Please enter search keywords!

Site Map

Site Map

High Precision Benchtop SMU

S2022H

8-channel High-precision SMU


The S2022H is a high-precision, compact and cost-effective benchtop Source/ Measure Unit (SMU) with the capability to source and measure both voltage and current. The S2022H is ideal for IV (current versus voltage) measurement tasks that require both high resolution and accuracy


Features

  • High Range

    Range: ± 30 V, ± 500 mA (DC/pulse)
  • High Resolution

    The minimum measurement resolution can reach 100 pA/60 μV
  • High Rampling Rate

    Supports up to 500K ADC sampling rate
  • Threshold Trigger

    Hardware high-speed IO, capable of threshold triggering, enabling efficient interaction between output measurement values and user systems

Functions and Advantages

  • I-V Output Capability

Voltage Source Specifications

Voltage Programming Accuracy

Range

Programming Resolution

Accuracy (1 year)

± (% reading+offset)[1]

Typical Noise (RMS)

0.1 Hz-10 Hz

± 30 V[3]

1  mV

0.03%+4 mV

1000 μV

± 6 V

200 μV

0.03%+1 mV

100 μV

Temperature Coefficient

± (0.15 x accuracy)/℃ (0 ℃ -18 ℃, 28 ℃ -50 ℃)

Channel[2]

CH1 to CH8

Output Power

Ch1 Ch4: 3 W per channel and 6W total

Ch5 Ch8: 3 W per channel and 6W total

Overshoot

< ± 0.1% (typical, normal, step is 10% to 90%, full range, resistive load)

Noise 10Hz-20MHz

6V voltage source, 0.5A resistive load,< 3 mVrms

[1] Accuracy calculation example: For testing the accuracy of a 6V range with a 1V output, the tolerance is

[2] All channel outputs are electrically isolated from ground, but the outputs of channels CH1-CH4 share a common ground (LO)


[3] This instrument has potentially dangerous high voltage (±31.5 V) output to the HI / Sense HI / Guard terminals. To prevent electric shock, relevant precautions must be taken before powering on. Safety precautions. Do not connect the Guard terminal to any output, including shorting it to chassis ground or output LO, as this will damage the instrument.


Current Source Specifications

Current Programming Accuracy

Range

Programming  Resolution

Accuracy (1 year)

± (% reading+offset)

Typical Noise (RMS)

0.1 Hz-10 Hz

± 500 mA[4]

20 μA

0.05%+100 μA+  Vo * 25 μA

10 μA

± 100 mA

4  μA

0.05%+10 μA+  Vo * 5 μA

1 μA

± 10 mA

400 nA

0.05%+5 μA+  Vo * 500 nA

100 nA

± 1 mA

40 nA

0.05%+500 μA+ Vo * 50 nA

10 nA

± 100 μA

4  nA

0.05%+50 nA+  Vo * 5 nA

1 nA

± 10 μA

400 pA

0.05%+20 nA+  Vo * 500 pA

150 pA

Temperature Coefficient

± (0.15 x accuracy)/℃ (0 ℃ -18 ℃, 28 ℃ -50 ℃)

Channel[5]

CH1 to CH8

Output Power

Ch1 Ch4: 3W per channel and 6W total

Ch5 Ch8: 3W per channel and 6W total

Overcharge

< ± 0.1% (typical, normal, step is 10% to 90%, full scale point, resistive load test)

[4] 500mA range is available only for 6V voltage range

[5] Channels are isolated from earth ground, but CH1 to CH4 share a common LO, and CH5 to CH8 share a common LO


Voltage Measurement Specificationgs

Voltage Measurement Accuracy

Range

Measurement Resolution

Accuracy (1 year)

± (% reading+offset)

± 30 V

300 μV

0.03%+4 mV

± 6 V

60 μV

0.03%+1 mV

Temperature Coefficient

± (0.15 x accuracy)/℃ (0 ℃ -18 ℃, 28 ℃ -50 ℃)

  

Current Measurement Specifications

Current Measurement Accuracy

Range

Measurement Resolution

Accuracy (1 year)

± (% reading+offset)

± 500 mA[6]

