advertisement
from the Automation Dept. department...
Comparison of FF, Profibus, HART, and 4-20mA?
Continuous process industries, DCS questions. topic
Posted by Kayvan Mahmoodifar on 13 August, 2005 - 4:26 pm
Dear All,
I will be happy if you answer this question.

What are the advantages and disadvantages of FF, Profibus, HART, and 4-20mA for Process Control System with following features:

1- will apply in a large gas refinery
2- 4000 analoge I/O
3- 6000 discrite I/O

Generally which system is preferred to use?

Thanks in advance

Kayvan Mahmoodifar
I&C Dept.
NIGEC
K.Mahmoodifar@nigec.com


Posted by Dick Caro on 14 August, 2005 - 7:50 am
This cannot be answered unless you can supply all of the constraints of the process in which it will be installed. This is the subject of my book: Consumer Guide to Fieldbus Network Equipment for Process Control. See the link below.

Each communications technology has its own advantages and disadvantages. Most new installations are currently using Foundation Fieldbus.

Dick Caro
===========================================
Richard H. Caro, CEO
CMC Associates
2 Beth Circle, Acton, MA 01720
Tel: +1.978.635.9449 Mobile: +.978.764.4728
Fax: +1.978.246.1270
E-mail: RCaro@CMC.us
Web: http://www.CMC.us
Buy my books:
Automation Network Selection http://www.isa.org/networkselect
Wireless Networks for Industrial Automation
http://www.isa.org/wirelessnetworks
The Consumer's Guide to Fieldbus Network Equipment for Process Control


Posted by Anonymous on 14 August, 2005 - 7:50 am
For a project of that magnitude, I would hope that your decision is based on more than a paragraph or two on a public web forum.

Two gentlemen who frequent this forum have written books on the subject:

Mr. Jonas Berg has written, "Fieldbuses for Process Control".

Mr Dick Caro has written "Automation Network Selection"

I recommend both.


Posted by Rao on 27 May, 2006 - 1:52 am
I would like to know if FO transmission or wireless transmission is possible for remote mounted 4-20 ma pressure or temperature transmitters. My question is how can we convert 4-20ma signal over to to a multidrop network via wireless or Fiberoptic cable. We plan to provide solar or battery power for the transmitters.


Posted by PaVee on 14 August, 2005 - 7:51 am
Kayvan,
The first two, FF and Profibus are propriety Digital systems, pushed eagerly by various groups of manufacturers. Hart is a system superimposed on the 4-20mA signal.They are all said to be "Smart". There are vendor champions for each system and for these digital systems they will extol their hi-tech virtues including the dubious attraction of "less wiring" and quicker installation.

Special devices (read more expensive) are needed for FF and Profibus; whereas 4-20Ma are "off the shelf" from any manufacturer. Most newer 4-20Ma devices can handle Hart, it is a time proven and reliable "add on" with many benefits including on line documentation and record keeping and 30 odd more data items.

A failure of the Hart communication will not effect the device at all, not so FF and Profibus.
As for applying to different numbers of I/O you'd have to price the cost of 4-20mA with the lower cost of equipment, against the cost of the digital equivalents plus cabling, junction boxes et al. Then price replacement field devices and other parts replacement. Take note of the horror of the maintenance personnel that have to fix (read replace) the FF/profibus devices.
Also note that you really don't want to be using FF/Profibus for Shutdown duty do you?
I have worked with FF over a long period, but not yet with Profibus. Hart enabled systems are my choice.


Posted by Armin Steinhoff on 16 August, 2005 - 1:12 am
>Kayvan,
>The first two, FF and Profibus are propriety Digital systems, pushed
>eagerly by various groups of manufacturers. <

Sorry, FF and Profibus are not proprietary systems, they are based on open
standards.Yes, they are used by various groups of manufacturers but FF and
Profibus are not the property of these companies.

