Skip to Content

SiePortal

The integrated platform for your product selection, buying and support workflow - bringing together Industry Mall and Online Support.

SiePortalDeutsch

Search

Home
Support Support
{{(apiData.Support | filter:{'rank':1}:true)[0].title}} Support
- Menu temporary not available
- Menu temporary not available
- Menue temporary not available
- {{item.title}}

Products & Services
- Menu temporary not available
- Menue temporary not available
- {{item.title}}
Support Support
{{(apiData.Support | filter:{'rank':1}:true)[0].title}} Support
- Menu temporary not available
- Menue temporary not available
- {{item.title}}
MySiePortal MySiePortal
{{(apiData.MySiePortal | filter:{'rank':1}:true)[0].title}} MySiePortal
- Menu temporary not available
- Menue temporary not available
- {{item.title}}
Cart ({{apiData.CartCount}})

Navigation

Modbus RTU Help

Created by: Ratan at: 2/26/2014 2:25 PM (10 Replies)

Rating (2)

Thanks 2

Result pages:: |; 1 | 2; |

11 Entries

Entries per page: 10 | all

2/26/2014 2:25 PM	Rate (0)
Ratan Experienced Member Joined: 12/7/2010 Last visit: 3/26/2025 Posts: 94 Rating: (0)	Dear All I need help regarding Modbus RTU Communiction. I have PLC S7-1200 & Communication Module RS422/485. I have Connected Pin 3 and Pin 8 for RS 485 + and RS485- Respectively. Main part is that i want to communicate a Modbus Register Data Type Float (Real), Is it possible to obtain 32 bit data in real format. Currently i am getting garbage value after communication. As done bit for MODBUS_MASTER instruction. Thanks In Advance.

Suggestion To thank Quote Answer

2/26/2014 2:56 PM	Rate (1)
smiffy Platinum Expert Joined: 1/17/2007 Last visit: 4/15/2025 Posts: 1560 Rating: (540)	It is possible to get REAL (floating point) data over a modbus interface. However, there is a caveat. The modbus protocol does not inherently understand REAL numbers - it only has the concept of BOOLEANS and 16-bit registers. REAL values are encoded in IEEE754 format, which are encoded into 32-bits. So each REAL value will have to be sent as 2 16-bit registers. This is the duty of the application layer, as already stated, the modbus protocol has no knowledge of how to handle REAL values directly. So sending partner must ecode the REAL value into 2 16-bit registers and then send these. The receiving partner must then reconstitute these 2 16-bit registers back into a REAL value format. There are issues in doing this due to the way that different systems enocde the word order for REAL numbers (endian issues). So the reason you are getting garbage could be down to:- 1) the word order of the data is mixed up 2) you have not converted the registers back into REAL numbers (e.g. took each register to be a REAL number which is nonsense as you would only be using have the REAL data). 3) The data really is garbage (wrong address or data is not REAL data). This question has been asked before and is covered in these threads:- /tf/WW/en/Posts/97287 /tf/WW/en/Posts/83233
	Programming today is the race between software engineers building bigger and better idiot proof programs, and the universe producing bigger and better idiots. So far, the universe is winning...
Suggestion To thank Quote Answer
This contribution was helpful to 1 thankful Users Ugljesa

2/26/2014 3:26 PM	Rate (0)
Piotr.M Platinum Member Joined: 6/24/2013 Last visit: 4/16/2025 Posts: 1678 Rating: (431)	Hi 1. what is range of the values? 2. if range is known You could convert it to integer in units x 10^n for example - temperature - you can send it in degrees x 10 (0..32767 --> 0..3276.7 degree) - percent - you can send it as % x 100 (0..10000 --> 0..100.00%) etc. 3. You can even send thing as a text value (this can be useful when chr<32 are used as control characters non-rtu modbus (when ascii used)

Suggestion To thank Quote Answer

2/26/2014 6:22 PM	Rate (0)
Ratan Experienced Member Joined: 12/7/2010 Last visit: 3/26/2025 Posts: 94 Rating: (0)	Is there any facilty to directly swap 2 word data recieved in REAL format. I mean to say any instruction.
	Last edited by: Ratan at: 2/26/2014 6:23 PM
Suggestion To thank Quote Answer

