VectorFridgeMagnetPowerSupplyNonlinearity

Comparison of the linearity of different superconducting magnet power supplies, connected to the 9T magnet on our 9/5/1T vector magnet.

Magnetic field was measured with a Hall probe from wafer W0191 using an SR830 lockin amplifier and SRS560 voltage preamplifier, 100nA at 77Hz. Fridge is at ~460mK.

Comparison of Cryogenic Ltd SMS-120C PSU and Oxford Instruments IPS-120

In this experiment we sweep B from -15mT to +15mT at 2mT/min (except for the SMS-120C in USB mode, whose slowest sweep rate is 6.4mT/min). All switch heaters are on.

Note: When switching the power supply from Cryogenic SMS-120C to Oxford Instruments IPS-120, we forgot to ground the sample, which caused a change in the carrier density and thus the Hall resistance.

Cryogenic Ltd SMS-120C PSU in DAQ mode

Here we monitor the voltage across a shunt resistor (actually a transducer) in the PSU to get a less noisy idea of what the current supplied to the magnet is.

In this experiment we sweep B from -15mT to +15mT at 2mT/min. All switch heaters are on.

• Note the steps in shunt resisitor (actually the transducer) voltage and Rxy (not in B output reported by the software), increasing with field.This is due to the limited resolution of the DAC which sets the requested magnetic field.
• Note the nonlinear shunt resistor (actually the transducer) voltage and Rxy around zero-field:

Zoomed-in:

To see how linear the power supply is we can compare the Bz output from the software with the field converted from shunt resistor (actually the transducer) voltage:

Now we plot the measured Hall voltage against the field determined from the shunt resistor (red), and also against the field reported by the Cryogenics software (blue)

Bottom line - the DAQ mode isn't great for low field measurements due to the limited DAC resolution and the non-linearity around B=0.

Cryogenic SMS-120C power supply in USB mode

This power supply has the 20-bit extended resolution option.

Note that the Cryogenic power supply only has 64 sweep rates in USB mode, and the lowest is 6.4mT/minute.

In this mode the sweep through B=0 is extremely non-linear:

Comparison of the Bz output with the field converted from the shunt resistor (actually the transducer) voltage:

Plotting the measured Hall voltage against the field determined from the shunt resistor (red), and also against the field reported by the Cryogenics software (blue)

Plotting the measured Hall voltage against the field reported by the Cryogenics software before (red) and after (blue) changing the variable in "PSU Z Get Outputs.vi" to double (no improvement on the reported field resolution, which is around 0.3mT):

Oxford Instruments IPS-120

• Used IPS 120 to power the magnet, Cryogenic power supplies for heaters.
• 9T, 1T heater at 2.7V, 5T heater at 3.2V. All switches are above 10K.
• Smooth sweep through B=0.
• The sweep actually goes to and from the requested field.

Access Control

##* Set ALLOWTOPICVIEW = QedGroup

Tags

create new tag

Comments

(actually the transducer)