I'm planning to use Ethernet as the backbone to link two nearby manufacturing factories, any redundant solution?

 

Check out the Fieldbus Foundation's High Speed Ethernet (HSE) protocol. It supports complete LAN redundancy.

Chris George
Southern Alberta Institute of Technology

 

Schneider Electric has a range of products they call Connexium. It's a family of switches, hubs, and transceivers which can be designed with a redundant fiber optic ring backbone. That way, if one path is broken, data will still pass in another direction, and no device should lose it's network connection.

Also try Hirschmann. They offer the same products (SE may be brand labeling theirs).

 

Try a Hirshcmann HIPER-Ring(r) Ethernet.
"http://www.hirschmann-usa.com":http://www.hirschmann-usa.com

 

You might look into Modicon's Ethernet Fiber-optic "Self-healing Ring." Possible through their Connexium line of Ethernet Switches, it provides multiple-path fiber Ethernet connectivity with both high performance capabilities and built-in diagnostics. They call this a "Redundancy Manager" which in addition to performing all the standard Ethernet switching functions, it allows a physical 200Mbps ring to be created by terminating both ends of the traditional Ethernet Bus (fiber or copper). These two sides to the Redundancy Manager continuously transmit and receive real-time diagnostic messages around "the ring." The result is a real-time report on the actual state of the network at any instant, which in the case of ring trouble will interpret the loss of diagnostic data as a network failure. Then the Redundancy Manager connects the two interfaces internally. This will return the network to full operational status in between 20 and 300 msec (depending on the size of the ring).

Combine this with their Hot-standby PLCs that can swap IP addresses on the fly (on PLC failure) and you have a fully redundant Ethernet enabled automation system. But that’s a different topic all together... Hope this helps.

 

Check out "www.scadaserver.com":http://www.scadaserver.com

 

You may consider a fiber optic link between the two factories for better noise immunity. As far as redundancy or reliability is concerned, check out either Hirschmann's or Moxa's websites. They both have solutions that provide ring redundancy.

"http://www.hirschmann-usa.com/an/ie_switches.asp":http://www.hirschmann-usa.com/an/ie_switches.asp
"http://www.moxa.com/IndustrialEthernet":http://www.moxa.com/IndustrialEthernet

 

Hirschmann have a product line called "ETHERNET Rail" that allows you to create an Etherner ring, which is fault redundant. You can find more info at "http://www.hirschmann.com/":http://www.hirschmann.com/ .

Carlos Mandolesi
Sao Paulo - Brasil

 

Check GarrettCom also. How about Spanning Tree?

 

I'm not sure if this is what you are looking for exactly but we've used a Ethernet board from Intel in one of my companies customer sites in
China. This allowed a redundant network but the SCADA computer only needed one IP address. This greatly simplified the configuration of the
SCADA software.

Intel PRO/100+ Dual Port Server Adapter
64bit PCI, it can be used in an 32bit PCI bus.

3COM also has a similar board: 3C982-TXM

Robert Ackerman
V.P. Engineering Development
automationX Canada Inc.
#202-1050 West 14th Street
North Vancouver, BC, Canada
V7P 3P3
Tel: ++1 (604) 983-8728
Fax: ++1 (604) 983-8563
http://www.mnrcan.com
mailto:roba @mnrcan.com

 

Look into using two routers, with redundant Ethernet connections.

Each Ethernet connection should be on a different slot of the router. Router should have redundant power supplies, one to "conditioned" power,
another source of UPS power (UPS do fail).

It was unclear on how "nearby" the factories are.
Are you planning on running fiber-optic between them - or leasing a line ?

Look at the reliability of your tele-comm provider. See what guarantees (penalties) they have for providing service.

If you have the right-of-way, then you may want to lay your own fiber. The fiber should be run into two entrances into your factory.

Most contractors use "Fiber-Optic seeking" back-hoes - that will shut you down
If you don't plan ahead.

One client of my, encased the fiber runs with concrete in trenches to reduce the probability of this occurrence.

Before looking into "redundancy" issues, it is advisable to look at the cost of downtime. A 10K problem has different constraints than a 100K or 1000K problem.

