Device Numbering Methods: A Poll


Recently I was involved in a discussion on the best method of numbering and labeling devices for systems and machinery. I have used several different methods in the past, depending on customer preference and the type of application; these methods were usually already chosen and were appropriate based on the job. In this case however, a method had not been selected yet, giving us lots of fuel for discussion.

After going through this analysis exercise, I have broken down device designation into three broad categories:

1. Labeling the device based on the I/O point it is connected to.

2. Labeling the device based on where it appears in the electrical schematics (i.e. line/page number)

3. Labeling the device based on mechanical or physical layout (i.e. P&ID or similar)

All of these methods have several things in common. They should show a technician where to connect the cable or wire on the terminal end and the device end; the cable label should have the same number as the terminal block label and a device label. They should allow the technician to look up the details of the device in the schematics or BOM. They should be unique. Along with the device description, the method should allow the technician to find the device physically.

The first method includes labels such as PX00205, CBL00314, SV00611A. The type of device can then be interpreted as PX for proximity switch, PE for photoeye, SV…A for solenoid valve and so on. The number then relates to the node, slot and point; i.e. PX030205 is connected to node 3 (distributed I/O), slot or card 2, input 5. Sometimes designers will even go so far as to ensure that the node or slot number lines up with the page number in the schematics.

The second method looks similar, but a number such as 0232703PX might indicate that the device is shown on page 23 of the schematics, line 27 and is connected as the third wire number on that line. Both this and the first method are often used for machinery. This is discussed further here.

The third method uses a designation based on the physical layout of the system. P&ID diagrams provide designations for physical groups of equipment, often a four digit number that can include a subsystem. For instance, device ZSL8705 might indicate a low-level or retracted sensor in group 8700, which may also include 8701, 8702, etc. This method is often used in process control systems and a P&ID diagram is used to provide details on the function and location of the device. I have only occasionally seen this method used in machinery outside of the process world.

Another method I have seen used that is related to the third method is designating the station number and device number in the label. An example would be PX01005, which indicates proximity switch number 5 in station 10. I am not a fan of this method since the switch number is arbitrary (simple numerical order) as may be the station number. Rather than being related to a standard document like a process and instrumentation (P&ID) diagram, the method depends on the vendor to provide detailed mechanical layout drawing showing station numbers and sensor locations. As for which method was used after the discussions I mentioned at the beginning of this post? You guessed it, it was this one. Despite my best efforts, I was unable to convince the end-user that this was not ideal.

I have also seen OEMs simply number terminal blocks in numerical order (1,2,3…x) and label the cable by whichever terminal the wire landed on. I don’t consider this a method at all, since it really provides no benefit. It indicates inexperience on the designer’s end to me.

I would like to poll readers as to which of the three methods above they use and why. I often share posts on Linked In and people choose to place their comments there; I will try to reflect those comments on this site also, but if you will comment here it will make things a little easier. Again, listed below for your convenience:

1. Labeling the device based on the I/O point it is connected to.

2. Labeling the device based on where it appears in the electrical schematics (i.e. line/page number)

3. Labeling the device based on mechanical or physical layout (i.e. P&ID or similar)

I hope to be able to share the results of this poll as they come in over the next few weeks. Please comment!

** Results from this poll have been tallied here…


Electrical Engineer and business owner from the Nashville, Tennessee area. I also play music, Chess and Go.

15 Comments on “Device Numbering Methods: A Poll

  1. #2 labeling to page and line number seems to be the preferred method that most customers want and the method I prefer unless a customer specifies different. It allows easiest way to locate devices in the schematics for maintenance troubleshooting.

  2. Any labeling is better than no labels!
    System 3 is the most useful since it can be informative with or without associated documentation. (Often documentation goes missing)
    The worst offenders IMHO are the IT networking dept. They often do not put any labeling on their cables/patch leads. With Vlans etc. this becomes as important as labeling IO points in a PLC system.
    I prefer your system of adhesive labels as pictured on your site.

  3. Hi Frank,

    I too have tried all of the above methods. My overall preference is for Option 2 where devices are numbered according to their primary location in the electrical schematics. Eg 200U1 would be power supply 1 on sheet 200. We tend not to use a strict line number or zone number reference for the device because the whole system collapses if the drawing has to change at a late stage. As long as you point the electrician to the sheet then he or she will find the device. The letter code we use comes from the Aus standard symbology for electro technology, based on IEC. This system also works for wire numbers eg 200/1 is wire 1 on sheet 200.

