Why can I not set the transformation ratio parameter to a non-integer value in the xPC NAII 76CL1 LVDT emulator card Simulink block

I am running a real-time model of an actuator using xPC Target and emulating the position of an LVDT which is connected to the actuator using an NAII 76CL1 LVDT Emulator card. I am feeding in an external AC excitation signal set to 7Vrms.

It appears that the card is only accepting the integer part of the value I am typing in for the Transformation ratio as I get no change in output if the TR is anywhere between 1.0 and 1.9. However, if I enter 2, the output is twice as large as the output when the TR was set to 1, as you'd expect.

If I set a TR value that is less than 1.0, the outputs are all set to zero and the D2 BIT status seems to report an error (I say "seems to" as I don't really know how to debug the status of the D2 BIT tests as neither the card documentation nor the MatLab help file explain it particularly well). The status that is reported is [3, 0, 0].

If anyone has used this card and understands what is going on here I would very much appreciate your help. Likewise if you have had this problem or not, it would be good to hear about it.

Many Thanks.

Answer 1

[Gordon Weast]

You found a bug. I can have a fix for you once you tell me which version of Matlab you're using.

It is usually easier to work through MathWorks tech support when you have a real bug like this.

[Kev Heaton]

Ah ok. I will be sure to do that next time if I find any more.

The version of MatLab I am using is 2013b and it's xPC 5.5.

When you implement this fix, will I have to create a new boot disk for the target machine with the fix on or should it be incorporated when builiding the model to the target machine?

Many Thanks for your help.

Regards,

Kev

[Kev Heaton]

Hi Gordon,

Do you have any news on the fix for this bug? I am waiting on this to begin development testing of our unit and need to factor in any timescales and look for short term alternatives if it is going to be substantial .

Many Thanks,

Kev

[Gordon Weast]

I've just posted a fix to the bug report database. See: http://www.mathworks.com/support/bugreports/search_results?utf8=%E2%9C%93&search_executed=1&keyword=&release_filter=Exists+in&release=240&selected_products=2679&commit=Search

The bug number is 1112598. That one came up at the top of the list when I just did a search.

This is just a fix to the driver for the board to avoid truncating the ratio. Once you unpack the zip file, you only need to rebuild your model. Please verify that it works for you.

[Kev Heaton]

Hi Gordon,

We have implemented the fix and it works perfectly. Thanks for your help on this. We are now ready to begin testing of our product.

Regards,

Kev