PH/ESE PSU test bech

Detailed description of the tests

The paragraphs below describe the sequence of actions in the various test procedures.

Wiener crate power supplies

Limit test

NOTE: The limit test is part of the initialisation routine for the PSU and not a separate test. Errors during the limit test will be shown on the screen of the test bench but will not be written to the test report.

  • Send a CAN message (ID =127) and wait for replies. If there is no reply terminate the test (PSU does not communicate with the PC)
  • Now check the parameters that are stored in the non volatile RAM of the PSU. There are checks for the minimum, maximum and exponent values
    • The parameters that will be check are (DB names):
      • voltage_nom
      • voltage_(max/min/exp)
      • current_limit_(max/min/exp)
      • underv_compare_(max/min/exp)
      • underv_compare_nom
      • overv_compare_(max/min/exp)
      • overv_compare_nom
      • overc_compare_(max/min/exp)
      • overv_protection_nom
      • overv_protection_(max/min/exp)
      • temp_warning_(max/min/exp)
      • temp_warning_nom
      • temp_limit_(max/min/exp)
      • temp_limit_nom
      • outv_adjust_(max/min/exp)
    • The general rule is: If the value read from NV-RAM of the PSU does not match the value stored in the DB the test will display an error and exit
    • Exceptions :
      • "voltage_max" and "overv_protection_max": The value in the PSU has to be within a range of 90% to 140 % of the value in the DB. (For all these exceptions togheter, display only one warning message to the operator with the list of values that generated the warning. Once operator has clicked ok continue the test.)
      • "underv_compare_nom", "overv_compare_nom", "temp_warning_nom" and "temp_limit_nom": The value in the PSU has to be within a range of 90% to 110% of the value in the DB. (For all these exceptions, display one warning message to the operator with the values that generated the warning (if possible include DB and PSU values). Once operator has clicked ok continue the test.)
  • Read the nominal values for the parameters mentioned above from the DB and write them into the PSU
  • Read the parameters back from the PSU and store them in the report file
  • Read the serial number of the PSU and compare it to the serial number that is stored in the DB for the given LCH-ID
  • Read and store: Operating time, AN S/W version. PSU S/W version

Sensor test

  • Get that status of the PSU (via a CAN message), mask the bits with 0x4c17 and check if the result is different fom 0x16
    • If this is that case the PSU must be in a bad shape and the test will be aborted immediately
  • Read the nominal current of the channel from the XML file (this is the only parameter that we don't store in the DB for historical reasons)
  • Safety check: Exit the test if the nominal current of the channel is higher than the current limit of the load
  • Set the slew of the load to 100 A/s
  • Set the load to 0A
  • Switch the PSU on
  • Wait 5 seconds
  • NOTE: The next thing in the S/W would be the temperature test but it is currently disabled as we are not sure how to handle it is a PSU has not got temperature sensors
  • Check the Voltage sensor
    • Read voltages from DVM, PSU and load
    • Read the value of "voltage_max" from the DB
    • The voltage measured by the PSU must not differ more than 0.6% (was 0.25%) of "voltage_max" from the nominal voltage (compare Vpsu <-> Vnominal)
    • The DVM and PSU voltages must not differ more than 0.6% (was 0.25%) of "voltage_max" (compare Vpsu <-> Vdvm)
  • Check if the load voltage is OK
    • The difference between Vload and Vdvm must not be larger than 1%
    • If not ok: Error message on the screen: " Important load and DVM voltage discrepancy. Possible cabling problem on the test bench.";
  • Set the load to the nominal current
  • Wait 4 seconds
  • Read voltages from: DVM, PSU and load
  • Read currents from: PSU and load
  • Check if the load current is OK
    • The difference between Iload (measured) and set current must not be larger than 4%
    • If not ok: Error message on the screen: "Load current largely out of range. Possible problem on the load "LOAD NUMER".";
  • Check if the voltage sensor of the PSU is OK
    • The difference between Vpsu and Vdvm must not be larger than 0.6% (was 0.25%) of "voltage_max"
  • Check if the current sensor of the PSU is OK
    • The difference between Iload and Ipsu must not be larger than 3.1% (was 1.1%) of max module current
  • Values that will be written (in bold if higher than tolerance) to the report file:
    • abs(Vpsu -Vdvm) at 0 A
    • abs(Vpsu -Vdvm) at nominal current
    • abs(Ipsu -Iload at nominal current
  • Set the load back to 0A
  • Turn the PSU off

Soak test

  • Compute the storage interval "SI". During the soak test the S/W will store values of PSU parameters in regular intervals. The "SI" is the duration of the nterval. "SI" is the soak time divided by 10. If the soak time is less than 300 seconds "SI" will be set to 30 seconds and less than 10 data sets will be stored.
  • Switch the PSU on and read the status bits "ST"
    • Abort the test if ("ST" & 0x4c17) = 0x16
  • For each channel:
    • Set the slew of the required load channel(s) to 100 A/s
    • Set the required load channel(s) to the nominal current of the respective PSU channel
    • Wait for 1 second
  • Read the status bits "ST" once more
    • Abort the test if ("ST" & 0x4c17) = 0x16
  • Enter into the soaking loop:
    • Read the status bits "ST"
      • Abort the test if ("ST" & 0x4c17) = 0x16
    • Each time a "SI" has ellapsed store these parameters for all channels:
      • Value of PSU internal reading "Vpsu"
      • Value of PSU internal reading "Ipsu"
      • Voltage measured by DVM
      • Current measured by load
    • Every minute, check the same 4 parameters for all channels and record the min and max values (if higher or lower than the min/max stored before).
    • In the report, print the min and max values of those 4 parameters for all channels.
  • Set all loads back to 0 A
  • Turn off the PSU
  • NOTE: The test is considered a success if none of the PSU channels report an error during the soaking time.

Mains test

  • Set the power box to 230 V
  • Set the slew of all required load channels to 100 A/s
  • Set all required load channels to the nominal current of the respective PSU channel
  • Turn the PSU on
  • For 60 seconds:
    • Check if the PSU has tripped
    • Once per second measure: Iload, Ipsu, Vdvm, Vpsu
  • Compute mean values and standard deviations (with the measurements from the previous loop ) for: Vdvm, Vpsu, Iload, Ipsu
  • Write these parameters to the report file:
    • Mean current from Load
    • Mean current from Power Supply
    • Standard Deviation current from Load
    • Standard Deviation current from Power Supply
    • Raw values current from Load
    • Raw values current from Power Supply
    • Mean voltage from DVM
    • Mean voltage from Power Supply
    • Standard Deviation voltage from DVM
    • Standard Deviation voltage from Power Supply
    • Raw values voltage from DVM
    • Raw values voltage from Power Supply
  • If the PSU has tripped switch it on again and continue the test
  • Set the input power to 207 V (230 -10%)
  • Wait for 2 seconds
  • If the PSU has tripped switch it on once more and continue
  • Set the input power to 253 V (230 +10%)
  • Wait for 2 seconds
  • If the PSU has tripped continue
  • Set the power back to 230 V
  • Switch the PSU on (just in case)
  • If the PSU does not trip immediately:
    • Wait for 5 seconds
    • Measure the power of the individual load channels and sum it up
    • Compute efficiency of the PSU (Pin / Pout)
    • Store these parameters in the report file
      • Input Power
      • Output Power
      • Overall Efficiency
  • Set the current of all loads to 0 A
  • Loop over all channels. For each channel:
    • Set the load to the nominal current
    • Wait 2 seconds
    • Read the power from the power box
    • Read the power from the load
    • Compute the efficiency
    • Store these values in the report file:
      • Input Power (as measured by the power box)
      • Output Power (as measured by the load channel)
      • Channel Efficiency
    • Set the load back to 0 A
    • The test result will be "failed" if the PSU tripped once or several times during the test

Current limit test

  • Set the load to the nominal output current of the PSU channel
  • Swich the PSU on
  • Wait for 0.5 seconds
  • Enter into a loop
    • Set the load to the current "C" (start with the nominal current)
    • Wait for 0.5 s
    • Terminate the loop f the PSU trips
    • Increment "C" by 1% of the nominal current
    • Terminate the loop if 200% of the nominal current have been reached and the PSU still did not trip
  • Set the load to 0 A
  • Compare "C" to the current limit of the PSU channel. The test fails if "C" differs by more than 3.1% (was 2%) with respect to the current limit
  • Store these values in the report file
    • Reference Current (Nominal current of the channel)
    • Trip Current
    • Current Limit of the PSU channel
    • Difference of "C" to current limiy in %

Overvoltage test

  • Tell the S/W to ignore a PSU crowbar
  • Get the value of the parameter "OvervoltageProtection" of the PSU channel
  • Get the value of the parameter "OutputVoltage" of the PSU channel
  • Turn the load channel off
  • Turn the PSU on and disable the "PSU off on trip" feature
  • Run the commands below in a loop:
    • Read the volatge with the DVM
    • Set the "Delta" to the starting voltage = "OvervoltageProtection" * 0.95
    • Close the concerned OVPS relay (relays 500 to 513 depending on the load channel connected) after the OVPS was set to the right value and not before.
    • Wait for 0.3 s
    • Read the status byte array of the PSU (via CAN bus)
    • Terminate the loop if:
      • Byte 7 is not 0
      • The PSU signals a trip
      • The DVM reads an unrealisticly low voltage
    • Increment the voltage of the "delta" by 0.02V
  • Turn the PSU off
  • Open the concerned OVPS relay (relays 500 to 513 depending on the load channel connected) before the OVPS is set back to 0.
  • Set the "delta" back to 0V, 0A
  • Re-enable the crowbar feature of the PSU
  • Values that will be stored in the report file:
    • Start Voltage
    • Trip Voltage
    • Protection Overvoltage (= value of "OvervoltageProtection")
  • The test fails if the voltage at which the PSU trips is more than 2% greater or smaller than the value of "OvervoltageProtection"

