GR 1217-B Unit Pulse Generator.
I have decided to take on writing up at least some of the GR kit in this collection. This unit was for sale at an attractive price given that it included a very neat GR 1201-B unit regulated power supply. It did not work properly, ceasing to pulse whenever I turned the pulse width control much either side of centre. I did not do well in diagnosing the problem, diving in too deep before simply studying the thing. The result was a cycle of picking it up and putting it aside. A very perplexing issue was that the pulse width was off by a factor of ten and I got into all sorts of trouble with that one, sure that it was acting as a frequency divider even though that made no sense. Finally I saw it, a previous muddler had (re)connected the pulse width pot incorrectly, damn! And so it now works quite well. The Pulse Repetition Frequency was low and again, manual in hand went through all sorts of component value conniptions (based on the manual) only to realise upon studying the circuit calmly, that the tube condition would affect the PRF. The manual makes no mention of this yet looking at the circuit it must be, and is. I simply went through a number of tubes until the PRF came in correct at the calibrated position of the PRF pot. Such dependence on the tube is not necessary and is surprising to me considering the reputation GR has.
I don’t have a lot more to say about it. It works well after many the self-made detours. The manual claims that it can deliver rise times of less than 18nS and fall times less than 10nS. I was able to see 18nS and 10nS on my Tek 475 (250MHz bandwidth) but not less. It does find quite a bit of use on my bench.
Repetition Rate, 2.5Hz to 500kHz with calibrated points in 1-3 sequence from 10Hz to 300kHz plus 500kHz. Continuous coverage with uncalibrated control.
Duration, 100nS to 1S in 7 decade ranges.
Pulse Output Levels: + and – 40mA pulses, each 40Vpk into internal 1k load. DC coupled with DC component negative wrt ground.
The 1201-B power supply provides a regulated floating 300VDC up to 70mA and 6.3VAC at 4A. In the pulse generator, the floating voltage is referenced to ground at -150V and +150V.
The pulse repetition generator is a Schmitt trigger with a RC circuit charge/discharge appended. The C of the time constant is charged from the plate circuit of the A section via the R of the time constant that is between the plate and the grid, thus the charge and discharge of the grid drives the A section. If the A section is off (plate high), the capacitor will charge through the R until the A section turns on*, turning the “B” section off. This action happens suddenly due to the regenerative switching action of a Schmitt trigger. The B section has an inductor in its plate circuit and the sudden release of energy due to the B section turning off produces a sharp pulse that is used to trigger the output pulse timing circuit. When the A section turns on, its plate voltage will fall and the capacitor will discharge until the A section turns off again. In this way, the oscillation swings takes place within the hysteresis of the trigger. The reason for PRF dependency on the tube as well as the RC values is that the hysteresis of the circuit will vary from tube to tube. The PRF ranges are controlled by switching in different values of C and continuous control is provided by making R variable from the front panel.
* The manual incorrectly states this as off. Here is the repetition generator circuit:
V101 serves as a current source when the circuit is in generate mode (switch position O), allowing the Schmitt trigger to freely swing around its hysteresis point. In position A, it acts as a driver for the Schmitt trigger to allow the circuit to be driven from an external source.
The positive trigger pulse from the repetition generator is applied to the pulse timing circuit and then the whole bloody thing goes belly up, or at least it did until I spotted the wiring blunder.
Q101 is normally on and with it, V103A and diode V103B, holding the voltage at the junction of R118 & R122 at a level determined by the setting of R125, the Pulse Duration control. Since Q101 is on, V105 is also on producing a current in the positive pulse output load resistor, R130.
A positive pulse trigger pulse from V102 (the pulse repetition generator Schmitt circuit) turns Q101 off and with it, V105 and V103A. Q102 turns on, and with it V106, producing a current in the negative pulse output load, R133. C2 begins to charge (ramp up) via R118 and the grid voltage of V104 rises until this tube conducts and the Schmitt circuit V104A and B changes state, sending a positive triggering pulse to Q102, turning Q102 off and Q101 on, re-establishing the initial circuit conditions. The circuit is now ready for the next cycle initiation pulse from the Pulse Repetition Generator. The pulse duration is therefore controlled by the time constant R118, C2 and the initial potential on C2 established by the Pulse Duration control.
The pulse output circuit consists of pentode V105 and V106 that are switched as described above by the transistor bi-stable, Q101 and Q 102. In the initial state, Q101 is off and with it, V105 causing the output pulse to rise to ground, going low when the circuit changes state. V106 operates in the opposite direction. The screen voltages set the zero bias currents at 40 to 45mA and since the output loads are 1k, the pulse amplitudes are 40 to 45 volts in maximum amplitude.