jmw
Industrial
- Jun 27, 2001
- 7,435
This is an aside from the "Jet Fuel Bulk Storage" thread and responds to a question by FredGarvin. I have no doubt that there are specific issues here which might be more relevant to the "Engine and Fuel Engineering" forum but since the original query arises in this forum, it is where I will place it.
Fred asked:
[blue]"I noticed back in JMW's post in the referenced thread about Bunker C, his concerns regarding viscocity as a reliable measure of quality with heavy oil fuels.
Does this hold true for other fuel types? We base our daily inspections on specific gravity. My thought is that if it is a good indicator, perhaps we should be testing for viscocity as well."[/blue]
I can only comment on the importance of viscosity in heavy fuels and to observe that viscosity (and density) vary for refied fuels. I do not know how significant these variations will be on performance. Hopefully others can inform.
Viscosity is a valuable process measurement in many areas because it is a very sensitive indicator of change, more so than many other process measured properties, but only where the viscosity is significant i.e. due to the resolving powers of the measurement instruments.
But there are other factors to consider as well.
Viscosity is a valuable measurement with heavy fuel oils because they are blended based on the volume or mass ratio necessary to achieve a specific fuel viscosity. Fuel blend calculators are used to preduct various quality parameters at this blend ratio.
The viscosity is the most sensitive of the indicators such that, within the instrument resolution, if the viscosity is right then the other parameters will be within spec also. If we measure the density instead, then, within the resolution of the density measurement, it is possible that the viscosity and other parameters may be out of specification.
A point with heavy fuels is that the viscosity of the blend can be 180cSt, 380cSt or 500cSt.
Diesels, aviation fuels and other refined fuels are blended for other properties.
Niether density nor viscosity is a controlled parameter. These properties can vary according to the refining history.
Two sequential batches of aviation fuel/diesel/petrol of the grade and from the same refinery can have very different densities and viscosities but still be in spec.
To produce fuels which do have controlled density and viscosity properties, in addittion to their other properties, will add cost. I am not in a position to comment on the benefit. When racing formula 1 cars the lower the density the better if, per litre of fuel, the performance of the fuel is unimpared. Consistent viscosity might also give consistent mixing and combustion properties.
But is it signficant? i.e. is the cost benefit ratio acceptable?
Viscosity is certainly an important factor in injector design, or rather, manufacture, to be able to know or correct for viscosity when testing injectors.
At low viscosities e.g. 1-4cSt, the current viscosity technologies lose their advantage of resolution compared to density, for example, in indicating other properties if they are linked. Density can be measured to within +/-0.1 to 0.15kg/m3 while at 4cSt, viscosity to +/-0.1cSt no longer has the same resolving power for a process measurement.
None the less, both density and viscosity, even though uncontrolled variables, are valuable process measurements in interface detection. For example, when transferring petrol (gasolene) from production to storage, the differences between the densities of different batches enables the detection of the interface between grades for proper diverting of the products to the correct storage and the control of ullage. Viscosity is far less useful for these applications though for improved confidence in product identity of diesels and aviation fuels measuring both the density and the viscosity together provide a much better confirmation that the product measured is the product expected as these fuels usually do exhibit good resolvable viscosity variation. Petrol, probably not. That is, the viscosity can be measured reasonably well but not well enough to discriminate between batches.
If the original values of density and viscosity are recorded either or both may be useful in detecting quality changes or for identifying if the product is not what is expected.
Viscosity, though uncontrolled for refined fuels, does have an effect on spray patterns and/or flow rates, hence the references cited to power factor correction in the original thread.
In the case of any engines where there are uncontrolled fuel properties, the design and performance specifications reflect the tolerance of the engine for the range of expected conditions.
In Heavy fuel engines the variation in viscosity is significant and is controlled because of the performance benefits.
Turbines for marine applications normally burn MGO or equivalent. There is no control of the viscosity. However, if the turbine burns IF30 (30Cst fuel) then the benefits can be significant but the viscosity variation with quality and temperature are also significant and must be controlled.
The fuel cost differential more than balances the added equipment cost. For a high speed ferry operating in South Amrerica somewhere (don't recall where) the fuel cost savings were estimated at US$5000 per day burning IF30 compared to MGO.
Work to burn heavy fuel oils in turbines is in progress.
So viscosity measurement and control that enables cheaper fuels to be burnt is cost effective.
For a refined fuel, the difference between viscosity measurement and control and no control is more difficult to assess. I am not qualified to comment. This is where i suggest the question might be more appropriately addressed in the "Engine and Fuels Engineering" forum.
I don't know who may have done studies on this or who may have any data. Until now most viscometers were insufficiently accurate to make such investigations worth while. Viscometers are evolving at a significant rate but it is not just accurate and affordable measurement, it is what is doen with the measurement that adds cost.
