Oil pressure and (probably) temps--arrgh

[Gansan]

Hello folks,

For quite a long time, I've followed the threads on oil pressure drops during track use, mainly because I've experienced them when I go to the track too.

This past Saturday, I was at Buttonwillow with SpeedVentures. I found it a frustrating day because I was continually plagued by dramatically falling oil pressures at high rpm. As a result, I was driving around with one fearful eye on the oil pressure gauge, and I was always short shifting to keep the rpms down. Not fun.

Now I know that this kind of stuff has been discussed at great length in the past, but I'm writing it down to contribute data to the group wisdom.

Car: 3.2L, completely stock engine. Running 10w30 Mobil1. Falken Azenis rt615.

Symptoms: When the engine is warm, but not hot (as in the morning session), the pressure drop occurs at high rpm, but never falls much below 4 which is halfway on the stock gauge. The pressure drop seemed to be really significant (falling to around 2 from 6-7) primarily between 7000 and 8000 rpm and when the engine is really hot from multiple laps late in the day. The drop seemed to get worse in proportion to the amount of time the engine spent at high rpm, which happens during acceleration on long straights at higher gears. The pressure would recover nearly instantaneously after an upshift and the corresponding engine speed drop.

My thoughts on the symptoms: The oil pressure drops I experienced seemed to be related to rpm, not really cornering Gs. The reason I think this is that although the pressure drops would often be observed at corner exit (when both rpm and cornering forces are high), I would see similar or worse drops during pure straight line acceleration. In both situations, the pressure would rise as soon as I upshifted and rpms dropped, which also makes it seem rpm related.

There are three primary reasons as far as I know for oil pressure problems (not including obvious ones like oil pump failure or the filter falling off).

1. Cornering causes oil pickup to starve.
2. Very high oil temps reduce viscosity, which reduce pressure.
3. Overfilling the oil causes crank to splash, causing foaming.

I must confess I was guilty of overfilling and causing #3. I had been trying to get the right amount in, but I was having a lot of trouble reading the level on the dipstick. Finally, I cleaned the dipstick completely with brake spray and checked again and found I had overfilled by about 1/2 quart. :frown: Halfway through the day I figured this out and I drained the extra out. In subsequent sessions, the pressure drop seemed better, but it was still significant enough to warrant concern.

What is somewhat confusing is that this problem does not happen at all tracks. For example, at Laguna Seca, the oil pressure never dropped below 3-4 even on the front straight. I was paying keen attention to the pressure because I had previously had pressure issues (again at Buttonwillow). I know the oil was not overfilled that day, but the weather was much cooler as well. On another day at Streets of Willow, I did not notice any issue at all, and that was a cold and rainy day with temps in the upper 50s. So far, my experience has been that warm days + high speeds = low pressure, although my data is limited.

So that leaves the basic question of the typical countermeasures the people take against these oiling issues. As I understand it, options include:

1. Baffled oil pan
2. Accusump
3. Oil cooler

I am somewhat torn on these options because I don't want to make my car a track car. It's an extremely clean Zanardi and I like the fact that it is stock (okay, I have some NSX-R parts, but it's all reversible :tongue. But, I want to take this thing out every so often, so maybe I should consider some oiling robustness measures. After all, spinning a bearing or blowing up is not worth trying to keep my car "stock and clean." Even just the stress of watching that gauge is just ruining what is otherwise a great day.

My first choice is actually the oil cooler (looking at the SoS one). It seems like the least intrusive thing that may even solve my immediate problem on its own. The NSX runs very hot oil temps as well documented in Best Motoring's "Endurance Battle" episodes where they show even ten laps on a hot day at Tsukuba made temps go up past 130 degress C (266 degrees F). In my Integra, which is equipped with an oil cooler, my oil temp is rock steady at 215-220 even on the hottest days and my oil pressure is always at 50-60 psi no matter what I'm doing.

I have no doubt the oil pan would help in corners, but my immediate problem seems unrelated to cornering Gs.