> Hart is a system superimposed on the 4-20mA signal.They are all said to
> be "Smart". There are vendor champions for each system and for these
> digital systems they will extol their hi-tech virtues including the
> dubious attraction of "less wiring" and quicker installation.
>
>Special devices (read more expensive) are needed for FF and Profibus;
>whereas 4-20Ma are "off the shelf" from any manufacturer. Most newer
>4-20Ma devices can handle Hart, it is a time proven and reliable "add on"
>with many benefits including on line documentation and record keeping and
>30 odd more data items.
>
>A failure of the Hart communication will not effect the device at all, not
>so FF and Profibus. <

This is not the case... FF and Profibus don't stop working if a device is
defective.

>As for applying to different numbers of I/O you'd have to price the cost
>of 4-20mA with the lower cost of equipment, against the cost of the
>digital equivalents plus cabling, junction boxes et al. Then price
>replacement field devices and other parts replacement. Take note of the
>horror of the maintenance personnel that have to fix (read replace) the
>FF/profibus devices.
>Also note that you really don't want to be using FF/Profibus for Shutdown
>duty do you? <

It depends on the target scan time... there are 4000 analog values and
6000 digital signals to scan.
What is the targeted sampling period?

Best Regards
Armin Steinhoff


Posted by PaVee on 17 August, 2005 - 3:17 pm
Armin,

Yes, they are "supposed" to be "open" systems. But as I said, they are "pushed" by different groups of manufacturers. Rosemount naturally promotes Fieldbus, Siemens with Profibus; of which I have no practical experience. I am purely basing my experience on FF, sorry profibus.

What I meant was, you lose comms with a device and then you have no absolutely no idea what's happening, "diagnostics" then does not work. With a normal 4-20 this is not a problem, it can be fixed quickly and cheaply.
The rest of it of course, goes on regardless. I believe hand held FF testers are now available, which may help.
For SIL rated equipment used in Shut down equipment, at the time, FF could not be used. I personally would never work in place that had a FF shutdown system.

I know there are die-hard fans of various systems, but having comissioned and then maintained a FF system,for 4 years and with a Hart equiped 4-20mA system running the SIS to compare it with, over a long period, the Hart won it by a long chalk.

I'm not entirely convinced about equivalent FF/Profibus systems being cheaper to purchase,install,train on and maintain over the life of the plant, than a 4-20 plus Hart.


Posted by Jonas Berge on 23 August, 2005 - 2:21 pm
Like Amir said, FF and Profibus are just as open as HART. The organization members and manufacturers are pretty much the same.

The premium for and H1 or PA device need not be more than US$70 if you look around.

HART does not use digital for control (too slow) so of course a communications loss have no impact. For a full comparison you would need to look at the impact of a failure of a 4-20 mA I/O card or multi-core cable.

True, HART does provide many benefits if fully supported in the control system. Not all control system supports device management with HART, and even fewer allow you to use the HART diagnostics to create shutdown interlocks (but once you do use HART as part of the interlocks, then a communication failure may have an impact).

Yeah. Don't use bus for SIS yet. Wait for FF-SIS and ProfiSafe-PA... For 4-20 mA the HART devices shall be write-protected since HART is not a safety protocol to IEC 61508 and IEC 62280.

Early implementations of FF into control systems was a bit crude by today's standards. Modern control systems such as the latest release of SYSTEM302 has built in features to make day-to-day tasks such as device replacement much easier than it was in the past (standards for transducer blocks has also helped). For example, replacing a device with one of a different version or different make is now easier than before.

Jonas Berge
SMAR
===========
jberge@smar.com.sg
www.smar.com
Learn fieldbus and Ethernet at your own pace: www.isa.org/fieldbuses
Learn OPC and automation software at your own pace: www.isa.org/autosoftware


Posted by Eben on 18 March, 2006 - 10:38 am
Your replies are very valueable and it is good to hear the facts from experienced people. What I would like to know is which of these "Smart" diagnostic capabilities are of value to the maintenance engineer and how does the capabilities differ between these bus topologies?

