[Nivek]

I made a very detailed video of comarison and modifying a k24 oil pump so that it can surpass a k20S or R pumps easyly.
This was all new to me aswell and many surprises to how much supirior a k24 oil pump is once you delete the balance shafts. Your feedback is highly appriciated

 

[Stefanzoh]

I'd love to see what lotus/lotuselise has to say about this.
Respect for this resourceful mod!

 

[Lotus]

The porting is always a good idea.

simple:
K24 rev limit is lower, hence higher oil pump speed and larger rotor for the same flow.
K20 revs higher, hence bigger pulley for slower oil pump speed in order to avoid cavitation and excessive losses at high rpm.

Even the “small” k20 oil pump exceeds the oil relieve pressure valve past 3000 rpm. Any more flow would just run around in a circle within the oil pump. ”So much more” is irrelevant from a lubrication perspective. This extra flow never enters the engine.
For a k24, the k24 oil pump is hitting this max. pressure at lower rpm as a k24 is usually operated at lower rpm anyhow. The k24 pump is a better match for a lower reving k24.

once you start reving a k24 to 8k or higher, not a good idea in my opinion if you want this engine to last, you might want to use a k20 oil pump. Less losses = more HP.

Removing the balance shafts reduces weight and reduces losses, but still does not prevent excessive oil pump rpm at very high engine rpm.

I would assume that most higher reving k24 builds would benefit from removing the balance shafts, but not do really benefit from a k20 oil pump unless they rev way beyond what Honda intended.

You could have the same argument with mechanically driven water pumps.
They need to spin fast enough to maintain reasonable flow at idle, but should not need excessive power near the rev limit. Nevertheless they spin way faster than they should cavitation a lot because of that idle flow. Once you build a race car engine, you can increase idle speed to say 3000 or 4000 rpm. This allows to fit a bigger pulley on the water pump. It is now better matched to the main operating regime of a race car engine, the upper third of the rev range. You can gain 1-2HP that way. In competition you’d kill for an extra 2HP. Just never try to sit in a traffic jam with that engine idling at 1000 rpm. It would overheat.

 

[Nivek]

The porting is always a good idea.

simple:
K24 rev limit is lower, hence higher oil pump speed and larger rotor for the same flow.
K20 revs higher, hence bigger pulley for slower oil pump speed in order to avoid cavitation and excessive losses at high rpm.

Even the “small” k20 oil pump exceeds the oil relieve pressure valve past 3000 rpm. Any more flow would just run around in a circle within the oil pump. ”So much more” is irrelevant from a lubrication perspective. This extra flow never enters the engine.
For a k24, the k24 oil pump is hitting this max. pressure at lower rpm as a k24 is usually operated at lower rpm anyhow. The k24 pump is a better match for a lower reving k24.

once you start reving a k24 to 8k or higher, not a good idea in my opinion if you want this engine to last, you might want to use a k20 oil pump. Less losses = more HP.

Removing the balance shafts reduces weight and reduces losses, but still does not prevent excessive oil pump rpm at very high engine rpm.

I would assume that most higher reving k24 builds would benefit from removing the balance shafts, but not do really benefit from a k20 oil pump unless they rev way beyond what Honda intended.

You could have the same argument with mechanically driven water pumps.
They need to spin fast enough to maintain reasonable flow at idle, but should not need excessive power near the rev limit. Nevertheless they spin way faster than they should cavitation a lot because of that idle flow. Once you build a race car engine, you can increase idle speed to say 3000 or 4000 rpm. This allows to fit a bigger pulley on the water pump. It is now better matched to the main operating regime of a race car engine, the upper third of the rev range. You can gain 1-2HP that way. In competition you’d kill for an extra 2HP. Just never try to sit in a traffic jam with that engine idling at 1000 rpm. It would overheat.

