The tough truth - Wear resistant coatings using HVOF
TWI Bulletin, January/February 1995
With a degree in Chemical Engineering, David Harvey joined TWI in 1986. His early activities covered process and materials problems related to MIG and TIG welding, followed by development and application of process control systems, such as backface penetration control. Since 1990, David has overseen the development of TWI's thermal spraying activities, in particular the installation of high velocity oxyfuel spraying facilities. He is currently a Technical Specialist in Surface Engineering in the ALS Department with responsibilities for promotion of industrial applications for HVOF coatings, process development and technology transfer. In May 1994, he became Chairman of the Surface Engineering Society.
High velocity oxyfuel (HVOF) spraying has become established as a leading technology for the deposition of wear resistant coatings, suitable for a wide range of industrial applications. David Harvey reviews recent developments in the following key areas:
- HVOF process basics, and the significance of spraying parameters
- Comparison with other thermal spraying processes
- Taguchi experimental design for optimisation of coating wear performance
- Industrial applications of HVOF spraying
- New developments in HVOF technology and coating materials
Note to readers
Since writing this article, TWI has acquired JP5000 and Diamond Jet systems, in addition to the TopGun HVOF equipment described.
Thermally sprayed coatings
Thermally sprayed coatings are generated by propelling fine, molten (or softened) particulate materials on to a prepared substrate. Coatings are normally characterised in terms of adhesion, porosity levels and oxide content. Generally, the higher the particle velocity the better the adhesion, the higher the density and the lower the oxide content of the coating.
The bond between the coating and the substrate is primarily a mechanical bond, although some coating materials are fused after spraying, and claims are made that some metallurgical bonding occurs with the higher velocity processes. Adhesion is generally enhanced by roughening the substrate surface by grit blasting or rough machining. Unlike welded coatings, there is no dilution of either coating or substrate by fusion, and the melting point of the coating material can be higher than the substrate. Other advantages of spraying include little or no pre- and post-heat treatment and minimum distortion of the substrate.
There are five principal thermal spraying process types in general use, listed here in order of decreasing particle velocity:
- Detonation flame spraying
- High velocity oxyfuel (HVOF)
- Air plasma
- Arc
- Flame
High velocity oxyfuel spraying
The most recent addition to the thermal spraying family, high velocity oxyfuel spraying has become established as an alternative to the proprietary, detonation (D-GUN) flame spraying and the lower velocity, air plasma spraying processes for depositing wear resistant tungsten carbide- cobalt coatings. HVOF spraying differs from conventional flame spraying in that the combustion process is internal, and the gas flow rates and delivery pressures are much higher than those in the atmospheric burning flame spraying processes. The combination of high fuel gas and oxygen flow rates and high pressure in the combustion chamber leads to the generation of a supersonic flame with characteristic shock diamonds. Flame speeds of 2000 m/sec and particle velocities of 600-800 m/sec are claimed by HVOF equipment suppliers. A range of gaseous fuels is currently used, including propylene, propane, hydrogen and acetylene. Depending on the fuel type and flow rate, up to 100kW is generated in the combustion chamber of an HVOF system. Five types of gas-fuelled HVOF system are currently available in the UK:
CDS
DIAMOND JET
HV2000
JET-KOTE
TOP GUN
Although very similar in principle, potentially significant details, such as powder feed position, gas flow rates and oxygen to fuel ratio, are apparent between each system. Powder is fed into the combustion chamber in the CDS, HV2000 and TOP GUN systems, and into the exhaust barrel in the JET-KOTE and DIAMOND JET systems at different distances from the chamber, Fig.l. TOP GUN and HV2000 HVOF systems are capable of spraying high melting point materials, eg. Al 2O 3 which clearly demonstrates that feeding powder into the combustion chamber maximises heat transfer to the particles.
The wide range of materials' melting points is accommodated by selection of powder size and combustion chamber length. In general, high melting point ceramic materials have a fine powder size (5-15µm) and require a longer chamber to soften the particles. Low melting point materials, have a coarser particle size (15-45µm) and use a shorter combustion chamber. An inert gas, usually argon (but occasionally nitrogen), carries the powder coaxially through the centre of the combustion chamber. High powder carrier gas flow rates are often used to reduce heat transfer to lower melting point materials. HVOF coatings are generally characterised by the highest bond strength (>70MPa), lowest porosity (<1%) and lowest oxide (1-5%) content of the conventional spraying processes, Table 1.
Table 1 Principal thermal spray processes and coating characteristics
| Process | Particle
velocity,
m/sec | Adhesion,
MPa | Oxide
content,
% | Porosity,
% |
|
| HVOF | 600-800 | >70 | 1-5 | 1-2 |
| Air plasma | 150-300 | 20->70 | 1-5 | 2-8 |
| Arc | 100 | 10-20 | 5-20 | 5-15 |
| Flame | 40 | 6-10 | 5-15 | 5-20 |
|
Wear testing of thermally sprayed coatings
A major application area for thermally sprayed coatings is for abrasive wear resistance. The two principal wear tests used are the ASTM G65 rubber wheel abrasion test, which measures high stress abrasion, and ASTM G99 pin-on-disc test which measures sliding wear rate. For certain materials, there is a ranking correlation between the results of the rubber wheel and pin-on-disc tests, but as with all wear testing, care should be exercised in interpreting the results.
Coating microstructure and wear resistance
For a tungsten carbide-cobalt coating, the optimum microstructure would appear to have a high level of retained WC particles, Fig.2. Coatings exhibiting poor wear resistance are often characterised by an amorphous, ribbon-like structure, with a high proportion of W 2C, Fig.3. The presence of W 2C is co-incident with low carbon content in the coating, and it would appear that WC is oxidised during spraying if the conditions are right (high oxyfuel ratio, high temperature/energy). Tungsten carbide should not melt at the temperatures encountered in HVOF spraying.
