FA20 oiling problems, study and solution

[solidsnake11]

Originally Posted by Joesurf79

You haven't lost volume though - the volume of gas (air for arguments sake) entrained in the fluid doesn't change - it was always there, always being pumped, down the line, evenly distributed. Nor does the volume of liquid in the system change. It's the drop below vapor pressure for the liquid that allows the gas to dissociate from the liquid into clumps (bubbles) that collapse as soon as the pressure on the fluid rises back above vapor pressure. That bubble collapse causes the damage to adjacent surfaces.

You'll see more of an effect on discharge pressure from a centrifugal pump due to cavitation than you will from a PD type pump due to the fixed relationship between RPM to volume moved in a PD (reciprocating piston, gear, lobe vane, peristaltic, etc) pump.

https://www.mdpi.com/1996-1073/10/9/1261/htm

The cavitation and aeration is a critical issue in all positive displacement machines, hence it is crucial that the simulation models could be able to reproduce these phenomena. Beside the effects related to the noise and to the wear, the presence of a gaseous phase implies a reduction of the volume of liquid displaced by the pump.

[solidsnake11]

Does anybody know where or what the lower galley from the passenger head goes to?
Also what about number 2?

[Ultramaroon]

Originally Posted by solidsnake11

Does anybody know where or what the lower galley from the passenger head goes to?
Also what about number 2?

I think that's the bank-1 chain tensioner.


#2 is for the A/T vacuum pump.

[Joesurf79]

Originally Posted by solidsnake11

https://www.mdpi.com/1996-1073/10/9/1261/htm

I hear ya. No argument that while the conclusion of that Italian guy's paper is that cavitation is of critical concern for positive displacement pump systems, I'll say that in my operations experience with fluid pumps of all sizes and configurations, cavitation has a much greater effect on (and affects with greater frequency) the flow rates of centrifugal pumps in viscous fluid service. Just an internet rando's $0.02

Carry on good sirs and figure out how to make these things live on track!

[Ultramaroon]

After tweaking my test gauge to get it in agreement in between two other references at 40, 60, 80, and 120 PSI, close enough for gubment work.

Below operating temp, while the oil is more viscous and the engine is tight, I'm unable to measure a pressure difference between the two points with my ghetto tools.





Then, at test temperature, I can see a slight drop at the downstream tap. Less than a bar, though.

[ETM_Shaman]

@Ultramaroon, @solidsnake11, @Irace86.2.0
Great information guys!

I've looked at and measured the lubrication system in this engine so much that I know it by heart, but that diagram helps a lot and I'm gonna populate it a little better:



As you can see the way from the stock pressure switch location to the main gallery is not that long but it's VERY convoluted with many sharp turns (all the drillings that meet have sharp round edges inside). The gallery that goes from the pump to the block is the thinnest part of the whole crank feed system. Also notice *10 is not straight like they picture it but tapered from the outside to the center (probably from the use of tapered pins for casting). *4 is not a return line but the gallery that takes oil to the no 1. chain tensioner, it only exists for that.

@Ultramaroon I did validate my gauges and as you have noticed, the pressure drop changes a lot with RPMs (because pressure drop in a gallery is tied to flow, with all other variables keeping the same). With my measuring setup at idle, both gauges read almost the same and the difference increased gradually as I increased the RPMs (which also validates the sensors readings)

I see you get about 10 psi drop (~ 0.7 bar) at 6,000 RPMs. I'd be curious to know what crank and rod clearances your engine has, is it stock?.

I figure the reason I see more pressure drop is from using looser crank and rod clearances than a stock engine (although barely out of the stock range).

As a side note, remember my engine only failed when taken to high RPMs and I think it's only there that the stock lubrication system is at the edge of collapse (particularly the pump's inlet tract). This means:

1. You have a stock engine with clearances on the tight side of the range: you never have this problem

2. You have a stock engine with clearances on the loose side of the range: you might damage a rod bearing if you spend time above 7,000 RPM under load and with hot oil (tracking, etc.)