With the Accusump, I feel like that is something that addresses oil starvation, not oil temperature and viscosity. I guess it will help push oil through the bearings when it's all hot and watery, but I'd rather cool the oil down first. Plus, it seems like a very intrusive mod with plumbing into the trunk. This is a street car after all.

What do you guys think? I appreciate any input.

 

[john@johnhudakracing]

Just replace your stock oil pressure sender.

 

[Tytus]

Oil pressure drop occurs when the Vtech system is engaged above 5800 RPM. That is when oil pressure drops as the oil drives the system to engage the high rev cam lobes. I may not be describing the sequence correctly but it is something to this effect. My car also drops oli pressure at high RPMs and recovers in the upshift when you fall out of Vtech range. I beleive this is normal.

Try a search on the subject in Wiki.

Tytus

 

[greenberet]

There have been threads about this in the past and I believe no one has definitively determined why the reading on the stock oil pressure gauge falls to one or two bar during extended high-rpm driving. If it were a problem of oil sloshing around the unbaffled pan, it shouldn't happen in a straight line. If it were a problem of temperature, the oil pressure shouldn’t recover so quickly when the revs drop. Replacing my oil pressure sender didn’t change the high-rpm readings but if I remember correctly, if the oil pressure were really that low, the EMS would disengage VTEC

My oil pressure gauge behaves like yours and the engine does not fall out of VTEC during extended straight line high rpm driving. The engine also still seems to be healthy. Based on that, I’d guess the oil pressure at the camshafts and bearings doesn’t really drop to almost 0 at top speed, as the gauge indicates. At least I hope it doesn’t and that for some reason, the gauge just gives weird readings.

 

[stuntman]

BW 13B CW is notorious for causing blown motors due to oil starvation. Since the NSX dosn't have the best oiling design, and since the sump is on the passenger side, the "riverside/talledega" as well as "sweeper" corners starve the motor of oil. This problem becomes worse the faster you go through riverside with stickier tires, and as you start picking up full throttle sooner.

Get an aftermarket oil gauge,
Get an oil cooler

And I would suggest an accusump and baffled pan, those aren't 'huge' modifications and IMO does not hurt the value of the car.

 

[NSXAMG996TT]

^^^well said Billy...

Kudos for tracking the Zanardi (which as I recall was made for the track).

 

[Gansan]

Thanks for the input guys.

My gut feeling is that this is probably a combination of factors. For example, I would find it plausible that the normal pressure drop from operating vtec would be made much worse if the oil was too thin because it got hot.

Okay, so I guess the first thing I will do is look into the oil cooler and maybe the baffled oil pan. It'll be interesting to see what effect they have.

It would be nice if someone developed a non-mechanical, vibration-resistant pressure transducer that was electrically matched with our stock units. Then you could just replace the stock one and enjoy accurate readings from the stock gauge. Someone should step up and sell it.

 

[Mac Attack]

I'm not an expert (I just play one on Prime like so many others), so if I mistate or misinterpret anything someone please correct me. Ironically, I've been studying the oil system since I would like to do a dry sump when going twin turbo, but here's why I think this pressure drop at very high RPM's (obviously above VTECH engagement at 5800-6000 RPMs) is a normal design feature:

From page 5-11 of the service manual, you can see the general oil path. I'll leave it to you guys to study it on your own, but what caught my eye was at the outlet of the pump, there is a relief valve pictured that will supposedly divert some of the oil flow back to the pump inlet when some pump discharge pressure is achieved. In actuality, this is probably open all the time to some extent and doesn't really function as a "relief path," more of a miniflow or recirculation line back to the pump suction. When designing any pumped system, this is a normal feature since the pump will never dead-head and helps prevent cavitation. Anyways, if you go to page 8-6, you can see that this "relief valve" is built into the pump housing....

Now, there is another true "relief valve" on page 8-10 on the oil cooler/filter base. It's not clear to me from the diagrams and pictures on whether this relief valve diverts oil from both the filter and cooler, but I would assume it doesn't. It would make more sense to just divert some of the oil from passing through the restrictive oil filter and still keep everything cooled by going through the cooler. Maybe not - that could be another reason why our oil heats up so much during extended high RPM use compared to other Honda engines.

