Oil/air sperators

[KiwiBacon]

Well i have been reading this and searching the terms mentioned but honestly im lost, do these not work for a diesel?
http://shop.ebay.com.au/i.html?_nkw=oil ... 86.c0.m359

cheers mark.

Those ones are extremely basic. They will catch big sloshes of oil but won't do much to seperate oil mist out.

[tweak'e]

with whatever you use, the hard part is making it so gas can't get around it. big problem i had with foam is that the gas and oil got past around the edges. it needs to be sealed so the oil has to go through the filter.

even a bit of cotton shirt would do, as long as it flows gas well enough. once its wet with oil the oil itself is what filters out the oil mist.

Woohoo - a get rich quick scheme. I'll sell my old jocks on ebay as an oil scrubber and claim the elastic as a new high performance oil seal.
The soaked up gas may even improve vehicle performance

Paul

and that bit of gas that leaks out of your old jocks is the performance enhancing lpg system

[tweak'e]

What about air filter foam?
Possibly too open to be honest.
Finer - visit the local rubber supplies. Just make sure it's not closed cell.

Alternately - what about cotton? I reckon an old T shirt would work.
So long as it's big enough to slow the gasflow and fine enough to "scrub".

As indicated - it prod does depend on mist size.

I need one for mine. Sick of ppl telling me my turbo leaks oil when it's the breather from the rocker cover dumping oil into the turbo intake.

Paul

both will work. someone on patrol forum got unifilter (i think) to make a foam replacement for the compressed air filter they where useing. personally i think its to small.

with whatever you use, the hard part is making it so gas can't get around it. big problem i had with foam is that the gas and oil got past around the edges. it needs to be sealed so the oil has to go through the filter.

even a bit of cotton shirt would do, as long as it flows gas well enough. once its wet with oil the oil itself is what filters out the oil mist.

Actually neither will work (well).

Filtration efficiency is inversely proportional to fibre diameter. Cotton fibres are large and foam is crap.

But don't listen to me. I just do filter R&D for a living so I wouldn't know anything.

who said anything about the filter catching all the oil ?
after all its the OIL that catches the oil not the filter.

i'm only running a cheap breather filter, its not 100% but certainly good enough with minimal risk of clogging up. its probably only 90%, theres only a slight film in the inlet pipe which vast improvement of the puddle that used to be in the intake.

[me3@neuralfibre.com]

What about air filter foam?
Possibly too open to be honest.
Finer - visit the local rubber supplies. Just make sure it's not closed cell.

Alternately - what about cotton? I reckon an old T shirt would work.
So long as it's big enough to slow the gasflow and fine enough to "scrub".

As indicated - it prod does depend on mist size.

I need one for mine. Sick of ppl telling me my turbo leaks oil when it's the breather from the rocker cover dumping oil into the turbo intake.

Paul

both will work. someone on patrol forum got unifilter (i think) to make a foam replacement for the compressed air filter they where useing. personally i think its to small.

with whatever you use, the hard part is making it so gas can't get around it. big problem i had with foam is that the gas and oil got past around the edges. it needs to be sealed so the oil has to go through the filter.

even a bit of cotton shirt would do, as long as it flows gas well enough. once its wet with oil the oil itself is what filters out the oil mist.

Actually neither will work (well).

Filtration efficiency is inversely proportional to fibre diameter. Cotton fibres are large and foam is crap.

But don't listen to me. I just do filter R&D for a living so I wouldn't know anything.

I'm listening.

Apart from the genuine (too expensive) item - could you suggest a reasonable faximile that might be common?

Is oil mist seperation based on the same principles as depth filtration?
(err - you said not a seive - so - I probably dont' know those principles real well)

I know filters are based on %ages, so in this case a 90% reduction would be significant. 99% would be brilliant. There is no multi-pass though (unless you count the "pulsing" of the gas)

Thanx
Paul

[ISUZUROVER]

Well i have been reading this and searching the terms mentioned but honestly im lost, do these not work for a diesel?
http://shop.ebay.com.au/i.html?_nkw=oil ... 86.c0.m359

cheers mark.

Those ones are extremely basic. They will catch big sloshes of oil but won't do much to seperate oil mist out.

What Dougal said.