Overvoltage trip test

  • Get the value of the parameter "OvervoltageCompare" of the PSU channel
  • Get the value of the parameter "OutputVoltage" of the PSU channel
  • Turn the load channel off
  • Turn the PSU on
  • Set the "delta" to: ("OutputVoltage" - 0.2) V and 0.5A
  • Close the concerned OVPS relay (relays 500 to 513 depending on the load channel connected) after the OVPS was set to the right value and not before.
  • Wait for 1 second
  • Turn the PSU on if it has tripped
  • Run the commands below in a loop:
    • Read the voltage with the DVM
    • Set the "Delta" to the new voltage (for the first run throught the loop: ("OutputVoltage" - 0.2) V)
    • Wait for 1 s
    • Read the status byte array of the PSU (via CAN bus)
    • Terminate the loop if the PSU signals a trip
    • Increment the voltage of the "delta" by 0.02V
  • Turn the PSU off
  • Open the concerned OVPS relay (relays 500 to 513 depending on the load channel connected) before the OVPS is set back to 0.
  • Set the "delta" back to 0V, 0A
  • Values that will be stored in the report file:
    • Start Voltage
    • Trip Voltage
    • Trip Voltage Internal Setting (= value of "OvervoltageCompare")
  • The test fails if the voltage at which the PSU trips is more than 2% greater or smaller than the value of "OvervoltageCompare"

Static regulation Test

  • The tests mentioned below will be executed 3 times with the input voltage set to:
    • 230 V
    • 230 V + 10 %
    • 230 V - 10 %
  • Read the nominal voltage "NV" of the PSU channel from the parameter "OutputVoltage"
  • Read the nominal current "NC" of the PSU channel from the XML file
  • Set the slew of the load to 5000 A/s
  • Set the load to 0 A
  • Switch the PSU on
  • If the PSU has tripped switch it on once more
  • Measure the load current and store it in the report file
  • Test fails if Vdvm differs from "NV" by more than 1%
  • Store in the report file (at 0 A requested current)
    • Current measured by load
    • Current measured by the PSU
    • Voltage measured by the PSU
  • Set the load current to "NC" and wait for it to stabilise (wait for 4 seconds)
  • If the RSU has tripped switch it on again
  • Test fails if Vdvm differs from "NV" by more than 1%
  • Store in the report file (at max. requested current)
    • Current measured by load
    • Current measured by the PSU
    • Voltage measured by the PSU
  • Compare the voltage at 0 A to the voltage at full load (both measured with the DVM). The test fails if they differ by more then 0.5%
  • Set the load to 0 A
  • Turn the PSU off

Ripple and Common mode

Ripple test is done for every channel individually.

  • Set the oscilloscope as follow:
    • Time base: 5us
    • Voltage range: 10mV
    • Channel 1 coupling : DC50ohm
    • BWL set to : 30MHz
    • Trigger to : Auto
    • Measurement/cursors set to : measurement, voltage
    • Statistics set to : ON
  • Read the nominal current of the channel
  • Switch the PSU on and get the PSU status
    • If not in good shape abort the test immediately
  • Set the load to 0A
  • Wait 1 second
  • Reset the oscilloscope's number of sweeps
  • Wait for the oscilloscope to have taken 50 sweeps minimum
  • Read the oscilloscope average peak to peak voltage measurement
  • The test is OK if the peak to peak value is below :
    • 50 mV for all channels except the 48V channel.
    • 100 mV for the 48V channel.
  • Write the value (in bold in case it is higher than limit) in the report as "ripple @ 0A :"
  • Set the load to 50% of nominal channel current
  • Wait 1 second
  • Reset the oscilloscope's number of sweeps
  • Wait for the oscilloscope to have taken 50 sweeps minimum
  • Read the oscilloscope average peak to peak voltage measurement
  • Check the peak to peak value: (see above)
  • Write the value (in bold in case it is higher than limit) in the report as "ripple @ xxA :"
  • Set the load to nominal channel current.
  • Wait 1 second
  • Reset the oscilloscope's number of sweeps
  • Wait for the oscilloscope to have taken 50 sweeps minimum
  • Read the oscilloscope average peak to peak voltage measurement
  • Check the peak to peak value: (see above)
  • Write the value (in bold in case it is higher than limit) in the report as "ripple @ xxA :"
  • Set the load back to 0A
  • Repeat for all channels
  • If any value had to be written in bold: Ripple test has failed
  • Turn the PSU OFF

Common mode measurement:
Common mode measurement is made for the entire PSU with all channels together. There is no pass/fail criteria for this CM test. The results are simply stored in the report.

  • Set the oscilloscope as follow:
    • Time base: 5us
    • Voltage range: 5mV
    • Channel 2 coupling : AC
    • BWL set to : 30MHz
    • Trigger to : Auto
    • Measurement/cursors set to : measurement, voltage
    • Statistics set to : ON
  • Read the nominal current of all channels
  • Set the load to 0A
  • Switch the PSU on and get the PSU status
    • If not in good shape abort the test immediately
  • Wait 1 second
  • Reset the oscilloscope's number of sweeps
  • Wait for the oscilloscope to have taken 50 sweeps minimum
  • Read the oscilloscope average peak to peak voltage measurement
  • Store the result.
  • Write the value in the report as “CM noise @ 0A : “
  • Set all loads to 50% of nominal channel currents
  • Wait 1 second
  • Reset the oscilloscope's number of sweeps
  • Wait for the oscilloscope to have taken 50 sweeps minimum
  • Read the oscilloscope average peak to peak voltage measurement
  • Store the peak to peak value.
  • Write the value in the report as “CM noise @ xx% of nominal current : “
  • Set all loads to nominal channel currents.
  • Wait 1 second
  • Reset the oscilloscope's number of sweeps
  • Wait for the oscilloscope to have taken 50 sweeps minimum
  • Read the oscilloscope average peak to peak voltage measurement
  • Store the result
  • Write the value in the report as “CM noise @ xx% of nominal current : “
  • Set all loads back to 0A
  • Turn the PSU OFF

CAEN LV power supplies

Sensor test

  • Set the slew of the load to 100 A/s
  • Set the load to 0A
  • Reset the channel of the PSU and check for errors
    • If there are error flags set the PSU must be in a bad shape and the test will be aborted immediately
  • Set the channel to the value of the DB parameter "test_voltage"
  • Determine the current for the test under load
    • Get the power limit of the channel from the DB parameter "power_max". Compute the corresponding current on the basis of "test_voltage"
    • Get the current limit of the channel from the DB parameter "current_max"
    • Select the smaller one of the two currents and multiply it with the arbitrary dreating factor of 0.85
    • Exit with an error if this current is above the limit of the load channel
  • Set the nominal current of the channel to the value of the BD parameter "current_limit_max"
  • Switch the PSU channel on
  • Wait 5 seconds
  • Read the voltages measured by the PSU, the load and the DVM
  • Check the voltage sensor
    • Compute the accuracy (Vpsu-measured relative to Vpsu-requested)
    • Compute the accuracy (Vpsu-measured relative to Vdvm-requested)
  • Check if the load sees the proper voltage
    • Vload and Vdvm mus be within 0.2V
  • Read the currents measured by the PSU and the load
  • Check the current (load still off)
    • Compute the accuracy (Ipsu-measured relative to Iload-measured)
  • Store these values in the report file:
    • abs(Vpsu - Vset)
    • abs(Vpsu - Vdvm)
    • abs(Ipsu - Iload)
  • Set the load to the current that was computed earlier
  • Wait 5 seconds and check if the load flags an error. In case of an error the test will abort on this channel
  • Read voltages from PSU and DVM
  • Read currents from PSU and load
  • Check if the current is OK
    • Compare Iload and Irequested. They must be within 0.5A
  • Check the voltage sensor
    • Compute the accuracy (PSU measured relative to DVM measured)
  • Check the currente sensor
    • Compute the accuracy (PSU measured relative to load measured)
  • Turn the PSU channel off
  • Set the load to 0A
  • Set the default operational parameters in the PSU
  • Reset the PSU

Limits for this test:

Value Description
0.2 Voltage accuracy load <-> DVM: 0.2V
0.003 Voltage accuracy parameter 0.3%. Margin = voltage * this parameter plus fixes offset (one of the two lines below)
0.031 Voltage accuracy parameter (31 mV) for A3006, A3009, A3016
0.051 Voltage accuracy parameter (51 mV) for A3025, A3050, A3100
0.5 Current accuracy load measured<->load set : 0.5A
0.021 Current accuracy parameter 2.1%. Margin = current * this parameter plus fixes offset (one of the two lines below)
0.1 Current accuracy parameter in [A] for A3006, A3009, A3016
1.0 Current accuracy parameter in [A] for A3025, A3050, A3100

Current limit test

  • Reset the PSU channel. If the channel reports en error do not start the test
  • Read the value "voltage_min" from the DB and program the PSU chanel to output this voltage
  • Read the value "current_max" from the DB and store it as "CM"
  • Read the value "currentcurrent_limit_max" from the DB and store it as "LM"
  • Set the output current limit of the PSU channel to "LM"
  • Exit the test if the current limit of the load channel is lower than "CM"
  • Set the slew of the load channel to 100 A/s
  • Set the load channel to "TC" = "LM" - "Current tolerance" (see below) - "CP" ("CP" is a compensation for the inaccuracy of the current sensor in the loads. The value is: 200mA * "NL" with "NL" being the number of loads connected to a PSU channel. As "NL" is difficult to get in the S/W we fix the value of "NL" to 2)
  • Switch the PSU channel on and check if it is OK
  • Wait for 30 seconds and check the status of the PSU channel
  • Enter into a loop. The loop terminates if the test current ("TC") reaches "CM" * 1.1 or when a trip has been detected. Start with "TC" = "CM"
    • Increment "TC" by 0.05 A and set the load to the new current
    • Wait for 0.5 s
    • Terminate the loop if the PSU has tripped or reports an other error
  • Set the load to 0 A
  • Reset the PSU channel and check the status
  • Terminate the test of the channel if no trip was seen
  • Record in the report file at which current (if any) the trip was seen. The current is the current measured by the PSU before the trip plus one current step
  • Pass/Fail criteria for this sub-test: ["TC" at which trip was seen = "LM" +/- "Current tolerance"]. Where "Current tolerance" is the same as in the sensor test current accuracy.
  • Set the trip time out of the PSU channel to 10 s
  • Switch the PSU channel on and chech the status
  • Set the load to "LM" * 1.1
  • Enter into a loop
    • Terminate the loop if no trip was seen after more than 20 s
    • If a trip was seen:
      • Compute the differentce between the actual time and the expected 10 s in %
      • The test fails if the deviation is larger than 20 %
      • Store the time after which the trip was seen in the report file
  • Set the load back to 0 A
  • Reset the PSU channel
  • Set thew PSU channel to the default parameters and turn it off