Anyways, I think this second relief valve at least diverts oil (or some portion of it) from going through the oil filter and being filtered at high pressures (high engine and therefore high oil pump speed). I can't be sure of the actual oil path until I take one apart, but this would definitely explain why all of our cars see an oil pressure drop once we're in VTECH near redline, and there really shouldn't be an issue with it. Obviously, if your oil pressure drops to zero you're screwed.

If others here say they don't notice a change in oil pressure near redline, they may not be paying attention to it :wink:



A few more thoughts about the oil system:

1) From looking at the coolant path, the oil cooler is the last heat load coolant picks up before it goes to the radiator. Obviously, the coolant is already very hot at this point, and our tiny oil cooler means that it can't be very efficient anyways. Instrumenting would confirm, but that is why an aftermarket oil cooler is a good idea.

2) People point to failed motors due to "inadequate lube" on certain tracks where high g's are sustained. Engine oil passages are very complicated. There are a lot of orifices in the engine to control the flow rate of oil to the various places it's needed. On engines that have been neglected or with very high-mileage ones, sludge can and will build up. That's why it's always a good idea to keep the oil changed on schedule and use the best you can justify. People destroy NSX motors on the track and blame it on poor design, but I just don't believe it without knowing the history and mechanical condition of the car. Sure, a baffled oil pan, Accusump, forged oil pump gears, oil coolers, etc are nice, but I think for 99.9% of us they aren't necessary. That being said, a baffled pan would be a simple reversible mod, as well as overfilling an extra 0.25qt of oil or so.

3) Honda engineers spent a lot of time and money on the oil system in our car. Surely this was extensively instrumented and tested as part of their new VTECH system as well as for a "halo" car that would see 8000RPMs of extended use on the street and track. So, if your Zanardi has been well maintained and you want to keep it original, I would just drive it and enjoy it without making any changes to the oil system.

Hope that helps.

Dave

 

[stuntman]

I think their's something to be said when many of the Japanese cars (in Best Motoring) use an enlarged, baffled ARC oil pan. I also think their is a design problem, or known problem when rod bearing #6 consistently goes from loss of oil pressure in track cars.

If you don't drive your car, or push it to the limit, the stock system is fine as you stated. I also agree that most cars aren't pushed enough to need these modifications and the majority of people who take their cars to the track should not worry. But their are people out there who do push the car to its true limit and do need to change what's mentioned above.

If you are doing 2:05 or lower at BW, I would look into said modifications.

 

[Mac Attack]

Were you aware of the two oil bypasses in our vehicles? From a quick search, it doesn't sound like that's been brought up before.


Don't you think Honda engineers subjected the car to thousands of miles on tracks before selling to the public? Did Senna blow any of his personal NSX engines? Did the NA1 NSX-R have oil issues? Do you think there is really a design flaw? It's only until people really start pushing the limits of older vehicle with slicks and maybe a few special turns on special tracks, that you see potential issues with oil.

Baffles only slow down the movement of oil as it slides around in a pan under acceleration (turning is accelerating too). Under a sustained high-G load for more than 2 seconds or so, baffles aren't doing anything really. Trap doors help, but they can also work against you as well.

Dave

 

[Mac Attack]

Don't get me wrong - people make mistakes and no design is perfect. My first post was to hopefully point out some new piece of information that I've never seen on here that would explain the OP's observation, and try to give some reassurance that it is a design feature.

My recommendation for the OP would be to continue enjoying his Zanardi since it is probably low-mileage and well-maintained. If the use of R-compounds is used in the future, then perhaps consider a larger, better baffled oil pan. IMO, an Accusump or oil cooler would not be necessary.

However, for those with older, high-mileage cars that aren't adverse to modifying, I would definitely do everything possible to beef up the system. Check oil pump tolerances, Accusump, oil pan, oil cooler, etc. That should help alleviate concerns due to stuff getting old and tired, along with gunk buildup.