It is quite difficult to find information explaining the capabilities of these instruments. Can someone direct me to a good information source to consult?

Regards - Eben


Posted by Jonas Berge on 31 March, 2006 - 12:20 am
I will take a crack at these questions based on my experience.

The goal is predictive, performance-based, and condition-based maintenance in place of corrective or preventive maintenance. To be of good value these intelligent devices must have permanent and continuous communication (not just a temporarily connected handheld) and they must be used together with "device management software". Applied System Technologies "Cornerstone" was really the first such software, but since then we (SMAR), Emerson, and Yokogawa etc. also make device management software. Apart from the permanent communication, another major feature that sets device management software apart from handhelds is that device management software has a database where historical records and supporting information is kept.

Device management software delivers the promise of information regardless if the underlying communication protocol: HART, Fieldbus, or Profibus. One protocol does not have better diagnostics than another because self-checks are done inside the device. A given device typically provides the same diagnostics regardless of the protocol option chosen because the sensor or output transducer is the same for all. The difference in sophistication of diagnostics primarily varies from vendor to vendor, not from protocol to protocol. However, HART devices only have 4 mA to operate on whereas Fieldbus and Profibus devices can consume much more. Theoretically this means that in the long run you may find that Fieldbus and Profibus devices will provide more sophisticated diagnostics, and faster.

For practical purposes I like to distinguish between the following maintenance triggers: - Diagnostics - Performance analysis - Operational statistics

Diagnostics: is found in most devices and is done at power on, continuously
during operation for non-disruptive checks, and on demand for checks that disrupt normal operation. The result is usually a simple pass/fail for each test. Example of simple self-diagnostics include sensor failure, output read-back failure, and memory failure etc. If some test fails, this is reported to the device management software through the communications. The diagnostics is used for condition-based maintenance scheduling. Advanced diagnostics include measurement of critical factors affecting the function of the device, such as pH sensor electrode impedance or build-up on the electrodes of an electromagnetic flow meter. The rate of deterioration of these factors is looked at for predictive maintenance.

Performance Analysis: is primarily used in valve positioners and is done continuously during operation for non-disruptive analysis, and on demand for analysis that disrupt normal operation. The result is a numerical value, a plotted curve, or a chart. Examples of performance analysis include friction, dead-band, open/close time, hysteresis plot, and step response. Performance analysis is used for performance-based maintenance. The rate of deterioration of these factors is looked at for predictive maintenance.

Operational Statistics: is primarily used in valve positioners and runs continuously during operation and is completely non-disruptive. The result is a count, time, or an accumulated total. Examples of operational statistics include number of reversals, total accumulated travel, and time spent near closed. Operational statistics is used for predictive maintenance.

It is easy to scoff at operational statistics like number of valve reversals or total travel because if you look at it with a handheld one valve at the time it is really totally useless. It is only in conjunction with continuous monitoring of this information from device management software that this begins to make sense and can play a part in the effective scheduling of maintenance. Similarly you need device management software to immediately capture diagnostic alarms and to monitor deterioration in parameters indicating wear-and-tear.

Really, you could have a very long discussion of what device management software can do and what the benefits are. We (SMAR) have a white paper on this topic centered around our AssetView software. Contact me directly on the address below if you are interested.

In the mean time, to learn more about device management, take a look at chapter 2 and 9 of the yellow book "Fieldbuses for Process Control: Engineering, Operation, and Maintenance" buy online: http://www.isa.org/fieldbuses

Jonas Berge SMAR
===========
jberge@smar.com.sg
www.smar.com
Learn fieldbus and Ethernet at your own pace: www.isa.org/fieldbuses
Learn OPC and automation software at your own pace: www.isa.org/autosoftware


