Varying oil pressure

[RogerH]

I see it as a stop-go process on each 4-stoke cycle.

 

Hi Pete,

that is how I see it also.

The greater the pressure the quicker the flow can be reintroduced and give the benefits of the flow.

 

Most of the 'contact' is on the upper shell of the big end for three of the 4 movements. It is only at the top of the exhaust stroke that the piston/con-rod keeps moving up when the crank is beginning to go down. This is where the new oil flow comes in to get things back in order. Very simplistic

 

Roger

[Peter Cobbold]

I see it as a stop-go process on each 4-stoke cycle.

 

Hi Pete,

that is how I see it also.

The greater the pressure the quicker the flow can be reintroduced and give the benefits of the flow.

 

Most of the 'contact' is on the upper shell of the big end for three of the 4 movements. It is only at the top of the exhaust stroke that the piston/con-rod keeps moving up when the crank is beginning to go down. This is where the new oil flow comes in to get things back in order. Very simplistic

 

Roger

Hi Roger,

This looks useful:

http://www.epi-eng.com/piston_engine_technology/engine_bearings.htm

- its for main bearings .

Peter

Edited January 1, 2018 by Peter Cobbold

[DaveN]

@Peter

That article made fascinating reading. Are crankshafts ground oval? I doubt it.

Also states one manufacturer uses moly/graphite blend in a PTFE substrate.....hhhhhmmm where have we seen that before ‘snake oil’?

[littlejim]

many years ago I was lent, by my Kiwi flatmates, their old bomb, while they went back to Kiwiland for the Uni holidays.

The big ends were rather worn, and after 35 miles of driving the oil pressure actually dropped to zero on the gauge.

by then the oil was obviously hotter and thinner, and I guess the big ends had heated up and expanded the gap so there was less/no resistance getting back to the oil pump.

Stopping and letting everything cool down meant I started off with good oil pressure again and the cycle repeated itself.

With reference to John's comment, I would have thought the "flow" through the bearings was at its peak at zero pump pressure, this and the the thinner oil, offered less resistance to the pump and the oil pressure gauge.

Edited January 1, 2018 by littlejim

[Motorsport Mickey]

@Peter

That article made fascinating reading. Are crankshafts ground oval? I doubt it.

Also states one manufacturer uses moly/graphite blend in a PTFE substrate.....hhhhhmmm where have we seen that before ‘snake oil’?

Dave,

 

You need to read the article again.

 

Mick Richards

[RogerH]

Hi Peter,

that was almost understandable.

Showing that enough oil flow will allow the wedge to continue to exist. And an adequate pressure is there to keep the flow going.

 

So a drop in pressure can be tolerated but not forever.

 

Roger

[Nick Jones]

I think the drop in oil pressure as the engine warms is mostly due to the falling oil viscosity - though I much admit that until reading the rest of this interesting thread I'd not appreciated how big the drop-off is. This simply means that the oils flows more easily through all the various designed system exit points (nearly every bearing surface in the engine) so the pump has to push less hard for a give flow rate.

 

However, one further possible factor not so far mentioned; where you have an oil pump with alloy outer casing (pretty much all the 6s do, as do most of the small dry-liner 4s, don't know about the wet-liner 4s), the casing will grow more with increasing temperature than the gears and increasing the the gear end-float and thus slip across the end will also increase.

 

At low engine speeds, and remember that the oil pump runs at half engine speed so it might only be doing 300 rpm, the slippage is a much more significant percentage of the total pump output. This makes oil pump tolerances more significant at idle. This all the more true as in order to minimise the over-capacity of the pump at higher rpm (pump output is directly proportional to rpm) it is intentionally sized to be only just adequate at idle speeds.

 

Nick

[Hamish]

I always thought that the multi grades we now tend to use are designed for variable temperatures.

 

 

This is not a bad site for some initial understanding.

http://www.driverstechnology.co.uk/oils.htm

 

 

This is their VISCOSITY explanation

 

Most oils on the shelves today are "Multigrades", which simply means that the oil falls into 2 viscosity grades (i.e. 10w-40 etc)

 

Multigrades were first developed some 50 years ago to avoid the old routine of using a thinner oil in winter and a thicker oil in summer.

