Sign in to follow this  
Followers 0
tolsen

More power out of your Cdi - for free!

302 posts in this topic

BOT: Got my parts in today. All stainless all the way! I have enough now for the AOS and the branch to the oil dipstick guide tube, but will wait to do that part of the mod later on. I went a different route, with the stainless: I found existing parts that -- when assembled together -- will make a complete cyclone.

Share this post


Link to post
Share on other sites

BOT: Got my parts in today. All stainless all the way! I have enough now for the AOS and the branch to the oil dipstick guide tube, but will wait to do that part of the mod later on. I went a different route, with the stainless: I found existing parts that -- when assembled together -- will make a complete cyclone.

Do tell, brotha!:senile:

Share this post


Link to post
Share on other sites

Kindly note that you will have to blank off supply to exhaust gas recirculation valve at turbo end to fit the air oil separator. Length of pipes have to be kept to a minimum so air oil separator can only be fitted where shown in below photo.

Posted Image

EGR pipe is in the right position to provide additional support for the separator. My separator is fairly light as made from thin wall copper tubing but I doubt it is advisable only to have it supported by the hoses.

My transparent drain pipe is only temporary. I'll change this one out soon with proper 3/8" or 10 mm bore rubber fuel hose.

Edited by tolsen

Share this post


Link to post
Share on other sites

You'll laugh. Flask funnels. :D

http://www.amazon.com/gp/product/B00236D0LA

http://www.amazon.com/gp/product/B003SS7AMC

The first (taller) one will be the "funnel" of my cyclone, it is about 1.5" diameter by 1.9" from the bottom of the cone. It will be brazed to a 3" x 3/8" OD 321 SS tube for the oil return. The second one is stubbier and about 1.4" diameter. I will cut off its tube and braze in a 1/2" OD 321 piece to be the centre column/gas output, bent 180° to feed into the TIK. The hardest part will be drilling the elliptical hole in the side of the lower funnel's wall and brazing on the 1/2" OD tube that will go into the breather outlet.

Share this post


Link to post
Share on other sites

Fuel consumption is improving. On previous tank I averaged around 3.9 liters per 100 km. On last tank I managed 3.807 liters per 100 km. Only local driving on both tanks. Narrow, windy and undulating roads. It seems like the accelerator pedal does not need to be depressed as much as before the big clean.The improvement can only be caused by the engine now breathing easier. I would certainly recommend a good spring clean of the inlet system. A time consuming job but well worth the effort. The air oil separator should hopefully keep inlet system free of oil and gunk. I will post detailed dimensional drawings soon so you can make your own gadget.

Share this post


Link to post
Share on other sites

Thanks for the update...it sounds like we should all be checking this once a year and not waiting for the bad weather to do it. Mind you the current temp here in Ontario is 33 degrees with a humidex of 42. This could be called bad weather by some people.

Share this post


Link to post
Share on other sites

Removed air inlet temperature sensor (IAT) last night and inspected outlet of intercooler for oil contamination. Found it bone dry. Have now driven 1150 km on the cyclonic air oil separator and after spring cleaning inlet ducting including intercooler and inlet manifold. Oil level on dip stick is 4 mm below max level. Checked the usual way with engine at operating temperature but allowed to drain a couple of minutes for oil to settle.

Share this post


Link to post
Share on other sites

Wonder if Sameguy or anyone else have made and fitted one of these cyclone separators yet? Making one is not particularly difficult but also not as easy as it would appear.The cyclone has to be made small, preferably same size as mine (28 mm OD) or smaller. Space is tight where is fits. Barbs for connecting to existing hose must be right size. EGR must be blanked off as EGR pipe is used to support cyclone to limit vibration.Another query: What is the asking price for a catch tank that will fit the Cdi?

Share this post


Link to post
Share on other sites

Summer storm season at the airport has me working crazy, unpredictable hours. Parts are still on the bench, waiting for a spare moment.

Share this post


Link to post
Share on other sites

Wonder if Sameguy or anyone else have made and fitted one of these cyclone separators yet? Making one is not particularly difficult but also not as easy as it would appear.The cyclone has to be made small, preferably same size as mine (28 mm OD) or smaller. Space is tight where is fits. Barbs for connecting to existing hose must be right size. EGR must be blanked off as EGR pipe is used to support cyclone to limit vibration.Another query: What is the asking price for a catch tank that will fit the Cdi?