2/26/2014 6:47 PM	Rate (1)
smiffy Platinum Expert Joined: 1/17/2007 Last visit: 4/15/2025 Posts: 1560 Rating: (540)	Best way is to put the 2 INTs into a DINT, them rotate the DINT right (ROR) or left (ROL) by 16 bit positions. It doesn't matter which way you rotate it as long as you rotate by 16 bit positions. This will effectively swap the words over. Then you simply convert the DINT to a REAL using the DINT_TO_REAL conversion instruction.
	Programming today is the race between software engineers building bigger and better idiot proof programs, and the universe producing bigger and better idiots. So far, the universe is winning...
Suggestion To thank Quote Answer

2/27/2014 8:08 AM	Rate (0)
Ratan Experienced Member Joined: 12/7/2010 Last visit: 3/26/2025 Posts: 94 Rating: (0)	Dear Smiffy I have done what you told me in your suggestion. I am getting data 24B1_43D4 in 32bit HEX fromat after Rotating of 16 bit or swaping of word getting 43D4_24B1 which is exaclty meter shows 424.2 which is conveted fromhttp://babbage.cs.qc.edu/IEEE-754/ but in PLC in CONV instruction Dint to Real i am not getting Exact data it is showing something 1.138057+e009. What can be remeady for same In WATCH TABLE in PLC software it is showing actual as meter, as i have seen display format floating.
	Last edited by: Ratan at: 2/27/2014 8:13 AM
Suggestion To thank Quote Answer

2/27/2014 3:46 PM	Rate (0)
smiffy Platinum Expert Joined: 1/17/2007 Last visit: 4/15/2025 Posts: 1560 Rating: (540)	Ratan Dear Smiffy ...after Rotating of 16 bit or swaping of word getting 43D4_24B1 which is exaclty meter shows 424.2 Yes, 43D4_24B1 looks about right, and the fact that it converts to the correct FP number bears this out. Ratan but in PLC in CONV instruction Dint to Real i am not getting Exact data it is showing something 1.138057+e009. Hmm 1.138057+e009r decodes as 4E87_AABE so it doesn't look like a word order. But saying that, if the word order is wrong, then the value could change by VERY large ammounts when the source number changes by VERY small ammounts. The reason is that the very small part of the source number (that usually changes quite a bit do to noise say if it is an analogue input) then becomes the very large part of the resultant number = strange numbers and very large variations in the number (e+20 directly to e-20 or so in a single step). Is the resultant number stable at the 1.138057+e009 value? Can you put a fixed hex number in the source regsiters that represents a known number (e.g. 12.345 = 4145_851F) and then check that the conversion is working.
	Programming today is the race between software engineers building bigger and better idiot proof programs, and the universe producing bigger and better idiots. So far, the universe is winning...
Suggestion To thank Quote Answer
This contribution was helpful to 1 thankful Users La mujer esponja

7/3/2015 11:13 AM	Rate (0)
La mujer esponja Posts: 2 Rating: (0)	Dear smiffy. I have a doubt about this. For example, i want to read register 801 of SENTRON PAC3200 type Double en C programming language with libmodbus library. 1. Read the registers 801 and 802 and append the result in an array tab[]. I have tab[0] = value_801 and tab[1] = value_802. 2. Read the registers 803 and 804 and append in the same array. tab[2] = value_803 and tab[3] = value_804 (In fact, i can read four registers together with libmodbus in one instruction) 3. Concatenate four values 4. Finally, rotate by 16 bit positions. Is that right? Thank you very much. Best regards from Seville (Spain)
La mujer esponja Posts: 2 Rating: (0)	Last edited by: La mujer esponja at: 7/3/2015 11:39:01 AM
Suggestion To thank Quote Answer

7/16/2015 1:14 PM	Rate (0)
Piotr.M Platinum Member Joined: 6/24/2013 Last visit: 4/16/2025 Posts: 1678 Rating: (431)	Hi You can use ready functions which handle 64-bit Real from Your PAC meter. Library attached. Attachment IS0034_LongReal.zip (758 Downloads)

Suggestion To thank Quote Answer

8/6/2015 10:49 AM	Rate (0)
La mujer esponja Posts: 2 Rating: (0)	I don't know how use that library. My program is written in C language programming. My solution is not valid? Really, i need to know how is stored the information. Which are the Big endian and little endian? With 32-bit register is what i do. Thanks you very much. Elvira
La mujer esponja Posts: 2 Rating: (0)
Suggestion To thank Quote Answer

|

Share this page:

Share this page on...

Follow us on