William F. Hullsiek

 

Dear Vincent,

How redundant do you want? There are plenty of solutions for media duplication. Typically these scheme takes one Ethernet port and split it into two and two media (UTP or fiber) then runs (often in a ring) to the other devices. This way if one section is broken the communication goes the
other way. This scheme will work with just about any Ethernet device. However, at the end of the day the basic device only has a single port and
if it has a problem the duplicated media does not help. Certainly you can argue that the media is the most vulnerable point, and this in many cases is all the redundancy you need. For applications that require high availability you may need devices that are redundant, i.e. the device is a pair consisting of primary and secondary. The FOUNDATION(tm) HSE Fieldbus protocol runs on IP/Ethernet with diagnostics to detect network as well as device failure and select the appropriate communication path. Devices may also have redundant ports (i.e. four ports for a redundant device with redundant ports). We (Smar) have a device based on this technology.

The below link (make sure you get the complete link if it wraps) is for a book that also covers these aspects of Industrial Ethernet although it
focuses on field-level fieldbus.
"http://www.isa.org/Template.cfm?Section=Shop_ISA&Template=/Ecommerce/Product
Display.cfm&ProductID=3036":http://www.isa.org/Template.cfm?Section=Shop_ISA&Template=/Ecommerce/Product
Display.cfm&ProductID=3036

Jonas

 

> I'm planning to use Ethernet as the backbone to link two nearby manufacturing factories, any redundant solution?

Yes there is !
Hello, pls. look at "http://www.ethernetrail.com/":http://www.ethernetrail.com/
and then at "Managed Switches" These devices support a redundancy ring technology (copper or fiber optics based).

Pls. feel free to contact our subsidary in the US,
contact is shon at the web page, or mail to me
"[email protected]", mailto:[email protected]
to provide you with further information.

Best regards
Stefan Gallmann, Phoenix Contact GmbH, Germany

 

Looking at the responses you have had already, I would agree that Hirschmann is the most effective and desirable solution for your application. You can set up 2 Hiper Rings and 2 links, one at each site both at 1 gigabit and then drop down to subnetworks in each building running Hiper Rings at 100Mb. In this way you will have redundancy at all levels and also enough band width to include VIP-T (Hirschmann Video/voice over IP) Don't be fooled by cheaper and bigger marketing promises from other companies. Hirschmann has been doing this for a long time now very successfully and offers the backbone switches as well.

Regards,
Kevin Tomkyns
Australian Utilities & Scada Technologies
Tel: +61 7 3342 7011 Fax: +61 7 3342 7022
Mobile Phone: 0408 427033
Email: [email protected]

 

Rafael,

Take a look at the SM 2501F from Shore Microsystems . The unit works for 100 BaseFX
and provides FDDI Level Fault Tolerance . The SM 2501F recovers from a failure to re-transmit data
in 20-40MS not just restore Link Signal . These figures are actual figures established by tests
at NASA Kennedy Space Center while conducting tests to upgrade their FDDI Backbone . They are
on the web at:

http://clcs.ksc.nasa.gov/erp/fddi_draft_c.ppt
See page 8 for results on SM 2501F

"www.shoremicro.com/sm-2501f.htm":http://www.shoremicro.com/sm-2501f.htm

Scott Bald
Regional Sales Manager
Shore Microsystems
ph:732-870-0800 ext#103
email:[email protected]
www.shoremicro.com

 

Please take a look at our redundant products for either copper or fiber .

"www.shoremicro.com/sm-2501.htm":http://www.shoremicro.com/sm-2501.htm
Highly programmable device failsafe relay based process for 100 BaseTX otr 10 BaseT -autonegotiation supported.

"www.shoremicro.com/sm-2501f.htm":http://www.shoremicro.com/sm-2501f.htm
FDDI Level fault tolerance for Switched Fast Ethernet . 20-40 MS failover rate . This is for data re-transmission not just Link Signal restoration .Numbers are from NASA Kennedy Space Center tests which they made available at available at :

"http://clcs.ksc.nasa.gov/erp/fddi_draft_.ppt":http://clcs.ksc.nasa.gov/erp/fddi_draft_.ppt
See page 8 for SM 2501F

"www.shoremicro.com/sm-2400.htm":http://www.shoremicro.com/sm-2400.htm
PCI device which can be added internally to a Gigabit /100 BaseTX/10 BaseT environment

For more information contact::

Scott Bald
Regional Sales Manager
Shore Micorsystems
(732)870-0800 ext#103
email:[email protected]