thanks for the valuable info. yes from what i discovered from this is that its not a must to have a k20s or above oil pumps for k24 engine revving may be upto 8200-8500. Just balance shafts del will be enough. It has become a norm combo to have just k24 with a type S pump although for a normal k24(not built) the limiting factor is not the oil pump at these rpms but rather the rods, valve to piston contact, valve springs etc etc.
But those balance shafts Do reduce the engine vibrations. I once used a Type S pump in my k24 swpaaed Fit and hated the ride quality which came after. Swapped backt to k24 oil pump with balance shafts in after a week

 

[Lotus]

It was nice to see the comparison of the two pump types.
Honda added the balanced shaft pump even to the later k20 models in later Civic Type Rs.
NHV can be an issue, especially with harder engine mounts.

 

[AWD]

Interesting. My K24 Element speedometer does not rev past 8000 rpm. Honestly, I rarely go past 6500 RPM in regular use.

 

[yaggs]

If in a pinch with $, could one potentially switch the cog from the k24 to the type s pump? This could save money with new chain and guide, albeit not much. This would also increase idle oil pressure, no?

 

[Lotus]

yes, that would increase idle pressure for a k24 running a k20 oil pump. it would compensate for the smaller rotor the k20 oil pump is using.
But is this really a problem?

 

[yaggs]

yes, that would increase idle pressure for a k24 running a k20 oil pump. it would compensate for the smaller rotor the k20 oil pump is using.
But is this really a problem?

Obviously not since the k20 pump has been tested as reliable. Just thinking of the cost savings with the chain and guide

 

[Nivek]

Obviously not since the k20 pump has been tested as reliable. Just thinking of the cost savings with the chain and guide

its not only chain and guide. you have to open crank pulley to replace chain and to do that whole head has to be removed and a new gasket and a new headstuds. thats a lot work. But if you are building the engine from scratch than your argument is valid that its just a chain and guide and k20 pump cost and to keep oil pan fittment issue too if you have a k24. For a k20S/R pump the oil pan should also be changed to that of typeS/R as it is deeper.

 

[MoHo]

its not only chain and guide.

It's hard to beat factory Honda. They do things for a reason. I've worked at a foundry and mass produced parts/assembly's could almost always benefit from deburing and what not. To change there enginering is questionable as to someone told you "it's a good idea?" Has that someone spent millions of dollars and thousands of hours coming up with this?
A few moons ago I read about the: K20a2 "EZ" head modification for improved top end oiling? I was unaware of this mod, searched and still have yet to find anything about this? I am in know way knocking the altering and or adding editions such as better breaks because your car does 11's in the quarter or stamps out 1.45 lap times at carlesbad. Case in point there are some seriously hard core OP's up in here. Then there's the average guy. All of this is important, in moderation. To much of a good thing, not alway's a good thing? Ok I need a drink.

 

[Lotus]

Joe Mccarthy posted pictures of that oil tower mod to increase oil pressure to the feed to the cam bearings. Usually it is not a problem, but I think he was pushing things at the time, cam design was not as good as today and he loved to tinker, mostly in an useful manner, anyhow.

 

[yaggs]

to replace chain and to do that whole head has to be removed and a new gasket and a new headstuds

This is news to me. Pretty sure you can remove the crank pully, timing cover and timng chain to get to oil pump chain and guide? Why would you have to remove the whole head?

 

[LotusElise]

I made a very detailed video of comarison and modifying a k24 oil pump so that it can surpass a k20S or R pumps easyly.
This was all new to me aswell and many surprises to how much supirior a k24 oil pump is once you delete the balance shafts. Your feedback is highly appriciated

Thanks @Nivek for sharing your recognition's about your K20 and K24 oil pump comparison 🤗.

I would have some questions:

1. In a first order the flow of a positive displacement pump = displacement x rotational speed. Hence my question, did you measure the displacement of both Gerotor's?
2. Do you have an idea how much power is consumed by an oil pump over engine speed and what the residual gear transmission losses would be of the K24 oil pump?