 

In a 10w-40 for example the 10w bit (W = winter, not weight or watt or anything else for that matter) simply means that the oil must have a certain maximum viscosity/flow at low temperature. The lower the "W" number the better the oil's cold temperature/cold start performance.

 

The 40 in a 10w-40 simply means that the oil must fall within certain viscosity limits at 100°C. This is a fixed limit and all oils that end in 40 must achieve these limits. Once again the lower the number, the thinner the oil: a 30 oil is thinner than a 40 oil at 100°C etc. Your handbook will specify whether a 30, 40 or 50 etc is required.

 

 

In my mind this helps to cover cold and hot running rather than just seasonal. ?

[RogerH]

Hi Nick,

the wet liner pumps are cast steel items.

 

Roger

[Peter Cobbold]

Hi Peter,

that was almost understandable.

Showing that enough oil flow will allow the wedge to continue to exist. And an adequate pressure is there to keep the flow going.

 

So a drop in pressure can be tolerated but not forever.

 

Roger

Hi Roger, I think the flow will pass though the arc above the crank so ensuring that fresh cool oil is pulled around into the wedge by the rotation of the crank. Someone will have measured the heating of the oil in that high pressure wedge.....

Peter

[john.r.davies]

The most telling point of that interesting article is that the pressure in the wedge is in thousands of psi. So no wonder that "the hydrodynamic pressure has no relationship at all to the engine oil pressure, except that if there is insufficient engine oil pressure to deliver the required copious volume of oil into the bearing, the hydrodynamic pressure mechanism will fail and the bearing(s) and journal(s) will be quickly destroyed."

My underline.

 

Kastner recounts in his Triumph Preparation Handbook how he had Spitfire bearings failing when tested on the dyno. His team devised a way of measuring the oil pressure in the crankshaft and found it was less than 10psi at 5000rpm. This was also true on his GT6 engines, and led to the development of the Kastner Octupus that bypassed the in-block oil gallery, and distributed oil via a short external rail and hoses to each main bearing drilling. That may not be essential, but the figure of 10psi, which would be more at the sensor, must be a minimum. And most OE sensors, in the absence of a gauge, don't trigger until the pressure is less than about 5psi.

 

John

 

PS the point about 'grooved bearings' that were used "Long ago" is ironic to us!

Edited January 1, 2018 by john.r.davies

[john.r.davies]

So, how much does an oil pump pump?

 

The 'gerotor' design that Triumph used has inner and outer rotors, with four lobes on the inner and five recesses in the outer. It functions as the oil fills the spaces between and is squeezed out as the space grows bigger, then smaller.

The maximum possible flow rate is the difference in volume of the spaces, by the rpm of the pump

 

I took the end off a spare, dressed the naked end with blue, took a 'print' of it onto millimeter graph paper and counted the squares, in the space at both maximum and minimum.

I measured the length of the outer rotor, and then the difference between those volumes.

See pic.

 

Turns out that my hand estimation gets a difference which is near as dammit to 1ml!

So at half crank speed and idle of 600rpm, the pump will deliver 300mls/min.

At 6000rpm, 3000mls, three liters a minute.

 

Of course, there are losses in the pump, as oil leaks around the sides of the rotor. This paper (http://www.hydraulicstatic.com/20121029_hydraulic-gerotor-pump-volumetric-efficiency-characteristic.html) describes the volumetric efficiency of a particular pump witha much larger displacement as falling with with the opposing pressure, but even more as speed rose. Between 1200 and 1800rpm the efficiency fell from about 92 to 80%, but other publications offer estimates of less efficiency at higher revs, so the real flow may be only 2L at 6,000rpm.

 

John

Edited January 1, 2018 by john.r.davies

[Peter Cobbold]

OK, I'll take up the batten.......

As the wedge pressure at 6000 psi is unaffected by the oil psi in the crank it must mean that the major role of the oil flow is to remove heat.

How much heat?

