OK--I believe in the centrifugal oil filter now!!

[Snakeoil]

Another concern over modern detergent oils, especially synthetics with their ability to loosen up even the oldest and most stubborn sludge deposits, is that it can take old dirt out of the sludge trap and circulate that thru the engine as well.

I guess I can understand that and it makes the need for a true filter more important than ever. Not having a true filter on these engines is one of the reasons I won't run syn oils in them. And I'm a big fan of syn oils.

There is an article written on oils in relation to motorcycle engines that I may have mentioned here before. I found it very enlightening and helped me better understand oil and how to select the right oil for our little Hondas. NOT STARTING AN OIL THREAD!! I've come to the conclusion that oil choice is a lot like religion and politics and better left undiscussed. But for those who would like to read that article, you can find it here.

http://www.calsci.com/motorcycleinfo/Oils1.html

regards,
Rob

[bonjour]

The biggest reason why Honda probably got rid of the mechanical filters was
less power loss
Less weight
quicker gas respons
less breakdowns
less maintenance
smaller engine design
cheaper to produce

And thus overall more reliability.

Better filtering was likely not the first reason why it had to go.

Overall im happy with my spinning filter, no worry's about buying new filters.
And I think it works great.

I just use modern full synthetic oil, no problems here.

[brewsky]

Another concern over modern detergent oils, especially synthetics with their ability to loosen up even the oldest and most stubborn sludge deposits, is that it can take old dirt out of the sludge trap and circulate that thru the engine as well.

Rob

Rob,
I have seen that exact thing happen to a friend who bought a used Audi, which hadn't had the oil changed in quite a while.

He immediately took it for an oil change and service after he bought it, and within a week, the engine locked up due to sludge loosening up and blocking oil passages. (He had previously bragged about how good a deal got on the car!)

I don't recall seeing that much sludge inside the cover of my Dream, but I haven't had the cases split either. So I don't know first hand how effective the spinner is at preventing sludge build up. I'm sure a lot has to do with how often the oil is changed also.

I do know that my 750 had quite a bit of sludge built up in the oil tank after 40,000 + miles, and I change oil and paper filter every 2,000 miles.

The spinner filters are more common on diesel truck fleets and construction equipment to "save money, extend service life, and be more environmentally friendly" if you believe the sales pitch.......which I have heard first hand, with all the handouts to "prove it".

Some companies actually use what I would call "oil dialysis" units, which come to the jobsite, clean the oil from a bulldozer, for example, thru a portable spinner unit, and then leave the equipment with freshly cleaned oil.

The pitch basically included changing all the fleet filters to centrifugal, extending the oil change life indefinately, and having routine oil sampling and analysis results tell you when to change oil, and to forecast when an engine would need overhaul.

We actually fitted a few "pilot units" for evaluation, but eventually abandoned the idea based on several factors, mostly the inconvenience and problems involving collecting and transmitting the oil samples.

It does seem interesting to me that the spinners are optional extra cost on some pretty high priced equipment.

[jensey]

Hi

I have the feeling everyone is running in circles, because in the oil thread, all of the above issue's are discussed.

This is what I wrote about it :

Hi Steve,

Maybe I wasn't clear why the spinner oil filter doesn't work as good as a paper filter, and I apologizes for that, I'll try to explain it in a better way.

The principal of a spinning oil filter seems to be simple but isn't when you look at the details.

The idea is that by centrifugal force particles are driven toward the edges of the filter where these particles will cumulate. Once in a while you clean the filter and the process starts over again.

The centrifugal force on the particles is depending on the circular speed of the filter and the weight of the particle and the distance of the particle from the center.

Fc= m*w2*r (2 is square)

Fc is the centrifugal force
m is the mass of the floating particle
w is the circular speed
r is the radius or distance from particle toward the centre

In this particular case (honda 305) the w (circular speed) is not constant, it changes linear with the rpm, it's influence in the formula (thus on the particle) changes with the square of that.

The first experiences with full synthetic oil (5W20) I had where with Honda SS50 / CD50 engine.
I experienced that, when I used the oil, the engine was cleaned within a few hundred miles, and everything what was lying around in the cases would accumulate in the spinning filter (part of the clutch). So, in this case the spinning filter is working fine.

Later I experimented with full synthetic oil on larger engines (CB72) and found the same results.
The first conclusion was that the spinning filter works fine with full synthetic oil, so I started using this oil in a rebuild engine (and a new oil pump). Every 1000 km I cleaned the spinning oil filter and every 4000 km I changed the oil. After 12.000 km I was curious how the internals of the engine would look like, so I decided to open it up. Everything was in as new condition and measurements after 12.000 km where comparable with the initial values before installation.

