Pros and Cons of Bend Insensitive Multimode Fiber
April 13, 2017 / General, Installation and testing, Upgrading and troubleshooting, Industrial Networks
Lately, I have been getting questions about why bend insensitive multimode fiber (BIMMF) can’t be used in test cords. The argument is that non-BIMMF (e.g., “traditional” fiber) is becoming more difficult to procure and installers need the flexibility to buy something off the shelf. Additionally, requests are being made to change one IEC testing standard to not mandate the use of non-BIMMF in test cords. Let’s examine this closer and explore the consequences.
How will you be using your Versiv?
First, a few technical or other related facts:
- Fluke Networks provides multimode test reference cords specifically designed to work with their multimode cable test equipment – all using non-BIMMF; source of supply for fiber is not a problem.
- Other test vendors use non-BIMMF exclusively with test cords and have no problems with their fiber supplier.
- To achieve the proper launch condition (e.g., encircled flux) using an LED, higher order modes must be stripped out – only possible, practically, with non-BIMMF.
- BIMMF may have a length dependency on numerical aperture and core diameter due to leaky modes propagating in the glass. The net effect is an attenuation measurement on installed cabling that may be different than if using a non-BIMMF launch test cord.
- Encircled flux compliance is difficult to verify on BIMMF because the far field, due to leaky modes, also needs to be measured, but only near field on non-BIMMF. In other words the launch changes along the fiber.
- It is not possible to use off the shelf BIMMF or non-BIMMF with current test equipment designed for encircled flux compliance unless one has a means of making near field launch measurements with special lab equipment.
About 8 years ago 50/125 μm bend insensitive multimode fiber (BIMMF) was introduced. Prior to that time, the major optical fiber vendors were producing non-BIMMF. Over the last several years, as BIMMF became more popular, some of the vendors stopped producing non-BIMMF. While this is beneficial for installed cabling, reducing workmanship problems due to tight bending, it puts the TIA, IEC, and ISO/IEC test standards at odds with the industry direction. To achieve the proper launch condition out of multimode fiber (e.g. encircled flux), non-BIMMF is required, which is what the standards still mandate.
Since standards are written by a variety of experts, no one wants to be locked out. Testing standards put most of the burden on the test equipment vendors to comply with the specific requirements spelled out in the document. Sometimes, the cabling vendors, fiber vendors, and test cord vendors can be at odds with the test vendors mandates to comply with a standard.
From IEC 61280-4-1, attenuation measurements of installed multimode fiber cable plant, under revision, the following is currently, subject to change, stated for the launch cord (LSPM - light source/power meter; Annex F is the encircled flux requirement):
When testing BIMMF or non-BIMMF cabling with a LSPM, a non-BIMMF launch cord meeting the requirements of the launch conditions defined in Annex F shall be used. Additional care is needed when selecting test cords. For optimized testing, the test cords supplied or recommended by the test equipment supplier should be used.
In order to maintain the proper launch condition, test equipment vendors have used different implementation strategies. One method is to apply a mandrel, tuned at the factory, on an external test cord. Another method tunes the launch within the light source and achieves the launch condition at the source’s bulkhead. An external test cord is then coupled to the bulkhead without a mandrel. However, this “modally transparent” test cord must have very strict optical and physical properties to maintain the proper launch. Regardless of the implementation method used, non-BIMMF is required.
If cable and fiber testing was done with an off the shelf test cord using BIMMF, a few things would happen. First, it is unlikely certification testing would be accepted – it would be very difficult to show encircled flux and standards compliance. Second, the measured link attenuation, if not using an encircled flux launch, could cause a link measurement to fail especially where loss budgets are small.
So even though the Datacomm world is increasingly going to BIMMF, the standards committees have determined that testers need to stick with non-BIMMF test cords.
*apologies to Shakespeare