Posted by Kevin on 14 August, 2005 - 5:44 pm
This is a very big question that cannot be answered easily. There are many engineering issues that need to be considered. Not least are expense, functionality, ease of installation, availability of support, qualified programmers, etc. For example FF offers the capability of an instrument such as a pressure or flow transmitter to control an output device such as a valve whilst being disconnected from the controller, for example; if the controller is down but there is still power on the bus. This may be of benefit to your process. ( although I personally have not seen this function). Another issue is the location classification. FF lends itself to hazardous locations better than the other busses that I have dealt with- although still quite cumbersome and expensive. I am no fan of Rosemount's DeltaV controller- the most widely used for FF apps. FF is slow. 4-20ma is almost instantaneous. It may be that you will require more than one bus. The process plants that I have dealt with use FF for continuous measurement and control (i.e. analog i/o) and Devicenet and/or ASi busses for discrete control (i.e. digital i/o). So for this size project you have your work cut out for you.

Best regards, Kevin


Posted by Joe Hohn on 15 August, 2005 - 11:49 pm
Kayvan-

Kevin's reply is by someone who has obvious implementation experience. We are a system integrator and also find that FFB is quite useful for analog points. We most often use DeviceNet or Profibus DP for motor control and discrete ins and outs.

Some things to look out for:
1. Design time goes way up when you elect to use bus technologies - expect these costs to double.
2. Material costs go up with bus technologies - special cable, special 'bricks' to facilitate the spurs or drops, special instruments.
3. Procedures and equipment for 'loop checks' are very different.
4. You will need to retrain maintenance personnel (all FFB devices need to be configured and calibrated thru the DCS HMI or a similar interface on a laptop) and will almost certainly have to rewrite your procedures.
5. To take full advantage of the buses, you need to accomodate data quality and diagnostic functions in your programming and HMIs.
6. FFB, Profibus and DeviceNet implementations are available for almost all of your instruments and final control elements. Vendors will attempt to dissuade you and I recommend that you disallow exceptions except in the case of your analyzers, where you may have to concentrate on functionality rather than bus compatibility.
7. Your design team and your system integrator must have experience with field buses. Ask for sample design documents and sample test reports. Call their references.

The wins:
1. Greatly reduced installation labor costs - particularly if you take the time to connectorize all devices and take advantage of the fact that the spur cables can be run in free air.
2. A huge reduction in startup and commissioning time - this is the single greatest savings in doing a bus implementation.
3. A large reduction in field acceptance testing time, in particular because of the reduction in the amount of troubleshooting required to complete the task.
4. A big reduction in the size and cost of the DCS because there is so little native I/O.
5. An opportunity for standardization and the consequent reduction in design, fabrication and testing costs. For instance, all control panels can be identical with a minimum of (or even an absense of) I/O.
6. A big improvement in data quality, both because the FFB/Profibus instrumentation is better and because eliminating the analog input circuits makes the uncertainty of the data = the uncertainty of the instrument itself.

In a system the size of the one you are contemplating, you will reduce cost, improve schedule and deliver a better product by using bus technologies (assuming that your design and integration teams have done this before).

Joe Hohn


Posted by ControlNovice on 16 August, 2005 - 12:54 am
I've installed HART and FF and I will try everything to avoid installing FF again.

FF is more expensive. Despite what the vendor's say on savings on installation and commissioning - the cost of the instrument is more and I've found that commissioning is MUCH more difficult than HART - and thus more time consuming.

If you lose a FF signal you lose the whole trunk line (up to 12 instruments). If you lose a HART signal, you only lose one instrument.

Several DCS manufacturers have Intrinsically Safe I/O cards for both HART and FF - so that should not be a determining factor.

A FF device can run without the controller, but the controller is probably the most reliable piece of hardware - so not really a problem.

Training was a big issue. Your maintenance department is probably very familiar with HART devices. FF is much more difficult to troubleshoot. In fact, at the plants where I've installed FF, the maintenance department does not troubleshoot the devices (even after training). The process engineer does the troubleshooting!! Do you want your engineer to become an instrument technician?

I can not recommend FF. Stick with HART - it's much easier and much more cost effective.


