Period check module - to make a Tacho from pulse input

[Glenn_United_Pumps]

Hi everyone,

I am trying to use the period check module in dasylab V9.00.00 to work as a tacho for testing motor speed on machinery. The input to dasylab is a pulse (one/rev) acquired from a light sensor (light bounces off reflective tape on shaft of motor and the sensors sends a pulse to the PC) I will need to be able to check frequencies from a couple of hundred Hz to around 2KHz. I have it working nicely if I use a sine wave from a generator, but when it is connected to a pulse input I can only set it to work at one frequency (adjusting scaling etc). I am wondering if there is an easy way around this or not?

Ideally I require this to have virtually no error (<0.1%) between approx 200Hz and 2KHz outside of that doesnt really matter.

I have attached the layout I am using. With this layout it will work at 2KHz but any change from that and it isnt happy.

I am using DasyLab V9.00.00 with Windows XP and a NI-PCI-4472 DAQ card.

Any ideas are appreciated, Thanx to you all in advance for your time and effort

[NatasaLingurovska]

Hi Glenn_United_Pumps,

The error that you are getting is a time-out error. The problem comes from inability to send the triggers between the DMM and the switch module for sending and receiving data. In order for the triggers to be read, the modules either have to be plugged in the same chassis segment (share the same bus) or they have to have specifically defined trigger lines in MAX across different chassis segment. Currently, the DMM (being in slot 12) is plugged into chassis segment 2, where as the switch module is in segment 3. Therefore, the trigger are not being send/received between the DMM and the switch. Just move the DMM to any of the slots 13-18 and the example should work. Each of the segments are labeled with a small line on the face of the chassis. If you don't wish to move the card, then you have to open MAX and configure the trigger lines independently.

This example is good implementation for your requirements. You might also want to look at using the NI DMM/Switch Express vi.

I hope this helps. Please let me know if you have any additional questions or concerns.

[Glenn_United_Pumps]

Thanks Natasa for your quick response.

Me being a relatively new person to DasyLab, maybe I didnt explain my prediciment properly first time round but I understand what you were saying in your last reply which will no doubt be handy to know for later

The problem i am having is completely inside dasylab (no external inputs) The issue at hand is that for up to approx 1000Hz I am getting accurate readings (within 0.1% of true signal), using the generator within dasylab to produce pulses for me, The errors I am obtaining are inaccurate readings ramping upwards from approx 1KHz onwards where there is around 0.2% error to approx 10% error at 2KHz (at 1KHz input it is reading: 1002 and at 2KHz it is reading 2205) . below 1000Hz the readings are fine, within tolerances. Most of the readings taken will be in the high 1000's and some in the 2000's. 

At the moment I havent connected any external (to dasylab) signals yet, in my mind, if I cant obtain accurate results in "a perfect world" with an internal generator then I probably dont stand a chance when using "real" signals from external sources. I am thinking of using the scaling module but I am not sure if I can use that conditionally, ie: set it so that for a given frequency range it would scale by a certain amount then for another range it would be different scaling etcetera, ie:1000-1200Hz scale by 99.5% then 1200-1500 scale by 94% or something of the sort. I dont know if this can be done in dasylab or not, then again maybe I am thinking the wrong way about it...

Thyanks again to everyone for your time and effort it really is appreciated

- Glenn

[Daniel_L]

Hi Glenn,

Natasa is actually out of town on business for the week, but I did want to follow up with you on your discussion forum post.  It sounds like your situation is that you have a piece of NI hardware (which currently is not hooked up), and some slightly problematic results with DASYlab. 

National Instruments actually does not directly support DASYLab itself, but I can redirect you to those who do support this product. 

-North American Customers can contact MCC Technical Support group at 508-946-5100, or techsupport@mccdaq.com

-Customers in Europe and Asia can contact MeasX

Additional DASYLab support

• Discussion Forums: DASYLab discussion forums at ni.com
• External Site: DASYLab discussion forums at dasylab.com
• External Site: Support contact form at dasylab.com You will need to log in to dasylab.com to use this support feature.

Regards,

Daniel L.

Applications Engineer

National Instruments

[dlabdev]

Hi Glenn,

what about simply counting the pulses? Just have a look at the worksheet below. The counter module is set to "One value per block". That is the right setting if you just want to display this value. If you need to do further computations with it you should set it to "One value per sample". So the output block would have the same length as the input block. Hope it helps.

Regards

[Glenn_United_Pumps]

thanks dlabdev for you "simple" approach to my problem.

It works a treat and as i thought there must have been an easier way than what I was trying but I just didnt know how to do it. "I knew I was barking up the wrong tee!!"

thanks again your assistance is really appreciated

Regards

Glenn