Museum Hours:

Saturday 10-6

Sunday 12-5

Mechanical Television

Getting the Daven Receiver to Operate

We are attempting to re-create broadcast reception of TV, using the Daven receiver and our Western Visionette scanning disk. 

Transmitter: First, we purchased a low power AM radio transmitter kit and assembled it. After assembly, we were concerned about the bandwidth of the modulation, which needs to be about 30 kHz to pass the 45 line image with decent resolution. We tested the bandwidth by connecting an audio signal generator to the input of the transmitter and a wideband scope to the RF output. We discovered that the signal started to roll off at about 8 kHz. By replacing the bypass capacitors in the audio preamplifier stages with smaller ones we were able to get good response to above 15 kHz.

The Standards Converter: We connected Darryl Hock's universal standards converter to the transmitter, and connected a video source (gray scale pattern generator) to the converter.

The Receiver: Next, we checked the tubes and resistors in the receiver. Several of the glass plug-in resistors were out of range, and a couple were open. For now, we have kept the out of range resistors in the receiver, and have hidden modern resistors under the old, open ones. 

There are a few paper capacitors in the receiver, which we replaced. There are also several mica types, which we left alone. 

We had to build a power supply for the receiver, since the commercially available battery eliminators have only 5 volt filament supplies, and our Daven tubes in the resistance coupled amplifier are 6 volt. Also, we need 250 vdc for the final amplifier stage and the neon tube.

The Scanner: Finally, we connected our Western Visionette scanner to the receiver's output. The Visionette uses a synchronous motor, so we simply plugged it in to the AC line.

Testing: The first thing that we noticed when we turned on the receiver and attempted to see a picture was that the image was negative. To solve this problem we added another amplifier stage to the transmitter to serve as an inverter. 

The image was the right polarity, but it was extremely smeared, with very low resolution. The gray scale looked like a continuous spectrum rather than 8 individual bands of different intensity. So we started looking at the reasons.

We connected the standards converted directly to the input of the resistance coupled amplifier, bypassing the receiver. The image was much better, but still somewhat blurred. Removing a .006 mfd capacitor that parallels the neon made the image much sharper. The capacitor was defective, so it was replaced.

Then we connected a signal generator to the receiver and a scope to the detector output. We varied the frequency of the generator and recorded the response of the receiver, and discovered that when the three tuning dials were peaked, the detector output was 6 db down only 6 kHz on either side of the center frequency.

We are now experimenting with stagger tuning of the three dials to get the required 15 kHz of bandwidth. If this doesn't work we will damp the three tuning capacitors to reduce their Q and broaden their response.

Stagger tuning doesn't seem to be repeatable, and the tuning drifts, so it wouldn't be a practical way to go. We've determined that damping the input tuned circuit with a 220 k resistor, and the next tuned circuit with 68 k broadens the bandwidth substantially and still retains enough gain.

The next item we experimented was installing high frequency peaking in the transmitter. We experimented with adding bypass capacitors from the emitters to ground of the two modulator amplifier stages, and found the values that would result in about a 6 db higher gain at 15 kHz than at 100 Hz.

The picture now looks reasonably good. However, there is a problem with the video of the three fields (the Visionette is 45 lines triple interlace) being displaced horizontally from each other, resulting in a straight vertical line being jagged. We have an early version of Darryl's standards converter, which may have had that defect. We will send ours back to him for an upgrade.

It should be remembered that this receiver dates from 1928, when 24 line transmission was the standard, and only about 8 kHz of bandwidth was required.