Call our Sales Team on 0208 912 2120

Sump sampling simple

4th December 1982
Page 44
Page 45
Page 44, 4th December 1982 — Sump sampling simple
Noticed an error?
If you've noticed an error in this article please click here to report it so we can fix it.

Which of the following most accurately describes the problem?

Send it samples of your engine's oil and Shell will tell you if anything's wrong with your engine. David Wilcox tests the service

"THE OIL is the lifeblood of the engine and so our service is just like a blood test and equally valuable."

That is how Shell's David Dewar described the Shell CARE service. The CARE part of the name stands for Condition Assessment Reporting for Engines, which summarises the object of the service.

The use of engine oil sampling as a way of monitoring engine condition is well-established in the aviation industry; indeed the Civil Aviation Authority makes it compulsory for most airliners.

Shell introduced its oil sampling service to road transport operators in France eight years ago and it has since spread through Shell companies in Holland, Belgium and Australia. It came to the UK two years ago having been suitably adapted for UK vehicles and operating conditions.

The service is open to any road transport operator; he does not have to be using Shell oils. David Dewar, who has overall responsibility for Shell CARE, told me that there are nearly 100 companies with 1,000 engines being monitored who subscribe to it and the list includes some well-known names in the industry.

There are freight operators such as Cory Distribution and Christian Salvesen, public sector fleets such as Cheshire Council and Thames Water Authority plus own-account companies like Corona Soft Drinks.

There is also a noticeably high proportion of customers in the psv sector such as National Travel, Strathclyde PTE and several NBC subsidiaries. And every new DAF vehicle sold in the UK gets the service free during its first year.

To find out why these and other subscribers use Shell CARE I visited the laboratory where the oil samples are analysed to see what information can be extracted.

The subscribers do not use the service for every vehicle in their fleets, just those that they are particularly interested in such as ones that are covering high mileages or operating arduous schedules. A monthly sample is normally recommended.

On joining the service the subscriber buys an oil sampling kit which contains a syringe, small (50m1) plastic sample bot

tles and a supply of plastic tubing. The tubing is cut to the length of the engine's dipstick plus 30cm (12in) and attached to the bottle and syringe arrangement.

The dipstick is removed and the tubing inserted instead; the syringe is designed so that the oil sample passes directly into the bottle and does not enter the syringe itself. The engine oil must be hot when the sample is taken and a fresh length of tubing is used for subsequent samples so that each one is uncontaminated.

The operator is provided with a supply of padded envelopes and pre-paid address labels and he simply sends the sample bottle, along with information such as mileage covered since the last oil change and total mileage, to the laboratory. Already on record will be details of the vehicle and engine, plus the brand and grade of oil used.

Shell chooses to use a commercial laboratory which specialises in the fast turnround of oil samples, Spectro Oil Analysis near Chobham in Surrey. The sample arrives here the day after it was taken and Spectro will do the lab work on it immediately, using spectometric analysis.

David Dewar stresses that the analysis and subsequent diagnosis is not primarily to establish the condition of the oil (although this information is also revealed) but to look at the engine itself and detect developing problems at an early stage.

Regular sampling (such as the recommended monthly frequency) is necessary so that a trend can be established; once the trend is identified any deviation can be spotted and preventative action taken.

The two main areas of analysis are the top and bottom of the engine. The top analysis deals with the combustion function, such as testing for fuel or water in the oil and checking the injection system by measuring soot and solubles in the oil. The bottom end work is looking at engine wear rates such as bearing, piston and liner wear.

Having discovered that certain types of wear are occuring, the Shell CARE service reaches its diagnostic stage. It has built-up over the years "critical wear rates" for most diesel engines in common use on UK roads and this information is stored on Shell's mainframe computer at Wythenshawe, Manchester.

When the chemists at Spectro have finished their analysis the results are keyed into a computer terminal at the lab and the mainframe computer compares each parameter with the critical wear rates for the engine type, taking into consideration the mileage covered.

If all is well the computer comes back with a "green" rating and a green card is sent to the subscriber, so that within two or three days of taking the sample he has confirmation that the engine is healthy.

If the computer reveals something that is out of the ordinary but not needing urgent attention, it will give an "amber", and the subscriber gets an amber card describing the abnormality and recommending the likely area to investigate for the cause. Finally, if a serious fault is picked up the subscriber immediately receives a telephone call or telex to warn him; within 24 hours of despatching the sample the subscriber knows not to send the vehicle out if he can avoid doing so.

Spectro's chief chemist Colin Stevens-Hoare showed me how the analysis is carried out. The first step is a relatively simple viscosity check where a small amount of the oil is injected into a viscometer which measures the time the oil takes to pass down a small bore tube.