Those things look nice and shiny, but that are basically what in engineering speak is called a "settling chamber" they rely on slowing down the airflow sufficiently so that the particles (oil droplets) will settle out through gravity before they have left the chamber.

They might work fine on an old-tech NA engine which doesn't breathe much, but modern turbo engines - and specifically turbo diesels - produce very fine oil droplets. At a guess they would only remove 10% of the mass coming out the breather from an engine like that.

Slightly better (but not much) are the steel wool filled chamber type devices.

More efficient are cyclones like Dougal has posted - and many vehicles have this sort of device fitted OEM. Cyclones work in the same way as cyclonic pre-cleaners on the top of snorkels. A breather of this type will be 50-60% efficient (in terms of % mass removed) at best.

The most efficient technology is a filter - but only if correctly designed. And I am not talking about gauze/cotton devices here. Breather filters can get up to 99% mass removal.

[ISUZUROVER]

who said anything about the filter catching all the oil ?
after all its the OIL that catches the oil not the filter.

i'm only running a cheap breather filter, its not 100% but certainly good enough with minimal risk of clogging up. its probably only 90%, theres only a slight film in the inlet pipe which vast improvement of the puddle that used to be in the intake.

In a dust filter, what you say is more or less correct. The filter fibres capture the initial influent particles, then once a porous dust cake forms, the dust captures further particles.

This means that a DUST filter becomes more efficient as it collects dust. The OPPOSITE is true for a mist filter. As it collects mist it decreases in efficiency, because the collected oil forms large, non-porous droplets, which do not aid filtration. Because the droplets are so large, the flow (and the oil mist) can easily bend around the big oil droplets, and instead flow through the gaps.

I am glad your filter works for you, but it wouldn't be 90% efficient. You may be removing the large droplets, which could otherwise settle out readily and leave oil residue.

[me3@neuralfibre.com]

who said anything about the filter catching all the oil ?
after all its the OIL that catches the oil not the filter.

i'm only running a cheap breather filter, its not 100% but certainly good enough with minimal risk of clogging up. its probably only 90%, theres only a slight film in the inlet pipe which vast improvement of the puddle that used to be in the intake.

In a dust filter, what you say is more or less correct. The filter fibres capture the initial influent particles, then once a porous dust cake forms, the dust captures further particles.

This means that a DUST filter becomes more efficient as it collects dust. The OPPOSITE is true for a mist filter. As it collects mist it decreases in efficiency, because the collected oil forms large, non-porous droplets, which do not aid filtration. Because the droplets are so large, the flow (and the oil mist) can easily bend around the big oil droplets, and instead flow through the gaps.

I am glad your filter works for you, but it wouldn't be 90% efficient. You may be removing the large droplets, which could otherwise settle out readily and leave oil residue.

The OEM breather in the Toyota TD's is not real flash, but Toyota obviously sees it as ok - any thoughts?

Thanx
Paul

[ISUZUROVER]

who said anything about the filter catching all the oil ?
after all its the OIL that catches the oil not the filter.

i'm only running a cheap breather filter, its not 100% but certainly good enough with minimal risk of clogging up. its probably only 90%, theres only a slight film in the inlet pipe which vast improvement of the puddle that used to be in the intake.

In a dust filter, what you say is more or less correct. The filter fibres capture the initial influent particles, then once a porous dust cake forms, the dust captures further particles.

This means that a DUST filter becomes more efficient as it collects dust. The OPPOSITE is true for a mist filter. As it collects mist it decreases in efficiency, because the collected oil forms large, non-porous droplets, which do not aid filtration. Because the droplets are so large, the flow (and the oil mist) can easily bend around the big oil droplets, and instead flow through the gaps.

I am glad your filter works for you, but it wouldn't be 90% efficient. You may be removing the large droplets, which could otherwise settle out readily and leave oil residue.

The OEM breather in the Toyota TD's is not real flash, but Toyota obviously sees it as ok - any thoughts?

Thanx
Paul

Most passenger cars (including diesel 4x4s) have pretty average breather systems. Mainly because the car companies want to save money, and they don't expect the engine to live long enough for the blowby to affect engine life.

