Skip to Content

Register Login

Industry Online Support

Technical Forum

Navigation

Regarding to Mres Function

Created by: TEBANCH at: 3/26/2011 3:19 PM (8 Replies)

Rating (3)

Thanks 10

9 Entries

3/26/2011 3:19 PM	Rate (0)
TEBANCH Silver Member Joined: 3/25/2008 Last visit: 1/28/2022 Posts: 232 Rating: (20)	Hello dear colleagues. I have a problem related to understand in full way the Mres function in a CPU; let me explain you what is this about: I have 3 S7-319 CPU's each of them with a CP341-1 Advanced. Each pair CPU-CP belongs to 3 diferent assembly lines, the CP'sare used to entablish a connection to a main "Server" by means of the FTP protocol, we send and receive .TXT files to/from the Server; i have writed a litle secuence using FB40 to entablish conecction to the server and to write/delete or read something on it. The normal sequence is: 1. The PLC excecutes FB40 in mode "Connect" (the server logs this activity and register it, every time that the server register an activity from the PLC a "supervision timer" is reset to 0; if the PLC is connected to the server and do not execute any action in 120 seconds, then the server kicksout the PLC and close the entablished connection). 2. Once the PLC is conected to server FB 40 is exceuted in mode "read" or "retrieve". 3. After this, if PLC found a new file, it copies to a DB and then FB40 is excecuted in "Delete" mode and the file is deleted from server. 4. Finally once the file has been deleted, the PLC executes FB40 in "Disconnect" or "Quit"mode. Some how the sequence gets stocked some times , and then the server close the connection with the PLC; this happens to the 3 lines and should no be a big deal, and actually it is ok in 2 of the 3 PLC's, but in the other (lets call it "PLC_X")when this happens, i'm not able to reach the server anymore . When PLC_X is disconnected by this time out event, is not possible to reach the server from this CPU, i have made some tests to it and what i found is that under this conditions when FB40 is executed his status word is W#16#8181 (from the online help this status means "No resources on CP"). So,how iget thePLC_X working again? well the first time i saw this fault i get in panic and i just download all program and HW config. again, but it did not work, soi decide to clear all the PLC, i delete every thing, including MMC; and after that i just downloaded the program again and the link start working again , but a few days later the fault was present again. This time i just did the following (method_1): 1. Stop PLC. 2. Set the selector switch on CPU to MRES pos; and hold it by 3 seconds, the Stop Led Flashes 2 times slow. 3. Get the switch to Stop Position. 4. Set the selector switch on CPU to MRES pos again, and hold it until the Stop led flashes Fast. 5. Hold it in that position until stop led stops flashing, and then set it to stop Position. 6. Select Run mode position. And thats all, the aplication was running again; but after a few days the fault was there again; this time what i did was (Method_2): 1. Stop PLC. 2. Set the selector switch on CPU to MRES pos; and hold it by 3 seconds, the Stop Led Flashes 2 times slow. 3. Get the switch to Stop Position. 4. Select Run mode position. And thats all, the aplication was running again . So by using the Mres key i can get the aplication running again, some times by means of method_1 and others with method_2. But i will like to know how is this possible? and more important i think: 1. What happens to MW's DB's, HW config, SDB's when i use method_1? 2. What happens to MW's DB's, HW config, SDB's when i use method_2? I have a clue about that but i will like to have your opinion, my intention is to find out wich DB's or variables or registersshould i reset to their default values when i have this "Disconnect" fault in this specific PLC_X. I will apreciate any comment, tip, or suggestion you can provide about this issue; thaks all you in advance; and have a very nice day . PD: Please excuse me for the large description, and the bad use of the technical words
	Keep working!
Suggestion To thank Quote Answer