Posted by Atlanta on 25 August, 2005 - 4:32 pm
FF is not more expensive then HART. And it is not true when you lose one signal you will lose all other instruments if you have use the right topology. I think that he most people do not understand the FF philosphy.

HART is a proven technology however iff you get A good explanation how to use FF you will see that FF will be a great advantage in your installation.

And your Process Engineers do not have to become an instrument technician.


Posted by ControlEng on 16 August, 2005 - 1:31 pm
I have worked with Profibus DP and PA and Hart. If you are considering using Siemens PLCs, I would definately use Profibus. I worked on about a 2000 point system and experienced no difficulties. The main bus is Profibus DP and is very reliable. You can install I/O, Drives, and other controllers on this bus. Profibus PA is used for devices such as transmitters and is intrinsically safe. We used Siemens PDM to configure, check, and calibrate the devices. We had some software issues early but it worked quite well after we became more proficient. I would not use Profibus with any other control system that required the use of third party adapters such as SST or Hilscher. We had issues with configuration and the software is quite bad.


Posted by Jonas Berge on 23 August, 2005 - 2:24 pm
That large amount of discrete I/O is not suitable for FF-H1, you need a discrete I/O subsystem on FF-HSE (FF protocol on Ethernet media and IP transport).

Look for a control system that supports both FF H1 and FF-HSE. We (SMAR) make such a system called SYSTEM302: http://www.smar.com/system302

For FF, PA, HART, and industrial Ethernet take a look at the yellow book "Fieldbuses for Process Control: Engineering, Operation, and Maintenance"
buy online: http://www.isa.org/fieldbuses

Jonas Berge
SMAR
===========
jberge@smar.com.sg
www.smar.com
Learn fieldbus and Ethernet at your own pace: www.isa.org/fieldbuses
Learn OPC and automation software at your own pace: www.isa.org/autosoftware


Posted by YKJARIWALA on 25 September, 2005 - 5:02 pm
Dear Kaywan
We have implemented fieldbus project which involved following buses

1] FF nodes : 235
2] HART : 64
3] PROFIBUS DI/DO : 15 nodes [ 2000I/O]
4] conventionalDI/DO : 200
5] MODBUS RTU : 25 nodes

This are all Zone 1 installation & WE Wanted this
plant easily & fast. Hence we have identified
critical & non critical loops.All critical loops
we implemented with HART transmitter , it is not
unreliability with other buses , but we wanted
guranted response for critical variables.
FF master card is minimum 1.0 second scan & more.
DO NOT IMPLEMENT CONTROL on FF BUS PLEASE

Avoid Profibus DP for analogue loops altogether,
for digital I/O this gives good perfomance.

FF bus/nodes requires careful study about voltages,resistance & current.

When you are installing various buses as far as
possible avoid all recommendations from Vendors.

HART was probably started with remote calibration
& multidrop functions but due to multidrop limitation , we have opted for only point to point
configuration.This limitation is eliminated in FF
bus.
We have also done HART multidrop for control & it
is still working with 3 sec time frame.

Let me tell you FF bus is really ideal for non-critical loops & this can not be ignored.

Cost of ownership for bus architecture is always
20 % higher than conventional system but still worth it.

Jari
iconcnl@vsnl.net


Posted by John on 22 October, 2008 - 11:45 pm
Hi,

Here is the question answered easily therefore in general terms.

FF
It is the best system, its transmission rate is 31.25Kbits/second. You get more diagnostics information and every device can give info without request. One H1 segment can support up to 32 devices--in reality 4-14 or 15 devices, but you get more if you use a repeater. It is Communist system (all device are equal peer to peer).

Disadvantages:
The transmission speed is not as high as Profibus (but it is still better) and if you lose a signal the loop goes down.
Money: It is either a little cheaper or slightly more expensive but you are getting a much better system.