Dave

 

[ScienceofSpeed]

The symptoms you describe happening in a straight line is unlikely oil starvation if the engine is topped off. The factory pressure sensor tends to amplify or deamplify pressure changes as it goes bad. The sensor's location on the head tends to show distorted readings when the VTEC solenoid is open. For a tracked car with R compound tires, a oil pressure gauge is a good buy. It can be placed off the factory pressure switch port area which feeds the rods & main bearings. We sell an adaptor block & gauges for this.

You may also consider a baffled oil pan which is cheap insurance.

http://www.scienceofspeed.com/produ...e_products/NSX/scienceofspeed/oil_pan_baffle/

Cheers,
-- Chris

 

[stuntman]

It was good seeing you last week Chris.

Even high mileage cars shouldn't have 'gunk' buildup if the car has regular oil changes with a good synthetic oil. My car had over 100K on it and the internals were very clean.

 

[Mac Attack]

The factory pressure sensor tends to amplify or deamplify pressure changes as it goes bad. The sensor's location on the head tends to show distorted readings when the VTEC solenoid is open.
Cheers,
-- Chris

In order to take good oil pressure readings at the head, it's a good idea to add an inline snubber to your fitting that will dampen the pulses during VTEC. Yes, a good place for an aftermarket oil pressure gauge is at the filter base.

Dave

 

[Mac Attack]

Even high mileage cars shouldn't have 'gunk' buildup if the car has regular oil changes with a good synthetic oil. My car had over 100K on it and the internals were very clean.

Keep in mind that a lot of these old NSX's were only driven an average of 4-5k miles per year when new. Owners may have only changed the oil once every 6months or maybe once a year since the recommended oil change interval was 7500miles. Synthetic oil wasn't around in those days either.

Can you see my logic for trying to rationalize why a 20YO car that is suddenly tracked may develop "issues" with oil and wear? That's why I take these "design flaw threads" with a grain of salt and would take extra precaution for an old NSX out on the track with an accusump.

Dave

 

[chang99]

Same thing happened to me at the track. I hooked up an external oil pressure gauge at the spool valve located at the top of the head. Ran the engine at 7K. dash gauge drop to 1-2 bar which it should read 5-7 bar. But the external oil pressure gauge read 6 bar. That is the only sure way to know if oil pressure really dropped. I agree with John, replace OIL PRESSURE GAUGE SENDING UNIT.

 

[Gansan]

Same thing happened to me at the track. I hooked up an external oil pressure gauge at the spool valve located at the top of the head. Ran the engine at 7K. dash gauge drop to 1-2 bar which it should read 5-7 bar. But the external oil pressure gauge read 6 bar. That is the only sure way to know if oil pressure really dropped. I agree with John, replace OIL PRESSURE GAUGE SENDING UNIT.

Hi Chang,

The question I have is, when you replaced your oil pressure gauge sending unit, did it make the dash gauge match the external gauge (approximately)?

 

[Mac Attack]

I wouldn't be surprised if it doesn't match. You are measuring pressure from two different points in the system (look at the oil flow diagram in the service manual). Also, the stock oil pressure gauge is notorious for not being very accurate anyways.

My stock oil pressure gauge drops approximately 1 kg/cm^2 at 7500RPM's or greater to about 4kg/cm^2 at redline. That's on a C30 with engine temps normal (not on a racetrack), Amsoil 10W30, and a NAPA oil filter.

You can clearly see the gauge suddenly drop one unit near redline. It's got to be a design feature of the bypass valve opening up.


STUNTMAN:
Sorry to hear about your engine - when you took it apart was the #6 bearing OK? I thought most problems were at the #4 bearing and journal. Did you see any wear there?

 

[stuntman]

Keep in mind that a lot of these old NSX's were only driven an average of 4-5k miles per year when new. Owners may have only changed the oil once every 6months or maybe once a year since the recommended oil change interval was 7500miles. Synthetic oil wasn't around in those days either.