    When we get to the PLC and SCADA domain we tend to switch to the PID or process terminology if the project is process related. This ties the functional spec and the code together, the drawing index links the process tag eg MV0801 to a drawing sheet eg sheet 300.

    Not perfect but it is a system that can be used across any machine or process control schematics.

    Ross Leamon

  4. Our machines are pretty small, so physical location isn’t as important. In the past, we’ve used #2, but after a decade of changes, it’s become rather jumbled up.

    I’m re-doing it now because we’re doing a major update (different control architecture, more custom PCBs). I’m working with our techs to move to a functional system, reserving number blocks for related functions (so break out boards 1->9 (WD1…WD9) are reserved for servo motor boards, 10->20 for PLC boards, etc), instead of sequential order or strict page numbering.

    But my case is probably a little different than normal: we primarily do customizable base machines, and the machines can have various options — and different types of base machines share a lot of components. So I don’t think strict page-based numbering works well for us, because I’d like a lot of common numbering between base machine types and more flexibility to update the schematic layout in the future.

    However, of the 3, I’m most likely to use #2.

  5. I am a strong supporter of #2 using page/line number. I have worked as an end user and now design control systems for custom machinery. Anything that helps the people on the floor where to find the correct information to fix the machine is extremely beneficial.
    I have seen #1 used but what to do with wires that do not connect to I/O points such as power distribution? It also does not lead the technician where to look in the prints. This can be a hassle on larger machines with hundreds of pages.
    I have not used #3 but have not done much in the way of process machinery. When we have used a PID we have added the device names on the wiring diagrams and the wiring diagram location for the device on the PID. This allows a quasi cross reference between the two documents. Unfortunately, it has always been a manual system that requires significant effort to keep correct as the design changes.
    As long as whatever system is used is clear and consistent it comes down to what the designer is comfortable with.

  6. We use a combination of #1 & #2 – devices (such as sensors, motors, etc) are labeled sequentially by device type (PE001 for photoeye, for example) but we use sheet/line number for terminal block and wire numbering. The cable labels for each device (usually on the quick disconnect cable just before the device) generally include the device number, the I/O point, and a short description of the device.

  7. Method #2 would be the most beneficial to a technician on a plant floor. Having a device tag, wire numbers and even a PLC tag using the same naming convention makes trouble shooting much easier. The important thing to remember is that the technician is the majority of the end users and this helps in reducing downtime($$$).

  8. I’m in favour of…

    Option 1 for ICA Sections
    Option 2 for Relay Logic and Complex Systems

    For me they are two separate beasts as if a PLC exists, a lot of the relay logic will be moved into software. Where as for smaller systems with no Automation, Option 2 makes perfect sense as with a 15-20 sheet schematic you can find the root of any wire or the location of any relay, timer etc.

    So when I design systems with any relay complexity, I always follow Option 2, and I find that most system manufacturers who produce to BS60204 also use this method.

    Relays will be “4K2” (Sheet 4 Column 2), 6K2 (Sheet 4 Column 2) for example.

    Good poll and interesting topic!

  9. In my experience and my preference, I like the page/line number/device number method. It lets the field techs and any other engineer easily find the device in the schematic. The I/O point is then easily found on the page as well. As an example CR42151 would be a control relay (or contact) located on page 42, line 15, the first device.

  10. Additionally, if there are multiple panels in a large system, the panel designation can be added to the front of the number. ie; MC-CR42151. This would indicate Main (control) Cabinet-Control Relay, page 42, line 15, device 1

  11. For the most part, I use the #2 method. My systems are typically less than 200 I/O, so I can keep numbers to mostly 3 digits. It takes time to print wire numbers. The smaller the better.

    I like to organize my schematics with the first pages being AC, then misc DC, discrete inputs, discrete outputs, analog I/O, and misc. The numbers are in dedicated ranges. For instance, all discrete inputs are in the range 400 to 599. DO’s are 600 to 799, etc. I can troubleshoot a 10 year old system without constantly needing to refer to the schematics.

