Skip to main content

Mix mode solid-state inline analogue line interface for asymmetric DIGItal Subscriber LInes.


Existing telephone infrastructure worldwide is being developed to offer communications capabilities to a wider audience and improve communication services to open up many new applications. ADSL is emerging as a key technological advancement increasing transmission bandwidths enabling new high-speed applications, improving quality of life and opening doors to electronic commerce. ADSL services today have limitations due to reliance on 50 years old technology to provide the required functions of safety isolation, impedance matching and signal filtering. This limits the distance over which ADSL can be transmitted and therefore restricts the availability of the service. Additionally, these old technologies are labour intensive and unreliable by the standards needed for the next 25 years. This project will overcome these inherited problems through the development of specialist new materials and manufacturing processes.

Current line interfaces implemented in ADSL equipment are based on 50 year old technology utilising a ferrite based line isolation transformer to provide an electrical safety barrier. The transformers typically are very large in volume and very heavy relative to the rest of the circuit and are very labour intensive in their manufacture. Specifically they have a volume of approximately 1.30sqcm, weigh 0.003kg and each take approximately 5 minutes to produce. As would be expected, a device which is virtually hand made also exhibits a fairly wide spread of key parameters such as inductance which is typically specified no better than +/-5% from nominal value.

In addition to the transformer, line interface circuits also include discrete inductor and capacitor components which combine to act as filters which attempt to improve the noise to signal ratio of the system and thereby the transmission capabilities of the system. The transformers and discrete components are traditionally mounted onto PCB's utilising conventional manufacturing techniques.

System performance is limited by the performance of the line interface circuit as a whole. The transformer is required by the signalling protocol to have a wide bandwidth extending from 30KHz to 1MHz+. Over this spectrum, all the properties of the transformer behaviour are evident some of which are undesirable. To achieve a wide enough bandwidth for example, low losses are critical consequently, a considerably high volume of copper is utilised. On the other hand as frequency increases the high volume of copper acts to degrade the performance of the device as Eddy currents generated within the copper wires act against the magnetisation currents. Similarly, a high volume of ferrite is required to transform the signals with minimum distortion but has the negative effect of increased size and costs. It is this manipulation of opposing effects that make the transformer only a compromise on the ideal and thus ultimately limits th

Work description:
Our research will be broken down into 7 stages covering a thorough definition of the required specification through to the management and exploitation of the new technologies created by the project. Initially we will define the required specification of materials according to analysis of the current state of the art. This will include an analysis of system requirements and will provide the key design criteria for the project. APC are ideally placed to carry out this work because of our experience and immediate access to the necessary ADSL equipment.

The second stage is focussed on creating models and simulations of 3 potential solutions, those being an improved solution based on optimising the current state of the art utilising a 3-D transformer, a new solution based on a novel "magnetic" approach which we are calling a 2-D transformer approach and also the third approach which is based on moving away from the state of the art completely by utilising Magneto-resistive elements to provide the electrical safety barrier. These models can then be used to assess the viability of each approach and to define the limiting factors which need to be overcome to create a suitable solution.

The work in the next three stages then use the results of the modelling to create prototype devices based on the three approaches. At this stage it will become clear which approaches are practical and those that are not. The focus will be set at this stage to the solution which offers the greatest chance of advancing the state of the art by the greatest degree.

The sixth stage will be totally focussed on a single approach and will consist of materials science work to create materials with the desired properties. By this stage of the project we will fully understand the requirements of the system and also the limitations of the chosen approach. Our work will consist of reducing the technical limitations of the solution to match the required specifications and meet the performance requirements of the system.

The final stage of the project will be the final verification of the device and complete analysis and reporting of its performance.

Funding Scheme

CRS - Cooperative research contracts


47 Riverside, Medway City Estate
ME2 4DP Rochester
United Kingdom

Participants (4)

Imperial College London
United Kingdom
Exhibition Road
SW7 2AZ London
Batalionow Chlopskich 1
83-000 Pruszcz Gdanski
United Kingdom
58 Clarendon Avenue, Leamington Spa
CV32 4SA Warwickshire
Birger Jarlsgatan 69 1Tr
113 56 Stockholm