Heavy duty diesel engines retained open breather systems for as long as they were allowed to legally. When they switched to closed crankcase ventilation, the soot in the blowby oil droplets started to detrimentally affect engine life - which is why serious breather systems like the Provent (and others) were developed.

You are continually losing the lighter fractions of your oil through the breather. Capturing that oil and returning it to the sump will keep your viscosity lower for longer - Regardless of whether it extends turbo life or not, or reduce the frequency at which you need to clean your intercooler out.

[tweak'e]

who said anything about the filter catching all the oil ?
after all its the OIL that catches the oil not the filter.

i'm only running a cheap breather filter, its not 100% but certainly good enough with minimal risk of clogging up. its probably only 90%, theres only a slight film in the inlet pipe which vast improvement of the puddle that used to be in the intake.

In a dust filter, what you say is more or less correct. The filter fibres capture the initial influent particles, then once a porous dust cake forms, the dust captures further particles.

This means that a DUST filter becomes more efficient as it collects dust. The OPPOSITE is true for a mist filter. As it collects mist it decreases in efficiency, because the collected oil forms large, non-porous droplets, which do not aid filtration. Because the droplets are so large, the flow (and the oil mist) can easily bend around the big oil droplets, and instead flow through the gaps.

I am glad your filter works for you, but it wouldn't be 90% efficient. You may be removing the large droplets, which could otherwise settle out readily and leave oil residue.

why would 90% of compressed air filters work on that principle if it doesn't work ???

i've tried empty can, foam filter, baffles etc and caught next to nothing. i don't get much in the way of large droplets.

[ISUZUROVER]

why would 90% of compressed air filters work on that principle if it doesn't work ???

i've tried empty can, foam filter, baffles etc and caught next to nothing. i don't get much in the way of large droplets.

Sorry, but they don't work on that principle. They work on the same principle. That is why compressed air filters also work best when new and clean.

I think we have a different definition of what a large droplet is. The human eye can only see down to about a size of 20 microns. Therefore you cannot say if the droplets coming out of your engine are 10 nanometres or 10 microns (or anything in between) - unless you have $150k of particle measurement equipment...

I don't doubt that even a crude filter/sock is better than any other option you have tried.

[tweak'e]

why would 90% of compressed air filters work on that principle if it doesn't work ???
.

Sorry, but they don't work on that principle. They work on the same principle. That is why compressed air filters also work best when new and clean.

can you explain that a bit better??

afaik compressed air filter works on the wet filter collecting the fine droplets into larger droplets which then can settle out.
thats pretty much how i'm using the breather filter. i can't see why you can't use foam, cotton or your undies the same way.

instead of useing lots of filter medium to catch the oil, simply use the oil held in a cloth/foam/whatever to catch the oil mist.

fine droplets to most people would be the fine oil mist that floats in the air, large droplets is what spits out and falls to the ground.

[ISUZUROVER]

why would 90% of compressed air filters work on that principle if it doesn't work ???
.

Sorry, but they don't work on that principle. They work on the same principle. That is why compressed air filters also work best when new and clean.

can you explain that a bit better??

afaik compressed air filter works on the wet filter collecting the fine droplets into larger droplets which then can settle out.
thats pretty much how i'm using the breather filter. i can't see why you can't use foam, cotton or your undies the same way.

instead of useing lots of filter medium to catch the oil, simply use the oil held in a cloth/foam/whatever to catch the oil mist.

fine droplets to most people would be the fine oil mist that floats in the air, large droplets is what spits out and falls to the ground.

To explain it fully I would have to give you a course in how filters work, however..

1 - filters are NOT sieves. Filters collect particles by interception, inertia and diffusion (mainly)
2 - all other things being equal, filtration efficiency is inversely proportional to fibre diameter - i.e. the smaller the fibres the better

The image below shows a series of ~7 micron fibres (as might be used in crankcase vent filters or air compressor separators) with liquid droplets on them - which they have collected from the air. As the large coalesced droplets grow, they behave as if they are really large fibres, which the flow can easily bend around. Almost all the influent oil particles are captured on the "bare" sections of fibre. They get transported to the droplets only once captured (by a phenomena called Plateu-Rayleigh instability - which we won't go into).

So the bottom line is - the coalesced droplets capture a negligible number of particles. The "bare" sections of fibre capture the influent particles (oil mist). The more the filter loads up, the smaller area of bare fibres, the fewer particles get captured.

