Can we Do the two differnent SMU cascade

[ram121188@a]

Hi Team 

        In higher voltage Load board design. Stack up i am using SMU 4137 & 4162 for getting 250V can i do two different instrument stack up

Thanks 

Ramkumar S

[santo_13 · Accepted solution]

I would not recommend doing so - the current limits and protection circuit are different, instead, get another 4137 for the cascading. It would be possible as long as you don't exceed the maximum voltage rating of each instrument channel. Note that in this connection method, none of the accuracy specifications is valid.

I believe this is how you would be planning to connect the 4137 and 4162 to achieve 250V (248V reality = 200 + 24 - (-24))

Some articles to refer to,

https://www.ni.com/en-us/support/documentation/supplemental/11/cascading-the-outputs-of-the-ni-pxie-...

https://knowledge.ni.com/KnowledgeArticleDetails?id=kA03q000000ww0yCAA&l=en-US

https://knowledge.ni.com/KnowledgeArticleDetails?id=kA00Z000001DcHPSA0&l=en-US

https://knowledge.ni.com/KnowledgeArticleDetails?id=kA00Z000001DbUNSA0&l=en-US

[ram121188@a]

sorry about it . In my project we are cascade the 4137 to get the 250 volt.is it recommanded two 4137 channel. 

[ram121188@a]

NI 4139 i have seen below suggestion . in this +/- 5V why we have to choose ?

For example, suppose you wish to source 1 mA with a limiting voltage of ±100 V. Each NI PXIe-4138/9 device can provide only 60 V, so two must be placed in series to reach the required 100 V.

Assume you decide to limit the voltage across each module to 50 V ± 5 V. This keeps the voltage delivered by either SMU comfortably under its 60 V limit. The current accuracy specification for the NI PXIe-4139 states that the output current is within 0.03% + 60 nA of the programmed value in the 1 mA range within 5 °C of the last self-calibration. For a 1 mA setpoint, the accuracy works out to ±(0.03/100 * 1 mA + 60 nA) = ± 360 nA. The worst-case situation would have one device 360 nA high while the other is 360 nA low, for a 720 nA difference between the two outputs. That current drives the common node between the two SMUs, and if you wish to limit the voltage error of the middle node to 5 V to ensure each device stays comfortably within its compliance limits, you need a resistance of 5 V/720 nA = 7 MΩ at the junction of the two devices. More resistance could work, but too much voltage difference between the two devices could force channels into compliance prematurely and waste power in some instances.

Thanks 

Ramkumar S

[santo_13]

Your schematic looks good, you have one SMU always GND referenced, the other one can either be GND referenced or cascaded with the first SMU based on a set of relay states.

+/-5V is chosen to restrict the SMU from hitting the 60V max limit in CC mode. This can also be +/- 8V too, still, it will be under the 60V max.

For example, though both SMU are in series and source the same current, the voltage required to force that current would vary with each SMU. It is likely that since in this example, you're dealing with compliance voltage which is close to maximum, they have chosen to limit it further to avoid hitting the maximum.

If you are trying to operate only in CV mode, you don't need to worry about these resistances and stuff.