tele-tone: 50C5 grid bias from 12BE6

Fellow Radiophiles,

A couple of years ago, I played with the idea of eliminating the cathode bias resistor and capacitor from the 50C5 in a Toptone Topline 5M-100 AC-DC set, by taking the bias for the 50C5 grid from the local oscillator grid of the 12BE6. This worked very well, but I had never seen it done in a commercial radio.

As I was looking inside my Teletone 990, I noticed that the cathode of the 50C5 was hard grounded. I had not noticed this before because I don't have a schematic. Then I noticed that the 50C5 grid was biased with a long white wire to a 470k resistor soldered to the local oscillator grid at pin1 of the 12BE6 pentagrid converter. This resistor is visible at the upper left of this photo. There is also a small red ceramic filter capacitor at the 50C5 grid pin2 to keep the LO signal out of the 50C5. Also note that wiring the 470k at the 12BE6 keeps parasitic capacitance to the oscillation grid pin1 to a minimum.

One concern with this efficient method of 50C5 grid bias, is that the voltage stay reasonably constant as the LO frequency is varied when the radio is tuned from one end of the dial to the other. This set did pretty well in this respect. The DC bias voltage provided by the 12BE6 grid varied only from -7.3V to -9V, as I changed the tuning from one end of the band to the other.

I made an adapter to replace the 12BE6 pentagrid converter with a 12BA7 with twice the conversion transconductane to improve the performance of this set. The DC bias variation over the tuning range, for the 12BA7 was even less, from -8.5V to -9.5V.

---

Flattening the LO DC self bias over the tuning range can be done by controlling the Q of the oscillating circuit with series or paralel resistance, which affect Q differently at high vs low frequencies. The self bias is proportional to the oscillation amplitude at the grid, and it is a little less than half the p-p amplitude.

If the oscillation is too strong at the high end of the dial, then some parallel resistance is needed to lower the amplitude at the high end. If the oscillation is too strong at the low end of the band, then some series resistance with the LO tuning capacitor or tuning coil is needed.

Another way to flatten the amplitude over the tuning range is to change the value of the usual 22k grid bias resistor. Increasing the value of the 22k resistor increases oscillations more at the high end than at the low end. Reducing the 22k resistor has the opposite effect.

The amplitude of oscillation can be manipulated over the tuning range because the reactance of the L and C components varies over frequency.

For a set that is capacitance tuned, the reactance of the fixed inductor is lowest at the low end of the dial, but the reactance of the tuned capacitor is also lowest here, because the capacitance is greatest at the low end. This means that the Q of this tank circuit will be more sensitive to series resistance at the low end of the dial.

At the High end of the dial, the fixed inductor reactance has doubled as the LO frequency increased from 1MHz (455kHzIF+540kHzRF) at the low end to 2MHz (455kHzIF+1540kHzRF) at the high end. The reactance of the capacitor has dropped much more, because the capacitance was also greatly reduced at 2MHz to 1/4 of the 1MHz value. Now we have a much higher impedance tank circuit, that is much more sensitive to shunt resistance.

Regards,

-Joe

To thank the Author because you find the post helpful or well done.