[me3@neuralfibre.com]

why would 90% of compressed air filters work on that principle if it doesn't work ???
.

Sorry, but they don't work on that principle. They work on the same principle. That is why compressed air filters also work best when new and clean.

can you explain that a bit better??

afaik compressed air filter works on the wet filter collecting the fine droplets into larger droplets which then can settle out.
thats pretty much how i'm using the breather filter. i can't see why you can't use foam, cotton or your undies the same way.

instead of useing lots of filter medium to catch the oil, simply use the oil held in a cloth/foam/whatever to catch the oil mist.

fine droplets to most people would be the fine oil mist that floats in the air, large droplets is what spits out and falls to the ground.

To explain it fully I would have to give you a course in how filters work, however..

1 - filters are NOT sieves. Filters collect particles by interception, inertia and diffusion (mainly)
2 - all other things being equal, filtration efficiency is inversely proportional to fibre diameter - i.e. the smaller the fibres the better

The image below shows a series of ~7 micron fibres (as might be used in crankcase vent filters or air compressor separators) with liquid droplets on them - which they have collected from the air. As the large coalesced droplets grow, they behave as if they are really large fibres, which the flow can easily bend around. Almost all the influent oil particles are captured on the "bare" sections of fibre. They get transported to the droplets only once captured (by a phenomena called Plateu-Rayleigh instability - which we won't go into).

So the bottom line is - the coalesced droplets capture a negligible number of particles. The "bare" sections of fibre capture the influent particles (oil mist). The more the filter loads up, the smaller area of bare fibres, the fewer particles get captured.

Is the above implying that the filter material won't "drain" the oil to the bottom of the canister?
I kindof got that impression from what you said earlier.

If yes - how then do you coalesce a "fine" oil mist back into a liquid? Condense it on a cold surface?

Paul

[tweak'e]

always love to learn new tech stuff

however does that apply? eg the filters we usually see in home air compressors would have fibres like tree trunks ! and you can see through the gaps.
i think the droplet sizes and quality of filtration that we are expecting is very different. yours are probably designed to catch every last drop right down to the smallest possible size. where as we are just after catching the bulk with an easy to make filter.

when i first tried the breather filter setup (which is certainly not a fine filter) a fair bit of oil mist got through. when it was wet there was very little getting through.
i always thought that instead of the mist going through the visible gaps of the filter, the gaps where now filled with oil, so it had to go through bits where the oil drained or got blown out. got more of a tortured route to go through. granted the really fine mist will go around it all buts thats not really a big deal as most of this mist would be big droplets.

[mud4b]

Well i have done searches again to see what these look like and i cant find anything, could i get one at a truck spares place?

Also are you saying people use one from a air compressor, these are very small, would the honestly work?

cheers mark.

[me3@neuralfibre.com]

Well i have done searches again to see what these look like and i cant find anything, could i get one at a truck spares place?

Also are you saying people use one from a air compressor, these are very small, would the honestly work?

cheers mark.

Dunno about work - the Patrol forum has a heap of dicussion on it.
There are larger air comp units, not just the small ones.

Paul

[ISUZUROVER]

Is the above implying that the filter material won't "drain" the oil to the bottom of the canister?
I kindof got that impression from what you said earlier.

If yes - how then do you coalesce a "fine" oil mist back into a liquid? Condense it on a cold surface?

Paul

No, not at all. Just that drainage doesn't occur until the filter is fairly heavily loaded with oil. e.g. for a provent on an average car, it will take ~15000km for it to collect enough oil for it to drain. But after that, drainage becomes more or less constant.

[ISUZUROVER]

Well i have done searches again to see what these look like and i cant find anything, could i get one at a truck spares place?

Also are you saying people use one from a air compressor, these are very small, would the honestly work?

cheers mark.

http://www.westernfilters.com.au/docs/Mann_Pro-Vent.pdf
This is one of the best on the market IMO.

I believe fleetgard, donaldson and parker/racor have one as well. I haven't tested the others, but I suspect this one is as good or better than the others.



here is an install thread - which should give you an idea of size, etc...

http://www.aulro.com/afvb/projects-tuto ... 110-a.html

I wouldn't use (or reccommend) a compressor filter. These are designed for much higher pressure drops.