  12. I would not use #1 since it fixes your system onto one particular PLC manufacturer. If you wanted to change PLC manufacturer for another based on geographic locations for example, you would have to retag all your devices. In addition if you use the Logix platform, the address is way too long and you can’t really use the tag.

    I would also not use the page number in the drawings. This makes changes and Maintenance very cumbersome if you need to start adding equipment and inserting pages. Then you are managing pages, you have not idea as to the relation of that component to the function that it is monitoring or performing on the machine. In addition you might know where it is on the page, but not where it is on the machine or which cabinet…

    I would recommend the IEC 81346-1&2 using a functional designation to identify the function of the device within a system (=Conveyor1) then a location to identify where is my component (+Cabinet1) and finally a product to know exactly what it is such as (-BT1) as a Temperature sensor.
    This will allow to automate your designs in the future as well.
    The mechanical engineers think in functions and using advanced software to quickly put systems together. Why can’t the electrical engineers do the same?

    Sean Mulherrin

  13. I posted on LinkedIn before reading the full article and discussion here.

    I’ll include the original comment but basically I generally favor a form of method #2 as I see most of the commentators have favored. The difference being that I don’t use page numbers. Instead I use column numbers leaving breaks in the column numbers to add more devices in the future. I prefer to generate just one drawing number for all of the electrical schematics generally with many pages (or sheets). I like to see all the incoming power shown on the first sheet after the title page and drawing index (if used). This is followed by all the 480VAC motors and other power loads, then the 120VAC loads, then the 24VDC loads and so on. The PLC I/O is another section or set of columns except that I intermix the inputs and outputs and arrange the I/O by the rack number and then by the slot order of the physical hardware.

    All this goes beyond the discussion about device numbering except that I use the column and line to generate the device number. In this way if the sheet numbers get changed it has no impact on the device and wire numbers. Typically the device number ends up being a 5 digit number but on larger jobs 6 digit numbers are used.

    Original LinkedIn Post Follows:

    For small jobs it would be simple to just start with the number “1” and increment by “1” with each succeeding device of the same type. If the simple system is to be included in a larger facility it can be beneficial to ask which device number to start with especially if the wiring might end up intermingled with other systems.

    For typical jobs tying the device number to the line and/or column of a drawing has been been my default method. The only downside to this method is when you might want to rearrange the drawings when a project upgrade is performed. Typically this means that that added equipment gets added to the end of the drawings or placed in the spaces provided for future expansion if the original drawings had this capability built in.

    Often it makes sense to organize the drawings so that similar pieces of equipment can be quickly replicated on the drawings. This means that a drawing (and device) scheme that allows for quick copy and paste operations or the use of template parts with a single numerical offset to create a new part or set of parts should be considered.

    Keeping the device numbering unique when installed into an existing facility should also be given consideration.

    As was already noted, the customer often specifies the drawing and device numbering scheme. In this case, ask about the importance of keeping the device IDs unique.

  14. Other Items not discussed – Source / Destination Numbering – In destination Numbering the connection point of the other end is identified at the local end. It does help a field tech understand where a wire goes to but if they remove wires from a terminal strip they have no reference to place them back.

    I have seen other methods that included building/floor/area designations.

    I have also seen where the Wire number remains consistent from source to destination through all Junction Boxes. I liked where a letter suffix was added as it went through each Junction Box (-a was at source -b after first field J-box, etc)

    Method chosen really depends on application – Machine or equipment based or plant wide process based.

    Power Distribution wiring & Cable labeling and IO based wiring have different challenges. (IO based wire labels do not work for Power Distribution)

    It will be a compromise of systems – Often IO internal wiring between the IO card and terminal strip is done on a rack/slot/point -rrsspp basis (this is generic for all plc vendors). From this terminal strip to the field would be based on the P&ID Tag with a suffix for each FJB, no Suffix = Home run.
    P&ID tags are generally based on Process Equipment destination plus IEC Designator – tank 100 pressure transmitter PT100 if multiple elements on a vessel PT101, PT102. The plc tag should follow the same format .

    Referencing Schematic Drawings (Page and line number) is good for the first pass but gets cumbersome with additions and modifications. Adding a safety interlock during a PHA could require a lot of rework. I would not recommend. This evolved during relay days and is the default for autocad electric.