Compactlogix analog out to VFD

[toddp65]

Scenario:
Scaling speed reference to 0-10 vdc speed reference on Mitsubishi D700 VFD
Controller is a 1769 L32
AO Module is a 1769 OF4 configured for 0-10 and raw units.

I've been trying to get this drive controlled by my analog output. I have some frequency control but its wrong.

Scaling number on my SCP AOI
Min in 0
Max in 90 (drive is setup for 90 maximum frequency)

Min out 0
Max Out 32700

When enabled, the 90 HZ brings my drive to 100% but it doesn't leave 100% (or 90 hz) until about 75 Hertz pushed to it proportional to the scaling raw value.
When 0 is pushed, I'm still getting 5 volts from my analog card despite the value being 0 on my push to the assigned analog output channel. (Local:4:O.Ch2Data).

Am I scaling this incorrectly? I am used to pushing values to speed reverence tag using ethernet IP and it's been a while since I used an actual AO module that wasn't using a PID instruction.

TIA

 

[toddp65]

So I changed the parameters in the SCP AOI and reconfigured that channel to engineering units 0-10500 and the same on the min in and max in.

Now the drive is corresponding the entire range but not identical. Like 50 input is 56hz and so on. It's not perfect but in reality they're just looking at speed and the actual number isn't too important. Thoughts?

 

[A_G]

Min out 0
Max Out 32700

i'm not sure this is correct, I always thought it was 32767

 

[A_G]

So I changed the parameters in the SCP AOI and reconfigured that channel to engineering units 0-10500 and the same on the min in and max in.

Now the drive is corresponding the entire range but not identical. Like 50 input is 56hz and so on. It's not perfect but in reality they're just looking at speed and the actual number isn't too important. Thoughts?

I'm not sure where you got the 10500 from but it sounds like the numbers you are getting are close

I would try to tweak the scaling a little and see what happens, 50 input to 56 actual is somewhat close, but see if adjusting the scaling a little can get it closer. For example try 10550 or 10450 instead of 10500 and see what happens. Again I'm not sure exactly how you came up with that number but it seems to be working somewhat, so try changing it a little and see what happens

 

[toddp65]

The 10500 was out of the modules pdf for engineering units for 0-10. It did get me the ball park-- it's more like 0-925 and it's within a hertz.

Thx.

 

[GaryS]

First you need to configure the module correctly
I would recommend select
Output Range select 0v-10 v
Data Format select Percent Range
O would be 0hz and 100 would be 90hz
Provided that your drive is setup correctly

Then in you program to you scale you date to the same scaling
0 % = 0hz and 100% = 90hz
Simple and clean setup and the resolution in more than adequate for a vfd application

Note here: most vfd’s don’t do well below about 10Hz so I would recommend setting the min in the vfd as 10hz (0v in = 10Hz ) and you should never have a speed reference of 0HZ or even near 0Hz while in RUN mode sometime the operator will think there is something wrong and poke around then the reference increases and the motor takes odd inexpertly

 

[toddp65]

Gary,

I tried your percentage setup but the analog card won't put out any voltage despite the fact the scaling does proportionately to the input. When I used the engineering unit it works but I have to play with the max out number to correlate with the drive hertz. Not sure why it's not sending any voltage out in the percentage configuration.

 

[GaryS]

It looks like you may have a bad output module because it doesn’t seam to be following the module scaling. Without the ability to see it for myself it hard to tell.
Having said that I assume that you have selected 0v to 10v module scaling
You said that 0 out to module will give you 5 v out to the drive
In raw data mode the scaling is -32767 to 32767 so with 0 to the module 5v would be the correct voltage out.
The scaling would be
-32767 to the module would be 0v out
0 to the module would be 5v out
32767 to the module would be 10v out
If you change your scaling in the program you should take care of it

 

[toddp65]

For raw data I was scaling 0-32767, not -32767 to 32767 so that may have been why it would still output 5v at 0. Using the engineer units and manipulating the output max I was about to get it within a hertz. I don't understand why the percentage configuration wouldn't work, though.