LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

NI 9485 buffered digital output

Has anyone tried to do buffered digital output with the NI 9485 (solid state relay)?  According to every bit of documentation I could find it appears that it should be able to. I have 2 of these plugged into the cDAQ chassis 9172 and I simply want to see a digital waveform reproduced. For example, say I want a single line to output 100 pts at 100 Hz sample rate with the first 50 pts as 1's and the last 50 as 0's. When I do this, I get 0.  If I put the first 50 pts as 0's and the last 50 as 1's, I get 1. If I put the first 99 pts as 1 and the last pt as 0, I get 0. If I put the first 99 pts as 0 and the last pt as 1, I get 1. Can you see the pattern? It appears that only the last value in the array gets generated. I am measuring the voltage across a resistor after the relay on a tektronix scope. The code is as simple as the shipping example Write Dig Chan-Ext Clk.vi. I'm not entirely sure if I'm using the right source for the clock (I'm using the /cDAQ1/100kHzTimebase or else /cDAQ1/20MHzTimebase) and I don't know how to probe the actual sample clock. These two clock sources are the only ones that don't give me errors. This clocking could be the source of my problems because I'm specifying a sample rate of 100 Hz and I don't know how or where the 100kHzTimebase gets divided down to give me the sample rate I want. Any comments or suggestions would be greatly appreciated.
 
DC
0 Kudos
Message 1 of 5
(3,671 Views)

Another thing, anyone know why my user forum icon would stop working? The URL seems to be fine.

https://mywebspace.wisc.edu/awcaswell/web/NIicon/RainbowSpectrum.jpg 

0 Kudos
Message 2 of 5
(3,670 Views)

Another thing, anyone know why my user forum icon would stop working? The URL seems to be fine.

https://mywebspace.wisc.edu/awcaswell/web/NIicon/RainbowSpectrum.jpg 

0 Kudos
Message 3 of 5
(3,670 Views)
Hello caz,

I think what you are seeing is a result of several things:

1)  The switching rate for the NI 9485 is slower than the 100 Hz you are going for (the spec sheet says 1 operation per second, however this additional specifications page gives graphs showing the actual rates).  This would explain why you are only seeing the last point (especially if you are providing less than or equal to one seconds worth of data)

2)  The rate you input to the timing VI is the maximum rate you are expecting from your external clock source.  So specifying at rate of 100Hz and providing a 20 MHz source does not do any divide down, but the values are used to configure the buffer sizes.

I believe the way to accomplish what you are trying to do is demonstrated with the Multi-Function-Synch Dig Read Write With Counter.vi example.  This is in the Example Finder under Hardware Input and Output»DAQmx»Synchronization»Multi-Function.  What this does is use a counter on the backplane of the cDAQ chassis to provide the sample clock (i.e. does the divide down). You'll have to get rid of the digital input part, but thats not too hard...

I do not have the hardware in front of me, but I simulated it and it ran fine.  You just need to make sure your module is in slots 5 or 6 to access the counters (or I believe you can use just the counter from the backplane in slots 1-4, e.g. cDAQ1/_ctr0).

I hope this points you in the right direction.  Please post back if you have any questions about my explination.

P.S.-  I don't know why your image will not work.  I tried it too (just to see if it would work for me) and it didn't...
Neal M.
Applications Engineering       National Instruments        www.ni.com/support
Message 4 of 5
(3,639 Views)

Thanks Neal for your insightful response. I have also gotten some feedback from another thread I had started.

http://forums.ni.com/ni/board/message?board.id=70&message.id=7997#M7997

I think you hit it dead on with the clocking. I had been very concerned with just specifying the 20MHz or 100 kHz clock sources and hoping to get some arbitrary specified rate. I will implement the counter clocking scheme you suggested as soon as I finish a software timed version of what I'm trying to accomplish as stated in the above link.

caz

0 Kudos
Message 5 of 5
(3,633 Views)