[ISUZUROVER]

always love to learn new tech stuff

however does that apply? eg the filters we usually see in home air compressors would have fibres like tree trunks ! and you can see through the gaps.
i think the droplet sizes and quality of filtration that we are expecting is very different. yours are probably designed to catch every last drop right down to the smallest possible size. where as we are just after catching the bulk with an easy to make filter.

when i first tried the breather filter setup (which is certainly not a fine filter) a fair bit of oil mist got through. when it was wet there was very little getting through.
i always thought that instead of the mist going through the visible gaps of the filter, the gaps where now filled with oil, so it had to go through bits where the oil drained or got blown out. got more of a tortured route to go through. granted the really fine mist will go around it all buts thats not really a big deal as most of this mist would be big droplets.

I have never seen an air compressor filter which you can see all the way through - I would like to see pics...

That said, there is a wide range of materials used for compressor filters.

If you have a very low efficiency filter, and large droplets (which I suspect you have, then collection efficiency for these very large droplets can increase with loading, as the large droplets have so much inertia they basically try to plough straight through the filter and impact. However if that is the case I am sure you wouldn't have more than ~60% mass-based efficiency from your filter.

[+dj_hansen+]

Ben, got any test units left over... for some Toyota motor testing

No questions, these are the ducks bajebus.... the price is a little off putting though

[shorty_f0rty]

some very interesting reading here... thanks for the tech guys!

can I ask if the amount of oil you're seeing in your blow by would affect any potential designs?

Currently with my 1hdt I'm seeing considerably more than a "fine mist" of oil coming out of the crank case.. this is more noticeable when doing highway speeds. So much so that after a 30min trip down the express way, I have a catch can (powerade bottle) full of oil?

I would assume in this instance that filtration isn't so much of an issue as being about to drain the oil away affectively?

I'll be attempting to make a catch can over the next few weekends and was considering implementing a baffle of sorts for the air from the rocker cover to hit before being able to vent, thus collecting the oil on the baffle, which would drain to the oil return feed line..

apart from the obvious issue of the amount of oil coming through to the catch can is there any other design elements I should take into consideration?

[KiwiBacon]

I'll be attempting to make a catch can over the next few weekends and was considering implementing a baffle of sorts for the air from the rocker cover to hit before being able to vent, thus collecting the oil on the baffle, which would drain to the oil return feed line..

apart from the obvious issue of the amount of oil coming through to the catch can is there any other design elements I should take into consideration?

My original (factory) Isuzu air/oil seperator was a baffle similar to the one you describe.
It was close to useless.

That's why I went to the vortex/centrifuge design. I built my own air filter box the same way.

[ISUZUROVER]

Ben, got any test units left over... for some Toyota motor testing

No questions, these are the ducks bajebus.... the price is a little off putting though

Sorry, no more.

Somone mentioned above that you can pick them up for ~$175 on ebay. Doesn't seem too bad compared to the cost of the shiny catch cans. The filter elements should last ~50000km.

Shorty40 - see my post above - a baffle is only slightly better than a settling chamber alone. Only useful for REALLY big droplets with lots of inertia.

[+dj_hansen+]

You can indeed...

http://cgi.ebay.com.au/PRO-VENT-200-CRA ... 3ca68153b3

I spose the questions begs, is something along the lines of KiwiBacon's and fester's design better than nothing? Even 50% of the mass removed is achieving something.

[KiwiBacon]

One more question along the same lines.

If a centrifuge/vortex takes out 50%. Is that 50% by mass or 50% of the droplets?

Can a 2 stage vortex/centrifuge be expected to remove 75%?

[tweak'e]

If you have a very low efficiency filter, and large droplets (which I suspect you have, then collection efficiency for these very large droplets can increase with loading, as the large droplets have so much inertia they basically try to plough straight through the filter and impact. However if that is the case I am sure you wouldn't have more than ~60% mass-based efficiency from your filter.

60%........mmmmmmm......doubt its that low. unless the engines is making oil, the amount the catch can collects is roughly similar to what the engine loses. its not accurate of course but i think if it was loosing another litre it would be noticeable. i have to do an oil change so i'll check.
it may be more like 70-80% but i can't measure it that well to tell.
it may just not make much very fine oil spray. don't forget it goes through factory baffle setup first.