Soak test

  • Compute the storage interval "SI". During the soak test the S/W will store values of PSU parameters in regular intervals. The "SI" is the duration of the nterval. "SI" is the soak time divided by 10. If the soak time is less than 300 seconds "SI" will be set to 30 seconds and less than 10 data sets will be stored.
  • Reset the PSU channel
  • Retrieve the Power limit of the PSU from the DB and devide by the number of channels. This is "PS"
  • Configuration of the channels:
    • Read the power limit of the channel ("PC") and select the smaller one of the two power values ("PS" and "PC") as "PT"
    • The test current "TC" is the value of "I0Set" (OPC) multiplied by the arbitrary derating factor 0.85
    • The test voltage for auto-calibration "TV" is the caluculated like this: "TV" = "PT" / (2 * "TC").
    • Set the PSU to output "TV"
    • Set the load to a slew of 100 A/s
    • Set the load to the test current "TC"
    • Wait for 1 second
    • Check the channel. If there is a trip terminate the test
    • Wait for another 10 seconds to let the PSU stabilize
    • Read the values od the OPC parameters "VMon" and "VCon"
    • The soaking voltage "SV" is the caluculated like this: "SV" = ("PT" / "I0set") - ("VCon" - "VMon)
    • Set the PSU channel to "SV"
  • Once all channels are calibrated and switched on enter into the soaking loop:
    • Each time a "SI" has ellapsed store these parameters for all channels:
      • Value of OPC parameter "VMon"
      • Value of OPC parameter "VCon"
      • Value of OPC parameter "IMon"
      • Voltage measured by DVM
      • Current measured by load
    • Every minute, check the same 5 parameters for each channel and record the min/max values (if higher or lower than the min/max stored previously)
    • In the report, print the min and max values of those 5 parameters for all channels.
    • Terminate the test is a channel reports an error
  • Turn all PSU channels off
  • Set all loads to 0 A
  • Set all PSU channels to the default parameters
  • Reset all PSU channels
  • NOTE: The test is considered a success if none of the PSU channels report an error during the soaking time.

Overvoltage trip test

The hard wired constants of this test are:

Name Value Description
cut_off 1.25 Never go beyond 125% of any nominal value +0.5V (this is in line with the tech.-spec.)
margin 0.25 Threshold for the OVV flag. This threshold of 250mV applies to all 6 types of CAEN PSUs
limit_1 0.003 Accuracy of the voltage sensor (0.3% of actual value), See CAEN documentation
limit_2a 0.05 Accuracy of the voltage sensor for 3025, 3050 and 3100, See CAEN documentation
limit_2b 0.25 Accuracy of the voltage sensor for 3006, 3009 and 3016, See CAEN documentation
hvmax_margin 0.15 Margin for HVMAX voltage in V
hvmax_offset 0.4 Offset for HVMAX voltage in V
start_offset 0.2 Set the PSU to output max voltage minus this constant
delta_offset 0.05 Start the scan at max output voltage minus this constant and plus "margin"
delta_step 0.01 Voltage increment for the Delta PSU
delta_curr1 0.5 Max. current for the Delta PSU for all but 3100
delta_curr2 1.5 Max. current for the Delta PSU for 3100

Description of the test:

  • Reset the channel. Do not start the test if the channel returns an error flag
  • Read the DB parameter "voltage_max". Store value as "MV"
  • Set the output voltage of the channel to "MV" - "margin"
  • Set the output current of the PSU channel to the value of the DB parameter "current_max"
  • Set the trip time out of the PSU to infinite
  • Turn the load channel off
  • Switch the "delta" on and set it to: ("MV" - start_offset + delta_offset) V and "delta_curr(1/2)" A
  • Close the concerned OVPS relay (relays 500 to 513 depending on the load channel connected) after the OVPS is set to the defined values and not before.
  • Switch the PSU node on
  • Enter into a loop:
    • Set the "delta" to the new voltage (start with: "MV" - start_offset + delta_offset) and delta_curr(1/2)" A
    • Wait for 3 seconds. Terminate the test if the PSU returns an error flag
    • If the PSU reports the OVV condition read the voltage from the DVM
      • In order to accelerate the test once the OVV is seen we jump to 8.2V and continue until we see HVMAX
    • If the PSU reports the HVMAX condition read the voltage from the DVM
    • If the PSU reports the OVP condition read the Vcon voltage from the PSU channel.
    • Exit from the loop if the PSU has switched itself off.
    • Exit from the loop if both the OVV and (HVMAX or OVP) flag have been detected
    • Increment the voltage of the "delta" by "delta_step"
    • Terminat the loop if the voltage of the delta is larger than "MV" * cutt_off
  • Open the concerned OVPS relay (relays 500 to 513 depending on the load channel connected) before the OVPS is set back to 0.
  • Set the "delta" to 0 V / 0 A
  • Reset the PSU channel
  • Program the default parameters into the PSU channel
  • Turn the PSU channel off
  • Check if the OVV was seen at the right voltage
    • Compute the OVV range: "range" = ("MV" * "limit_1") + "limit_2"
      • Note: "limit_2" is "limit_2a" or "limit_2b" depending on the PSU type * The test fails if the difference between the voltage at which the OVV flag was seen and ("MV" - "start_offset" + "margin") is larger than "range"
  • Check if the HVMAX or OVP were seen at the right voltage *IF HVMAX WAS SEEN: The test fails if the difference between the voltage at which the HVMAX flag was seen and ("MV" + hvmax_offset) is larger than "hvmax_margin" *OR
    • IF OVP WAS SEEN: The test fails if the difference between the voltage (Vcon) at which the OVP flag was seen and ("MV" + 0.5V) is larger than "range" (same tolerance as for OVV).
  • In the report file store:
    • The voltage at which the OVV flag was seen
    • Which of OVP or HVMAX was seen.
    • The voltage at which the HVMAX or OVP flag was seen

Ripple test

  • Close the relay no: 418 (this is only used for ripple test of Caen PSU)
  • Set the oscilloscope as follow:
    • Time base: 5us
    • Voltage range: 5mV (might be 10)
    • Channel 1 coupling: DC50ohm
    • BWL set to: 30MHz
    • Trigger to: Auto
    • Measurement/cursors set to : measurement, voltage
    • Statistics set to: ON
  • Switch the PSU channel on
    • If not in good shape abort the test immediately
  • Read the nominal current ("NC") of the channel
  • Set the output voltage to the 1st test voltage (4 V).
  • Repeat the following steps for a load current of 0 A, 0.5 * "NC" and 0.85 * "NC"
    • Set the load to the new current
    • Wait 1 second
    • Check the status of the PSU channel
      • If not in good shape abort the test immediately
    • Reset the oscilloscope's number of sweeps.
    • Wait for the oscilloscope to have taken 50 sweeps minimum.
    • Read the oscilloscope average peak to peak voltage measurement.
    • Compare the peak to peak value with :
      • Caen A3006, 3009, 3016 : 25 mV pk-pk
      • Caen A3025, 3050, 3100 : 15 mV pk-pk
    • Write the value (in bold in case it is higher than limit) in the report as “ripple @ 0A : “
  • After the last loop the PSU runs at 4 V and 0.85% of "NC". At this moment:
    • Measure the voltage measured by the PSU = "VMON"
    • Measure the voltage measured by the PSU at the connector = "VCON"
    • Calculate the second test voltage "VT2" = (nominal power / (nominal current * 0.9)) - ("VCON" - "VMON").
    • Set the PSU channel to "VT2"
  • Repeat the following steps for a load current of 0 A, 0.5 * "NC" and 0.85 * "NC"
    • Set the load to the new current
    • Wait 1 second
    • Check the status of the PSU channel
      • If not in good shape abort the test immediately
    • Reset the oscilloscope's number of sweeps.
    • Wait for the oscilloscope to have taken 50 sweeps minimum.
    • Read the oscilloscope average peak to peak voltage measurement.
    • Compare the peak to peak value with :
      • Caen A3006, 3009, 3016 : 25 mV pk-pk
      • Caen A3025, 3050, 3100 : 15 mV pk-pk
    • Write the value (in bold in case it is higher than limit) in the report as “ripple @ 0A : “
  • Set the load back to 0A
  • Open the relay no: 418
  • Turn the channel OFF
  • If any value had to be written in bold: Ripple test has failed

Common mode measurement:
Common mode measurement is made for the entire PSU with all channels together. There is no pass/fail criteria for this CM test. The results are simply stored in the report.