3/27/2011 2:01 PM	Rate (1)
fritz Diamond Expert Joined: 10/7/2005 Last visit: 9/17/2024 Posts: 3020 Rating: (1054)	Hello TEBANCH here's my take on things: Method 1: It is a "proper" Memory resetas per chapter 8.4.3 of the SIMATIC S7-300 CPU 31xC and CPU 31x: Installation manual,resulting in clearing of the work memory and reloading ofeveryting from the MMC load memory. MW's would be all zero andDB's get loaded with the offline "actual" values from the last download of them (or what as lastlast saved via SFC 84 from the user program) Method 2: Never tried this, but it is (should be) no different than doing a normal "Warm restart" since the proper MRES procedure is not fully completed. DB's should still have their last online actual values, MW contents will depend on how large you set their retentivity etc. (see HERE for more on Warm restart behaviour). You should inthe case of method 2 also NOT get the fast flashing Stop LED, can you confirm this? I hope this helps
	Cheers fritz The advice and opinion given in this thread is that of the author and does not necessarily reflect the views of I S Systems Pty Limited. To find out how I S Systems can help you with your automation needs please visit www.issystems.com.au.
Suggestion To thank Quote Answer
This contribution was helpful to 2 thankful Users TEBANCH zeemens

3/28/2011 8:53 AM	Rate (0)
Aret Posts: 2348 Rating: (264)	Some how the sequence gets stocked some times You should pay your attention here first of all. Check busy/done bits handling. No CP resources could mean you don't resume connection on PLC side - can you check it?
Aret Posts: 2348 Rating: (264)
Suggestion To thank Quote Answer
This contribution was helpful to 1 thankful Users TEBANCH

3/28/2011 4:58 PM	Rate (0)
TEBANCH Silver Member Joined: 3/25/2008 Last visit: 1/28/2022 Posts: 232 Rating: (20)	Thanks both of you for your valuable answers. Dear Fritz in answer to your question: Yes, the Stop led does not flash fast when i use method_2. Please, would you confirm or deny my folowing ideas: If the value of DB1.DBB0 was 10 before a MRES, and his declared unitial value in the editor was 5; then after the MRES, the value of DB1.DBB0 will be 5? Lets we suppose that the cause of my problem (apart of the sequence time out event) is that DB1.DBW2 is 2, but it should be 1, wich SFC can i use to set only an specific DB to his initial value (read it from MMC), do you have any trick you can share? . Dear Aret, the busy/done status bits in FB40 do not tunr on after this event, my first test was try to re-connect the PLC_X to the server, forcing the FB40 to work in Connect mode, every time the result was the same, no done/busyor error bitsturning ON and W#16#8181 status displayed on status word of FB40 . At this point, i'm changing the secuence to avoid the "Time out" event; but even so i can not find out how this fault do not afect the other 2 PLC's (Since the logic is almost the same in them, the only diference is the size of the files we send/recive from server). Do you have any experience/advise you canplease share about the use of FB40? What about the "No resource in CP" message-status, do you think that if i take a loock into the instance DB of FB40 perhaps i will find some usefull data ? Once again thanks a lot for your help; and have a very nice day
	Last edited by: TEBANCH at: 3/28/2011 5:04 PM Forgot to mention error bit on FB40 Keep working!
Suggestion To thank Quote Answer