[fester2au]

with whatever you use, the hard part is making it so gas can't get around it. big problem i had with foam is that the gas and oil got past around the edges. it needs to be sealed so the oil has to go through the filter.

even a bit of cotton shirt would do, as long as it flows gas well enough. once its wet with oil the oil itself is what filters out the oil mist.

Woohoo - a get rich quick scheme. I'll sell my old jocks on ebay as an oil scrubber and claim the elastic as a new high performance oil seal.
The soaked up gas may even improve vehicle performance

Paul

Trouble is I think your jocks have already soaked up their fair share of particle rich gas. Nobody likes to buy an already oil stained oil filter.

The only reason I like steel wool or foam (of the right consistency) is the possibility of having a sealed system assuming they actually work. Kiwi system is good in that respect being totally mechanical. With anything like the provent filter they seem to need replacing or cleaning. I'm not usr eif the foam is totally maintenence free or for that matter if your "T" shirt material would be either. As far as stopping the fumes form bypassign the medium I think that was the aim of the design I found in that the foam is cut to shape with a punch of slightly larger diameter than your tube so it's a tight fit inside and the air has to travel through the foam as soon as it enters the chamber and has to travel 180 degrees around and back through it again if it had bypassed partially initially.

I'm still out to lunch a little on this idea but I'm liking something like Kiwis but with maybe some medium inside as back up, I was also tossing up putting a form of "S" bend in my piping from the rocker cover to make it that little bit harder to push the heavier oil through the piping. Basically if possible all the piping should have some uphill sections.

[ISUZUROVER]

One more question along the same lines.

If a centrifuge/vortex takes out 50%. Is that 50% by mass or 50% of the droplets?

Can a 2 stage vortex/centrifuge be expected to remove 75%?

All the values I have given are by mass.

A 2-stage cyclone will likely have too much pressure drop. Parallel cyclones are usually preferable. Either way, filters are usually better in terms of efficiency vs pressure drop.

[ISUZUROVER]

If you have a very low efficiency filter, and large droplets (which I suspect you have, then collection efficiency for these very large droplets can increase with loading, as the large droplets have so much inertia they basically try to plough straight through the filter and impact. However if that is the case I am sure you wouldn't have more than ~60% mass-based efficiency from your filter.

60%........mmmmmmm......doubt its that low. unless the engines is making oil, the amount the catch can collects is roughly similar to what the engine loses. its not accurate of course but i think if it was loosing another litre it would be noticeable. i have to do an oil change so i'll check.
it may be more like 70-80% but i can't measure it that well to tell.
it may just not make much very fine oil spray. don't forget it goes through factory baffle setup first.

A provent starts at about 99% when new, and drops to ~90% (mass based) - though more efficient elements are available. All of this is engine dependant. An engine emits droplets from ~10 nanometres to ~1 mm. The human eye can see down to 20 microns. It is very difficult to judge the efficiency of any separator by eye.

[fester2au]

One more question along the same lines.

If a centrifuge/vortex takes out 50%. Is that 50% by mass or 50% of the droplets?

Can a 2 stage vortex/centrifuge be expected to remove 75%?

All the values I have given are by mass.

A 2-stage cyclone will likely have too much pressure drop. Parallel cyclones are usually preferable. Either way, filters are usually better in terms of efficiency vs pressure drop.

Sorry Ben I was a little premature in my response before as my bowser supposed took me to the latest post and I didn't see all these others. Whilst I mentioned the price of the Provents on ebay and it certainly is better than prices I have it also makes the filter replacement price a fair bit out of kilter. So back to the simple but as effective as possible home brewed filter for those than need to save some bucks and as much as anything love a bit of a play in the shed - How about somethign around kiwibacons design with maybe some form of readily available and cheap filter medium in the middle as a last stage scrubber/filter or are we really best justusign the centrifuse design and living with it being the best it can. Whilst the tech background on all this is interesting I'd certainly like a fairly simple answer/explanation to that end and am certainly interested in following your professional opinion. Or there simply no suitable easily available meduim other than buying somethign like the provent inner.