  • Set the oscilloscope as follow:
    • Time base: 5us
    • Voltage range: 5mV
    • Channel 2 coupling : AC
    • BWL set to : 30MHz
    • Trigger to : Auto
    • Measurement/cursors set to : measurement, voltage
    • Statistics set to : ON
  • Switch all PSU channels ON
    • If not in good shape abort the test immediately
  • Read the nominal current ("NC") of the channel
  • Set the output voltage of all channels to the 1st test voltage (4 V).
  • Repeat the following steps for a load current of 0 A, 0.5 * "NC" and 0.85 * "NC" for all channels
    • Set all loads to the new currents
    • Wait 1 second
    • Check the status of the PSU
      • If not in good shape abort the test immediately
    • Reset the oscilloscope's number of sweeps.
    • Wait for the oscilloscope to have taken 50 sweeps minimum.
    • Read the oscilloscope average peak to peak voltage measurement.
    • Store this result.
    • Write the value in the report as “CM noise @ xx % of nominal current and xxV : “
  • After the last loop the PSU runs at 4 V and 0.85% of "NC". At this moment:
    • Measure the voltage measured by the PSU = "VMON"
    • Measure the voltage measured by the PSU at the connector = "VCON"
    • Calculate the second test voltage "VT2" = (nominal power / (nominal current * 0.9)) - ("VCON" - "VMON").
    • Set all PSU channels to "VT2"
  • Repeat the following steps for a load current of 0 A, 0.5 * "NC" and 0.85 * "NC" for all channels
    • Set all loads to the new currents
    • Wait 1 second
    • Check the status of the PSU
      • If not in good shape abort the test immediately
    • Reset the oscilloscope's number of sweeps.
    • Wait for the oscilloscope to have taken 50 sweeps minimum.
    • Read the oscilloscope average peak to peak voltage measurement ad store this result.
    • Write the value in the report as “CM noise @ xx% of nominal current and xxV : “
  • Set all loads back to 0A
  • Turn the PSU OFF

Static Regulation Test

  • Reset the PSU channel and exit the test if an error is reported
  • Set the slew of the load to 100 A/s
  • Run the commands below for two voltages ("TV"): "voltage_min" and "test_voltage" (as defined in the DB)
    • Get "current_max" = "MC" for the channel
    • Get "power_max" = "MP" fhr the channel and compute the current limit "MC2" = "MP" / "TV"
    • Select the minimum of "MC" and "MC2" = "TC". Then multiply "TC" with the derating factor 0.85
    • Set the output voltage of the channel to "TV"
    • Set the nominal current of the channel to "current_limit_max"
    • Set the load current to 0 A
    • Turn the PSU channel on and exit the test if an error is reported
    • Wait for 6 seconds
    • Measure the voltage with the DVM = "VD"
    • Test fails if: (("TV" - "VD") / "TV") > 0.01
    • Set the load current to "TC"
    • Wait for 6 seconds
    • Abort the test if the PSU has tripped
    • Measure the voltage with the DVM = "VD2"
    • Test fails if: (("TV" - "VD2") / "TV") > 0.01
    • Measure the current seen by the load ("LC")
    • Measure the current seen by the PSU ("PC")
    • Test fails if: (("PC" - "LC") / "PC") > 0.05
    • Test fails if: (("VD" - "VD2") / "VD") > 0.005
    • Turn the PSU channel off
    • These values will be stored in the database
      • "current_limit_max"
      • "TV"
      • "VD"
      • Voltage measured by PSU at 0 A load
      • Current measured by PSU at 0 A load
      • Current measured by load at 0 A load
      • "VD2"
      • Voltage measured by PSU at full load
      • Current measured by PSU at full load
      • Current measured by load at full load
      • (("VD" - "VD2") / "VD")
  • Program the default values into the PSU channel
  • Set the load back to 0 A

Wiener LV PSUs (distribured and monolithic)

Applies only to the monolithic PSU.
Applies only to the distributed PSU.
Applies to both maraton PSU types.

Parameters

  • Tell the operator to insert the right configuration file in the RCM board. This has to be made via the USB using the Muhse software.
    Alternatively, this will possibly be made via the test bench SW if we get the information on how to perform that.
  • No need to read and store the parameters as those are in the RCM.
  • Read all OPC parameters of the PSU and store them in a temporary file. This file will be used at the end of the test to set the parameters back to their original values.
  • Write those parameters as they were set by the user (or as they were when the PSU was received) in the report.
  • Read the test default parameters from the technical DB and write them via OPC to the RCM / to the power supply. Write these parameters with which the power supply has been tested in the report.
  • If PSU is a limited one (with the L extension in the manufacturer part number):
    • Ask the operator to enter the voltage limitation as written on the PSU for each channel and store these values.
    • Replace the SupervisionMaxSenseVoltage of each channel with this limitation value + 0.2V (margin for all test except for the overvoltage test).
    • If the limitation value + 1.5V is smaller than “voltage max” (as specified in the DB), replace the SupervisionMaxTerninalVoltage of each channel with the limitation value + 1.5V (1.5V is the maximum voltage drop we can allow). In the other case (limitation + 1.5 larger than voltage max), keep the current value (default value).
  • If required (TBD): Check that these parameters have correctly been entered.
  • List of the concerned parameters:
Muhse name DB name OPC name Type Test value if not in DB
Sense voltage voltage_max Voltage Float -
Current Limit current_nom Current Float -
Ramp up - VoltageRiseRate Float 100
Ramp down - VoltageFallRate Float 100
min sense voltage svmin SupervisionMinSenseVoltage Float -
max sense voltage svmax SupervisionMaxSenseVoltage Float -
max terminal voltage svmax SupervisionMaxTerminalVoltage
Not there for the distributed
Float -
max current scmax SupervisionMaxCurrent Float -
max power spmax SupervisionMaxPower Float -
group number - GroupNumber Integer TBD
On failure min sense voltage - SupervisionBehaviourMinSenseVoltage Integer 1
On failure max sense voltage - SupervisionBehaviourMaxSenseVoltage Integer 1
On failure terminal voltage - SupervisionBehaviourTerminalVoltage Integer 1
On failure max current - SupervisionBehaviourMaxCurrent Integer 1
On failure max power - SupervisionBehaviourMaxPower Integer 1
On failure max temperature - SupervisionBehaviourMaxTemperature Integer 1
On failure communication timeout - SupervisionBehaviourCommunicationTimeout Integer 0
  • Apply the appropriate settings (multipliers), according to the power supply LHC number, in the MSB (marathon setup box).
  • Turn all PSU channels ON (this is required to be able to read the correct values using the MSB).
  • Wait 15 seconds.
  • Using the MSB, read the following parameters (once they are stabilized, no variations of more than 0.5 %) for every channel: OVP, Vset and Iset.
    • Vset: set voltage at the sense lines (output voltage is fixed and set by this, no SW setting).
    • I set: maximum output current. This is the HW limitation of the current. When reached, the PS acts as a fixed current source.
    • OVP: Voltage at which the PSU trips (HW trip, no SW flag via OPC). This value replaces “Supervision max terminal voltage”.
  • Make sure:
    • Vset is within svmin and svmax.
    • Iset is within 0 and “supervisionMaxCurrent” + 20%
    • OVP is within svmin and svmax + 20%
  • Turn all channels OFF.
  • Report an error is any of these is out of range.

Sensor test

  • Set the slew of the load to 100 A/s
  • Set the load to 0A
  • Check the PSU for errors.
    • If there are error flags set the PSU must be in a bad shape and the test will be aborted immediately
  • …voltage is fixed, nothing to do…
  • Set the channel to the value of the DB parameter "voltage_max" - 10%.
  • Determine the current for the test under load
    • Get the power limit of the channel from the DB parameter "spmax". Compute the corresponding current on the basis of "Vset - 10%" / "Voltage max - 10%"
    • Get the current limits of the channel; “Iset", “scmax” and from the DB parameter "current_nom".
    • Select the smaller one of the four currents (calculated, “Iset", “current nom” or “SupervisionMaxCurrent”) of the two currents and multiply it with the arbitrary derating factor of 0.9
    • Exit with an error if this current is above the limit of the load channel
  • Switch the PSU channel on
  • Wait 5 seconds
  • Read the voltages measured by the PSU, the load and the DVM
  • Check the voltage sensor
    • Compute the accuracy (Vpsu-measured relative to Vpsu-requested)
    • Compute the accuracy (Vpsu-measured relative to Vdvm-requested)
  • Check if the load sees the proper voltage
    • Vload and Vdvm must be within +/- 0.2V
  • Read the currents measured by the PSU and the load
  • Check the current (load still off)
    • Compute the accuracy (Ipsu-measured relative to Iload-measured)
  • Store these values in the report file:
    • abs(Vpsu - Vset)
    • abs(Vpsu - Vdvm)
    • abs(Ipsu - Iload)
  • Set the load to the current that was computed earlier
  • Wait 5 seconds and check if the load flags an error. In case of an error the test will abort on this channel
  • Read voltages from PSU and DVM
  • Read currents from PSU and load
  • Store these values in the report file:
    • abs(Vpsu - Vdvm)
    • abs(Ipsu - Iload)
  • Check if the current is OK
    • Compare Iload and Irequested. They must be within 0.5A
  • Check the voltage sensor
    • Compute the accuracy (PSU measured relative to DVM measured)
  • Check the current sensor
    • Compute the accuracy (PSU measured relative to load measured)
  • Turn the PSU channel off
  • Set the load to 0A
  • Set the default operational parameters in the PSU

Limits for this test (same as for the VME PSU):

Value Description
0.2 Voltage accuracy load <-> DVM: 0.2V
0.0025 Voltage accuracy parameter 0.25%. Margin = sv max * this parameter
0.0065 Voltage accuracy parameter 0.65%. Margin = sv max * this parameter
0.5 Current accuracy load measured<->load set : 0.5A
0.011 Current accuracy parameter 1.1%. Margin = sc max * this parameter
0.031 Current accuracy parameter 3.1%. Margin = sc max * this parameter

