• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 47 subscribers
  • Views 1669 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

XTR116 Arduino Problem

Thales Maia
Prodigy 80 points
Other Parts Discussed in Thread: XTR116, XTR117

I've build a simple circuit to control a Spirax Sarco valve using the arduino PWM as shown.

2425.transmissor_420.pdf

http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0CBwQFjAA&url=http%3A%2F%2Fwww.cloriuscontrols.com%2Finstructions%2F99.66.02-GB_EPL.pdf&ei=yGA0VKuJFLfIsATFuoHQCQ&usg=AFQjCNHRGjDoem_JXMbUDbQfdirsZ3nu-w&sig2=zGDO4_h5hX_2k1qhm7iC8g&bvm=bv.76943099,d.eXY


The problem is that the current goes very high to 29mA.

I should also mention that I have the negative of my 24V power source grounded.

I really appreciate any help, I really don't know what is happening...

  • 0
    TI__Guru 63965 points

    Hello Thales,

    We have a few comments on your circuit.

    First, since you're not utilzing the "VREF" function of the XTR116 we suggest you take a look at the XTR117 which does not include the VREF output.

    Second, it's recommended to add an external capacitor to the VREG output for best transient load regulation.  Place a 47nF capacitor from VREG to IRET.

    Third, we're not sure that the output of the optocoupler will properly make the PWM signal you're desiring.  For proper operation, the PWM will need to swing between IRET and VREG.  However, the way you have the optocoupler transistor connected, the left side of R4 swings from VREG to a Hi-Z state, so I believe that the average voltage will simply be VREG.

    To test this, can you please use a floating hand-held meter and connect the negative probe to IRET and the positive probe to the node between R4 and Rin?  Please let us know what voltage you measure as the PWM duty-cycle varies between low duty-cycle to high duty-cycle as this will help us understand what's going on.

  • 0 in reply to Collin Wells
    Prodigy 80 points

    First, since you're not utilzing the "VREF" function of the XTR116 we suggest you take a look at the XTR117 which does not include the VREF output.

    Couldn't find the XTR117 in Brazil so I tried with XTR116 that I found on local Farnell Store

    Second, it's recommended to add an external capacitor to the VREG output for best transient load regulation. Place a 47nF capacitor from VREG to IRET.

    Thank you for the advise. Added for the next sch.

    Third, we're not sure that the output of the optocoupler will properly make the PWM signal you're desiring. For proper operation, the PWM will need to swing between IRET and VREG. However, the way you have the optocoupler transistor connected, the left side of R4 swings from VREG to a Hi-Z state, so I believe that the average voltage will simply be VREG.

    I share the same feelings at principle, but I tested and it worked. After filter I got a smooth 0.8 to 4.2V, but, since I am asking for help, I will test again tomorrow and post the values and scope forms.

    One thing that I can not understand is that it work very well with an 470ohm resistor. I can swing from 4 to 20mA and made a smooth calibration, but when I change the resistor, my whole calibration changes. Let's say, change it from 470 to 220 seems to change everything. I have no idea what does that mean but, it shouldn't, unless somehow IREG changes and my PWM output change ass well.

    How do you suggest to build this application?

  • 0 in reply to Thales Maia
    Prodigy 80 points

    I made the test and there follow the results.


    PWM means the value of the dutty cycle in int (from 0 to 254)

    Current means the output value with an resistor

    Multi means the voltage value between Vref and Iref.

    PWM    CURRENT    MULTI
    1    2.7    0.561
    4    4    0.818
    34    12    2.25
    91    20    3.29
    125    22.4    3.595
    180    24.6    3.90
    210    25.4    4.01
    255    26.2    4.13

    I could not post the scope results because I found lot of noise (above 31kHz) signals, and the PWM was not visible...

    For the last test, I took out the PWM wth filter and put an analog power source and it worked very well. The valve opened and closed.

    I added the optical again with the capacitor and it started to work (33nF). Still, I have variations between the load and the PWM value...

    Should I increase or decrease the capacitor?

    Any better suggestions on how to wire this PWM?

  • 0 in reply to Thales Maia
    TI__Guru 63965 points

    Hello,

    Please confirm that the "Multi" values are measured at the node between R4 and RIN referenced to IRET, not between Vref and IREF as stated in your post. 

    It concerns me that you weren't able to make measurements with a scope because of the noise.  What is the frequency of your PWM signal?  The RC filter formed between R4 and C1 will need a time constant based on the frequency of the PWM to properly convert the PWM into a DC level with minimal ripple. 

    I'm also not sure what you meant in your previous post about a 470 Ohm vs. a 220 Ohm resistor.  I do not see these resistors in your schematic.  Is this the value of the load resistor placed between IOUT and the system GND (RTN)?  If so, I agree that changing the load resistor should not change the output.  Increasing the load resistor will eventually cause the circuit to reach a compliance level that will prevent the output from reaching the proper level.  Decreasing the resistor will cause higher power dissipation across the Q1 transistor but should not affect the output current level. 

    Finally, I wonder if a system like this may work better with a push-pull opto-coupler like one shown in this application report from Vishay:

    http://www.vishay.com/docs/83701/83701.pdf

    You'll need to find a model with a quiescent current below 3.5mA but a product like this will drive the output to the HIGH / LOW signal levels which we believe is more appropriate and common for PWM inputs to the XTR116.

  • 0 in reply to Collin Wells
    Prodigy 80 points

    Hi, thank you for your help again.

    You are right, the values were measured between R4/RIN refereed to Iret.

    The PWM frequency is 1kHz, the noise frequency is higher than 30kHz (with a very high amplitude +- 0.5V). The noise increase with the increase of Iin.

    The filter was build for the PWM carrier frequency.


    Yes, RTN. Somehow it change the value Iout for the same PWM duty-cycle. I also noticed that the change of RTN change the noise amplitude value.

    I seems to be high level noise problem. I will try to track it and eliminate it...


    I will post more news soon...