As well as I would like to add some thoughts:

• the limits of mass balancing system integrated oil pump are rotational speed
• cutting the balancing masses off open that door in favor of the mass balancing functionality. Not to forget, the mass balancing demand increases with engine speed with the inertia related power to two. As the K20 oil pump equipped 99 mm stroke engines prove the increase of engine speed without balancing mass is not harming the engine in short or mid therm on race track with an 8500 rpm limiter on a stock K24A3 longblock, this approach is definitely an alternative way to reduce the pressure on the K20 oil pump market.
• the higher friction and inertia of the pump would be reduced, but a good part of it would be left
• I wouldn't change the drive wheel before it is clarified how much the displacement difference is

 

[LotusElise]

...but I think he was pushing things at the time, cam design was not as good as today and he loved to tinker, mostly in an useful manner, anyhow.

I have two customers with 4Piston RR4 cams with worn low speed cam lobe and partly worn high speed lobes. There are still companies in business, which don't understand the engineering part of cam design as well as not the customer application consultation.

According customer these cams saw only 15,000 km. Oil, break in are done according 4Piston, the engine is a 4Piston crate engine. The whole engine is a mess, but the cam topic fits here. There are still companies, which still don't know how to specify valve springs and cam lobe designs which don't stand a year in a DD engine.

 

[Lotus]

but the cam tower mod was for the cam journals, not the lobes. These are just splash lubricated as they use roller rockers. Still should not happen today anymore.

 

[MoHo]

but the cam tower mod was for the cam journals, not the lobes. These are just splash lubricated as they use roller rockers. Still should not happen today anymore.

Good discussion. These things should be hashed out? There's miss information out there and ones needs might not be that of anothers? Again I'm not knocking modifying things, for cirumstance that anything would be put into. I'd soup up a weed wacker if it looked at me right or wrong ? Tell me you don't hate stringing those things? There, we can all get along. 😁
I've just learned over time that some things that are being feed to me doesn't always mean it's tried or true. It just may be, some B.S. Those cams ain't looking to sexy? Could you please expand as to what went wrong. I'm in the middle of a rebuild and trying to put the money in all the right places.

 

[LotusElise]

but the cam tower mod was for the cam journals, not the lobes. These are just splash lubricated as they use roller rockers. Still should not happen today anymore.

The VTEC control is a safety linked feature and has therefore 1st order priority. Thence, once the VTEC is not engaged a greater part of the oil flow is kept away from the head. This is the fail safe setup:

• VTEC solenoid has 0 V and is closed
• VTEC pins are in unlocked position

The consequences of that is that a greater part of the oil flow through all head oil sink systems (rocker shaft, tower bearings) is reduced as the oil supply over the VTEC solenoid is closed partely. About a 10th of the pressure can be found at the non-VTEC operation at those sinks, flow is lower accordingly. Now the rocker shafts are feeded by #2 tower. What Joe did, was plugging those #2 saddle rocker oil sources and separate tower and cam lobe oiling separately over the #5 tower oil source modification. So that the tower oiling and the rocker oiling is supplied by two independed lines. VTEC is still working, while oil flow to the rockers shaft is increased massively.

So his oil modification was not only for the cam bearings, it was mainly for the rocker shaft supply, which is the basis for the cam lobe oiling. His intention was to suppress lobe wear which cam up once valve acceleration and lift increased massively by e.g. the first versions of the S2 T2 cams, which was compensated by huge valve spring forces. What really killed these lobes was the Hertzian pressure in a oil lacking system around idle. Those casted cam lobes worn down fast and did their thing on the rollers of the finger followers too.

The oil pump relief spring mod you have hit the same mark too. The increased oil pressure allows an higher oil flow under non-VTEC into the head, which is not necessary if oil temp and oil grade as well as warm up operation are done right. Of course the game changes if oil grades are increased and or or oil temps are lowered. For what ever reason some race guys (KTuned, ...) think, this would be a good idea .