JMW
Eng-Tips: Pro bono publico, by engineers, for engineers.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.
Fred asked:
[blue]"I noticed back in JMW's post in the referenced thread about Bunker C, his concerns regarding viscocity as a reliable measure of quality with heavy oil fuels.
Does this hold true for other fuel types? We base our daily inspections on specific gravity. My thought is that if it is a good indicator, perhaps we should be testing for viscocity as well."[/blue]
I can only comment on the importance of viscosity in heavy fuels and to observe that viscosity (and density) vary for refied fuels. I do not know how significant these variations will be on performance. Hopefully others can inform.
Viscosity is a valuable process measurement in many areas because it is a very sensitive indicator of change, more so than many other process measured properties, but only where the viscosity is significant i.e. due to the resolving powers of the measurement instruments.
But there are other factors to consider as well.
Viscosity is a valuable measurement with heavy fuel oils because they are blended based on the volume or mass ratio necessary to achieve a specific fuel viscosity. Fuel blend calculators are used to preduct various quality parameters at this blend ratio.
The viscosity is the most sensitive of the indicators such that, within the instrument resolution, if the viscosity is right then the other parameters will be within spec also. If we measure the density instead, then, within the resolution of the density measurement, it is possible that the viscosity and other parameters may be out of specification.
A point with heavy fuels is that the viscosity of the blend can be 180cSt, 380cSt or 500cSt.
Diesels, aviation fuels and other refined fuels are blended for other properties.
Niether density nor viscosity is a controlled parameter. These properties can vary according to the refining history.
Two sequential batches of aviation fuel/diesel/petrol of the grade and from the same refinery can have very different densities and viscosities but still be in spec.
To produce fuels which do have controlled density and viscosity properties, in addittion to their other properties, will add cost. I am not in a position to comment on the benefit. When racing formula 1 cars the lower the density the better if, per litre of fuel, the performance of the fuel is unimpared. Consistent viscosity might also give consistent mixing and combustion properties.
But is it signficant? i.e. is the cost benefit ratio acceptable?
Viscosity is certainly an important factor in injector design, or rather, manufacture, to be able to know or correct for viscosity when testing injectors.
At low viscosities e.g. 1-4cSt, the current viscosity technologies lose their advantage of resolution compared to density, for example, in indicating other properties if they are linked. Density can be measured to within +/-0.1 to 0.15kg/m3 while at 4cSt, viscosity to +/-0.1cSt no longer has the same resolving power for a process measurement.
None the less, both density and viscosity, even though uncontrolled variables, are valuable process measurements in interface detection. For example, when transferring petrol (gasolene) from production to storage, the differences between the densities of different batches enables the detection of the interface between grades for proper diverting of the products to the correct storage and the control of ullage. Viscosity is far less useful for these applications though for improved confidence in product identity of diesels and aviation fuels measuring both the density and the viscosity together provide a much better confirmation that the product measured is the product expected as these fuels usually do exhibit good resolvable viscosity variation. Petrol, probably not. That is, the viscosity can be measured reasonably well but not well enough to discriminate between batches.
If the original values of density and viscosity are recorded either or both may be useful in detecting quality changes or for identifying if the product is not what is expected.
Viscosity, though uncontrolled for refined fuels, does have an effect on spray patterns and/or flow rates, hence the references cited to power factor correction in the original thread.
In the case of any engines where there are uncontrolled fuel properties, the design and performance specifications reflect the tolerance of the engine for the range of expected conditions.
In Heavy fuel engines the variation in viscosity is significant and is controlled because of the performance benefits.
Turbines for marine applications normally burn MGO or equivalent. There is no control of the viscosity. However, if the turbine burns IF30 (30Cst fuel) then the benefits can be significant but the viscosity variation with quality and temperature are also significant and must be controlled.
The fuel cost differential more than balances the added equipment cost. For a high speed ferry operating in South Amrerica somewhere (don't recall where) the fuel cost savings were estimated at US$5000 per day burning IF30 compared to MGO.
Work to burn heavy fuel oils in turbines is in progress.
So viscosity measurement and control that enables cheaper fuels to be burnt is cost effective.
For a refined fuel, the difference between viscosity measurement and control and no control is more difficult to assess. I am not qualified to comment. This is where i suggest the question might be more appropriately addressed in the "Engine and Fuels Engineering" forum.
I don't know who may have done studies on this or who may have any data. Until now most viscometers were insufficiently accurate to make such investigations worth while. Viscometers are evolving at a significant rate but it is not just accurate and affordable measurement, it is what is doen with the measurement that adds cost.
JMW
Eng-Tips: Pro bono publico, by engineers, for engineers.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.