Can you see my logic for trying to rationalize why a 20YO car that is suddenly tracked may develop "issues" with oil and wear? That's why I take these "design flaw threads" with a grain of salt and would take extra precaution for an old NSX out on the track with an accusump.

Dave

You should take everything with a grain of salt.

I do agree with you there can be issues with immediately tracking a dormant (or barely driven) car without running it for a while and changing out the fluids. Within a oil change or two, the engine should be pretty clean from the detergents in modern oils.

Mac, if you are referring to the thread where my crank snapped in half, yes rod bearing #6 was fine. Heck, main bearing #1 barely went though the first layer! If I had a new main bearing and crank, I could have put it back together.

The 'design flaw' (at least for me) is due to pushing the car to its limits on wide R-compound tires. Since Senna was driving the car back when it had 205/255 (I'm not sure what the Type R had) and stock suspension, the demands are not as high as a car with 235/275 R-compound tires and coilovers, albeit some people with those modifications and more aren't pushing their cars to the point Senna was on the stock car -further agreeing that the stock system is fine for most people.

Rod #6 has gone over 8 times on Time Attack and track cars that I know of, the problem is known/prevalent to those people.


Like Chris said, you shouldn't take your oil pressure reading (especially with an aftermarket gauge) in the head, you should take it from the bottom-end of the motor.

 

[Gansan]

I wouldn't be surprised if it doesn't match. You are measuring pressure from two different points in the system (look at the oil flow diagram in the service manual). Also, the stock oil pressure gauge is notorious for not being very accurate anyways.

My stock oil pressure gauge drops approximately 1 kg/cm^2 at 7500RPM's or greater to about 4kg/cm^2 at redline. That's on a C30 with engine temps normal (not on a racetrack), Amsoil 10W30, and a NAPA oil filter.

You can clearly see the gauge suddenly drop one unit near redline. It's got to be a design feature of the bypass valve opening up.


STUNTMAN:
Sorry to hear about your engine - when you took it apart was the #6 bearing OK? I thought most problems were at the #4 bearing and journal. Did you see any wear there?

Mac, thanks for pointing out the diagram in the service manual. My 1997 manual does not even have that informational section on the engine systems. After looking at it, I think it is legitimate that the oil pressure in the head drops when VTEC is active and that this pressure is not the same as what is inside the engine block. The reason is that the diagram shows an oil control orifice, which would serve to reduce the volume and pressure seen in the head and isolate the oil passages in the block from losing too much due to the demands of the VTEC system.

I could see it being easily possible that under high temperature conditions, the oil pressure loss from activating VTEC would sap a lot of the standing pressure in the head and affect the gauge, but there would be much less pressure loss in the critical areas of the bottom end because the oil control orifice holds the oil back.

If this is the case, I would think that getting the oil cooler would be a good choice for oil life and help keep pressure in the head (leading to less stress for me because the gauge wouldn't go down to zero), but the engine would not be quite as close to blowing up as a lot of us feared.

 

[s-zero]

Chris and Billy are right. The oil pressure switch (for the low oil light) is at the oil filter housing. The oil pressure sending unit is on the cylinder head (for the gauge). I have tried teeing in a mechanical oil pressure gauge at the pressure switch near the filter housing to check pressures and at the track, when hot, the pressure drops off there as well - bottom end oil pressure . No engine failures in 4 years with r compounds and hundreds of laps though . Pressure drops in my car to about 25psi at 8000rpm and comes back on shifts to 80 psi so i shift a bit slower and sometimes short shift which helps. This happens whether straightline or cornering. The baffled pan will help a bit i'm sure but i would take Billy's suggestion and try an accusump. If anyone knows if this will fix it for sure i'm the next in line for one. Cheers all.