So nothing wrong here ? Yes and no. I also inspected the oil pump and I found wear at the oil pump gears (as stated it was new 12.000 km ago). I checked with the oil pumps used in the
small displacement engines and there was no sign of wear. Time for looking for differences.

It took a while, but in the end I found the few reasons for the wear on the oil pump in the CB72 engine. The mean reason is the difference in construction between the oil filters, and I will explain with the formula above.

When I compared the values in this formula for the two engines I found a few differences.

Mass off the floating particles in both engines are (assumed) the same.

The radius of the area where the particle is (toward the centre) in the centrifugal filter is depending of the radius of the filter itself, and here I found a difference in construction. The radius of the small displacement engine is a little bigger then in a CB72, so the force on the particle is higher.

The circular speed (w) is depending on how fast the filter is turning around, and here I found the also construction differences. In the small displacement engine the oil filter is attached to the clutch (part of the clutch) and this clutch is directly attached to the crank. Therefore the speed of the filter is the same as the crank speed. In the CB72 it’s not directly attached to the crank but the power is transmitted by a chain. The chain gears on the crank and oil filter would be the same there isn’t any difference, but this is not the case. The filter runs with a higher speed then the crank, not much, just a little (24/22 teeth).

The difference in driving styles are also responsible for a difference in speed of the filter.
Since the crank speed is depending on the rpm, the rpm of the engine dictates the force (Fc) on the floating particle. I ride the small displacement engines with much higher revs as the CB72 engine, because of the lack on power at certain speeds. Where I tend to run my CD50 at a constant speed of 55 km/hour the engine is running over 7500 rpm. I calculated with an average of 7000 rpm. At the CB72 I’m not running this rpm as often as I do on the CD50. I calculated with an average rpm of 4500 for the CB72.

So, when we look at the impact on the force on the particle, with the data above :

Fc for the CB72 compared to the Fc for the CD50 :

Mass of particles is the same, so m (cb72) = m (cd50)
Radius of the CB72 is 0.8 of the CD50, so 0.8 r(cb72) = r (cd50)
Circular speed of the CB72 is 0.64 of the speed of the CD50, so 0.64 w (cb72) = w (cd50)
The construction factor between CB72 and CD50 is (24/22), so 1.09 w (cb72) = w (cd50).

When filled in :

Fc (cd50) = m.w2.r and Fc (cb72) = m.((1.09*0.64)w)2.0.9 r
Fc (cd50) = 1 m.w2.r and Fc (cd50) = 0.43 m.w2.r

This means that the centrifugal force on a floating particle in the CB72 is a halve of the centrifugal force on the same particle in the filter of the CD50

A full synthetic oil is a very powerful detergent, it will clean the engine very good and drop the particles in the filter due to the above explained principal. The very small particles (like cool) are not filtered out by the filter, but don’t do any harm to the engine either.
But if the detergent power of the oil exceeds the influence of the Fc on the particles in the filter, these particles will not be held in the filter and start floating around in the oil.

This means that when the running speed of the engine is low (rpm) the oil starts cleaning the filter itself, and this is probably why I found wear to the oil pump gears. In the breaking-in process I never push the machine to it’s red line, so dirt collected in the filter was washed out and went through the pump.

The conclusion of the oil analyse was that there where a lot of particles floating around in the oil, which should be filtered out.

After this ‘discovery” I drained the oil more often, found less particles in the drained oil, and the oil pump stayed within specs.

This is the reason why I developed an oil pump with a paper filter, sometimes I like to ride low rpm’s (as a tourer) without the knowledge of ruining my oil pump, and I don’t want to change the oil every 1000 km or so.

I analysed the oil and after 6000 km, the oil is still good enough to use.

Like I stated earlier, with this oilfilter solution I can use all the benefits of a full synthetic oil without the problems it causes at low revs.

So before you put a half synthetic / full synthetic oil into you engine, make sure that your oil filter is clean. If there's any sludge in your engine cases it will be washed out by the new oil and dropped into the spinner oil filter. Needless to say you have to check the oil filter within a very short period after using the new oil, depending on how clean the internals are.

I hope this explains a little more about the choices I made.

Jensen

And I wrote this :

I explained the conversion elsewhere on the forum, but here the text :

In theory it's a simple conversion, practically it was a little more difficult.