Soak test

  • Compute the storage interval "SI". During the soak test the S/W will store values of PSU parameters in regular intervals. The "SI" is the duration of the interval. "SI" is the soak time divided by 10. If the soak time is less than 300 seconds "SI" will be set to 30 seconds and less than 10 data sets will be stored.
  • Retrieve the Power limit of the PSU from the DB. This is "PS".
  • Retrieve the individual power limit of all channels and sum them up. This is “PC”.
  • Compare PS with PC. If PC is larger than PS, compute the derating factor by doing PS/PC and apply this derating factor to each individual channel power. Those will be PT [0 to 12].
  • Configuration of the channels (to do for every channel):
    • The test current "TC" is the smaller value of these two: “Iset" and “scmax” multiplied by the arbitrary derating factor 0.8.
    • The test current "TC" is the value of "Current limit" (“Current” in OPC) multiplied by the arbitrary derating factor 0.9.
    • The voltage is fixed.
    • The test voltage for auto-calibration "TV" is the calculated like this: "TV" = "PT"(chan) / (2 * "TC"). Compare the TV with the value of SupervisionMaxSenseVoltage. If the latter is the smaller value set TV to SvMaxSenseVoltage. This is only an issue for limited PSUs
    • Set the PSU to output "TV"
    • Check that the theoretical output power (TC * Vset) is smaller than PT [for this channel] * 0.8. Report a warning if this is out of range and calculate the new value of TC as follow: TC = PT[of this channel]*0.8/Vset.
    • Set the load to a slew of 100 A/s
    • Set the load to the test current "TC"
    • Turn the channel ON.
    • Wait for 1 second
    • Check the channel. If there is a trip terminate the test
    • Wait for another 10 seconds to let the PSU stabilize
    • Read the measured values thru the OPC: "Terminal Voltage" and "Sense Voltage"
    • Turn the channel OFF.
    • The soaking voltage "SV" is calculated like this: "SV" = ("PT"(chan) / "Current limit") - ("Terminal Voltage " - "Sense Voltage")
    • Set the PSU channel to "SV"
    • Switch the channel ON.
    • Wait for 1 second
    • Check the channel. If there is a trip terminate the test
  • Enter into the soaking loop:
    • Each time a "SI" has elapsed store these parameters for all channels:
      • Value of OPC parameter "MeasurementSenseVoltage"
      • Value of OPC parameter "MeasurementCurrent"
      • Voltage measured by DVM
      • Current measured by load
    • Every minute, check the same 4 parameters for each channel and record the min/max values (if higher or lower than the min/max stored previously)
    • In the report, print the min and max values of those 4 parameters for all channels.
    • Terminate the test is a channel reports an error
  • Turn all PSU channels off
  • Set all loads to 0 A
  • Set all PSU channels to the default parameters
  • NOTE: The test is considered a success if none of the PSU channels report an error during the soaking time.

Current limit test

First the current limit (Iset) is tested and then the current max (RCM limitation, SupervisionMaxCurrent) is tested. The first setting should simply limit the current going out of the power supply and decrease the voltage (sort of constant current mode). The second limitation triggers an “Imodmax” flag and makes the channel trip.

First the current limit is tested and then the current max is tested. The first setting should simply limit the current going out of the power supply and decrease the voltage (sort of constant current mode). The second limitation triggers an “Imodmax” alarm in the PSU.

First part.

  • If the channel reports en error do not start the test
  • If “Iset” * 1.1 is > “scmax” (as in the DB) then set “SupervisionMaxCurrent” to “scmax” (as in the DB), store the original Iset value and prompt a message to the operator to manually (using the trimers) set “Iset” 10% below “scmax”. Once the Iset is adjusted by the operator, read Iset using the MSB.
    Else: Set the “SupervisionMaxCurrent” to: “Iset” * 1.1. (This is to make sure we will be able to reach “Iset”.)
  • Read the value "voltage_min" from the DB and program the PSU chanel to output this voltage
  • Read the value "current limit" in the PSU (this is called current_nom in the DB)
  • Read the value "max current" in the PSU (this is called scmax in the DB)
  • Compare those two values and make sure “max current” is > than “current limit”. This shall be the case after the parameters initialization.
  • Set the slew of the load channel to 100 A/s
  • Set the load channel to "Iset" / "current limit" - 1.1% / 6.4% of “current nom” - "CP" ("CP" is a compensation for the inaccuracy of the current sensor in the loads. The value is: 200mA * "NL" with "NL" being the number of loads connected to a PSU channel. As "NL" is difficult to get in the S/W we fix the value of "NL" to 2).
  • Switch the PSU channel on and check if it is OK
  • Wait for 5 seconds and check the status of the PSU channel
  • Enter into a loop. The loop terminates if the test current (current at which the load is set to) reaches "Iset" / "current limit" * 1.1 / 6.4% or when the voltage starts to decrease.
    • Increment the load current by 0.05 A and set the load to the new current
    • Wait for 0.5 s
    • Measure the current with the PSU and store this value.
    • Check the output voltage of the channel (PSU voltage readout).
    • Measure the current with the PSU and terminate the loop if the PSU output voltage went down by more than 25% of its original value (Vset value) (voltage set value) or if the PSU reports another type of error. Once the V drop is seen, add one step to the stored current. This is the value for the pass/fail check.
  • Set the load back to 0 A
  • Check the PSU status
  • Terminate the test of the channel if no trip was seen
  • Record in the report file at which current (if any) the voltage decrease was seen. The current is the current measured by the PSU before the trip plus one current step
  • Pass/Fail criteria for this sub-test is: tolerance of +/- 6.4% of scmax. The reference for the trip is: “Iset”.
  • Pass/Fail criteria for this sub-test: same tolerance (1.1% of scmax) on the current as for the sensor test. The reference for the trip is: current limit
  • If, at the beginning of this test, “Iset” * 1.1 was > “scmax” (as in the DB), ask the operator to adjust the Iset trimmer back to its original value.
  • Switch the channel off

Second part.

  • Set the “SupervisionMaxCurrent” of the PSU channel 10% below “Iset” / “current limit”.
  • Set the slew of the load channel to 100 A/s
  • Switch the PSU channel on and check if it is OK
  • Set the load channel to “SupervisionMaxCurrent” - 1.1% of “current nom” - "CP" ("CP" is a compensation for the inaccuracy of the current sensor in the loads. The value is: 200mA * "NL" with "NL" being the number of loads connected to a PSU channel. As "NL" is difficult to get in the S/W we fix the value of "NL" to 2)
  • Wait for 5 seconds and check the status of the PSU channel
  • Enter into a loop. The loop terminates if the test current (current at which the load is set to) reaches “Iset” / “current limit” or in case another error is seen by the PSU.
    • Increment the load current by 0.05 A and set the load to the new current
    • Wait for 0.5 s
    • Measure the current with the PSU and store this value.
    • Check the channel status for any error (the error we are looking for is called “Imodmax” in the muhse SW). OPC: "Status.FailureMaxCurrent". Once seen, add one step to the stored current. This is the value for the pass/fail check.
    • Terminate the loop if the PSU channel has tripped or if the PSU reports any other type of error.
  • Set the load back to 0 A
  • Pass/Fail criteria for this sub-test is: tolerance of 1.1% of the sc max (same as for the sensor test). The reference for the trip is: “SupervisionMaxCurrent”.
  • Set the PSU channel to the default parameters and turn it off

Overvoltage trip test

First part of the test: Max sense voltage test. We are looking for a flag (max sense voltage status) to be set by the RCM.
  • Do not start the test if the channel returns an error flag
  • Set the “SupervisionMaxSenseVoltage” to: 0.9 * OVP. If this new value is larger than “svmax” (as entered in the DB), set “SupervisionMaxSenseVoltage” to this “svmax” value.
  • The set voltage is fixed.
  • Check that the Vset * 1.05 is still < “SupervisionMaxSenseVoltage”. If this is not the case, warn the operator that this part of the test cannot be done (also report that in the written report) and continue on with the next test.
  • For limited PS (“L”), set the “supervisionMaxSenseVoltage” to the limitation entered by the operator - 0.2V.
  • For the non-limited PS, set the “supervisionMaxSenseVoltage” to the svmax value (as in the DB) - 0.2V.
  • Read the DB parameter "svmax". Store value as "MV"
  • Set the output voltage of the channel to "MV" - 0.4 V (for the limited version to the limit - 0.4V)
  • Turn the load channel off
  • Switch the "delta" on and set it to: (“SupervisionMaxSenseVoltage” - 0.2) V and 0.5 A
  • Switch the "delta" on and set it to: ("MV" - 0.4) V and 0.5 A
  • Switch the PSU node on
  • Enter into a loop:
    • Set the "delta" to the new voltage (start with: "SupervisionMaxSenseVoltage" - 0.2 "MV" - 0.4) and 0.5 A
    • Close the concerned OVPS relay (relays 500 to 513 depending on the load channel connected) after the OVPS is set to the defined values and not before.
    • Wait for 3 seconds. Terminate the test if the PSU returns an error flag
    • If the PSU reports the "Status.FailureMaxSenseVoltage" flag, read the sense voltage from the PS (measuredSenseVoltage) and store this value. This event terminates the loop
    • Exit from the loop if the PSU has switched itself off
    • Increment the voltage of the "delta" by 0.01 V
    • Terminate the loop if the voltage of the delta is larger than "!SupervisionMaxSenseVoltage" / "MV" * 1.25
  • Switch PSU channel off
  • Tolerance for this test (pass/fail) is the same as for the sensor test, 0.25% / 0.65% of the voltage max. The reference voltage being the "supervisionMaxSenseVoltage" / "MV"-0.2.
  • Open the concerned OVPS relay (relays 500 to 513 depending on the load channel connected) before the OVPS is set back to 0.
  • Set the "delta" to 0 V / 0 A
  • Set the “supervisionMaxSenseVoltage” back to its nominal value
  • For limited PS (“L”), set the “supervisionMaxSenseVoltage” back to the limitation + 0.2 V (margin).
  • For unlimited PS , set the “supervisionMaxSenseVoltage” back to the nominal value

Second part of the test: OVP (called max terminal voltage with the monolithic PS ) voltage test. We are not looking for a flag but for the PSU channel to switch itself off (trip and/or output voltage dropping).