Did you have any idea of the blow-by flow in sizing the cyclone? As a design engineer, I designed many industrial cyclones. The diameter of the body as well as the nozzle sizes are important in achieving efficiency. They should ideally be mounted vertical. Cyclones are not efficient in removing mist sized droplets. For this type of application, I would more likely have used just a cylindrical knock-out drum with tangential inlet to remove large droplets and a demister pad in the upper section to remove fine mist sized droplets. the drum would have a cylindrical baffle that extends below the tangential inlet. To design such a device (or a cyclone), would need a knowledge of the gas flow. This should not vary too much in order to maintain efficiency yet it probably does. But I am no auto engineer. Nevertheless, I would be concerned about adding back pressure to the crankcase. Doesn't the crankcase normally vent to the suction side of the turbo where the pressure could be slightly negative? If the crankcase pressure is increased in order to drive the vent gases through the separator, is there not a chance that the engines vent flow is restricted with possible build up of deposits within the engine? Might also show as external leaks around filler cap and elsewhere.Or maybe I have this wrong? Do you still connect the cyclone outlet to the air inlet?Seems like you are happy with the results so far, so don't let the technical remarks of an old retired engineer deter you :) Edited by Graham

Share this post


Link to post
Share on other sites

For this type of application, I would more likely have used just a cylindrical knock-out drum with tangential inlet to remove large droplets and a demister pad in the upper section to remove fine mist sized droplets. the drum would have a cylindrical baffle that extends below the tangential inlet.

That is essentially what I have except that it is a conical shaped cyclone and there is no demister pad. I did consider fitting a demister but there is not sufficient space. Anyway, tests so far have shown there is no need for any demister.

Posted Image

Posted Image

Pressure build up is too small for my DIY manometer to measure so I regard it as negligible.

The cyclone is fitted vertically. Inlet is tangential coming directly from rocker cover. Outlet goes vertically up, then turned 180 degrees through two 14mm bore street elbows before being sent to original inlet at top ot TIK pipe before turbo. Oil is drained back to engine through a branch brazed to lower part of sounding tube. There is no more oil or even any trace of sweating in intercooler. The cyclone feels quite warm to touch. Will measure its surface temperature as a matter of interest.

Collection rate is equivalent to a large bottle of whisky (0.7 liter) per 10,000 km.

Share this post


Link to post
Share on other sites

That is essentially what I have except that it is a conical shaped cyclone and there is no demister pad. I did consider fitting a demister but there is not sufficient space. Anyway, tests so far have shown there is no need for any demister.

A demister wouldn't normally be installed in a cyclone, but rather in a knock-out pot. These have larger diameter so that gas moving upwards toward top outlet does not carry much liquid with it before it passes through the demister pad. Something like this:

Posted Image This one doesn't have a tangential inlet - it also doesn't need teh level control just so long as the gas can't escape out the bottom.

Pressure build up is too small for my DIY manometer to measure so I regard it as negligible.

The cyclone is fitted vertically. Inlet is tangential coming directly from rocker cover. Outlet goes vertically up, then turned 180 degrees through two 14mm bore street elbows before being sent to original inlet at top ot TIK pipe before turbo. Oil is drained back to engine through a branch brazed to lower part of sounding tube. There is no more oil or even any trace of sweating in intercooler. The cyclone feels quite warm to touch. Will measure its surface temperature as a matter of interest.

Collection rate is equivalent to a large bottle of whisky (0.7 liter) per 10,000 km.

I guess I wasn't paying attention - having the outlet going to the TIK pipe should alleviate my concerns about higher back pressure - Should be minimal through cyclone. It may be more through a demister, depending on type and sizing.

Now I understand this better, it does seem like an excellent mod. Unless the designers wanted oil to enter teh turbo, but I can't imagine why. Spraying oil into a heat exchanger is definitely not good!

I did find a reference to a cyclone separator for crankcase blow-by. For a bigger engine, but its interestingly used small diameter multi-cyclones sized it seems much like yours:

http://www.hengst.de/conpresso/_rubric/ind...al_Separator_EN

They seem to have some type of valve so as make system work over a wide range of flows (presumably because of the affect of gas flow on separation efficiency)

Share this post


Link to post
Share on other sites

Thanks for that link Graham. What I found interesting is how similar the Hengst cyclone separator is to mine. Very small diameter. Hengst's cyclone is seen on the left side of that strange contraption.I like to make one point regarding back pressure. Catch tanks will in my opinion create a higher back pressure than a free flowing cyclone. Reason: Catch tanks have demister elements and an air filter at the outlet.

Share this post


Link to post
Share on other sites

My experience in this application is that demister pad/element quickly becomes obstructed and creates far too much back pressure.

B :senile:

Share this post


Link to post
Share on other sites

I suppose the question might be - what causes the pad to be obstructed? Presumably solids or heavy tars? But how do they get into the pad. For a demister something like the one I posted to work, the lower part must separate out solids and heavy droplets. The upwards gas velocity must be low enough so that none of these get carried up into the demister. Demisters have the advantage that they can separate smaller droplets than a cyclone. Often the outlet from a cyclone is passed through a demister to remove fine mists. But, I am talking on an industrial scale. Tolsens little cyclone seems to be working and that is what is important! Seems like something that could be sold as a kit! I don't think it even needs to be a conical shape. A straight piece of pipe with two end caps and a tangential inlet and an outlet much like Tolsens might be easier to make. My Mercedes 300D, has a conventional round air filter. But in the center, there is a separator - just a pot with a baffle or two. The valve cover vent is connected to this separator. It used to drain back to the crankcase, but current engine does not have the drain point. So I have outlet blocked off. I collect about a cup between 10k km oil changes. Some oil likely gets picked up by the air and is burned. But no intercooler to get fouled up.