The theory / wishes / engineering considerations stated in 2003 where as follows :

1. All the oil should be filtered before entering the engine, no unfiltered oil entering the engine.
2. As less as possible machining on crank cases, it’s difficult and I can’t change it back in the original state.
3. Because the oil pump is rather a volume pump than a pressure pump the oil filter should have a low flow resistance and a large surface.
4. Changing the oil filter should be really easy, without opening the engine or removing important parts.
5. Should be able to work with 10W50 full synthetic oil, and the oil filter should be easy to buy.

From a logical point of view there is only one place that meets the criteria above and that is a place before the oil pump, so I decided that the oil filter should be placed underneath the pump itself, as a part of the pumphousing.

Another advantage was that the the oil is drawn down by gravity to the lowest point (oil filter) and thus the oil pump itself is not working different than in any other standard engine.

I bought several used oil pump housings to experiment, and finally I machined a three part oil pump housing where the top (part in the engine) stays the same.

I removed the centrifugal oil filter, replaced the oil filter shaft by a direct pipe with a diaphragm to regulate the flow between crank, clutch and head and dismissed the oil filter chain.

After years of optimizing and further development (I destroyed / machined at least 12 oil pump housings) I got what I wanted, I even included a pressure sensor and a temperature sensor. On top of that the engine also gained an extra horsepower due to removing the oil filter and the chain (and less noise from the engine).

I hope that this answers the question,

Jensen

Been there, done that.

In the mean time I took the engineering a step further.

With an oil pump in new condition this solution works fine. The oil pump delivers enough pressure to overcome the resistance of the filter, and the pressure drop isn't giving any problems.

But when the engine is cold, it takes longer for the oil pump to overcome the pressure of the cold oil in combination with the paper oil filter (cold oil has a higher viscosity). For that reason I had an extra electrical oil pump (low flow, but high pressure) attached which pumps oil to the head before the engine is started. I works fine for me, even better then what will be possible in another way.

But a few CB72 owners here in the region showed interest in this conversion, and of course they don't have a new pump, and neither an added electrical pump. So I was asked if I was able to produce a simple add-on for the CB without all the hassle.

There are a few solutions, using a bypass, or using another type of filter or make the filter surface bigger. I looked around and found non paper filters with a much lower resistance.

At the moment I'm testing this conversion, and I am curious what it will do with the temperature around zero degrees. I'll keep the forum informed if interested.

Jensen

And this :

The pro's of (modern) synthetic multi grade oils are (I guess) clear to everyone who's interested in life expectancy of there beloved piece of mechanics.

But I think that there are two main reasons why people hesitate to use modern oils

1. Viscosity.

multi grade or not, 30W isn't 5W or 10W. As stated in modern manuals it's important (with modern engines) to stick to the manufacturers advice , as I explained, it has to build up pressure in a therefore designed system to lubricate the non roller bearings.

2. Detergent or non-detergent.

Everyone wants a clean engine, so everyone should use a detergent oil, at least , that seems logic to me. Why would someone use a not cleaning (non detergent) oil ? The only reason I could think of is that the oil filter wasn't working optimal, it's designed in a time when there weren't detergent oils.

There are no other reasons (I can think of) why someone should hesitate not using modern oils in a CB72/77 or a C72/77.

Leaking packings, engine noise and many other reasons are fairy-tales or worn engine reasons.

I didn't care about the viscosity difference between a 5W or a 30W, multigrade / monograde. There are no non-roller bearings, and the oil pump is working fine when the oil pump is within specifications (I wanted to be sure so I tested it). There are not many (mechanical) changes between the oil pump of a 305 and a oil pump used in modern engines (gear pump, trochoide pumps).

But I did care about the way the oil is filtered, because in today's engines (automotive industry) you won't find this filter principle any more. Why not ? Is it only for economical reasons ? Is it because of maintenance reasons ? Or is it because it's not optimal from a technical perspective?

Maybe it was a little bit of the reasons mentioned above, but I wanted to be sure, as no one could tell me why precisely. However, in the non- automotive industry the spinning filter method is used in many forms, but mostly applied with a fixed rpm. Why ?

I tried to explain the reasons in this thread and I used physics for that, and in my opinion it's the only way to understand what's happening, because the spinning oil filter works because it's acting by physical lows.

The differences in force applied to the particles in the oil within the rpm range between tick-over (1000 rpm) and the red line (10.000 rpm) is the square of 10, so 100 times bigger. That would not matter if the nominal force applied to the particle at tick-over would be enough to keep it in the filter, but the filter isn't running fast enough to accomplish that in combination with the detergent power of the oil.

My conclusion is that it works fine within a certain rpm range. It's not working optimal for low rpm's, but does it harm the engine when sticking to the periodical oil changes ? NO, at least not worse then non detergent oils.

