Shifting an array (Ratcheting?)

[HoldenC]

Greetings,

I am trying shift data in an array (such as Array[0] is shifted into Array[1], Array[1] is shifted into Array[2]......all the way to Array[11].

I found a post by Ron Beaufort explaining how to accomplish this, but I cant seem to make it work. When i run the logic, all of the data turns to zero.


The tag "Press_Downtime" is an UDT, and has an array size of 11 (0-10).
The tag "Press_Downtime_Temp" is the same UDT with an array size of 11 (0-10).
The tag "Press_Downtime_Empty" is a single instance of the UDT that has all zeros in the data.

Would anyone be able to tell me where my mistake lies ?

Thanks.

 

[dmroeder]

What triggers the COPies? If it is unconditional, you're going to shift shift the data through in 10 scans, clearing it out. You should probably use a oneshot to trigger that rung so it only happens one scan.

 

[HoldenC]

I use a tag that identifies when the machine is down to trigger that rung. I was thinking the COP instructions would only execute 1 time.

Adding the ONS fixed it.

Thanks.

 

[TheWaterboy]

Could that be shortened if you first shifted 10 to 11, then 9 to 10 then 8 to 9, and ripple it up the numbers. that would eliminate the 'temp' step.
Nah never mind, you got this right.

 

[HoldenC]

@TheWaterboy

I had seen another example of the way that you explained. That would allow me to skip the temp step. I'm sure ill try that method next time I use this logic.

 

[joseph_e2]

If it's ok to reverse the order of the array (newest data in element 9 instead of 0) you can use a single COP instruction:
Source Press_Downtime[1]
Destination Press_Downtime[0]
Length 11
Then store the new data in Press_Downtime[10]

 

[TheWaterboy]

If it's ok to reverse the order of the array (newest data in element 9 instead of 0) you can use a single COP instruction:
Source Press_Downtime[1]
Destination Press_Downtime[0]
Length 11
Then store the new data in Press_Downtime[10]

I was just thinking about that. if a COP copies one element at a time (i.e. 0 to1 then 1 to 2) then the value at element 0 will just ripple through all of the elements. I don't think COP was intended to use the same element in both operands. It's easy enough to verify. I'll have to try that.

 

[joseph_e2]

With it reversed like that (source "below" destination), it will copy [1] to [0], then [2] to [1], then...[10] to [9].

 

[TheWaterboy]

It makes sense that would behave that way, but you go to agree that's a little less intuitive. You would have to remember to add new values at the bottom element each time purely because of how the instruction functions.

Need a FILO instruction here. Or poke and pop

 

[joseph_e2]

Agreed. It's not the best in every situation but I encountered a time or 2 with certain HMIs that it made more sense.

Edit:
I don't remember the reason as it's been a while, but it seemed like a good one at the time...

 

[TheWaterboy]

I don't remember the reason as it's been a while, but it seemed like a good one at the time...

Reminds me of the code comment:
When I wrote this only Me and God knew how it worked... now only God knows