Mineral / Semi-Synthetic / Synthetic Oil ..... Viscosity Selection

[RichardS]

We've enjoyed many threads over the years focusing on the mineral/semi-synthetic/synthetic oil debate and my view has always been that modern fully synthetic oils are preferable to any other oil.

However, until recently, I have used semi-synthetic oil in older engines like classic cars and the boat because they usually require higher viscosity oils at the "hotter/summer" end of the scale such as 40 or 50 weight and such synthetic oils have been difficult to obtain and/or expensive. Those days seem to be over and I am now running my classic cars on 20W-50 synthetic and my boat on 15W-40 synthetic.

I was recently discussing oil viscosity specs with my Son as I sometimes ring the changes with modern engines although using gut-feel about what might be acceptable based on the manufacturers specs which usually show large amounts of overlap at different notional ambient temperatures, and he pointed out the following which I think is an interesting way of looking at it.

Imagine an engine where the ambient temperature recommendation, say 0 - 30 degrees for the UK, is the common oil of 10W-30 spec.

Now, that oil starts off on a winter's morning at viscosity 10W, but as it warms up it thins and goes to viscosity 5W and then viscosity 0W so 0W-30 and 5W-30 are both within the range of viscosity that the engine actually sees, even when you start off with 10W-30. However, the engine will never normally have to work with a cold viscosity of 15W so going high at the W end is not the best idea.

Similarly, at the other end of the scale, the oil which is at 30 at engine temperature will have previously been at 50 before it drops to 40 and eventually, when the engine is up to temperature drops to 30. However, the engine will never normally have to work with a hot viscosity of 20 so going low at the "hot" end is not the best idea either.

In summary, in this example, going down to a 0W / 5W or up to a 40 / 50 is still going to be within the range of what your engine operates under so if you can't get exactly what is recommended at the right price, then going low at the cold end or high at the hot end is not going to cause a problem, although the closer to recommended, the better.

Richard

 

[TernVI]

A lot of people never get their heads around the idea that the bigger the difference in the two numbers, the less the viscosity changes with temperature.
A 10W60 is more constant than a 20W50.
A straight SAE30 oil is treacle when cold, thinner than water when hot.

Some air cooled bikes are using 10w60, because the engine temperature varies more, and they have big thumpy pistons which are demanding on the oil in the big end.

5W30 synthetic is mostly about getting a bit less oil drag to emit a couple of g less CO2 here and there.
0W20 is taking that to extremes IMHO.

For my bikes, I go for a fully synth from a 'good' manufacturer.
If not needing to re-shim the valves is anything to go by, this seems to be working OK so far.

The car gets a Motul oil which meets the manufacturer's specific long life spec, although as the car is rapidly reaching 'shed' status, I picked up some bargain stuff to the same spec at Tesco a while back.

Tidying my garage, I find I have 5 different, less than half full, flavours of Fork OIl, from 0 upwards.
Some fork designers seem to use the plan of speccing the thinnest possible oil so it can't get much runnier when hot

 

[Stewproud]

I have been advised by Nanni not to use synthetic oil in my Nanni N4.60 engines. The oil must be straight mineral oil.

 

[reeac]

Careful now. The notation is such that, for example, a 10w40 multigrade has the same cold viscosity as a straight monograde 10 oil and the same viscosity as a straight 40 oil at about 100C. The numbers do not denote actual viscosity, they compare with straight monograde oils at the same temperature. In all cases the actual viscosity falls rapidly with increasing tempersature but less rapidly in the case of multigrades. Graphs of actual viscosity versus temperature are available on the internet.

 

[Jmc1764]

On the theory that 95% of engine wear is on first start up, consider what gets around an engine faster , sewing machine oil or peanut butter ?

 

[RichardS]

I have been advised by Nanni not to use synthetic oil in my Nanni N4.60 engines. The oil must be straight mineral oil.

I find that very surprising. As Reeac says, it is almost inevitable that a straight weight oil will be either too thick to be optimal when cold or too thin when hot to be optimal. The engines I have worked on have specified multigrade oil since the 1960's.

All the web references I can find to Nanni N4.60 engine specify multigrade.

Just a thought ..... maybe you mean straight as in ordinary mineral oil, in which case your engine would be better off with the same grade of synthetic oil but Nanni have simply not tested it. It's the same with my Yanmars and, indeed, many boat engines and many car engines older than 20 years. However, they would all be better off with synthetic oil but the manufacturer will never say that.