3. You rebuild your engine and use looser than stock clearances with increased power and a lot of heavy high RPM usage: welcome to my world

Why use looser clearances? To increase oil flow to the crank and rods. That's no black magic and the benefits are widely proven and used in the perfomance engine building world. Would my engine have survived sustained 300 ft-lbs of torque at all RPMs on track if I had used 0.02 mm crank and 0.03 mm rod clearances? Who knows but I would consider that tight for this engine under that use, I wouldn't have built it like that for a customer.

[Ultramaroon]

Originally Posted by ETM_Shaman

I've looked at and measured the lubrication system in this engine so much that I know it by heart, but that diagram helps a lot and I'm gonna populate it a little better:



As you can see the way from the stock pressure switch location to the main gallery is not that long but it's VERY convoluted with many sharp turns (all the drillings that meet have sharp round edges inside). The gallery that goes from the pump to the block is the thinnest part of the whole crank feed system. Also notice *10 is not straight like they picture it but tapered from the outside to the center (probably from the use of tapered pins for casting). *4 is not a return line but the gallery that takes oil to the no 1. chain tensioner, it only exists for that.

@Ultramaroon I did validate my gauges and as you have noticed, the pressure drop changes a lot with RPMs (because pressure drop in a gallery is tied to flow, with all other variables keeping the same). With my measuring setup at idle, both gauges read almost the same and the difference increased gradually as I increased the RPMs (which also validates the sensors readings)

I see you get about 10 psi drop (~ 0.7 bar) at 6,000 RPMs. I'd be curious to know what crank and rod clearances your engine has, is it stock?.

I figure the reason I see more pressure drop is from using looser crank and rod clearances than a stock engine (although barely out of the stock range).

Thanks for updating that chart. Makes way more sense.


I wouldn't put much faith in my absolute numbers but they're ok for comparative analysis. Mine is bone stock with a tad over 50 Kmi now. I am super gentle when it's cold but otherwise am not afraid of beating the piss out of it. I can run sustained high RPM using 0w20 without pressure drop because I have a massive oil cooler. Flow is manually controlled with a diverter valve to keep things in balance while I daily. Max pressure drops like a rock above 90*C and although I'm sure I'm being paranoid, I feel good about my setup. UOA helps to confirm that I'm doing it right.

I forgot about the deliberate extra clearance at your main and rod journals. Taking that into account, our respective numbers jive nicely! Note your comment in bold - I gotcha. Your point about the twisted path is well noted. I feel like I've learned quite a bit this past couple days.

[ETM_Shaman]

Originally Posted by Irace86.2.0

@ETM_Shaman

Have you considered testing the stock pickup tube pressure or inlet pressure with 0w20 or even 5w30 instead of your 10W40? Perhaps the heavy grade oil was the problem all along.

Skip to 1:20. If the cavitation you are trying to resolve is from starving the pump then couldn't higher viscosity oil be the problem?

When I had the failure using 10W60 I was sure it had to do with the problem (after all, main gallery pressures changed a lot when going from 10W40 to 10W60). Then I found the GT86 CS-V3 User Manual:

https://www.racingbytmg.com/wp-conte...anual_EN-1.pdf

If you go to page 108 you'll see they use Elf HTX 825 10W-60 oil.
These cars have stock engines and are well proven so I guess the 10W60 oil is not an FA20 engine killer.
Even so I think thinner oil will allow better (closer to atmospheric) inlet pressures. I think the tests you suggest would be very interesting but to tell you the truth I don't think I'll go through the hassle of changing three different oils (including the oil cooler flush to get it all out). I had problems with 10W40, 15W50 and 10W60 oils. Always with the same clearances. I wouldn't run 0W20 oil with those clearances either, but it would also be an interesting experiment. I would've loved to have thought of the inlet pressure tap from day 1 of the first rebuild and I probably would've learned a lot and saved time and money.

