Let us know a little more of these lamps before going into the filament & the wire manufacture,. The general public often mistakes the fill gas to be "Halogen". The fact is that the gas is usually Argon, Iodine & or Bromine which contain traces of Halogen. The conventional light source has a tungsten filament & so do these lamps. The tungsten atoms boil off the filament & are deposited on the glass envelope in case of ordinary incandescent lamps. In case of these lamps, there is a process of Vapour Phase Transport, which occurs due to the evaporated tungsten reacting with the "Halogen gas" and forming tungsten halide. This tungsten halide evaporates easily & breaks down when it comes back in contact with the intense heat of the filament releasing the tungsten back on to the filament. The tungsten hence gets re deposited on to the filament rather than collecting on the glass bulb. This process is conventionally called the Halogen cycle. The halogen cycle extends the life of the filament & allows it to burn hotter & also keeps the inner surface of the bulbs clean. In having the surface clean, we have close to full brightness through out the life span of the lamp and we do not see any ageing effect. The name is thus derived as a Halogen lamp.

In order that the halogen cycle work, the bulb surface temperature must be hot & generally averages over 250 Deg Celsius below which the Argon fill gas may not adequately vaporise and or fail to react with the condensed tungsten on the bulb wall. This indirectly means that the bulb has to be small & made of either quartz or high strength heat resistant grade of glass known as Hard Glass for rated wattage’s upto 50 W. To improve performance the bulbs can be filled with gas at comparatively high pressures of around 20 bars. This slows down the evaporation of the filament. In addition, the small size of the bulb sometimes makes it economical to use premium fill gases such as the KRYPTON & XENON instead of the cheaper Argon. The high pressures and better fill gases can extend the life of the bulb and or permit a higher filament temperature that results in higher luminous efficacy of 5 to 10 % & colour temperature by about 100 Deg Kelvin. The luminous efficacy increases with higher tungsten temperature. Use of premium fill gases can also result in less heat being conducted from the filament by the fill gas; meaning more energy leaves the filament by way of radiation; meaning a slight improvement in the efficiency. The xenon gas fill can only be used for low voltage lamps as the lower ionising energy of xenon would lead to an electrical discharge with higher voltages.

Halogen lamps are 10 to 20 percent more efficient than an ordinary bulb of similar voltage & life expectancy sometimes doubled & or also tripled.

The filament geometry & dimensions depend on many factors in a complex way. The lamp structures itself are various. The failures of these lamps are often due to the hot spots, which can develop in the filaments. These are as a consequence of improper wire drawing & or non-homogeneous material. The current through the wire is in direct proportion to the applied voltage & the resistance inversely proportional to voltage. The re deposition of the tungsten is almost uniform but for certain cold zones where it can be slightly higher. The re deposition of tungsten does not mean that the weak spot in the filament shall be corrected. On the contrary, the evaporation loss at this spot could be higher and the failure early. The uniformity of the wire and the filament geometry in case of these lamps is hence of utmost importance. This is why the wires for these lamps are to be treated with caution.

WIRE DRAWING

The quality of wires used shall tilt the scales between quality lamps & others.

Tungsten, atomic number 74 in the periodic table, has a high melting point of 3410 deg cen. The atomic weight is 183.85 & the density is 19.26 g/cm*3. The crystal structure is a Body Centred Cubic.

People in the industry say that wire drawing is an art whereas I would like to mention that it is a science. Individual operators art for drawing wires fail to get consistent quality wires when you are operating on a three shift basis & more so in case of the cleaned and straightened tungsten wires. It was at this juncture that we decided to take prune up the existing process & make it operator independent.

This process is suitable for pure metals with high specific resistivity where we can harness the self-heating property of the material with a small electric current passed through it. The wire is to be heated to just above the DBTT Value. The rises in wire temperature for a given diameter, reduction percentage, length and the operating line speed are theoretically calculated. The preheat temperature requirement varies again with wire diameter, reduction ratio, die temp, line speed and the percentage increase in tensile strength required at each stage. The re crystallisation temperature required to anneal the wires vary for different metals & also depend on the wire diameter. Presently, we would like to restrict the usage of this technology to draw wires between 650 to 185 Microns and in case of annealing, straightening, electrolytic cleaning and inert gas cleaning of surface, to wire sizes from 300 to 100 Microns.

A few aspects which ought to have to been implemented but have not been, in most hot drawing machines, are

-The die heating to be electrically controlled. This ensures more uniform heating and precise control of the temperature is possible. Our observation is that the die wear is also more uniform.

-I would prefer that the colloidal graphite used, as lubricant to be circulated and not kept in a POT, which is constantly topped. The variation in the lubricant density can cause variation in the wear & tear of the die’s and consequently affect the wire sizes. It has also got to be ensured that the lubricant is free from dust particles & or is not exposed to dusty atmosphere.

-The enclosure made to avoid the wind has to embrace the die box also. The wind can vary the die temperature too.

We have tested the wires at various conditions of Temperatures -die and wire, speeds & reduction percentages. Any one of the parameter changes can bring about the change in the final wire specification.

To ensure that the die ovality is at the minimum, we propose to use the Poly crystalline diamond dies & not natural diamond dies. Well, the selection of the grain sizes required for different finishes plays a very important role. Well now you know about one of the contact points – natural diamond which is the purest form of carbon, does not conduct electricity.