Richard

 

[Halo]

Careful now. The notation is such that, for example, a 10w40 multigrade has the same cold viscosity as a straight monograde 10 oil and the same viscosity as a straight 40 oil at about 100C. The numbers do not denote actual viscosity, they compare with straight monograde oils at the same temperature. In all cases the actual viscosity falls rapidly with increasing tempersature but less rapidly in the case of multigrades. Graphs of actual viscosity versus temperature are available on the internet.

Absolutely correct !

I think the killer for many engines is leaving them idling and thereby glazing the bores.

 

[RichardS]

Careful now.

You are correct about cold and hot viscosities but I'm not sure whom you are counselling?

Richard

 

[rotrax]

We've enjoyed many threads over the years focusing on the mineral/semi-synthetic/synthetic oil debate and my view has always been that modern fully synthetic oils are preferable to any other oil.

However, until recently, I have used semi-synthetic oil in older engines like classic cars and the boat because they usually require higher viscosity oils at the "hotter/summer" end of the scale such as 40 or 50 weight and such synthetic oils have been difficult to obtain and/or expensive. Those days seem to be over and I am now running my classic cars on 20W-50 synthetic and my boat on 15W-40 synthetic.

I was recently discussing oil viscosity specs with my Son as I sometimes ring the changes with modern engines although using gut-feel about what might be acceptable based on the manufacturers specs which usually show large amounts of overlap at different notional ambient temperatures, and he pointed out the following which I think is an interesting way of looking at it.

Imagine an engine where the ambient temperature recommendation, say 0 - 30 degrees for the UK, is the common oil of 10W-30 spec.

Now, that oil starts off on a winter's morning at viscosity 10W, but as it warms up it thins and goes to viscosity 5W and then viscosity 0W so 0W-30 and 5W-30 are both within the range of viscosity that the engine actually sees, even when you start off with 10W-30. However, the engine will never normally have to work with a cold viscosity of 15W so going high at the W end is not the best idea.

Similarly, at the other end of the scale, the oil which is at 30 at engine temperature will have previously been at 50 before it drops to 40 and eventually, when the engine is up to temperature drops to 30. However, the engine will never normally have to work with a hot viscosity of 20 so going low at the "hot" end is not the best idea either.

In summary, in this example, going down to a 0W / 5W or up to a 40 / 50 is still going to be within the range of what your engine operates under so if you can't get exactly what is recommended at the right price, then going low at the cold end or high at the hot end is not going to cause a problem, although the closer to recommended, the better.

Richard

I think - respectfully - that you have the wrong end of the stick here and there in your appreciation of what the specs do as the tempreture changes.

The way viscosities are measured is to time a set volume of oil through a sized orifice at differing tempretures.

Ergo, good old fashioned 20/50. When cold, 20 viscosity, when hot, the same viscosity as a 50 weight at the same tempreture.

This was achieved with long chain additives.

Your supposition that the 10/30 weight starts at 10 viscosity and then drops to 5 viscosity cannot hold water because the tempreture has changed from the tempreture the 10 was measured at.

Multigrade specs are with the lower number cold and the higher at normal operating tempreture.

You can assume the 10/30 drops from 10 to 5 as it warms up, but so what? We all know oil thins as it is heated. The viscosities between cold start and operating tempreture are pretty meaningless IMHO.

What is important is the viscosity at normal operating tempreture. The 10/30 started at ten cold and ends up at the equivilent of a 30 at operating tempreture.

I drove, in 1988, to deepest darkest Czechoslovakia, researching for a book I have not yet written. I left the UK in perhaps 6 degrees C, in Pardubice it was minus 17 and falling! Next morning, no start. The oil on the dipstick was like jam, good old Castrol GTX 20/50.

The local Autoservis - where my mate Jiri Smida worked as the auto electrician got me going and put in stuff like hot piss, suitable for Czech conditions. The car, an early Hyundai, was perfect after.

Until I got into UK conditions where it sounded very 'loose'. When drained it was like paraffin!

You are supposeing the viscosities are measured at the same tempreture, they are not when talking about multigrade oils.

My whizz bang, a Ford Fiesta 1.6 Diesel takes 5/30 synthetic all year round.

Low internal friction is an important eco requirement and this oil provides it.