Slide 14 right panel shows the crank journals create 10% of the mechanical friction at wot and 5000rpm

http://web.mit.edu/2.61/www/Lecture%20notes/Lec.%2019%20Friction%20and%20tribology.pdf

FMEP typically accounts for around 10% of BMEP. So for an engine delivering 150hp the friction power loss in crank will be about 1.5hp.

1.5hp = 1.1kW of heat to be removed by the oil flow through the crank journals.

So in one second the energy dissipation will be 1,100 joules per second

For a oil flow of 2000 ml/min ( from John, above) , that's 33 ml of oil per second.

The specific heat of oil is ca 2.2 J/g/C (http://www.dynamicscience.com.au/tester/solutions1/chemistry/energy/specificheatcapacity.htm )

So roughly the oil would heat up by 1100/73 or 15C as it flows through the crank journals.

 

Peter

Edited January 1, 2018 by Peter Cobbold

[RogerH]

Wow - it has turned into quite a good thread.

 

Roger

[Peter Cobbold]

Wow - it has turned into quite a good thread.

 

Roger

and we haven't started on the Stribeck curve....yet.

Peter

[john.r.davies]

Real world, or part real world check. This paper from two Uppsala students, at Volvo, https://uu.diva-portal.org/smash/get/diva2:825739/FULLTEXT02.pdfis a study of the performance of oil coolers. The graph below matches their simulation of a proposed system witjh real data from a Volvo supplier (Mahle). It shows that at 2L/min of oil flow the cooler removes 15kW of heat, which if crank heat is 10% of total engine waste heat fits well with Peter's calculation.

 

Interesting paper too!

 

JOhn

 

Edited January 2, 2018 by john.r.davies

[Peter Cobbold]

Crank heat will less than 10% of total friction heat, as the rings add a lot of friction, and total friction heat will be less than 10% of total waste heat, around 5% , but depends upon design. Heat flows in an engine can be summarised as Sankey diagram, eg:

http://www.sankey-diagrams.com/wp-content/gallery/x_sankey_212/nissan_power_train.gif

The oil will also pick up heat off the bores, partly from ring friction but mostly form the water jacket I suspect. Without a oil cooler the oil can loose heat from the sump walls, and when the oil temperature exceeds the coolant , to the radiator. Cant find a diagram that describes the heat fluxes to/from the oil.

 

A 150hp engine at wot will consume about 400hp of energy: 150hp usefull and 250hp waste heat, if thermal effic is approx 35 %.

At part-throttle cruise with useful output say 20hp the TE falls, could fall from 35 to 10%, so waste heat is still high, ca 200hp.That's 150kW. The Sankey diagram will look pretty sick at cruise !

 

Peter

Edited January 2, 2018 by Peter Cobbold

[oldtuckunder]

Interesting thread.

 

Biggest lesson I have learn't over last two years in (hopefully) solving my #5 BE problem, is fit an oil temp gauge! as others have said its lack of flow (either continuous or momentary "probably too quick to register") not pressure that will kill an engine.

 

With an oil temp gauge you will quickly learn that (in the UK) oil takes a long time to warm up, that you are abusing the engine when the oil is too cold, and a lot of the time the oil is only occasionally reaching its designed working temp. Unless your racing and know the oil is up to temp, if you have an oil cooler fitted without a thermostat most of the time you are just robbing the engine of power and necessary lubrication, If you have an oil stat fitted and it opens any less than 85C then you are doing the same! In the UK a lot of the time you want to be putting heat into the oil not taking it out! Why do you think so many moderns designed to run on lower viscosity fully synthetic oils now fit oil/water heat exchangers and don't bother with oil coolers. A heat exchanger will warm the oil up quicker (when all the damage is being done) and will cool it when the oil temp starts rising above coolant temp.

 

Even a 0w oil is too thick to lubricate an engine properly at 75F, so on a warm summers day when you first start your classic filled with Classic Mineral 20w50 just take it real easy for a bit! Oh and its cools down quick in the sump also, I can finish a run with the oil approaching 100C let it stand for an hour or so and it will be back down below 50C and too low to push the engine hard.

