3517 RALEIGH AVENUE,   ST. LOUIS PARK, MN  55416
952/926-8198
HOURS
Wednesday – Saturday 10am - 5pm
Closed Sunday - Tuesday and holidays
ADMISSION
Adults $7
Students $5
 
 
 

The 2013 Radio Workshop

Another standing room only crowd showed up at the Museum Sunday, February 17, 2013, to watch the Annual Radio Workshop, co-sponsored by the Northland Antique Radio Club.

Those who came to share the camaraderie of radio collecting, learn a few new techniques, and see how many mistakes I would make were not disappointed. We managed to provide plenty of each element.

The two stars of the show were Mike Nickolaus’s top-of-the-line 1957 Bell & Howell Series 850 (later marketed as the Columbia Masterworks Series) and Allen Lein’s custom-made characteristic curve tracer.

Bell & Howell Series 850

Although Bell & Howell was not known for hi-end home audio, this one is really quite remarkable. Aside from having every feature available at the time, including a built-in two-inch oscilloscope, twin 12-inch speakers, a horn tweeter, and more push buttons than a ‘58 Cadillac, it has a Williamson-style amplifier circuit that we looked at in some detail.

Aside from building some of the most interesting test equipment in the world, Allen Lein has energized the hobby by rebuilding the best radios of the vacuum tube era and analyzing them with his unique devices.

Lein’s Characteristic Curve Tracer creating a family of curves for a 12AU7 dual triode.

The tracer is perched atop the scope on the left that displays the four grid-bias voltage steps that are being applied to the tube under test. Grid voltages represented, from top to bottom, are 0, -5, -10, and -15 volts. The actual curves are on the Tek 455 at the right. The horizontal axis represents plate voltage (50 volts per division), while the vertical axis represents plate current (10 mA per division). Both halves of the dual triode are displayed at once, making it easy to see that the two are badly matched. The curves with the solid lines are pretty good, but the dashed lines show that half of this tube is a little tired. A perfect tube would show only four curves, since both the dashed and the solid curves would be identical.

The foundation of the curve tracer is shown on the right. The variable grid and plate voltage power supplies produce plate current changes in the tube under test. The tracer starts with a fixed grid voltage and zero plate volts, then logs the increase in plate current as plate voltage is increased. Each plate curve shows the increase in plate current for a specific grid voltage. This would be a tedious process to graph manually.

Lein’s tracer solves the problem by electronically producing the sequence of events shown below. It has the option of creating either four or eight curves; we’re showing four.

Just in case you want to build your own tracer, Al has graciously provided a schematic. This is not a project for the faint of heart. Al even wound some of his own transformers.

schematic


The big question is, “How does it compare to a good quality tube tester like a Hickok 539?” The truth is, tubes can be very accurately matched with a mutual conductance tube tester if you pay close attention to the readings. The disadvantage of the Hickok is that it can only supply 150 plate volts, while Lein’s tracer will supply either 250 or 500 volts to the plate. Lein’s tracer can also provide a much wider range of grid voltages.

As always, not everything went as predicted. We fully expected to find a mismatched pair of output tubes in Mike’s amplifier that, once replaced, would allow us to re-balance the set and show a dramatic drop in distortion. Well, the output tubes were perfectly matched and the output balance adjustments were dead on. The tracer did reveal problems with the 12AU7 dual triodes that we replaced and, after some minor tweaks, we managed to get the overall distortion down to a respectable 0.26 percent.

amplifier schematic