The die’s play a very important role in the process of wire drawing. The entry cone angle and the bearing length’s have to be specified depending on the application to get the best out of the die. In case these parameters are not taken care, we shall land up spending more on the die’s and also wasting the process time as the wire shall not be as per our requirement for Halo coils.

The die temperature & the wire temperatures have to be synchronised to ensure that the wire gets drawn in the die and the die temperature does not increase with the wire temperature as the wire shall be at comparatively high temperatures

The manner in which tungsten has been processed prior to your having received it also play’s a very important role and we have to have it from units which can guarantee you the consistency. Why we say this is because tungsten, unlike other metals, even after annealing does not make up for the inconsistencies in the drawing stages. In colloquial terms - it as a memory metal.

WIRE DRAWING COMPARISION

Comparison of Tungsten Wire drawing on a single draw machine.

WIRE STRAIGHTENING

The straightness in a wire depends on the stress concentration in any given cross section of the wire. The wire tends to curl in the direction where the stress accumulation is high. Well, this is assuming the die through which it is drawn is round and with the least ovality as this in itself can lead to difference in stresses.

In the conventional processes, the wire is drawn to a high tensile value, annealed and further drawn again to the required size and TS. The bent tube & stretching after cleaning using external heating are the two most common methods adopted and have their own disadvantages. Well, even the eddy current systems used for testing the splits can be in a way used to heat & relieve the stresses from the surface. The limitation in this system is the size of the wire till which this principle can be used effectively. At lower sizes, the RF currents cancel out. There are no alternatives to the system we have proposed.

The self-heating of the wire automatically distributes the stress uniformly across the cross section of the material when stretched under controlled back tension to get the required straightness.

The system proposed uses the self-heating property of the material when a small electric current is passed through it. At reasonable speeds and with uniform diameter wire, we have been able to achieve the best and consistent results. The basic advantage of using this system is that the current heats the wire without discrimination of any side and or the core. The concept is nothing new as people have been trying to perfect these techniques for ages and a few have been using it for the wrong application. Well, in all the previous processes that we have been told and a few of which we have tried ourselves - using dry tungsten wool, mercury pool and or pulleys, there are disadvantages of their own. Dry tungsten wool, we have arcing seen as the wire leaves the contact and in case of the pulleys the temperature rise at the hot end can form groves within no time. In case of mercury - contact with and or exposure to open atmosphere even at room temperatures -let alone high temperatures can be very harmful and hazardous. In the system proposed we have been able to overcome all these problems. In our system the contact even at the "hot end" does not see the high temperature of the wire as the heat gets quenched.

There are no carbon brushes used and the mercury used as an auxiliary contact is in an enclosed chamber away from the heat, not exposed to air and in no way comes in contact with the operators functions. We have worked on this system for nearly a year and have found that there is literally no wear and tear on the system.

The prerequisite in this process is that the tensile strength of the tungsten wire has to be 20 percent above that required of the final wire size. Well, yes, the tensile strength of the material required has to be controlled during the entire process of drawing and cleaning. We insist that the wire is drawn and in case the material has got stretched unevenly outside the die, the self-heating of the wire shall only indicate hot spots throughout the length.

SURFACE CLEANING

The wires required for Halogen lamps are to be thoroughly clean. The process involves removal of the drawing lubricant from the surface through a process of electrolytic etching. The amount of removal of the material depends on the cleanliness required. In the ordinary course 6% by mg weight is the cleaning adopted in the industry. Even higher cleaning can be done in case it is warranted –but preferably in two stages. In the system configured we have two sets of electrodes. The electrolyte is - Potassium hydroxide, which is continuously circulated.

The electrolyte concentration, the temperature rise in the bath with time, the line speed and the operating currents are factors which can decide the reduction and or variation in the size of the cleaned wire. One other factor is that the wire has to be consistently at the centre of the electrode and that we should not have any oscillation occurring during the process to ensure uniform etching. The chemical flow rates are also critical and have to be uniform. The use of specialised pumps can be of great help. To ensure consistent flow of electrolyte within the electrodes, we have the tube flooded with electrolyte always. Tungsten carbide inserts are used at either end to ensure that the wire is held at the centre of the electrodes. The electrodes themselves are of special material to ensure that the electrode material & or impurity is not passed on to the wire.

The chemically etched wire is washed in water. The wire then passes through a heated inert chamber to vaporise any residual chemical and or water on the surface of the wire. Both these operations are done in a single pass there by avoiding need for additional table and process.

OTHER APPLICATIONS

INSIGHT

It all started as a small-scale unit for reprocessing the wires. The intricacies involved in the processes were beyond the comprehension of the staff with more than 80 years of combined experience. When they failed to achieve the requisite quality, I started working on each of the processes involved with the manufacture of the CAST Wires.

I have a prototype machine on which the different variations can be seen.

The data sheets on the precise current requirements at different stages of drawing have been compiled in excel to facilitate the computation at different speeds & reduction percentages.

The overall project feasibility has also been computed considering certain sizes of wires generally used in the industry.

I am a free lance chartered engineer with work experience at Kirloskar Electric Co - power electronics division and then on with ABB Ltd in the metallurgical projects group. At KEC it was R&D & application engineering and at ABB system engineering and process automation.

No 623 7^th Main ISRO Layout

Vikramnagar Bangalore – 560 078

Telefax: 91-080-6662316

E mail: [email protected]