I would have been no good at all in my old series 2 Landrover - it would have come straight out of the exhaust pipe as blue smoke if it was poured in with it running.............................

If you knew all this, apologies, but it was unclear to me - and apparently others - that you did.

 

[penberth3]

I think - respectfully - that you have the wrong end of the stick here and there in your appreciation of what the specs do as the tempreture changes.

The way viscosities are measured is to time a set volume of oil through a sized orifice at differing tempretures.

Ergo, good old fashioned 20/50. When cold, 20 viscosity, when hot, the same viscosity as a 50 weight at the same tempreture.

This was achieved with long chain additives.

Your supposition that the 10/30 weight starts at 10 viscosity and then drops to 5 viscosity cannot hold water because the tempreture has changed from the tempreture the 10 was measured at.

Multigrade specs are with the lower number cold and the higher at normal operating tempreture.

You can assume the 10/30 drops from 10 to 5 as it warms up, but so what? We all know oil thins as it is heated. The viscosities between cold start and operating tempreture are pretty meaningless IMHO.

What is important is the viscosity at normal operating tempreture. The 10/30 started at ten cold and ends up at the equivilent of a 30 at operating tempreture.

I drove, in 1988, to deepest darkest Czechoslovakia, researching for a book I have not yet written. I left the UK in perhaps 6 degrees C, in Pardubice it was minus 17 and falling! Next morning, no start. The oil on the dipstick was like jam, good old Castrol GTX 20/50.

The local Autoservis - where my mate Jiri Smida worked as the auto electrician got me going and put in stuff like hot piss, suitable for Czech conditions. The car, an early Hyundai, was perfect after.

Until I got into UK conditions where it sounded very 'loose'. When drained it was like paraffin!

You are supposeing the viscosities are measured at the same tempreture, they are not when talking about multigrade oils.

My whizz bang, a Ford Fiesta 1.6 Diesel takes 5/30 synthetic all year round.

Low internal friction is an important eco requirement and this oil provides it.

I would have been no good at all in my old series 2 Landrover - it would have come straight out of the exhaust pipe as blue smoke if it was poured in with it running.............................

If you knew all this, apologies, but it was unclear to me - and apparently others - that you did.

To be absolutely clear, the "20" isn't the viscosity is it? It's a grade number - and as you explain, the viscosity of the "20" or whatever will vary with temperature.

 

[TernVI]

I have been advised by Nanni not to use synthetic oil in my Nanni N4.60 engines. The oil must be straight mineral oil.

The term 'striaght' should be avoided at all costs.
It can mean two or more completely different things:
Monograde, e.g. SAE30
Detergent free
Neither of which are likely to be right for a modern yacht engine.

 

[RichardS]

I think - respectfully - that you have the wrong end of the stick here and there in your appreciation of what the specs do as the tempreture changes.

The way viscosities are measured is to time a set volume of oil through a sized orifice at differing tempretures.

Ergo, good old fashioned 20/50. When cold, 20 viscosity, when hot, the same viscosity as a 50 weight at the same tempreture.

This was achieved with long chain additives.

Your supposition that the 10/30 weight starts at 10 viscosity and then drops to 5 viscosity cannot hold water because the tempreture has changed from the tempreture the 10 was measured at.

Multigrade specs are with the lower number cold and the higher at normal operating tempreture.

You can assume the 10/30 drops from 10 to 5 as it warms up, but so what? We all know oil thins as it is heated. The viscosities between cold start and operating tempreture are pretty meaningless IMHO.

What is important is the viscosity at normal operating tempreture. The 10/30 started at ten cold and ends up at the equivilent of a 30 at operating tempreture.

I drove, in 1988, to deepest darkest Czechoslovakia, researching for a book I have not yet written. I left the UK in perhaps 6 degrees C, in Pardubice it was minus 17 and falling! Next morning, no start. The oil on the dipstick was like jam, good old Castrol GTX 20/50.

The local Autoservis - where my mate Jiri Smida worked as the auto electrician got me going and put in stuff like hot piss, suitable for Czech conditions. The car, an early Hyundai, was perfect after.

Until I got into UK conditions where it sounded very 'loose'. When drained it was like paraffin!

You are supposeing the viscosities are measured at the same tempreture, they are not when talking about multigrade oils.

My whizz bang, a Ford Fiesta 1.6 Diesel takes 5/30 synthetic all year round.

Low internal friction is an important eco requirement and this oil provides it.