Fig.2. Micrograph of WC-Co with high proportion of WC
Fig.3. SEM photograph of WC-Co coating showing WC and amorphous phase
Taguchi experimental design
TWI has recently assessed the Taguchi experimental design method as a means of optimising thermal spraying procedure with a minimum of experiments. A good example is the use of the L 9 array, Table 2, which permits the investigation of four parameters at three levels, Table 3. As the array suggests, this can be completed with only 9 experiments, in contrast to a complete matrix of 3 x 3 x 3 x 3=81 experiments. The example used in this paper refers to the use of the TOP GUN HVOF system with a tungsten carbide-cobalt consumable.
Table 2 Parameter level settings for L 9 array for WC-Co coatings using propylene fuel gas
|
| Setting |
| Parameter | Label | 1 | 2 | 3 |
|
Fuel gas flow,
litre/min | FG | 45 | 55 | 65 |
Carrier gas flow,
litre/min | CG | 15 | 25 | 35 |
Oxygen to fuel
gas ratio | OF | 3.0 | 3.5 | 4.0 |
Combustion
chamber length,
mm | CC | 22 | 19 | 12 |
|
Table 3 L 9 orthogonal array
Trial
N°. | Parameter setting |
| FG | CG | OF | CC |
|
1
2
3
4
5
6
7
8
9 | 1
1
1
2
2
2
3
3
3 | 1
2
3
1
2
3
1
2
3 | 1
2
3
2
3
1
3
1
2 | 1
2
3
3
1
2
2
3
1 |
|
This example examined the influence of spraying parameters on pin-on-disc wear resistance and deposition efficiency. The responses are plotted to indicate trends graphically, Fig.4 & 5, and using an analysis of variance (ANOVA) programme it is possible to determine the relative strength of each parameter's influence, Table 4. The next step is to select the combination of spraying parameters, which will produce the 'best' coating, Table 5, and predict the test results of the confirmation coating, Table 6.
Fig.4. Parameter influence on pin-on-disc wear performance
Fig.5. Parameter influence on deposition efficiency
Table 4 Relative strength of parameter influence
Response,
% | FG | CG | OF | CC | Error |
|
| Pin-on-disc wear | 11 | 6 | - | 0.2 | 83 |
Deposit
efficiency | 75 | 1 | 21 | 2 | 1 |
|
Table 5 Summary of most influential parameter and settings to give best response Optimum
response
setting | FG | CG | OF | CC |
|
Minimum
wear rate | 1 | 2 | (1) | (3) |
Maximum
deposit
efficiency | 1 | (1,2,3) | 1 | (2) |
Predicted
best
performance | 1 | 2 | 1 | 3 |
|
Table 6 Predicted and measured characteristics of confirmation coating | Characteristic | Predicted
value | Measured
value |
|
Wear rate,
mg/hr | 5.7(±18.0) | 19.1 |
Deposit
efficiency, % | 63.3(±0.7) | 63.0 |
|
The variance between the actual and predicted values for the confirmation coating wear test suggests there is significant scatter in the wear test data - and that a much larger test sample is required. (It may also suggest that the wear test is unreliable!) The prediction for the deposition efficiency is remarkably accurate, and reflects the small scatter in the experimental data and reproducibility of the HVOF process. It is general experience that the Taguchi method is a useful technique for optimising deposit efficiency.
The Taguchi method has certain limitations, eg. it can only predict an optimum procedure which uses parameter levels explored in the array. In addition, there is often a strong interaction between parameters, which can be difficult to eliminate.
Application of HVOF spraying
Volume applications for the HVOF process include aeroengine and aircraft parts, Fig.6, gate valves, Fig.7, and engineering components such as wear plates, journals and bearing surfaces. The materials widely used are tungsten carbide-cobalt (WC-Co) and NiCr-Cr 3C 2, at elevated temperatures 800°C and for nuclear applications where cobalt is unacceptable.
Fig.6. Aircraft wing flap track, bearing surface coated with WC-Co
Fig.7. Gate valves coated with WC-Co
Other materials used include NiCrBSi, a spray and fuse composition often used on glass moulding components, Fig.8, ceramic coatings such as alumina and chromia for print/analox rolls, Fig.9, and hydroxylapatite (HA) for prostheses, eg. hip joint replacements, Fig.10.
Fig.8. Glass plungers coated in NiCrBSi
Fig.9. Printing rolls coated in Al 2O 3 or Cr 2O 3
Fig.10. HVOF spraying of artificial hip joint with hydroxylapatite coating
Fig.11. Schematic of liquid fuel combustion HVOF system (Model JP5000)
New developments in HVOF spraying
Hobart-Tafa has recently launched a liquid-fuel (kerosene) JP5000 HVOF system, Fig.11, which is capable of much higher deposition rates than the conventional gas-fuelled units. WC-Co coating microstructures are promising, and the other leading thermal spraying equipment manufacturers will not be far behind in developing their own versions. At the same time there is a drive towards improved quality control with increasing use of mass flow control systems.
Novel carbide compositions such as vanadium and titanium carbide are currently being assessed as alternatives to tungsten and chromium carbide. Another material recently sprayed is yttria stabilised zirconia (YSZ), a material with a very high melting point (2700°C), and not readily sprayed by the HVOF process. This material has traditionally been deposited on to aero and diesel engine components by plasma spraying for thermal barrier application, but the improved density achieved with HVOF may improve resistance to hot gaseous erosion, Fig.12.
Fig.12. Piston crowns coated with yttria stabilised zirconia (by HVOF)
