A secondary function of this particular Heathkit GD-1B grid dip meter is the DIODE mode. This allows a pair of headphones to be inserted into the jack. Tuning the grid dip meter to the resonant frequency of a nearby oscillator will produce a beat tone in the headphones. This can be useful when neutralizing a transmitter.
Other functions naturally include finding the value of unknown inductor or capacitor components against a known capacitor or inductor. Determining the selectivity or "Q" of a circuit is also possible, by observing the sharpness of the grid dip. This function requires the aid of a high impedance volt meter.
Here's the schematic. It draws current from the mains, and has a set of coils. At a swap meet, I was able to obtain a FULL set of coils including the low frequency coils. I knew what they were immediately and snapped them up for a reasonable price. It sure beats trying to re-wind a set of coils yourself. The coil socket is safe due to the capacitors coupling the tank circuit to the vacuum tube terminals, so there is no high voltage DC present to the external connections.
The DC filter capacitors were replaced summarily, and a few of the carbon resistors were replaced as well as they had drifted away from their nominal values. The original dial scale was in decent shape, despite the actual dial being cracked from over-tightening the set screw. It wasn't cracked in half at least. The problem was the that indicated frequency wasn't really that accurate. There was opportunity to make it better.
I removed the scale, and scanned it in. This could then be imported into Inkscape and used as a template to make a new scale. Drawing with Inkscape is a breeze once you get the hang of it. The beauty is that it is all vector drawing, so any line can be re-adjusted with ease. Aspects of the drawing can be separated by layers and even locked or hidden from view. The first step was to create a relative scale, so I put a ring of the alphabet around the dial. This relative scale was used to create a chart of data for making a calibrated dial scale as compared to a frequency counter.
The lower frequencies could be inductively coupled to the frequency counter using a similar coil. The higher frequencies would not couple this way and had to be capacitively coupled into a short coax "antenna" into the frequency counter. Once the table of data was filled in for each coil, then I could finished drawing all the other band scales on the new dial label. Yes, it was tedious, but it was worth it. To get markers on the scales, I added intermediate nodes after the circles were converted to vector paths. This made it easy to sub-divide points between places where I wanted a numeric label. The text was written out in a straight line and then applied to the path. This is a very clever method that I have employed before to get things to look just right. I was able to eliminate the interior relative band scale, and put scales for the two additional low band coils. I made sure 455khz was clearly marked.
By using the original scan as an image to trace over top, the size and shape was easily determined. The scanned image opacity was reduced to show a ghost image of what I was starting with. You can see the calibration of some of the coils was quite poor, while others were not too far off.
The Cly Institute would like to thank KA3LJL for the donation of this grid dip meter. It will be a good addition to our array of vintage test equipment.