I would have been no good at all in my old series 2 Landrover - it would have come straight out of the exhaust pipe as blue smoke if it was poured in with it running.............................

If you knew all this, apologies, but it was unclear to me - and apparently others - that you did.

I certainly knew all that but I'm afraid that you've misunderstood my point.

I totally agree that the viscosity specs are measured at standard temperatures and I'm comparing viscosities at different temperatures, but for the overall conclusion that does not matter.

Let's take the 15W oil. That 15 represents its viscosity at 0C (assessed using actual viscosity units) so let's assume that we are starting the car at 0C. When you start the engine and the oil warms up a little, and the viscosity drops to the same viscosity as a 10W oil, albeit that the 10W oil would have been at 0C at that viscosity. This principle works whatever temperature you start at.

The engine which has a recommended 15W oil therefore has effectively had time running with an oil of 10W viscosity (once again, using actual viscosity units) whereas the reverse is not true.

Richard

 

[penberth3]

I certainly knew all that but I'm afraid that you've misunderstood my point.

I totally agree that the viscosity specs are measured at standard temperatures and I'm comparing viscosities at different temperatures, but for the overall conclusion that does not matter.

Let's take the 15W oil. That 15 represents its viscosity at 0C (assessed using actual viscosity units) so let's assume that we are starting the car at 0C. When you start the engine and the oil warms up a little, and the viscosity drops to the same viscosity as a 10W oil, albeit that the 10W oil would have been at 0C at that viscosity. This principle works whatever temperature you start at.

The engine which has a recommended 15W oil therefore has effectively had time running with an oil of 10W viscosity (once again, using actual viscosity units) whereas the reverse is not true.

Richard

I think I know what you're trying to say but that's as clear as mud. If you start talking about "grades" and stop putting viscosity alongside the number it might make more sense. 15 grade is 15 grade whatever the temperature. Its viscosity varies with temperature, but 15 grade never becomes a 10 grade.

 

[RichardS]

I think I know what you're trying to say but that's as clear as mud. If you start talking about "grades" and stop putting viscosity alongside the number it might make more sense. 15 grade is 15 grade whatever the temperature. Its viscosity varies with temperature, but 15 grade never becomes a 10 grade.

I've never said that 15W grade becomes a 10W grade. I have consistently referred to the 15W viscosity or the 10W viscosity. Please read the OP again.

At least you agree with me that an engine with 15W oil does experience, during running, the viscosity which relates to a 10W oil but please try and be a little less confrontational so we can make this an interesting discussion about a point that I, and I suspect most others on here, had never realised in such clear terms.

Richard

 

[penberth3]

I've never said that 15W grade becomes a 10W grade. I have consistently referred to the 15W viscosity or the 10W viscosity. Please read the OP again......

You've said it again "10W viscosity". 10W is not a viscosity.

 

[Spirit (of Glenans)]

The term 'striaght' should be avoided at all costs.
It can mean two or more completely different things:
Monograde, e.g. SAE30
Detergent free
Neither of which are likely to be right for a modern yacht engine.

"Modern" yacht engines are generally derived from basic industrial diesels that have been around for some time.

 

[JumbleDuck]

You've said it again "10W viscosity". 10W is not a viscosity.

Does this help?

 

[rotrax]

To be absolutely clear, the "20" isn't the viscosity is it? It's a grade number - and as you explain, the viscosity of the "20" or whatever will vary with temperature.

AFAIK - and I certainly used to know when involved in both the Motor and Motorcycle industries - the grade number is an oil industry standard for setting the specification of lubricating oils.

A measured amount of the oil to be measured is allowed to pass, by gravity, through a sized orifice. The time taken sets where it is graded.

The same measurement at differing tempretures gives real world comparison for multigrade oils. IE, 20 when cold for startup but same as a straight 50 at operating tempreture.

See the table above.................................

 

[RichardS]

You've said it again "10W viscosity". 10W is not a viscosity.

It's simply my shorthand for the "actual viscosity in N-s/m2 or whatever unit you prefer applicable to the oil that is nominally called 10W".

I didn't think that was particularly confusing but, if it is, please substitute the sentence above.

Richard

 

[JumbleDuck]

It's simply my shorthand for the "actual viscosity in N-s/m2 or whatever unit you prefer applicable to the oil that is nominally called 10W".

"10W" is a maximum viscosity at a given temperature, not a value of viscosity.