two dc power supply precisely timed switching

[jetfire]

Hi,

i am trying to controll switching  between two E3631A DC Power supply with different period,in my programm i am using to timed switching  case structur for controlling DC Power switch,but because of kommunication overhead doesn‘t  timing result  always agree with specified period,how can i modify my code to precisely controll this Switching？if f anyone can give me some ideas that would be appreciated.

Note：Labview 2009 sp1

[LV_Pro]

What are you trying to accomplish? How closely in timing does the switch over have to occur. I doubt that there is much that you can do to make sure that the two supplies switch on precise timings, regardless of how you configure your communications. Also, as the actual power supply vi's aren't included, we can't see how your program is changing the power supply settings, we only have blank boxes where those vi's would be.

[jetfire]

---

@LV_Pro wrote:

What are you trying to accomplish? How closely in timing does the switch over have to occur. I doubt that there is much that you can do to make sure that the two supplies switch on precise timings, regardless of how you configure your communications. Also, as the actual power supply vi's aren't included, we can't see how your program is changing the power supply settings, we only have blank boxes where those vi's would be.

---

 i want to controll two DC Power Supply like this：Case1：DC Power #1 enable one channel output 3.5 V ,and DC Power #2 disable all outputs,run for specified time ,then switch to case 2： DC Power #1 disable all outputs,DC Power #2 enable three channel outputs 1 V,+18V，-18V，run for specified time,like this  case switching with specified period ....

[GerdW]

Sounds like you want to implement a statemachine with a special "wait" state...

[LV_Pro]

How critical is the timing? Trying to trigger two power supplies, with multiple settings, via GPIB (or serial, tcp/ip, etc.) is non-deterministic, meaning that you can rarely predict when the actions will occur. I would also suggest moving the setting of the power supply parameters out of the time critical part, having only the "output ON/OFF" commands, which will save a lot of communications overhead, and is probably a lot quicker function for the supplies to execute. For really critical timing it might be necessary to have some kind of relay between the supplies and whatever they are supplying. Of course this has its own problems, like how much current they are supplying, etc.

[apok]

dont expect that turning power supplies off and on and expecting voltage to be there precisely is an option, due to settling factors and what not's....i think that power rails should be on at all times and as suggested using relays, fast fets or transistors to switch your power in.....digitally controlled by the daq by some sort of timming trigger?

[jetfire]

---

@apok wrote:

dont expect that turning power supplies off and on and expecting voltage to be there precisely is an option, due to settling factors and what not's....i think that power rails should be on at all times and as suggested using relays, fast fets or transistors to switch your power in.....digitally controlled by the daq by some sort of timming trigger?

---

thanks, relay switch is new to me,i must reading helps about it

[jetfire]

---

@LV_Pro wrote:

How critical is the timing? Trying to trigger two power supplies, with multiple settings, via GPIB (or serial, tcp/ip, etc.) is non-deterministic, meaning that you can rarely predict when the actions will occur. I would also suggest moving the setting of the power supply parameters out of the time critical part, having only the "output ON/OFF" commands, which will save a lot of communications overhead, and is probably a lot quicker function for the supplies to execute. For really critical timing it might be necessary to have some kind of relay between the supplies and whatever they are supplying. Of course this has its own problems, like how much current they are supplying, etc.

---

thank you for your help, i am starter to labview,i have another questions：i am using only NI plug und play driver to controll DC Power,but have it no  like "output ON/OFF" function or VIs,how to complete output ON/OFF" function？

[Dennis_Knutson]

The on/off is part of the Configure Output function with the Enable input. There is going to be some lag between enabling the supplies because the commands will execute sequentially. It might be possible to set them both to software triggering and then issue a GET (Group Execute Trigger) bus command. That would be as close to switching them on at the same time as you are going to get. It has been a while since I've done this. I know the lower level GPIB functions has the TriggerList function for doing this but I'm not sure about VISA.

[JÞB]

---

@Dennis_Knutson wrote:

The on/off is part of the Configure Output function with the Enable input. There is going to be some lag between enabling the supplies because the commands will execute sequentially. It might be possible to set them both to software triggering and then issue a GET (Group Execute Trigger) bus command. That would be as close to switching them on at the same time as you are going to get. It has been a while since I've done this. I know the lower level GPIB functions has the TriggerList function for doing this but I'm not sure about VISA.

---

A hardware switching solution does make more sense here since you seem to desire deterministic timing.  Windows is not going to get you reliable switch timing nor is the devices communications interface (The device needs to receive and process the command before the action is taken.)  Many power supplies also have  hardware trigger inputs and outputs and can be configured to execute actions on these line state transitions.  This gets around all the communications and processing lag of the controller and allows a rapid response.