  • Set the “SupervisionBehaviour.MaxSenseVoltage” to: 0 (this is to ignore the max sense voltage error).
  • The set voltage is fixed.
  • If PSU type is limited: Set the “!SupervisionMaxTerminalVoltage” to the limitation value (as entered by the operator at the beginning of the PSU test).
  • Set the channel output voltage to “svmax” (value defined for this test) - 0.2V.
  • Switch the "delta" on and set it to: ("OVP" / "svmax" - 0.2) V and 0.5 A
  • Close the concerned OVPS relay (relays 500 to 513 depending on the load channel connected) after the OVPS is set to the defined values and not before.
  • Switch the PSU node ON.
  • Enter into a loop:
    • Set the "delta" to the new voltage (start with: "OVP" / "svmax" - 0.2) and 0.5 A
    • Wait for 3 seconds. Terminate the test if the PSU returns an error flag
    • If the PSU voltage reading decreased by more than 25% (this is a consequence of the PSU channel self trip which is HW in this case; no relation with the RCM), store the previous DVM value and add one voltage step to it. This terminates the loop
    • If the PSU reports the "Status.FailureMaxTerminalVoltage" flag, read the terminal voltage from the PSU (MeasuredTerminalVoltage) and store this value (normally the previous value is stored and then one step is added). This terminates the loop.
    • Exit from the loop if the PSU has switched itself off or if it reports another type of error. Report an error message in case this happens. Note: No such test for distributed PSU…
    • Increment the voltage of the "delta" by 0.01 V
    • Terminate the loop if the voltage of the delta is larger than "OVP"/ "svmax" * 1.25 and no trip is seen.
  • Turn the PSU channel OFF
  • Open the concerned OVPS relay (relays 500 to 513 depending on the load channel connected) before the OVPS is set back to 0.
  • Set the "delta" to 0 V / 0 A
  • Tolerance for this test (pass/fail) is 6.5% of the svmax. The reference voltage being the value of “OVP”.
  • Tolerance for this test (pass/fail) is (same tolerance as for the sensor test) 0.25% of svmax. The reference voltage being “svmax” for unlimited PSU / “limitation value” for limited PSUs.
  • Enter all default test parameters back in the PSU.
  • Set the load back to 0A.
  • In the report file store:
    • The voltage at which the StatusMaxSenseVoltage flag was seen
    • The voltage at which the OVP PSU trip was seen
  • In the report file store:
    • The voltage at which the StatusMaxSenseVoltage flag was seen.
    • The voltage at which the StatusMaxTerminalVoltage flag was seen.

Ripple test

  • Relay 418 is not needed.
  • Set the oscilloscope as follow:
    • Time base: 5us
    • Voltage range: 5mV
    • Channel 1 coupling: DC50ohm
    • BWL set to: 30MHz
    • Trigger to: Auto
    • Measurement/cursors set to : measurement, voltage
    • Statistics set to: ON
  • The test current "TC" is the smaller value of these two: “Iset" and “SupervisionMaxCurrent”.
  • Switch the PSU channel on
    • If not in good shape abort the test immediately
  • Voltage is fixed
  • Set the slew of the load to 100A/s.
  • Read the nominal current ("current_nom" in DB) of the channel
  • Set the output voltage to the 1st test voltage (compute the middle point between min and max voltage; this would be 5 V for a 2-8V channel).
  • If the PSU is a limited one (“L”), the first test voltage is the minimum voltage as defined in the DB (voltage min in the DB).
  • Repeat the following steps for a load current of 0 A, 0.5 * “TC” / "current nom" and 0.85 * “TC” / "current nom" and for all channels
    • Set the load to the new current
    • Wait 1 second
    • Check the status of the PSU channel
      • If not in good shape abort the test immediately
    • Reset the oscilloscope's number of sweeps.
    • Wait for the oscilloscope to have taken 50 sweeps minimum.
    • Read the oscilloscope average peak to peak voltage measurement.
    • Compare the peak to peak value with :
      • Channel with nominal voltage (“voltage_max” in DB) <= 8V : 20 mV pk-pk (10mV included for the load ripple and noise)
      • Channel with nominal voltage (“voltage_max” in DB) > 8V : 25 mV pk-pk (10mV included for the load ripple and noise)
    • Write the value (in bold in case it is higher than limit) in the report as “ripple @ 0A : “
  • After the last loop the PSU runs at the 1st test voltage and 0.85% of "Current nom". At this moment:
    • Measure the voltage measured by the PSU = "Sense voltage"
    • Measure the voltage measured by the PSU at the connector = "Terminal Voltage"
    • Calculate the second test voltage "VT2" = (nominal power / (nominal current * 0.9)) - ("Terminal voltage" - "Sense voltage"). If VT2 is larger than the voltage max (as defined in the DB), then VT2 is equal to voltage max.
    • If the PSU is a limited one (“L”), VT2 is the limited voltage as entered by the operator.
    • Do that for all channels and set all PSU channels to "VT2".
  • Set the PSU channel to the calculated voltage (VT2)
  • Repeat the following steps for a load current of 0 A, 0.5 * "Current nom" and 0.85 * "Current nom" and for all channels
    • Set the load to the new current
    • Wait 1 second
    • Check the status of the PSU channel
      • If not in good shape abort the test immediately
    • Reset the oscilloscope's number of sweeps.
    • Wait for the oscilloscope to have taken 50 sweeps minimum.
    • Read the oscilloscope average peak to peak voltage measurement.
    • Compare the peak to peak value with :
      • Channel with nominal voltage < 8V : 20 mV pk-pk
      • Channel with nominal voltage > 8V : 25 mV pk-pk
    • Write the value (in bold in case it is higher than limit) in the report as “ripple @ 0A : “
  • Set the load back to 0A
  • Turn the channel OFF
  • If any value had to be written in bold: Ripple test has failed

Common mode measurement:

Note:To be seen later on: we might implement a pass/fail criteria with a certain tolerance which might be different depending on the presence or not of the CM filters in the PSU…
  • Set the oscilloscope as follow:
    • Time base: 5us
    • Voltage range: 5mV
    • Channel 2 coupling : AC
    • BWL set to : 30MHz
    • Trigger to : Auto
    • Measurement/cursors set to : measurement, voltage
    • Statistics set to : ON
  • The test current "TC" is the smaller value of these two: “Iset" and “SupervisionMaxCurrent”.
  • Switch the PSU channel on
    • If not in good shape abort the test immediately
  • Voltage is fixed
  • Read the nominal current ("Current nom" in the DB) of the channel
  • Set the output voltage to the 1st test voltage (compute the middle point between min and max voltage; this would be 5 V for a 2-8V channel).
  • If the PSU is a limited one (“L”), the first test voltage is the minimum voltage as defined in the DB (voltage min in the DB).
  • Set the slew of the load to 100A/s.
  • Repeat the following steps for a load current of 0 A, 0.5 * “TC” / "Current nom" and 0.85 * “TC” / "Current nom" and for all channels
    • Set all loads to the new currents
    • Wait 1 second
    • Check the status of the PSU
      • If not in good shape abort the test immediately
    • Reset the oscilloscope's number of sweeps.
    • Wait for the oscilloscope to have taken 50 sweeps minimum.
    • Read the oscilloscope average peak to peak voltage measurement.
    • Store this result.
    • Write the value in the report as “CM noise @ xx % of nominal current and xxV : “
  • After the last loop the PSU runs at the 1st test voltage and 0.85% of "Current nom". At this moment:
    • Measure the voltage measured by the PSU = "Sense voltage"
    • Measure the voltage measured by the PSU at the connector = "Terminal Voltage"
    • Calculate the second test voltage "VT2" = (nominal power (as “nominal power”, take the channel max power and, if the sum of all channel power is higher than the PSU max power, multiply by a calculated derating factor) / (nominal current * 0.9)) - ("Terminal voltage" - "Sense voltage"). If VT2 is larger than the voltage max (as defined in the DB), then VT2 is equal to voltage max.
    • If the PSU is a limited one (“L”), VT2 is the limited voltage as entered by the operator.
    • Do that for all channels and set all PSU channels to "VT2"
  • Set the PSU channel to the calculated voltage (VT2)
  • Repeat the following steps for a load current of 0 A, 0.5 * "Current nom" and 0.85 * "Current nom" for all channels
    • Set all loads to the new currents
    • Wait 1 second
    • Check the status of the PSU
      • If not in good shape abort the test immediately
    • Reset the oscilloscope's number of sweeps.
    • Wait for the oscilloscope to have taken 50 sweeps minimum.
    • Read the oscilloscope average peak to peak voltage measurement ad store this result.
    • Write the value in the report as “CM noise @ xx% of nominal current and xxV :“
  • Set all loads back to 0A
  • Turn the PSU OFF

Static Regulation Test

  • Exit the test if an error is reported
  • Set the slew of the load to 100 A/s
  • Run the commands below for two voltages ("TV"): "voltage_min" (as defined in the DB) and "test_voltage" (defined as follow: TV = ("voltage_min" + "voltage_max") / 2. In case of a limited (“L”) power supply, compare TV with the limitation entered by the operator and take the smaller of the two.
  • Run the commands below at Vset
    • Get "Iset" / "current_nom" = "MC" for the channel
    • Get "SupervisionMaxCurrent" = "MC2"
    • Get "spmax" = "MP" for the channel and compute the current limit "MC2" = "MP" / "TV"
    • Select the minimum of "MC" and "MC2" = "TC". Then multiply "TC" with the derating factor 0.9
    • ..fixed voltage..
    • Set the output voltage of the channel to "TV"
    • Set the load current to 0 A
    • Turn the PSU channel on and exit the test if an error is reported
    • Wait for 6 seconds
    • Measure the voltage with the DVM = "VD"
    • Test fails if: (("Vset""TV" - "VD") / "Vset""TV") > 0.01: Let's keep that value for the moment and see how this works with the few first PSU we will get…
    • Set the load current to "TC"
    • Wait for 3 seconds
    • Abort the test if the PSU has tripped
    • Measure the voltage with the DVM = "VD2"
    • Test fails if: (("Vset""TV" - "VD2") / "Vset""TV") > 0.01: Let's keep that value for the moment and see how this works with the few first PSU we will get…
    • Measure the current seen by the load ("LC")
    • Measure the current seen by the PSU ("PC")
    • Test fails if: (("PC" - "LC") / "PC") > 0.05: Let's keep that value for the moment and see how this works with the few first PSU we will get…
    • Test fails if: (("VD" - "VD2") / "VD") > 0.005: Let's keep that value for the moment and see how this works with the few first PSU we will get…
    • Turn the PSU channel off
    • These values will be stored in the database
      • "Vset""TV"
      • "VD"
      • Voltage measured by PSU at 0 A load
      • "VD2"
      • Voltage measured by PSU at full load
      • Current measured by PSU at full load
      • Current measured by load at full load
      • (("VD" - "VD2") / "VD")
  • Program the default values back into the PSU channel
  • Set the load back to 0 A