Share this post


Link to post
Share on other sites

I think the problem Bil describes is related to locally vented catch tanks. Due to its size, a catch tank need be fitted some distance away from engine requiring a long hose for connecting to rocker cover. This means breather gas is being cooled to such an extent that steam condenses out forming an emulsion when mixed with oil vapor. I reckon the emulsion or gunk will effectively block up any demister pad or baffles in the catch tank including air filter at its outlet.

Share this post


Link to post
Share on other sites

Having a slow day at work, I was able to read through the full post, all 197 entries.Two questions.1. This post is about removing the oil from the air leaving the rocker cover and keep it from fouling the EGR and intercooler, and the the whole purpose of the cyclone is to reclaim and keep this oil from poluting our wonderful planet as opposed from using the "mother of all catch tanks"....right?2. According to estimates, 7/10ths of a liter of oil is recovered. This part confuses me as this oil would never have been recovered with the stock set up, it would have been passed through the engine with the intake air and burned. Yet with a good running, tight engine, our oil consumption is nowhere near this amount between oil changes....Am I mising something?Thanks

Share this post


Link to post
Share on other sites

1. Yes

2. Much of it would not have been passed through and burned. That is the whole point.

That is what the goal of a properly working CCV system is supposed to be. But in this engine, it does not work that way. A whole lot of oil condenses inside the intake hoses, the intercooler, and the intake manifold, eventually leaking out holes in the intercooler, and gumming up the insides of the hoses and manifold. Increasing soot, plugging up the EGR mix valve (also a design failure), and obstructing the catalytic system.

That is why we owners have been forced to devise our own working systems and/or defeat the OEM systems - because the MB/smart engineers failed in the efficient design and implementation of their mandated emissions controls.

Bil :senile:

Share this post


Link to post
Share on other sites

... According to estimates, 7/10ths of a liter of oil is recovered. This part confuses me as this oil would never have been recovered with the stock set up, it would have been passed through the engine with the intake air and burned. Yet with a good running, tight engine, our oil consumption is nowhere near this amount between oil changes....Am I mising something?

It is according to measurements (not estimates) on my Cdi, that approximately 0.7 liters of oil is recovered over 10,000 km of driving. When changing oil, level will be near its maximum mark. There is 1.0 liter between max and min marks so you can loose 0.7 liters and still be above min mark. I used to top up around 0.4 liters between oil changes but had a leaking sump as well, which did not help much.Have a look inside your intercooler and report back what level of oil you have. Access is easy, just remove plug for IAT sensor, then IAT sensor itself and insert a bent cotton bud and sound oil level. You will be amazed by the amount of oil contamination. Then look up through hole for IAT sensor into the intercooler tubes. The tubes have fins inserted. Tubes and fins will be covered in black oil.The engine will breathe easier if you can keep inlet system clean of contamination. The benefit is slightly more power and better fuel economy.I am going on a 7,000 km journey with my Smart soon and will have ample opportunities to test out my new cyclone air oil separator. Hope to be saving 50 pounds worth of fuel on this journey thanks to my contraption.

Share this post


Link to post
Share on other sites

There has been a lot of discussion on correct oil level for the CDI engine here, and a general consensus that right up to the MAX line is a bit too much, and will increase oil entering the breather air. I'd love some more measurements showing if that is true. I generally run 2/3 up the dipstick level and warm, change my oil at 5000 Km and it is down to a hair over halfway, so my oil consumption is only about a third of Tolsen's. 145K on the engine, lots of highway, favourable coastal climate, no EGR problems although I've done a precautionary clean-out twice. I've been on this change schedule since about 30K and no increase in consumption. (Yet, it'll come!)

Share this post


Link to post
Share on other sites

Perhaps my oil level is too high.

Wonder how you read oil level? Engine at operating temperature, level surface and engine sat for about 5 minutes? Do you read at the front or at the back of the graduation marks? The answer to this is not found in the owner's manuals.

Posted Image

Oil volume between high and low marks is 1.0 litre according to my yellow tipped dipstick. The difference I read off between front and back of dipstick is around 8 to 10 mm. This is considerably greater than the theoretical difference "h" which is:

h = t x cot(α), where t is thickness of dipstick (3.o mm) and α is angle of business end of dipstick relative to the horizon (approximately 60 degrees).

Share this post


Link to post
Share on other sites

If this method recovers 7/10ths of a liter on a typical engine, what is the quality of the oil being recovered?If it has been vaporized, would it not be beginning to break down and not of a quality that we want to add back to the engine....Just a thought

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0

    Chatbox
    You don't have permission to chat.
    Load More