Prototype Cryo Relay Driver - Delivery Notes ---------------------------------------------- Original Rev. 21-Nov-2019 Current Rev. 2-Dec-2019 - Safety: 120 V AC, 40 V DC, Fire, Fuses, Splinters, Ran the driver for many hours before delivery. - Better to measure and design but in this case its OK to search a 2 dimension space to minimize the 1/2 C V sqrd because we have a starting points at room temp and at He temp. - Could have measured: Inductance, DC Resistance, Coupling Constant between the two windings, and Winding Phases. I expect the Q to be very low because of the high DC resistance and because of the core losses from Eddy current and Hysteresis. - Is the Q low enough so that there are no LC cycles ? - Datasheet Specification for Room Temperature Operation: 28 V DC, 225 Ohm DC Resistance, Operating Time Max 15 msec. --> 125 mA full current after a few L/R time constants --> 67 uFd with 225 Ohms for 15 msec RC time constant - We expect the DC Resistance to fall by about a factor of 10 when at He temperature. Thus we need a bigger C to keep a 15 msec RC time constant (to match the expected operating time). - The power supply can provide voltages under 28 V to match the lower DC resistance at He temperature. The power supply can provide voltage above 28 V to charge the relay coil inductance faster. - Scope BNC connectors show the Voltage waveform across the driven and floating coils - not the current through these coils. - Need to minimize 1/2 C V sqrd and the fraction of this energy that is dumped into the He temperature: Driving hard at start of a short pulse may be useful but moving the relay armature very quickly will require extra energy. Relay armature may bounce if it hits the travel stops very hard. If the armature is a permanent magnet then must not allow LC oscillation cycles. - Can use a different snubber circuit, e.g. could use back to back zener diodes about 30 V. This would prevent a forward voltage coupled from coil to coil from dumping heat at the He temp and prevent the coupling from slowing the rise of the magnetic field from the driven coil. When the field collapses the induced spike in the coil is clamped at 30 V and puts much of the heat at room temperature. - If the long tail of the RC discharge only makes heat then could cut off the tail of the exponential with a series zener. - I would not fire into a short circuit load. The prototype driver has good aluminum electrolytic capacitors vs junk ones: +-20%, 105 deg C, 5000 hrs, low ESR, low leakage. - I can probably find contacts for the relay pins so that you do not need to solder to them (at least for the room temperature tests). - If needed to it is easy to modify this prototype driver (do the work yourself if you want to) e.g. different capacitors or a different voltage range. - The following table shows the color code of the Relay Driver's output cable and the signal names and pin numbers of the relay's coil connector. See pages 105 and 107 of the Radial catalog for information about the coil connector on the R577433000 relay. Driver Relay Relay Output Cable Pin Name Pin Number ------------ -------- ---------- Red +1 1 Black -C 3 White +2 2 Green -C 3 Shield Cryostat Ground as a first Drain Wire small step for noise control Yes - both the Black and Green cable wires should be connected to the common pin, "-C", pin #3, on the relay. - Decode the relay's part number: R577433000 R577 --> DPDT 4 --> SMA Connectors 18 or 20 GHz 3 --> Latching 3 --> 28 Volt Coils 0 --> No TTL Driver 0 --> No Options 0 --> Solder Pins and Bracket - The URL for this document is: https://web.pa.msu.edu/people/edmunds/Cryo_RF_Relay_Driver/ - Start by verifying that everything works OK into 20 Ohm and 220 Ohm resistive loads. then operate the microwave relay at room temperature in a condition that should work, e.g. 200 uFd and 28 Volts. How different are the waveforms compared to a resistive load ? What is the minimum 1/2 C V sqrd at room temperature ? Then move to testing at He temperature.