[solidsnake11]

Originally Posted by ETM_Shaman

When I had the failure using 10W60 I was sure it had to do with the problem (after all, main gallery pressures changed a lot when going from 10W40 to 10W60). Then I found the GT86 CS-V3 User Manual:

https://www.racingbytmg.com/wp-conte...anual_EN-1.pdf

If you go to page 108 you'll see they use Elf HTX 825 10W-60 oil.
These cars have stock engines and are well proven so I guess the 10W60 oil is not an FA20 engine killer.
Even so I think thinner oil will allow better (closer to atmospheric) inlet pressures. I think the tests you suggest would be very interesting but to tell you the truth I don't think I'll go through the hassle of changing three different oils (including the oil cooler flush to get it all out). I had problems with 10W40, 15W50 and 10W60 oils. Always with the same clearances. I wouldn't run 0W20 oil with those clearances either, but it would also be an interesting experiment. I would've loved to have thought of the inlet pressure tap from day 1 of the first rebuild and I probably would've learned a lot and saved time and money.

Pretty sure that closes the too heavy of oil argument.

Elf HTX 825 10W-60
Density at 15°C 0.857 g/ml ASTM D-1298
Viscosity at 40°C 160 mm²/s ASTM D-445
Viscosity at 100°C 24.5 mm²/s ASTM D-445
Viscosity HTHS 5.5 mPa.s ASTM D-4741
Flash point 242 °C ASTM D-92

Castrol TWS Motorsport (aka Edge 10w60) sold at Advanced Auto
API SJ/CF
Density: 0.864
KV@40: 161
KV@100: 24.4
VI: 179
Flash: >200C
Pour: -51C
TBN: ?
HTHS: 5.4

[Irace86.2.0]

Originally Posted by ETM_Shaman

When I had the failure using 10W60 I was sure it had to do with the problem (after all, main gallery pressures changed a lot when going from 10W40 to 10W60). Then I found the GT86 CS-V3 User Manual:

https://www.racingbytmg.com/wp-conte...anual_EN-1.pdf

If you go to page 108 you'll see they use Elf HTX 825 10W-60 oil.
These cars have stock engines and are well proven so I guess the 10W60 oil is not an FA20 engine killer.
Even so I think thinner oil will allow better (closer to atmospheric) inlet pressures. I think the tests you suggest would be very interesting but to tell you the truth I don't think I'll go through the hassle of changing three different oils (including the oil cooler flush to get it all out). I had problems with 10W40, 15W50 and 10W60 oils. Always with the same clearances. I wouldn't run 0W20 oil with those clearances either, but it would also be an interesting experiment. I would've loved to have thought of the inlet pressure tap from day 1 of the first rebuild and I probably would've learned a lot and saved time and money.

See this in page 114 lol...

Always let the engine warm up to an oil temperature of at least 65°C before a race otherwise a reduction in performance can be expected (no damage to the engine is possible).

I’m not surprised they run a thicker oil with no upgrades like a larger radiator or oil cooler, especially as a dedicated track car.

[nikitopo]

Originally Posted by Irace86.2.0

I’m not surprised they run a thicker oil with no upgrades like a larger radiator or oil cooler, especially as a dedicated track car.

They have an oil cooler as an option. They didn't include it as default among with other things to keep the initial purchase price low.

[Irace86.2.0]

Originally Posted by nikitopo

They have an oil cooler as an option. They didn't include it as default among with other things to keep the initial purchase price low.

Missed that. Thanks.

[Ultramaroon]

Originally Posted by ETM_Shaman

Also notice *10 is not straight like they picture it but tapered from the outside to the center (probably from the use of tapered pins for casting).

I just caught this after a few reads. Does your gut say that deserves a finish bore? Is there enough material in the block?

[CSG Mike]

Originally Posted by Irace86.2.0

Wouldn’t fans limit flow/cooling at speed? Is there a lot of hot idling time during track days?

Not if the fans pull with enough pressure, which mine do. The blades also minimally impede flow at high speed.