Due to filtering characteristics of the fine mazed oil filters (not depending of the rpm of the engine) these are working better for the whole rpm region, and makes it unnecessary to change the oil often.

Jensen

I did a lot of research myself, and not only a few oneliners on a forum, so I would appreciate that others apply there TESTED data, and their conclusions.

"If", "else", "or", "I think", "probably" are empty words and add nothing to the subject or our knowledge.

Like I stated, test in your own engine, because then you really have to take risks, and if it fails, it cost money, real money.

The SG is important but that doesn't mean that it says it all, at low revs the spinner filter method will work with a bigger tolerance between the SG's between oil and foreign particles. The problem is that the centrifugal force on particles is not linear with the oil flux (flow), that's the reason why it doesn't work in all situations.

It means that at low revs the spinning filter works less good then at high revs, for all materials it isn't a 100% solution, depending on the rpm of the filter in combination with the detergent power of modern oils.

Like I said, it's all discussed already.

My final conclusion is that spinner filter method works fine as long as the centrifugal forces overcome the force of the flowing oil itself, pushing the particle through the filter (also depending on shape and resistance). Since the filter is driven by the crank, and the oil pump too, the rpm of the oil pump is linear with the rpm of the filter, but not with the centrifugal force applied to the particle itself.

I found that at low revs the force that works on a particles a certain time (the time that a particle travels through the filter), is just not enough to push it to the wall, it travels through the filter.

Thus, in the case of the CB72/77 and C72/77 or CB450's it doesn't work in all situations, at all times and on all particles, therefore I switched to another filter method in all my classic bikes were this filter method is used.

These conclusions are based on my own mathematics, experiences and tests, not on what other people believe, say or think.

Jensen

[brewsky]

Jensen, your statement here........

"My final conclusion is that spinner filter method works fine as long as the centrifugal forces overcome the force of the flowing oil itself".......

Makes much more sense than your initial statement here...........


"as long particles don't have the same specific gravity as the oil you use"

I do appreciate your detail and your initiative to improve, and I actually searched for your initial filter conversion thread without success.

HOWEVER....asking me to crush some glass and put it into the engine oil is a little over the top!!!!......You first!!!

[jensey]

Hi,

Both statements are connected via the forces on the particle while travelling through the filter, so both statements are correct.

The only thing is that "the same" should be interpreted as "around", because the bigger difference between the SG of foreign material compared to the SG of oil is, the faster the particle will be forced to the outer radius, thus the filter surface.

In the case of glass, this difference in SG, thus indirectly the mass / volume ratio thus flow resistance thus force isn't high enough to push the particle to the outer radius of the filter within in the time the particle travels through the filter, so being transported through the filter and destroy your crank bearings, big ends and all bearings in the cylinder head.

Glass has a SG around the SG of oil being 2.4 versus 0.9, the difference in SG is to small to make the spinner filter work for for example glass. Steel, on the other hand
has an SG of 8 or 9, and in most cases the difference is enough.

Try it, and after a few hundred miles you will notice both statement are correct, and are conflicting with the title of this threat and opinion from some of us.

HOWEVER....asking me to crush some glass and put it into the engine oil is a little over the top!!!!......You first!!!

The test is for the believers, for those who are convinced the spinner filter is working good.
Not doing the test is conflicting with your own believe, and makes it clear that you are not sure of what you're saying.

I know that the spinner filter isn't working in this case (glass) and I can predict what will happen, so I won't perform the test, that would be very stupid.

Actually I think no one really believes that the spinner filter is working 100% in these bikes, at least not people who opened op an engine.

Jensen

[jensey]

Hi,

The function of a filter should be :

Filtering all parts out which can harm an engine and / or block oil ways, in most cases being relative large particles. A sieve does that, that's the power of a paper filter, or a metal screen filter.

You know what you get, the sieve will filter out a percentage of particles as a function of the filterhole size.

Conclusion 1 :

A paper filter filters as a function of sieve hole size.


In case of a spinning filter one doesn't know what the function is exactly, it filters out a lot of particles, but this time based on SG / volume / shape / pressure / flow/ type of oil / construction and in case of our hobby, the RPM of the filter itself.


Conclusion 2 :

A spinning filter in our bikes filters as a function of RPM, radius, SG differences, type of oil, SG of oil, shape, flow, pressure etc.

When a spinning filter is optimized in a fixed situation it works very good, indeed it can even filter out ingredients of oil, it's that good. But in our bikes there isn't a fixed situation, therefore it will not work as good in every (daily) situation, there are just too many parameters.

These were my initial conclusions before replacing the spinning filters for a sieve filter.

Jensen