As I said, a bigger portion of the aftermarket misses basic engineering standards, coming up with wrong solutions for misunderstood customer issues. The lower thermostat balance temperature is a good example. They see customer ECT goes through the roof, solution start earlier big circuit cooling to raise some time until that point. Totally wrong solution if driving around is also an application. Same with the RR4 cams of 4Piston, more VE found by harder acceleration ramps and lower durations, but it comes with valve bounce in higher engine speed. Solution by 4Piston, increasing the valve pressure like NASCAR did it 30 years long. But we are not at a circle race or 1/4 mile, their customers also drive from home to track and back or even commute with it. They totally missed the fact that their valvetrain system is only proper working in VTEC for short time only. That is a totally mismatch of application and part development. An engineer would know that these two always need to be overlayed. They are still the garage guys with huge 1/4 mile experience, but not capable to transfer that knowhow to other applications as they let miss basic engineering standards.

 

[MoHo]

while oil flow to the rockers shaft is increased massively.

It would seem that under the right driving conditions and use you support the head mod for increased oil flow to the head. One must note that you can suck the pan dry....If your running a turbo pressure and flow characteristics also change.? To this day running cams takes a toll on the valve train. Tensioners and cam wear have dogged us since day one. I am curious to your insite on spring pressure I'm gonna run 06-08 tsx cams, I have blox duel springs ( 85lbs open and 220 closed, I might have to double check that!) with factory Valve train (2002 k20a2) With the exception of Guides and keepers that are curtiesy of supertech.
Spring pressure seems to be a very fine line between float or wear...please elaborate on this? Springs wear out. I;m also running new factory lost motion springs. A lot of people overlook this. I'm noticing a trend as to how people "look" at cams. You don't hear much about lobe separation for instance. I'm running a 383 stroker in a camaro with 108 degrees of lobe seperation. Hard to get enough vacume to even run the break booster! The powerband is peaky but if driven right, it rips. Cam values as advertised tells us very little when it comes to imports. Nobody is sharing much about it and industry terms like @.050 is like what? Like it's some vodoo, you can pretty much find anybody to complain about any damn aftermarket cam there is. Most problems seem to be tensioner failure? I am in no way trying to beat the way dead horse as to cam selection. To each his own. So I invite you to delve into this slippery slope. My valve train can handle more cam That day may come. Tear it up, theres a couple of points to make here.

 

[LotusElise]

It would seem that under the right driving conditions and use you support the head mod for increased oil flow to the head.

I am not sure if I understood you correctly. You ask if I support the head oil mod of Joe McCarthy if driving conditions and application are met? Not intentionally. I am a fan of the OE system, which is quite effective and efficient. All performance improver has to meet these given requirements. If one want to increase the oil mass in the head, this is a risky way. Joe understood finally the balance point of oil supply by pump, oil bypass to the 5th tower and VTEC capability. If you balance it too much to the 5th tower mod the VTEC capability will be reduced, which can cause harmful driving situations, if balanced too low the cam lobe/rocker oil supply can be too low. The tolerance of the system is greatly depended on the pump performance, oil grade and temperature. I don't recommend doing this for those who don't have a glue about the background as they won't be lucky if lobe wears or VTEC engagement loss broke their engine down and start a new trial for a different balance point.

One must note that you can suck the pan dry....If your running a turbo pressure and flow characteristics also change.?

Sure, any oil storage like a oil cooler or oil bypass sink like a rotational aggregate (turbo charger or supercharger shaft) changes the oil flow into the head area, as that is the last in the line of supply chain for oil. After the head oil system, the oil flows pressureless back to the sump. So it is crucial to think about if more stand still oil volume (not a higher oil filling level) does the job or if the pressure at the source has to be increased. But a turbocharger won't run your pan empty, as only the oil pump and the backflow balance point give you the oil nievau level in the oil pan. Of course more engine speed increase the mass of oil moisture/droplets, not available in the pan, but that is valid for any engine setup.