 

If you have an oil temp gauge plus an oil pressure gauge you can watch exactly the correlation between the two, there is very little correlation between coolant temp, oil temp, and oil pressure (unless you have a heat exchanger) If you have all three gauges you will have a ery good view of what is happening in your engine. If I had to junk two of the gauges it would be pressure and coolant temp, if you only have an oil temp gauge and a pressure idiot light you will have a better information than the other two put together.

 

Guess I'll get flamed for this!

 

Alan

[RogerH]

Hi Alan,

some good pragmatic stuff.

I like the idea of the oil heater/water cooler. We have this on the aircraft but use the fuel rather than water - Fuel cooled oil cooler.

 

Roger

[Peter Cobbold]

There is indeed high wear when the engine in is being started, whether oil is hot or cold. The rpm are so low that ther'es no dynamic lubircation, no wedge of high pressure oil separating crank from shells. Bore wear is also high as the rings are runnign initially on bores dry of oil splash. Also around TDC and BDC the ring motion is so slow that the hydrodynamic wedge is lost and boundary lubication by a very thin film of oil or anti-scuff coating ( ca 1/250 thou) remains, hot or cold. Hence the 'wear lip'. This dead centre wear occurs on every revolution !

The next worse thing to very low rpm is high rpm.

This behaviour is summarised in the Stribeck curve. Lots of web sites,eg:

https://www.researchgate.net/profile/Javier_Sotres/publication/259357833/figure/fig1/AS:297123783430159@1447851238693/Figure-1-Model-Stribeck-curve-where-the-friction-coefficient-ant-the-fluid-film.png

( V velocity can be crank rpm or piston instantaneous speed )

This one includes info on ZDDP too:

http://www.sae.org/events/pfs/presentations/2005spikes.pdf

Stribeck shows hIgh friction losses occur at very low rpm. And also at high rpm, where presusmably the power lost in rapidly squeezing the oil inot that wedge is reported as a friction loss. The effect of oil temperature can be seen as reducing the friction losses, whcih is why Alan likes his oil hot - more power !! ZDDP increases friction, but reduces wear ( but at very low concentrations increases wear ! ). If reducing wear is a priority over power, cold oil works better. [ However ZDDP needs higher temperatures to deposit and maybe replenish - not shown in the talk ]

 

Peter

Edited January 3, 2018 by Peter Cobbold

[oldtuckunder]

We have this on the aircraft but use the fuel rather than water - Fuel cooled oil cooler.

 

 

Ah a bit on the lines of the XJS that had an aircon/fuel heat exchanger to take the heat out of the injection return fuel line to the tank. Always thought that was a clever idea to keep the fuel tank cool.

[oldtuckunder]

There is indeed high wear when the engine in is being started, whether oil is hot or cold. The rpm are so low that ther'es no dynamic lubircation, no wedge of high pressure oil separating crank from shells.

 

I have found a pre-oiler helps with this, or at least makes it less painful emotionally when starting the engine to see the oil pressure around 50psi before you push the button

 

I'm fairly certain Marcus had a thread on here a year or so back on implementing one on his GT6, I just slavishly ripped off his design with a few mods to suit my application.

 

Alan

 

Alan

[Steves_TR6]

Wow, the second of Peter’s attachments :

 

http://www.sae.org/e.../2005spikes.pdf

 

Is very informative ( didnt understand a lot of it !)

 

Steve

[RogerH]

 

 

 

I'm fairly certain Marcus had a thread on here a year or so back on implementing one on his GT6, I just slavishly ripped off his design with a few mods to suit my application.

 

Alan

 

Alan

Yes he did. Knocked it u for just a few pounds. They are £450 in the shops.

 

Roger

 

PS - Maybe Rodbr was there first http://www.tr-register.co.uk/forums/index.php?/topic/54218-can-ya-guess-what-it-is-yet-can-ya-can-ya/

I thought it was Marcus.

Edited January 3, 2018 by RogerH

[Ian Vincent]

Im pretty sure it was Marcus.

 

Rgds Ian