The oil should still meet its original viscosity specification; if it is thicker this can suggest oxidisation of the oil due to overheating. Oxidisation means the depletion of the additives in the oil, reducing its lubricity at high temperatures and possibly causing piston ring stick.

Then, some of the sample is heated to 180 degrees Centigrade in a small enclosed chamber and a flame put to the air space above it. If there is a flash this signifies that there is some fuel in the oil and hence the possibility of an injection fault. The next test measures any water that may be present in the oil, expressed in parts per million or as a percentage.

A test for dispercency follows and this is the major indication of condition of the oil itself. Drops of oil are put on some specialised, high grade blotting paper and baked in an oven for an hour. The way in which the drop spreads is compared with a control sample of clean Shell Myrina oil.

Finally comes the major test using the spectrometer. A small sample of the oil is put in the instrument's chamber and a high current passed through it to a terminal a couple of millimetres above the oil. This produces a spark which burns the oil. The light given off is split up into the various wavelengths and the spectrometer then converts the amount of light at each wavelength into an electrical current which is easily measured.

Each element in the oil gives off a different light wavelength and so the spectrometer readings correspond to the amount of each element in the oil.

The spectrometer detects 15 different elements. The first four are the oil's additives; zinc, phosphorus, calcium and barium, and the exact proportions of these vary according to the • brand and grade of oil. The other 11 elements are the crucial ones; they are the impurities that can occur to indicate the engine wear.

They are measured in parts per million proportional to the 1,000 miles of service.

These elements are magnesium, aluminium, iron, chromium, copper, lead, tin, silica, sodium, boron and vanadium. The abnormally high presence of any one of these elements when compared with the previous months' samples and the overall trend is a warning sign.

For instance, a high iron reading suggests cylinder liner wear while high copper and lead figures point to bearing wear. Silica present in the oil comes from dust particles in the air and so higher than normal silica readings are likely to be caused by inefficient air filtration.

David Dewar gave an example of very high silica readings showing up in a new coach engine. It turned out that the air cleaner trunking had been disconnected while the chassis was being bodied and never reconnected. The fault does not have to be that obvious — a split in the trunking would be detectable.

All the results are keyed into the lab's computer terminal and the Wythenshawe computer does the necessary comparisons to come up with the green, amber or red classification mentioned earlier.

To see the system in operation I took two samples with me. The tests could not be duplicated exactly because the lab had no previous history of the particular individual engines and so could not examine the trend, just examine the one-off sample. I merely labelled the samples A and B.

The first sample was from a Leyland Leopard coach that had covered 259,000km (161,000 miles) and it was just 4,000km (2,485 miles) since the oil was changed. The analysis correctly identified the oil as 15W 40 grade in good condition (as one would expect with a relatively new oil). A small amount of fuel was found in the oil which is not unusual for a Leopard engine and was not considered serious. The silica level was commendably low, indicating good air filtration.

Water was detected in the oil at a level of just 898 parts per million, less than 0.1 per cent and so was passed as satisfactory. The amounts of copper and lead were slightly on the high side but not too bad and this reflected the bearing wear to be expected on an engine that had covered 259,000km.

The second sample was from a Dodge 50-Series with a small Perkins diesel engine that had covered just 27,350km (17,000 miles). The oil was 2,900km (1,800 miles) old. As one would expect, the analysis gave this engine a clean bill of health all round.

As always, the 64,000 dollar question is how much does the service cost? The subscribersigns a contract for a year and each sample costs £12, adding up to a bill of £144 for the 12 monthly samples for one engine in one year. This includes the supply of sample bottles, tubing, etc. Price of the initial sampling kit containing the syringe is £20.

David Dewar says this means that the annual cost of analysing the oil and diagnosing the engine's condition is roughly the same as the annual cost of lubricating that engine.

Approximately five per cent of the samples result in a "Red" classification, 18 per cent are "Amber", the remainder are "Green". Each red sample is potentially an engine saved — an obvious benefit.

The detection of less immediate points in the amber classification is equally important Small amounts of fuel, water or silica, for instance, in the oil reduce its ability to lubricate and so lead to greater engine wear which is confirmed in the analysis by the presence of the impure elements in the sample.

It is therefore not unreasonable to claim that the early detection and highlighting of these points should mean reduced operating costs and longer engine life.

Shell CARE is growing at a similar rate in the UK now as it did in France several years ago and there are plans to expand the service into other countries. Shell also intends to extend Shell CARE into the analysis of transmission oil.

comments powered by Disqus