciphered the RMA code for the pinout and such. After reading an article on ballast tubes I realized it wasn't just as simple as figuring the correct resistor values. They're not given and in the correct circuit shown in the diagrams, I lose the math upstairs in the brain department. Plus if I understand this correctly, we want the resistance value to change as the set comes up to temperature. This is a six tube set with 1800mA of filament and (2) 250mA pilots. R18 is the ballast. Any kind shove in the right direction would be much appreciated. Came here as a AK guy. Very glad this is here!
|
#2
|
||||
|
||||
Quote:
As you so rightly stated, it isn't just about a raw, fixed resistance value. During the initial power up and warm up of the tubes, the ballast changes it's resistance to regulate the current flowing through the tubes. Were this mine I'd do one of two things; #1, the most preferable, I'd find a new ballast. If your search around they can still be found, and the circuit will function as intended. #2, the hard way, use a 12SA7, 12SK7, 12SQ7, 12J5, 35Z5, and 35L6, and eliminate the ballast altogether. The 35Z5 has a tap for the dial light, so you won't loose that. I realize this is a fairly involved modification for someone new to this, but trust me it works, I've done it a few times to not only eliminate the ballast, but also on so called "curtain burner" sets as well (sets with the heater's dropping resistance built into the AC line cord). |
#3
|
|||
|
|||
Quote:
Mike |
#4
|
||||
|
||||
So looking at the datasheets for the tubes you can think of the tube heaters as a single resistor that is supposed to have 75.2V across it and 0.3A running through it.
One bit of information you didn't include is what voltage dial lamps you have 6.3V, 2.5V or 12.6V.... I'm going to assume 6.3V for my math. So if the tube heaters are 0.3A=300mA and the dial lamps are 250mA we need the resistors in parallel with the dial lamps to give the extra 50mA somewhere to go so it doesn't burn out the bulbs. R=V/A so the resistance of the 2 resistors in parallel with the dial lamps is 6.3V/0.05A=126 Ohm's...Those resistors need to be rated at P=VA or 6.3V x 0.05A= 0.315W so half watt resistors would work but 1W would run cooler. That still doesn't account for the resistor between the tubes and dial lamp....To find that guy we can say that the tubes and dial lamps (with parallel resistance previously calculated) drop 87.8V and thus we need to drop 117V-87.8V= 29.2V across that resistor and flow 0.3A through it so by previous formulas it should be a 97.3 Ohm resistor at 8.76W (I'd use a 15-20W resistor). If you want a soft start like the original ballast I would go with a CL-90 or similar thermistor in series with the power cord feeding the radio, or use a capacitor in place of the 97 Ohm resistor. That film cap would need to be about 27uF....a film cap would be preferable, but a motor run cap, a non-polar lytic or a couple polar lytics of double value connected in series with the polarity of one reversed (as in connect the cap positives together then connect one negative lead to the dial lamps and the other dial lamp to the tubes) would work.
__________________
Tom C. Zenith: The quality stays in EVEN after the name falls off! What I want. --> http://www.videokarma.org/showpost.p...62&postcount=4 |
#5
|
|||
|
|||
Yes Mazda 44 6.3 volt lamps. There’s the math that was throwing me. Calculating the third resistor. Thanks for straightening me out.
|
Audiokarma |
#6
|
|||
|
|||
Well the ballast search didn’t pan out so I started experimenting. Currently I have a 25 watt 50 ohm resistor and (2) 120 ohm 2 watts in parallel with the pilots. Without current limiting upon power up it blows the lamps. If I power with the dim bulb, the lamps survive and dim slightly once the tubes are conducting. Getting closer but I’m out of lamps and wondering if b+ will start to suffer if I keep increasing the parallel resistors. Is there another way to quash the inrush that’s eating bulbs?
|
#7
|
||||
|
||||
Are you using a CL-90 thermistor for inrush current limiting, if not one of those might be a good idea.
__________________
Tom C. Zenith: The quality stays in EVEN after the name falls off! What I want. --> http://www.videokarma.org/showpost.p...62&postcount=4 |
#8
|
|||
|
|||
I am not. I have heard of them. So this would be installed in series on the ac line in? Certainly willing to try.
|
#9
|
||||
|
||||
Yes.
__________________
Tom C. Zenith: The quality stays in EVEN after the name falls off! What I want. --> http://www.videokarma.org/showpost.p...62&postcount=4 |
#10
|
|||
|
|||
Quote:
|
Audiokarma |
#11
|
|||
|
|||
Quote:
Anyway, I would use the dropper cap without hesitation! Using a cap has a cushioning effect on the inrush current, so you don't need any other component. |
#12
|
||||
|
||||
Quote:
The caps are a better solution operationally.
__________________
Tom C. Zenith: The quality stays in EVEN after the name falls off! What I want. --> http://www.videokarma.org/showpost.p...62&postcount=4 |
#13
|
|||
|
|||
You might've used the 50CY column.
|
#14
|
||||
|
||||
Actually the online calculators take numerical input for frequency and capacitance and or desired impedance do the math and give numerical output...
I think I used the same calculator to try and whip up a ballast for my TS-18, but the value given didn't produce the right results on the variac and experimentally finding the value showed me I didn't have the caps I needed on hand...I ended up temporarily making one out of resistors I had, and plan to order the caps I need next time I stock up.
__________________
Tom C. Zenith: The quality stays in EVEN after the name falls off! What I want. --> http://www.videokarma.org/showpost.p...62&postcount=4 |
#15
|
|||
|
|||
Quote:
|
Audiokarma |
|
|