3/29/2011 2:30 AM	Rate (2)
fritz Diamond Expert Joined: 10/7/2005 Last visit: 9/17/2024 Posts: 3020 Rating: (1054)	Hello again TEBANCH TEBANCH If the value of DB1.DBB0 was 10 before a MRES, and his declared unitial value in the editor was 5; then after the MRES, the value of DB1.DBB0 will be 5? The short answer is: It depends. Before we get to the longer answer, let's clarify a things first regarding DB values as theSiemes terminology of "inital" value can be confusing and even some Siemens decription use the term "initial" with different context: DB view: Declaration View: Only Initial value is visible. This purpose of the "initial" valueis simply to have"initial" offline values that can be used to "initialise" the offline actual values (via "Initialize Data Block" command from the "Edit" menu when in "data View"). Also, as and when you create your DB and add/declare new variables, the initial value that you type in becomes the default offline actual value. The values in the "initial value" columnare NEVER usedby the PLC program though nor come into affect when an MRES takes place. DB view: DataView (offline): Initial value and Actual value are visible (Initial value is the same as above). The "actual" value is the offline actual value. It may/can have the same value as the initial value but can also be set to a different offline actual value. A download of a DB will download this offline actual value into the PLC where it then becomes the online actual value (until overwritten/changed by the PLC program). The DB with is lastdownloaded offline actual values resides on the MMC, it is theseoffline actual values that will become the online actual values after aMRES. DB View: DataView (online): Initial value and Actual value are visible (Initial value is the same as above). The "actual value" column now shows theonline actual value. This may be different to the offline actual value (e.g. changed by the PLC program). A download of the DB will transfer the offline actual values into the PLC and overwrite whatever latest online actual values existed. You can use SFC84 to overwrite the offline actual values in the MMC with the current online actual value (which eseentially achievesthe sameascopying the online DB into your offline S7 program and download it to the PLC) To sum it up: It is the last DB download relatedoffline actual values that become the online actual values after an MRES (or last SFC84 saved values if used). TEBANCH Lets we suppose that the cause of my problem (apart of the sequence time out event) is that DB1.DBW2 is 2, but it should be 1, wich SFC can i use to set only an specific DB to his initial value (read it from MMC), do you have any trick you can share? Any MRES is followed by a PLC restart, so a good place forany"initialisation" (comms related or sequence resets etc. etc.)is the startup OB(simply transfer whatever value you need into your DB). One more thing: In case of your method 2 (incomplete MRES), you should still have the last online actual valuesin your DB's. Asmethod 2 also fixes your FB40 "stuck" related problem I think it is unlikely tobe related to or caused by DB values. I haven't used FB40 for FTP related tasks yet, but wonder if perhaps FC10 "AG_CNTRL" can help with your problem or at least aid in diagnosing it (see HERE and HERE for more on it) Last but not least, below is a link with more on this (note that the second oneis S7-400 and PCS7 related and be aware in which contextthe term "initial value"is applied) What is the significance of the initial and actual values of a data block? What does the terms Initial value (declaration view), Initial value (load memory) and Actual value mean in data blocks when a controller is loaded / restarted? I hope this helps and please let us know what you find out
	Cheers fritz The advice and opinion given in this thread is that of the author and does not necessarily reflect the views of I S Systems Pty Limited. To find out how I S Systems can help you with your automation needs please visit www.issystems.com.au.
Suggestion To thank Quote Answer
This contribution was helpful to 2 thankful Users TEBANCH Marcjan

4/7/2011 3:45 PM	Rate (0)
TEBANCH Silver Member Joined: 3/25/2008 Last visit: 1/28/2022 Posts: 232 Rating: (20)	Dear fittz, i want to thank you for your valuable advices and suggestions, they are very ussefull. I have tryed to test FC10, but i'm not able to "Force" the PLC to get the "time out fault" so i'm not able to check if FC10 will help to fix this issue; but let me explain all you what i did: The normal secuence of PLC-Server comunication is Connect-Read-Delete-Disconect, and the problem comes in the Conect-Read transition, so i just force the PLC to get Stuck in this step, i wait 2 minutes until the Server shut down the PLC connection, and then i wait a couple of minutes to retry to get online with the server again, for my big surprise i was able to go online again (i mean to reach the server again! ). So in resume: If i force PLC to do not excecute the complete secuence, it will be a time out event in the server, but under thid conditions i will be able to go online again. If the PLC executes the secuence with out restrictions, some times with no special conditions present (i mean normal production settings), it fails to be executed complete (the fail is in the Connect-Read transition), and under this conditions it will be a time out error on server and the PLC will not be able to reach the server anymore, until we execute a Master Reset in the PLC. Onedetail that i have to mention is that when i make the Mres, the CP goes to stop and then to run, i wonder if that is why after the Mres the PLC is able to reach the server again? At this point i'm thinking that perhaps the defective element is the CP, but i need to find more evidence about. I wonder wich diagnostics tools can i use in order to find out if the CP is in a faulty state? What will be your advices respect this issue? and Finally What's your oppnion about this, do you think that is a logic related problem or a CP configuration problem? Thanks a lot for your help, any suggestion or tips will be very usefull, have a great day .
	Keep working!
Suggestion To thank Quote Answer