10 μA

0.05%+100 μA+  Vo * 25 μA

± 100 mA

1 μA

0.05%+10 μA+  Vo * 5 μA

± 10 mA

100 nA

0.05%+5 μA+  Vo * 500 nA

± 1 mA

10 nA

0.05%+500 nA+  Vo * 50 nA

± 100 μA

1 nA

0.05%+50 nA+  Vo * 5 nA

± 10 μA

100 pA

0.05%+20 nA+  Vo * 500 pA

Temperature Coefficient

± (0.15 x accuracy)/℃ (0 ℃ -18 ℃, 28 ℃ -50 ℃)

[6] 500mA range is available only for 6V voltage range



Typical Output Setting Time

Source

Range

Output setting time[7]

Condition

Fast [8]

Normal

Slow

Voltage

30 V

< 400 μs

< 1.5ms

< 2.8 ms

Time required to reach within 0.1 % of final value at open load condition.
Step is 10 % to 90 % range

6  V

<250 μs

<780 μs

<2.8 ms

Current

± 500 mA

<50 μs

<330 μs

<2.5 ms

Time required to reach within 0.1 % of final value at short condition.
Step is 10 % to 90 % range

±100 mA

<50 μs

<270 μs

<2.5 ms

±10 mA

<50 μs

<270 μs

<2.5 ms

±1 mA

<100 μs

<290 μs

<2.5 ms

±100 μA

<150 μs

<5 ms

<2.5 ms

±10 μA

<250 μs

<3 ms

<2.5 ms

[7] Output transition rate: Three modes of Fast, Normal, and Slow are available, allowing users to adjust according to load characteristics to achieve appropriate setup time or stability. This feature only supports switching the output transition rate when OUTPUT is set to OFF

[8] The Fast mode may produce a large overshoot under different ranges or load conditions. For overshoot-sensitive devices, it is recommended to use normal or Slow mode. mode


Similar recommendation

High Speed Communication Test
High Speed Communication Test

High speed communication test plays an important role in the rapid development of big data, cloud computing, 5G communication and other markets.
Semight offers various of instruments for optical Transceiver/Component testing, including wide bandwidth sampling oscilloscope, NRZ/PAM4 bit error ratio tester , burst error ratio tester, fast wavelength meter, high precise source measure unit, 400G network analyzer ,optical power meter, optical attenuator, optical switch etc. We provide cost-effective, complete solutions for optical testing.

Details
Electronic Measurement
Electronic Measurement

The high-precision source measure units integrates the functions of voltage source, current source, voltmeter, ampere meter, and electronic load in one, which is widely used in high-precision IV test and measurement for various discrete components, photovoltaic, green energy, battery and other industries. Semight provides high-precision benchtop SMU and plug-in PXIe SMU of standard PXIe chassis, fully meeting the application of various test scenarios.

Details
Optical Chip Test
Optical Chip Test

Burn-in testing of laser is an important method to ensure the reliability of laser. Through the test of CoC or bare die, the early failure of laser caused by the defects in the process of laser production can be screened out in advance. Semight provides a complete solution from bare die to CoC, from high temperature(150℃ or higher) to low temperature (-40℃), with CoC automatic loading and unloading system, forming a complete test solution, Semight's laser chip burn-in/load/unload test system has been widely recognized by the market.

Details
Power Semiconductor Test
Power Semiconductor Test

The semiconductor front-end test is mainly used in the wafer processing to check whether the processing parameters of the wafer products meet the design requirements or there are defects affecting the yield after each step of the manufacturing process. The semiconductor back-end test equipment is mainly used after wafer processing to check whether the performance of the chip meets the requirements, which belongs to the electrical performance test. Semight provides solutions such as Wafer Level Burn In system and Known Good Die handler for SiC testing, offering the value to customer in test efficiency improvement and test cost reduction.

Details
Log in

Account number

Password

Register an account

Name

Please enter your name *

E-mail

Please enter your email address *

E-mail verfication code

Please enter your email verification code

Phone

Please enter your contact number

Password

Please enter your login password *

Confirm Password

Please enter your login password again *
Reset password

E-mail address

Please enter your email number *

E-mail verification code

Please enter your email verification code

New Password

Please enter your login password *

Confirm Password

Please enter your login password again *