program help

Hi. I am relatively new to PLC programming and I am wondering if somebody can point me in to right direction on this project.
I have four compressors in a geothermal heating and cooling application. I’m using a DL06 c/w a DV1000. I have a 4-20 ma temperature sensor on a storage tank sensing water temp and storing in v-mem location. I have set points cut-in Comp #1 and cut-out Comp #1 and so forth to Cut-in/out #4 stored in v-mem, displayed and changeable on the DV-1000.
My problem: #1 and 2 compressors continue to get long hours while #3 and 4 are hardly getting any.
Can someone advise me on how to use timers and/or counters to ensure all compressors get hours, and if only 1 or 2 compressors are needed that they operate only 12 hours and move to the next one.
Would another option be to use the hours accumulated on compressors to change compressors and get even hour’s on-time?
Any help will be appreciated.

[barnschop]

Hi Roy

Nots ure on the PLC you are using but here is how I would approach the problem.

You need to have a timer to run for 12 hours.

Create 4 shift registers four bits long.

Initially set the four registers such that each as a different bit set.

When the timer elapes have the shift resister rotates round to the right. This makes bit 0 move to bit 3. Hence the last compressor on duty now becomes the least likely to be started.

For each compressor look at the bit position and then use this to determine the pressure at which these cut in.
Remember to include steps to raise the operation pressure of stanby compressors in case of duty compressor failure.

[Terry Woods]

Man... I had to kick-in the "Way-Back-Machine" to find this one.

It's not exactly what you asked for but I think you should find it close!

http://www.plcs.net/dcforum/DCForumID5/1050.html

That post, made by me so long ago, is for a "Bubble Sort" mechanism which tracks "Most-Time", "Least-Time", "Standby" and "Faulted".

I made an error in that post - it is corrected in my following post.

What do you think?

Am using Automation Direct, dlo6 plc.
Thanks for the quick responses,looks a little more complicated than I thought it might be.sorting through the info.

[barnschop]

Hi Roy

Nots ure on the PLC you are using but here is how I would approach the problem.

You need to have a timer to run for 12 hours.

Create 4 shift registers four bits long.

Initially set the four registers such that each as a different bit set.

When the timer elapes have the shift resister rotates round to the right. This makes bit 0 move to bit 3. Hence the last compressor on duty now becomes the least likely to be started.

For each compressor look at the bit position and then use this to determine the pressure at which these cut in.
Remember to include steps to raise the operation pressure of stanby compressors in case of duty compressor failure.

Hi Roy,
I was wondering why you wanted all your compressors to have similar run times. I have come across situations where standbye pumps / compressors have been rotated in this way and have all faulted / worn out / due for servicing together. your existing system with short proving runs at regular intervals on #3 & #4 sounds ok to me.

What do you / the group think?

ian.

be kind to me I'm only an electrician.

[barnschop]

Hi Ian

I would subscribe to your theory on the failures happening at around the same time, however:-

Modern machine reliability techniques show that any device that is used as a standby must be available to take over from the duty unit. The only way to prove this is to run the unit on the full duty. If we waited until the duty unit broke down it is possible that the standby could also be faulty due to lack of use or may have become faulty and simply overlooked.

From the original question the compressors would be proven in full operation every two days. Should the duty unit fail the standby is
less likely to be in a failed state.

Now as to whether you need the four compressors in the first plave is another story.

[Tom Jenkins]

You have three valid ways to change lead/lag, all of which could be implemented using hte basic logic in Terry's reference.

1) Operator sets lead and lag equipment - my choice, but it does require some confidence in operator ability.

2) First on First Off alternation. If you frequently have a second compressor starting, and subsequently only need one, then you can have the second compressor to start become the lead, and the first one running becomes the lag and shuts off when only one is needed.

3) Time based alternation. Use the real time clock to trip a once a day or once a week alternation.

The reason I prefer the first technique is I do trust operators, and if you frequently have equipment out of service for maintenance or failure the automatic alternation can get ugly.