Mains Test

No Mains test for the distributed maraton unless we want to test the PFC but this will be done later on.
  • Set the power box to 230 V
  • Retrieve the Power limit of the PSU from the DB. This is "PS".
  • Retrieve the individual power limit of all channels and sum them up. This is “PC”.
  • Compare PS with PC. If PC is larger than PS, compute the derating factor by doing PS/PC and apply this derating factor to each individual channel power. Those will be PT [0 to 12].
  • Configuration of the channels (to do for every channel):
    • The test voltage "TV" is the calculated like this: "voltage_min" + (("voltage_max" - "voltage_min")/2). In case the PSU is a limited one (“L”), TV must be compared with the limitation entered by the operator. TV is then the smaller of the two values.
    • Take the nominal current as entered in the DB and compare it with the result of (PT (chan) / TV). Take the smaller of the two, this is SC.
    • The test current, TC, is the value of "SC” multiplied by the arbitrary derating factor 0.9.
    • Set the PSU to output "TV"
    • Set the load to a slew of 100 A/s
    • Set the load to the test current "TC"
  • Turn all PSU channels ON
  • For 60 seconds:
    • Check if the PSU has tripped
    • Once per second measure: Iload, Ipsu, Vdvm, Vpsu
  • Compute mean values and standard deviations (with the measurements from the previous loop ) for: Vdvm, Vpsu, Iload, Ipsu
  • Write these parameters to the report file:
    • Mean current from Load
    • Mean current from Power Supply
    • Standard Deviation current from Load
    • Standard Deviation current from Power Supply
    • Raw values current from Load
    • Raw values current from Power Supply
    • Mean voltage from DVM
    • Mean voltage from Power Supply
    • Standard Deviation voltage from DVM
    • Standard Deviation voltage from Power Supply
    • Raw values voltage from DVM
    • Raw values voltage from Power Supply
  • If the PSU has tripped switch it on again and continue the test
  • Set the input power to 207 V (230 -10%)
  • Wait for 2 seconds
    • Measure the power of the individual load channels and sum it up
    • Compute efficiency of the PSU (Pin / Pout)
    • Store these parameters in the report file
        • Input Power
        • Output Power
        • Overall Efficiency
    • Individual channel efficiency is also computed but this is not documented…
  • If the PSU has tripped switch it on once more and continue
  • Set the input power to 253 V (230 +10%)
  • Wait for 2 seconds
    • Measure the power of the individual load channels and sum it up
    • Compute efficiency of the PSU (Pin / Pout)
    • Store these parameters in the report file
        • Input Power
        • Output Power
        • Overall Efficiency
    • Individual channel efficiency is also computed but this is not documented…
  • If the PSU has tripped continue
  • Set the power back to 230 V
  • Switch the PSU on (just in case)
  • If the PSU does not trip immediately:
    • Wait for 5 seconds
    • Measure the power of the individual load channels and sum it up
    • Compute efficiency of the PSU (Pin / Pout)
    • Store these parameters in the report file
        • Input Power
        • Output Power
        • Overall Efficiency
  • Switch all PSU channels OFF.
  • Set the current of all loads to 0 A
  • Loop over all channels. For each channel:
    • Switch this channel ON.
    • Set the load to the previously determined TC current for this channel.
    • Wait 2 seconds
    • Read the power from the power box
    • Read the power from the load
    • Compute the efficiency
    • Store these values in the report file:
        • Input Power (as measured by the power box)
        • Output Power (as measured by the load channel)
        • Channel Efficiency
    • Set the load back to 0 A
  • The test result will be "failed" if the PSU tripped once or several times during the test

Wiener PFC (Maraton AC-DC/PFC module)

Parameters

(This is not really part of the test for the AC-DC module but it is required for the ref PSU to work properly) See detailed description under distributed maraton psu.

Dynamic output voltage test

(Can somehow be seen as a sort of sensor test)

  • Get the PSU status
    • If not in good shape abort the test immediately
  • Set all loads to 0A and slew 100 A/s.
  • Turn the ref ‘maraton" PSU on (close relay no.: 516).
  • Wait 5 seconds.
  • Check the output voltage of the AC-DC PFC module
    • Close relay no.: 435 (this connects the DVM on the PFC output).
    • Read the DVM voltage.
    • If DVM reading has not yet stabilize (variations higher than 0.2 V), wait for stabilization (variations: < 0.2 V).
    • Read the DVM voltage.
    • Multiply the DVM reading by 10.51 (was 10.66 until 31.3.2016. Changed to the new value on the request of S. Mico in order to compensate for a drift of the voltage divider) to be corrected with the right value as in the code (voltage divider)
    • Check the result: Tolerance is 385 5 Vdc.
    • Write this value in the report.
  • Retrieve the Power limit of the PSU from the DB. This is "PS".
  • Retrieve the individual power limit of all channels and sum them up. This is “PC”.
  • Compare PS with PC. If PC is larger than PS, compute the derating factor by doing PS/PC and apply this derating factor to each individual channel power. Those will be PT [0 to 12].
  • Configuration of the channels (to do for every channel):
    • The test current "TC" is the smaller value of these two: “Iset" and “scmax” multiplied by the arbitrary derating factor 0.9.
    • The voltage is fixed.
    • Check that the theoretical output power (TC * Vset) is smaller than PT [for this channel] * 0.9. Report a warning if this is out of range and calculate the new value of TC as follow: TC = PT[of this channel]*0.9/Vset.
    • Set the load to a slew of 100 A/s
    • Set the load to the test current "TC"
    • Turn the channel ON.
    • Wait for 1 second
    • Check the channel. If there is a trip terminate the test
  • Verify the ‘maraton’ PSU status.
    • If not in good shape abort the test immediately
  • Read the voltage and current parameters of all channels.
  • Compute the total power going out of the ‘maraton" PSU.
  • Write the total PSU power in the report.
  • Check the output voltage of the AC-DC PFC module (same process as before, same tolerance).
  • Set all loads back to 0 A.
  • Open relay 435.

If at any time, the PFC output voltage is out of range, test has failed. If at any time the ‘maraton’ PSU trips or reports an error the test is not valid (not failed)… One should then investigate on the reason of the trip.

Soak test

  • Compute the storage interval "SI". During the soak test the S/W will store values of PFC output voltage as well as the total output power of the reference PSU at regular intervals. The "SI" is the duration of the interval. "SI" is the soak time divided by 10. If the soak time is less than 300 seconds "SI" will be set to 30 seconds and less than 10 data sets will be stored.
  • Retrieve the Power limit of the PSU from the DB. This is "PS".
  • Use a fixed value of 3600 W for the power limit of the PFC and multiply it by 0.75 (total inefficiency of the system). This is "PF".
  • Compare PF and PS and take the smaller of the two. Store this smallest value as PS.
  • Retrieve the individual power limit of all channels and sum them up. This is “PC”.
  • Compare PS with PC. If PC is larger than PS, compute the derating factor by doing PS/PC and apply this derating factor to each individual channel power. Those will be PT [0 to 12].
  • Configuration of the channels (to do for every channel):
    • The test current "TC" is the smaller value of these two: “Iset" and “scmax” multiplied by the arbitrary derating factor 0.9.
    • The voltage is fixed.
    • Check that the theoretical output power (TC * Vset) is smaller than PT [for this channel] * 0.9. Report a warning if this is out of range and calculate the new value of TC as follow: TC = PT[of this channel]*0.9/Vset.
    • Set the load to a slew of 100 A/s
    • Set the load to the test current "TC"
    • Turn the channel ON.
    • Wait for 1 second
    • Check the channel. If there is a trip terminate the test
  • Enter into the soaking loop:
    • Each time a "SI" has elapsed store these parameters for all channels:
      • Computed value of total output power by summing up the product of all channels "MeasurementSenseVoltage" and "MeasurementCurrent".
      • AC-DC output voltage measured by DVM (relay 435 and coefficient to be applied).
    • Every minute, check the same set of parameters and record the min/max values (if higher or lower than the min/max stored previously)
    • In the report, print the min and max values of these parameters for all channels.
    • Terminate the test is a channel reports an error
  • Turn all PSU channels off
  • Set all loads to 0 A
  • Set all PSU channels to the default parameters
  • Open relay 435.

If at any time, the PFC output voltage is out of range (same tolerance as for the previous test), test has failed but this event does not stop the test. The test will continue until the duration set in the DB is reached. If at any time the ‘maraton" PSU trips or reports an error the test is not valid (not failed)… One should then investigate on the reason of the trip.