4/10/2011 3:06 PM	Rate (0)
fritz Diamond Expert Joined: 10/7/2005 Last visit: 9/17/2024 Posts: 3020 Rating: (1054)	Hello TEBANCH I could really only speculate (or guess) as to what is or could be the source of the problem (more so as I unfortunnately still lack the hands on experience with FB40). I have a fewthings to note and/or check though: 1.) Have you had a look at the CP343-1 Diagnostics (aka NCM Diagnostics, right click on the CP in Simatic Manger, select "Object properties", go to the Diagnostic tab, startthe "Special Diagnostics" it and don'tforget toclick on the"Glasses" icon to get live data)? 2.) You say in yourfirst post that "when FB40 is executed his status word is W#16#8181 (from the online help this status means "No resources on CP")." My FB40 onlinestates for W#16#8181: Job active. If 8181H is indicated permanently: The CP is not released for FB40 (an illegal command for the firmware version (CMD6 or CMD7) was called.) Questions: 2.a.) Are you sure it is W#16#8181 that you aregetting when the problem occurs? 2.b.) Is your CP343-1 perhaps an older version and NOT a GX30? (FB 40 is only released as of the GX30, FC40-44 need to be used for older CP343-1 Advanced cards) 3.) FB40 uses SFC58&59 internally. This means the "limits" of consecutive SFC58/59 calls apply (8 in your case, see HERE for more). Do you perhaps have a number of other communication jobs going on that use SFC59/59 and//or SFB52/53? 4.) It could a firmware issue, have you checked that all your three319 CPU's and CP343-1 have the same firmware and/or upgrade them all to the latest versions? I hope this helps and please let us know what you find
	Cheers fritz The advice and opinion given in this thread is that of the author and does not necessarily reflect the views of I S Systems Pty Limited. To find out how I S Systems can help you with your automation needs please visit www.issystems.com.au.
Suggestion To thank Quote Answer
This contribution was helpful to 1 thankful Users TEBANCH

5/9/2011 6:59 PM	Rate (0)
TEBANCH Silver Member Joined: 3/25/2008 Last visit: 1/28/2022 Posts: 232 Rating: (20)	Hello. First to all thanks a lot for your valuable help and suggestions. In ansewer to those suggestions: 1. Well. i have tryed but i do not know exactly what to watch over there (since the plant is now in normal production and the fault comes in every 2 or 3 weeks it is very difficult to have enogh time to make the analisis). 2. I have to correct my post; the code you mention is the right one, and yes i'm sure that the code displayed is W#16#8181; the CP is a GX30 version. 3. No, the sequence is designed in order to only use one instance of FB40 at one time; and there is no other FTP links/connections to other devices differents than the "main server". 4. Well, i know that all CPU's and CP's have the same firmware version; but due to the present production requirements it is not posible to make an upgrade of them. (Actually at the moment i'm giving remote production support; so i'm not at site and the fault has not appeared since 2 weeks ago). I will keep searching for answers for this problem and i will let you know about futher information; also i willpost a new question in the Communicationsconference. Thanks a lot for your help and have a nice day .
	Last edited by: TEBANCH at: 5/9/2011 10:13 PM Last edited by: TEBANCH at: 5/9/2011 7:09 PM Keep working!
Suggestion To thank Quote Answer
This contribution was helpful to 1 thankful Users fritz

6/13/2011 4:57 PM	Rate (0)
TEBANCH Silver Member Joined: 3/25/2008 Last visit: 1/28/2022 Posts: 232 Rating: (20)	Hello dear colleagues. Here a litle update about this problem: We have contact with Siemens Product support, and their suggestions was to upgrade the firmware of the CP to V1.17(old firmware was 1.0), and also to run a network "tracker" like wire-shark software in order to capture the information packets exchanged between server and CP. Well we did both, and until now (3 weeks) the problem has not appeared anymore, so i think it was enought just upgrade the firmware of the CP (but maybe is to early to be happy ) we will continue monitoring thePLC<>Server data exchange for a while. Any way, i just want to share this with you, if in the future you have some similar problem, and if every thing goes OK/NOK in the next weeks i will update this thread. Thanks all of you for your help, and have a very nice day.
	Keep working!
Suggestion To thank Quote Answer
This contribution was helpful to 3 thankful Users fritz zeemens judith611

|

Share this page:

Share this page on...

Follow us on