Profibus:
The transmission rate is 9.6 to 12 Mbit/second, however you need a coupler to convert DP to PA and it does nothing else, if you lose one signal the loop stays up. Master-slave setup so the device can only speak when spoken to.

Disadvantage: Polling system not real time, cost.
Money: More expensive and commissioning takes much longer.

HART:
Super imposed digital signal on analog signal.
Disadvantages:
Slower than FF and Profibus, one way communication and the device can only give info when asked by host. Less diagnostics.
Money:
In theory more expensive due to wiring and slower response.

4/20
Pure analog signal slow, you can barely consider this communication compared to the above.
Enough said.


Posted by Dick Caro on 23 October, 2008 - 7:37 am
This post contains too many errors to let stand.

Why is FF "best"? True, the shared wiring for field instruments can reduce cost, and it is faster than most other field digital communications, but the power is that it supports true digital feedback loop control between field devices on the same H1 segment. If one of the loop components fails, the control may be lost, but the rest of the instruments on that segment can continue to communicate. The communications is a parallel multidrop link, not a serial daisy-chain. A linking device is available to Foundation Fieldbus HSE (High Speed Ethernet) to aggregate H1 segments, AND allow host-free communications between instruments on different H1 segments.

Profibus-PA is electrically identical to FF H1 including speed, but the protocol is only master-slave (client server). PA instruments are permitted to use their intelligence to perform signal processing, but not control in the field device. A coupler to Profibus-DP or to Profinet is available to aggregate PA segments.

HART uses the analog 4-20mA link for the primary variable - a continuous signal that is always available, but requires an analog to digital converter, signal processing, and control in a host system. The digital communications of HART (and WirelessHART as well) is bi-directional using a master-slave protocol.

All three protocols support a full range of diagnostics for the field instrument, independent of the protocol selected.

Dick Caro
===========================================
Richard H. Caro, Certified Automation Professional, CEO, CMC Associates,
2 Beth Circle, Acton, MA 01720
Tel: +1.978.635.9449 Mobile: +.978.764.4728
Fax: +1.978.246.1270
E-mail: RCaro@CMC.us
Subscribe to the CMC Wireless Report at http://www.CMC.us
Web: http://www.CMC.us
Buy my books:
http://www.isa.org/books
Automation Network Selection
Wireless Networks for Industrial Automation
http://www.spitzerandboyes.com/Product/fbus.htm
The Consumer's Guide to Fieldbus Network Equipment for Process Control
Buy this book and save 50% or more on your next control system!!!


Posted by wboyes on 23 October, 2008 - 11:30 am
It would be nice if you were also accurate, "enough said."

HART 7.1 has some bi-directional features, as well as a strong wireless component. You are comparing HART 5.0, which is two generations obsolete.

Profibus is more expensive than what? Foundation Fieldbus? Show your data, because everything I've seen indicates that Profibus/Profinet installations are less expensive than FF (not much, to be sure).

Do you have any data to substantiate your claim that commissioning a Profibus system "takes much longer." Longer than what? How much longer?

And if 4-20 were so awful, why are we still building new plants using it?

Sometimes simple isn't as old fashioned and clunky as we think.

Walt Boyes
Editor in Chief
Control and Controlglobal.com
www.controlglobal.com
Mailto:wboyes@putman.net
Read my blog SoundOFF!! At www.controlglobal.com/soundoff

Your use of this site is subject to the terms and conditions set forth under Legal Notices and the Privacy Policy. Please read those terms and conditions carefully. Subject to the rights expressly reserved to others under Legal Notices, the content of this site and the compilation thereof is © 1999-2014 Nerds in Control, LLC. All rights reserved.

Users of this site are benefiting from open source technologies, including PHP, MySQL and Apache. Be happy.


Fortune
If a 6600 used paper tape instead of core memory, it would use up tape
at about 30 miles/second.
-- Grishman, Assembly Language Programming
Advertise here
Advertisement
our advertisers
Help keep our servers running...
Patronize our advertisers!
Visit our Post Archive