Mains Test

  • Set the power box to 230 V
  • Not needed anymore: Retrieve the Power limit of the PSU from the DB. This is "PS".
  • Retrieve the individual power limit of all channels and sum them up. This is “PC”.
  • Compare the fixed value of 2475 W (this takes into account all inefficiencies and the 230V - 10%) with PC. If PC is larger than 2475 W, compute the derating factor by dividing 2475W by PC and apply this derating factor to each individual channel power. Those will be PT [0 to 12].
  • Configuration of the channels (to do for every channel):
    • The test current "TC" is the smaller value of these two: “Iset" and “scmax” multiplied by the arbitrary derating factor 0.9.
    • The voltage is fixed.
    • Check that the theoretical output power (TC * Vset) is smaller than PT [for this channel] * 0.9. Report a warning if this is out of range and calculate the new value of TC as follow: TC = PT[of this channel]*0.9/Vset.
    • Set the load to a slew of 100 A/s
    • Set the load to the test current "TC"
  • Turn all PSU channels ON
  • For 60 seconds:
    • Check if the PSU has tripped
    • Once per second measure: AC-DC output voltage with the DVM (coefficient to be applied)
  • Compute mean values and standard deviations (with the measurements from the previous loop) for the measured voltage.
  • Write these parameters to the report file:
    • Min, max and mean voltage from DVM
    • Standard Deviation voltage from DVM
  • If the PSU has tripped switch it on again and continue the test
  • Set the input power to 207 V (230 -10%)
  • Wait for 2 seconds
  • For 60 seconds:
    • Check if the PSU has tripped
    • Once per second measure: AC-DC output voltage with the DVM (coefficient to be applied)
  • Compute mean values and standard deviations (with the measurements from the previous loop) for the measured voltage.
  • Write these parameters to the report file:
    • Min, max and mean voltage from DVM
    • Standard Deviation voltage from DVM
  • If the PSU has tripped switch it on once more and continue
  • Set the input power to 253 V (230 +10%)
  • Wait for 2 seconds
  • For 60 seconds:
    • Check if the PSU has tripped
    • Once per second measure: AC-DC output voltage with the DVM (coefficient to be applied)
  • Compute mean values and standard deviations (with the measurements from the previous loop) for the measured voltage.
  • Write these parameters to the report file:
    • Min, max and mean voltage from DVM
    • Standard Deviation voltage from DVM
  • If the PSU has tripped continue
  • Set the power back to 230 V
  • Switch the PSU on (just in case)
  • If the PSU does not trip immediately:
    • Wait for 5 seconds
    • Measure the power of the individual load channels and sum it up
    • Compute efficiency of the PSU (Pin / Pout)
    • Store these parameters in the report file
  • Input Power
  • Output Power
  • Overall Efficiency
  • Switch all PSU channels OFF.
  • Set the current of all loads to 0 A

The test result will be "failed" if the PSU tripped once or several times during the test or if the output voltage is out of range (same tolerance as for the sensor test).

Wiener RCM

TBD

Functionality

VMEbus Wiener LV CAEN LV
Current limit yes yes yes
Mains yes yes (only monolithic) no
Over voltage yes yes no
Over voltage trip yes yes yes
Probe (control channel) yes yes yes
Sensor accuracy not yet (planned for next release) yes yes
Static regulation yes yes yes
Ripple yes yes (planned) yes (planned)
Ramp no not yet no
Soak no yes yes

Open issues

  • Guide for the use of the test bench to be written (PH) In progress
  • We need an interlock to rule out that two PSUs are plugged in at the same time (PH, PF)
  • Test reports
    • Document the parameters that are currently stored in the database (MJ)
    • Review the list of parameters (PH) Done (below)
  • Tester for the isolation of the mains
    • The device has arrived. Now we need to write down when and how it has to be used (PH) Will be included in guide (above)
  • OPC server for Wiener LV PSUs
    • New server release and documentation required (MJ to contact M. DuTour)
  • VMEbus PSUs
    • Specifications to be provided for sensor test (PH)Which sensors exist in the VMEbus PSU (just I, C, Temp. or is there more)? Which of these sensors do we want to test? What are the conditions for the test (i.e. load, duration, etc.)? What are the acceptance margins?

V, I & t (but we have no accurate ref for t, so a very approx room temp comparision will suffice)
Conditions - No load & full load, after warmup
pass/fail criteria should be based on 'reasonable application of B1186 tesch spec & manufactuers product spec (TBD in detail)

    • Paul mentioned a yet undocumented per-channel parameter for VMEbus PSUs: power-on delay. We need documentation. The parameter should probably go into the tech. DB (PH, CS)
    • We have to find a scope that can do the verification of the ripple and learn how to set it up for that test (PH)
  • Wiener LV PSUs
    • We have to find a scope that can do the verification of the voltage rise / fall and learn how to set it up for that test (PH)
    • The work on the oscilloscope will be done after the initial functionality is complete
  • Technical database
    • Default parameters for VMEbus and CAEN PSUs to be added (PH) Work in progress
    • New fields to be added for system parameters of VMEbus PSUs (e.g. F/W version, operating hours) (PH, CS)
  • Serial box for distributed Wiener LV PSUs:
    • Improved F/W to be provided (current F/W needs a 1s delay and may not yet provide all features) (EDS)
    • Some documentation should be written about this box (EDS)
    • The scaling factors have to go into the tech. DB (PH, CS)
  • Modules with negative channels may need additional consideration (PH) A3006 causes a problem - solution under investigation
  • Install suitable cooling for the Easy crate (PF)

Parameters in the test report for CAEN PSUs

Not mentioned here are various error messages that will show up in the test report if a test failed

Current limit test

Parameter name Description (in the Java code) Additional information
refcurr Reference Current = Current limit Value of "current_max" in the DB
tripcurr Trip Current Current at whihc the trip was detected
dpc Percent difference from internal setting abs(tripcurr - refcurr) * 100 / refcurr
trip_seen_after Trip time out The number of seconds after which the trip was seen. The accepted range is 10s +/- 5%

(PH) ok

Overvoltage trip test

Parameter name Description (in the Java code) Additional information
ovv_volt OVV flag Voltage Voltage at which the OVV flag was seen
hvmax_volt HVMAX flag Voltage Voltage at which the HVMAX flag was seen
ovcompare Trip Over Voltage Internal Setting Value of "voltage_max" in the DB + 0.25 V
dpc OVV percent difference from internal setting abs(ovv_volt - ovcompare) * 100.0f / ovcompare

(PH) ok

Probe test

Parameter name Description (in the Java code) Additional information
OPC_Name OPC tag Value of the OPC parameter of the given channel
OPC_Status OPC tag Value of the OPC parameter of the given channel
OPC_Trip OPC tag Value of the OPC parameter of the given channel
OPC_VCon OPC tag Value of the OPC parameter of the given channel
OPC_SVMax OPC tag Value of the OPC parameter of the given channel
OPC_GlbOffEn OPC tag Value of the OPC parameter of the given channel
OPC_GlbOnEn OPC tag Value of the OPC parameter of the given channel
OPC_V0Set OPC tag Value of the OPC parameter of the given channel
OPC_I0Set OPC tag Value of the OPC parameter of the given channel

(PH) ok

Sensor test

Parameter name Description (in the Java code) Additional information
ov0psu output voltage at 0 A load current (measured by PSU) -
ov0dvm output voltage at 0 A load current (measured by DVM -
dev Deviation in percent -
oc0psu Oputput current at 0 A load current (measured by PSU) -
oc0loa Oputput current at 0 A load current (measured by load) -
ov1psu output voltage at full load current (measured by PSU) -
ov1dvm output voltage at full load current (measured by DVM) -
dev1 Deviation in percent -
oc1psu Oputput current at full load current (measured by PSU) -
oc1loa Oputput current at full load current (measured by load) -
dev2 Deviation in percent -

Soak test

Parameter name Description (in the Java code) Additional information

This test currently does not store any parameters in the DB. It only creates error messages in case of a problem. (PH) Should do the following:
at regular periods during soak (10s?) store the following values:
- time
- DVM V, Caen V
- Load I, Caen I
If trip, should show what caused it.
Values could be graphed in report to show stablisation vs time
Working on it. The graph may be tricky.graph could come later.

Static regulation test

Parameter name Description (in the Java code) Additional information
cmax0, cmax1 Node Max Current Value of "current_limit_max" in the DB
novolt0 Nominal output voltage of the PSU channel Value of "voltage_min" in the DB
novolt1 Nominal output voltage of the PSU channel Value of "test_voltage" in the DB
c0, c1 Current from load at 0% -
cps0, cps1 Current from CAEN PS at 0% -
vps0, vps1 Voltage from PS at 0%) -
cf0, cf1 Current from load at 100% -
cpsf0, cpsf1 Current from CAEN PS -
vpsf0, vpsf1 Voltage from PS at 100% -
v10, v11 Voltage from DVM at 0% -
v20, v21 Voltage from DVM at 100% -
ratio0, ratio1 abs((v1(0/1) - v2(0/1)) / v1(0/1) -

(PH) ok

Note:
Parameter suffix "0" -> Test at min. voltage
Parameter suffix "1" -> Test at max. voltage

Database parameters for the monolithic Wiener LV-PSUs

Database name of the parameter Corresponging OPC parameter default Purpose of the parameter in the test bench S/W
systempower_max n/a n/a Used in MainsTest and SoakTest to avoid requesting more power than the PSU can deliver
voltage_max n/a n/a Used in the SensorTest to program the output voltage of the channel
power_max n/a n/a Used in MainsTest and SoakTest to avoid requesting more power than the channel can deliver
voltage_min Voltage n/a Voltage for the CLimTest
Used in the SensorTest to program the output voltage of the channel%BR (sensor test is done at both min and max voltage)
Voltage for the StatregTest
current_max Current n/a Just for initialisation
svmax SupervisionMaxSenseVoltage n/a Just for initialisation
svmin SupervisionMinSenseVoltage n/a Just for initialisation
svmaxt SupervisionMaxTerminalVoltage n/a Just for initialisation
scmax SupervisionMaxCurrent n/a Current limit for the CLimTest
n\a OnOffChannel false Enabling / disabling a channel. As a general rule a channel is only enabled if it is required for a test procedure
n\a MinSenseVoltage 1 Just for initialisation
n\a MaxSenseVoltage 1 Just for initialisation
n\a MaxTerminalVoltage 1 Just for initialisation
n\a MaxCurrent 1 Just for initialisation
n\a MaxTemperature 1 Just for initialisation
n\a MaxPower 1 Just for initialisation

Note: All parameters that have a OPC name (and only those) will be used for the initialisation of the PSU. If a value is available from the DB then this value will be written to the respective OPC parameter. In the absence of a value from the DB the default will be used. If a default is not available the test S/W will exit with an error message.

-- MarkusJoos - 1 Oct 2008
-- PaulHarwood - 10 Oct 2008

Edit | Attach | Watch | Print version | History: r92 < r91 < r90 < r89 < r88 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r92 - 2016-07-14 - SylvainMico
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Sandbox All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback