PXI-4110 individual channel power-up sequencing

[GSinMN]

Hello All,

  I have a hardware device that requires a very specific power-up/power-down sequence in order to operate correctly, and not be damaged.  According to the datasheet for the PXI-4110, it is capable of meeting the power requirements, but I would have to use Labview to create a power sequencing VI for each of the channels.

  After investigating the options, and studying the provided DCPower examples,  I understand how to configure each channel, but from what I have seen, the actual initialization (niDCPower Initiate.vi) will enable the power supply based on the "instrument handle" alone, rather than initiating each channel of the instrument individually.

  Is there a method to use Labview VIs to separate the individual channels of the PXI-4110 and to control the power-up sequence of each channel? 

Thanks for your time.

Regards,

GSinMN     

[TobiasG]

GSinMN,

Yes, you can use the Initialize With Channels VI to specify which channels are within a session. You can specify a single channel per session, and then initiate them separately. Instead of depending on Initiate, however, you may want to Commit and Initiate first, then later use Configure Output Enabled, Configure Voltage Level, or Configure Current Limit to set each channel. This will be more precisely timed, as there is a lot going on when you call Commit/Initiate.

All this said, how precisely timed does your power source need to be? The 4110 only supports software-timed commands, so precise timing will be difficult. If you were to use one of our newer devices (for instance, the PXIe-4143), you would be able to control your output sequence in a hardware-timed fashion, by using triggers and/or the Set Sequence VI. This means jitter of milliseconds for the 4110 (due to the software-timed nature) compared to microseconds (for hardware-timed).

As our various devices differ in their voltage and current capabilities, what are your power requirements?

[GSinMN]

Hello Tobias,

  Thanks for the quick response.  Our power requirements are for +5VDC, +15VDC and -15VDC.  The +5VDC rail must reach it's end voltage before the other two pass +5/-5, and all three must reach full voltage within 200 msec of each other.  Also, a smooth ramp-up is required, rather than a abrupt switch.  

  Your suggestion to "initiate then configure" sounds like it would make sense.  All we would have to do is initiate all channels to 0VDC, then ramp them up or down as needed.

  We have not committed to a specific power supply as of yet, I just chose the 4110 base on our minimum requirements, and to avoid overkill.

Thanks again,

GSinMN        

[Brandon_G.]

Hey GSinMN,

<200 ms seems reasonable to handle with software timing especially if you commit/initiate before setting the output state as Tobias suggested.  Can you describe what sort of "smooth ramp-up" is required for your DUT?  The rise time will not be able to be precisely controlled because this will be dependent on the DUT and the transient response settings of the 4110 itself.  You maybe able to tweak the current limit to limit in-rush current thereby slowing down your voltage ramp-up, but this may not work depending on how your DUT behaves on startup-up.  Another option, would be to use a 414x SMU which will allow custom transient response and also allow hardware timing (you can synthesis a ramp wave)  Do you have a certain rise-time in mind?

Also, do you know your current requirements for each rail?

Thanks!

Brandon G

[GSinMN]

Hello Brandon,

  Thanks for the response, and welcome to the conversation. 

  I do not have specific details related to the type of load, other than the max total power would be 5.5 watts (max current draw for each channel is 500mA, 100mA and 100mA).

  The requirement for the smooth ramp-up is related to the fact the device contains a laser gyro, and voltage/current spikes would (presumably) damage it.  I am not aware of a specific ramp time, other than the 200 msec total. 

  Would it be a reasonable implementation to use a loop to repeatedly re-configure the channel voltage levels?  It could be considered a psuedo-ramp, if the increments were small enough.  The data sheet for the 4110 indicates minimum increments of 120 uV and 400 uV. 

Thanks,

GSinMN

[Brandon_G. · Accepted solution]

Hey GSinMN,

Yes, you would be able to run a loop to increment the output in steps, but of course the more steps you have the more time it will take to reach your final value.  My concern would be that you would exceed the 200 ms time if you had too many steps.  Also, your total rise time will vary depending on your CPU and what other processes you are running in the background.  

If you benchmarked the rise time on a certain CPU with certain processes running, you will likely not have the same rise-time when porting your code to another machine.  

It maybe worthwhile to ask the laser manufacturer if they have any guidance on what the ramp turn on time should look like.  The 4110 is designed to have a fast rise time with minimal overshoot for most loads, but it would be good to understand what they require for their laser.  Perhaps it's acceptable to have a fast rise time so long as the output doesn't exceed x% of the final value?  Any insight they have on this would be helpful.

Please let us know how we can help.

Thanks!

Brandon G

[GSinMN]

Thanks for all of the assistance.  I will attempt to gather more info related to our hardware, and possibly try a few of the options we discussed.  I'll report back if I find any surprises.

Thanks again,

GSinMN

[Rathan123]

Hi Tobias,

I want to force at one SMU-4143 channel and measure at other SMU-4143 channel by loop back. Please help me out to solve this issue.

Thanks&Regards,

Rathan

[Rathan123]

I want to force at one SMU-4143 channel and measure at other SMU-4143 channel by loop back. I tried, but it seems to be not possible as both are sources. Is this possible?? Please help me out

Thanks,

[TobiasG]

Rathan, Please refer to Jared's reply to http://forums.ni.com/t5/Digital-Multimeters-DMMs-and/4143-Loop-Back/m-p/3288660 Basically, you want to use the second channel to take a measurement (voltage or current), which you can do by configuring the measurement channel to use the other Output Function at its smallest range. For instance, if you want to measure voltage on channel 1, configure the Output Function to be DC Current and set the Current Level and Current Range to the smallest possible. That will produce a minimal amount of current draw by channel 1 while still allowing you to measure voltage. Please let us know if you have any further questions.