The analysis of a drawing

[Tom749 0]

Hi, I have some questions about a P&ID drawing such as the attachments. The questions are as follows. Attachment (1) : What is the action process in the drawing? (For example the relationship between TS-51 and Motor, the relationship of interlocks...) Attachment (2) : How can I make a logic and ladder program using the drawing? or using the control logic? (For example this is for the control of some valves, How can I control them by some control logic?) Thanks for your answers. _1__P_ID_analysis.bmp _2__The_way_of_movement_01.bmp _2__The_way_of_movement_02.bmp Edited 11 Oct 2009 by Tom749

[PLCMentor.com 60]

Whoa! I actually had to power up my laptop at home when I saw this come across. I need to start with your last question first. The P&ID is one of several documents necessary to put together a control system. It gives all involved an overview of the process, but does not define how the process should run. Some people will try, but it is simply impossible. You need to generate or someone needs to generate a functional specification that defines how the process should run, interlocks, automatic and manual system actions, etc. Just as the P&ID lets all involved understand an overview of the process, the functional spec lets everyone understand the details of how the system will run even for those that do not understand programming. Ask for the FS or the ability to generate one. Now let me look at your first questions. The way that ts51 is shown is confusing to me. I would guess it is some type of temperature switch. If it is the temp switch on the motor then I would expect it to be in closer proximity to the motor on the drawing. I really dont understand the S/D(F). somehow that is interlocked with the TS and they are both interlocked with the function of the motor. You will have to have more discussions on this operation with whoever is defining the system functionality. Once again make sure it is written down (FS) when you understand it so it can be passed around for agreement. For attachment 2 I couldnt tell you. I could guess, but that is all anyone can do with this information. Once again you need to sit down and discuss the operation of these control loops. It looks like you have a modulating valve controlling temperature and maybe some mass control? I dont have my chart with me and I cant ever remember using an M for the first letter on a P&ID. Either way you cant really determine the operation for this system without a discussion with someone who knows. Of course then write it down. Ok the third diagram adds a little more info. It looks like your control loops are split control. one set operates off of temperature setpoint. Under temp setpoint your heating valve is controlled. above setpoint your cooling valve is controlled. On the other you are monitoring humidity and your cooling valve is controlled above the setpoint - your humidity? valve is controlled below setpoint. Once again sit down and talk with whoever understands the system operation. All kinds of other stuff is going to come out when you start examining the actual control. OK, I just looked up the M variable and it is user defined. I would assume from the other information that this is being used for humidity. Just too many assumptions having to be made here. Edited 11 Oct 2009 by PLCMentor.com

[paulengr 44]

It is very common to see Mxxxx or MSxxx for "motor" or "motor starter". There's supposedly a standard for P&ID diagrams but I've never seen any of them following the drawing to the letter. Second, although I agree with you about the fact that a P&ID isn't anywhere close to a functional specification, it almost is for certain cases. For instance, if you have a burner management system, the combustion codes define almost everything already. One is left with a small number of choices to make about a specific implementation and that's about it. The sequence of operations is defined by the code. It is very common in process industries to run into a P&ID with no other documentation supporting it. This is what you typically get from chemical engineers that look at things as a continuous system and aren't concerned with interocks and sequences of operation. You get something similar (even worse) from discrete manufacturing where often they have a sequence of steps involved in accomplishing a task and although it is intended to be an assembly line, they do not consider alternate paths and interactions among process steps, or pipelining (separating process steps so that multiple steps can be running concurrently). I'm very fortunate in that I spent roughly half my career so far as a process engineer and half as an electrical/controls engineer. I'm "bilingual". Trying to demand other specification documents often leads to a very unhappy relationship. You will usually find it will be much easier to draft your own functional specification and submit it for review..."OK, I will implement it like this. Is that what you want?" If you don't do this, be prepared to have a lot of screaming matches in the future.

[PLCMentor.com 60]

OK I can agree with the M as a motor or motor starter designation. I dont think I have seen it in a bubble like that tho. No no, dont get me wrong. There are probably more processes out there without documentation than with. The poor guys left holding the bag sometimes dont have a clue how the system runs. Becomes like a secret club with the controls guy and the process guy that designed it having it all in their heads (if the process guy actually knows all the small details). I was trying to describe how a system should be put together. It's too bad it doesnt happen more often. So controls is your second language... lol I've seen your posts and I'd say you must speak both fluently. You are right. Generally the best way is to offer up the func spec and work from there.

[Bill Wheeler 0]

I'm a new member and have been in the same boat. I recommend that you brush up a bit by reading through ISA standard 5.1 through 5.4. These documents can be found without too much problem. In any case a P&ID can never by itself give much detail as to the control logic. It can only define the ins and outs of a process. What I have found that works for me is to document the logic in the simplest form, a flow chart. I then create a document link name to the specific logic segment flow chart (i.e. L3.1 - Logic diagram 3, segment 1) and include that name in a diamond symbol on the P&ID. ISA 5.3 (I think) does a good job in defining binary control logic diagrams, but fails to provide a method for documenting more complex integer variable process controls where as a flow chart can be used very easy. Once in flow chart form, the logic can be created very easilyis very easy Standards are a good way to start, but todays controllers far exceed those available at the times the standards were written.