Usb 6009 5V output

[n00bzor]

Hey

I'm trying to use a Ni USB6009 with a optocoupler (4N25). Problem is i think the normal dig outputs of the board do not provide enough power to the led inside the optocoupler. I'm thinking of using the 5v 200ma output (that seems to work) but i'm not really sure on how that should be done. Any help would be greatly apreciated. 

[Anders Björk]

You need to put an additional relay there, I do not think the module who give enough current.

[PaulG.]

I have worked with a 6008 and I don't understand why you cannot drive a simple octocoupler with it. You are just driving an LED. The digital outputs should be sufficient to do that. But then again, my experience with (the really cheap) NI hardware is that they are pretty wimpy and you may need to use the digital output to drive a FET that will in turn drive the coupler.

[n00bzor]

QUOTE (PaulG. @ May 22 2009, 04:01 AM)

I have worked with a 6008 and I don't understand why you cannot drive a simple octocoupler with it. You are just driving an LED. The digital outputs should be sufficient to do that. But then again, my experience with (the really cheap) NI hardware is that they are pretty wimpy and you may need to use the digital output to drive a FET that will in turn drive the coupler.

Well, looking through the datasheet for the 4n25 i observed that all examples are given with a 10mA power source connected to the optocoupler's LED, that given, i'm thinking over 10mA is too much and under is too little. Any ideas on how much a dig output for the 6009 pulls out? From its datasheet i'm readling very small values(0.8uA)..

[ShaunR]

According to the specs. The U6009 is digital rated at 5mA and according to the current transfer characteristic of the 4N25, it should work down to about 0.5 mA (although you wouldn't get much out of it). It may be that the digital output is detecting the LED as a short (don't know if it has a crowbar or not...but might). Try putting a 1K5 in the digital line to fix the current to about 3.3 mA and see if that works.

[n00bzor]

QUOTE (ShaunR @ May 24 2009, 02:51 AM)

According to the specs. The U6009 is digital rated at 5mA and according to the current transfer characteristic of the 4N25, it should work down to about 0.5 mA (although you wouldn't get much out of it). It may be that the digital output is detecting the LED as a short (don't know if it has a crowbar or not...but might). Try putting a 1K5 in the digital line to fix the current to about 3.3 mA and see if that works.

I used a 0.12k and that didn't seem to do anything. I'm going to try it again for a larger value.

Also, what do you mean by "crowbar"?

[ShaunR]

QUOTE (n00bzor @ May 24 2009, 08:40 AM)

I used a 0.12k and that didn't seem to do anything. I'm going to try it again for a larger value.

Also, what do you mean by "crowbar"?

0.12 is still far too low. It equates to about 41mA (V=IR). Try a 1K5. Then measure the voltage across the digital output to ground. If its at 5V (or very close) then that isn't the problem.

Many output sources (digital/analogue outputs, power supplies etc) have a system of protecting the circuitry from short circuits (a fuse for example). The term "crowbar" in protection circuits is analogous to "being hit over the head with a crowbar" i.e instantly halting the cause of the problem to protect the curcuitry.

[n00bzor]

QUOTE (ShaunR @ May 24 2009, 01:19 PM)

0.12 is still far too low. It equates to about 41mA (V=IR). Try a 1K5. Then measure the voltage across the digital output to ground. If its at 5V (or very close) then that isn't the problem.

Many output sources (digital/analogue outputs, power supplies etc) have a system of protecting the circuitry from short circuits (a fuse for example). The term "crowbar" in protection circuits is analogous to "being hit over the head with a crowbar" i.e instantly halting the cause of the problem to protect the curcuitry.

Hey, just got it running, can't really tell what the problem was.At any rate, thank you for your response. Still laughing at the explanation for "crowbar", good one.

[ShaunR]

QUOTE (n00bzor @ May 24 2009, 11:57 AM)

Hey, just got it running, can't really tell what the problem was.

Obviously "operator error" then