 

[stuntman]

Chris and Billy are right. The oil pressure switch (for the low oil light) is at the oil filter housing. The oil pressure sending unit is on the cylinder head (for the gauge). I have tried teeing in a mechanical oil pressure gauge at the pressure switch near the filter housing to check pressures and at the track, when hot, the pressure drops off there as well - bottom end oil pressure . No engine failures in 4 years with r compounds and hundreds of laps though . Pressure drops in my car to about 25psi at 8000rpm and comes back on shifts to 80 psi so i shift a bit slower and sometimes short shift which helps. This happens whether straightline or cornering. The baffled pan will help a bit i'm sure but i would take Billy's suggestion and try an accusump. If anyone knows if this will fix it for sure i'm the next in line for one. Cheers all.

25psi from the oil filter at 8,000rpm? What are your oil temps?

Glad to hear you're driving your car!

 

[titaniumdave]

I'll leave it to you guys to study it on your own, but what caught my eye was at the outlet of the pump, there is a relief valve pictured that will supposedly divert some of the oil flow back to the pump inlet when some pump discharge pressure is achieved. In actuality, this is probably open all the time to some extent and doesn't really function as a "relief path," more of a miniflow or recirculation line back to the pump suction. When designing any pumped system, this is a normal feature since the pump will never dead-head and helps prevent cavitation. Anyways, if you go to page 8-6, you can see that this "relief valve" is built into the pump housing....

Now, there is another true "relief valve" on page 8-10 on the oil cooler/filter base. It's not clear to me from the diagrams and pictures on whether this relief valve diverts oil from both the filter and cooler, but I would assume it doesn't. It would make more sense to just divert some of the oil from passing through the restrictive oil filter and still keep everything cooled by going through the cooler.

First relief valve in pump is strictly pressure related, over pressure is diverted back into the pan. This is controlled by a spring loaded plunger valve, very simple. The second valve in the filter bracket is there in case either the filter or cooler is blocked completely, oil by passes and flows directly to the main bearings.

Your symptom does not sound abnormal, just oil pressure dropping after a period of time in VTEC, worse with higher engine temps. The stock location for oil pressure measurement on your dash gauge, as mentioned already, is at the end of the oil path on the front head. This is after the spool valve which controls the VTEC, if oil pressure drops too low, the motor will fall out of VTEC and give you a code.

I agree with Stuntman, baffled oil pan and Accusump are cheap insurance, reversible and don't hurt the value of the car. An aftermarket oil pressure/temp gauge like an SPA are a good investment too, allows you to monitor things at the pump and in the pan so you know what is going one!

I can help you with all of those mods, if you want.

Dave

 

[Mac Attack]

First relief valve in pump is strictly pressure related, over pressure is diverted back into the pan. This is controlled by a spring loaded plunger valve, very simple. The second valve in the filter bracket is there in case either the filter or cooler is blocked completely, oil by passes and flows directly to the main bearings.

Your symptom does not sound abnormal, just oil pressure dropping after a period of time in VTEC, worse with higher engine temps. The stock location for oil pressure measurement on your dash gauge, as mentioned already, is at the end of the oil path on the front head. This is after the spool valve which controls the VTEC, if oil pressure drops too low, the motor will fall out of VTEC and give you a code.

Thanks for clarifying where the oil gauge pickup is from, now I see it on page 8-3. I just thought it would be right at the oil filter base - seems like the most logical place for it and where the service manual says to hook up a portable gauge to check oil pressure (page 8-5).

The second relief valve in the oil base is also a spring-loaded relief valve. As such, it will also start to divert oil flow from the filter and cooler once oil pressure reaches a certain value. Like a simple check valve, there is a cracking pressure and then a full-open pressure. When the oil pressure drops on my car by 1 kg/cm^2 (about 15 psi) at 7500 RPM's or higher, I was assuming that was due to that spring-loaded relief valve starting to open a little bit.

Oil pressure will of course drop with higher temperatures as it's density decreases and viscosity changes.

Another thing to consider is running 50W oil on those hot track days for extra protection.

Dave

 

[chang99]

Hooking up an external oil pressure gauge at the bottom end of the motor on the oil filter housing is not necessary because it has shut off the idiot light on your dash when you start the engine. This location is the beginning of the oil passage. Hooking up at oil pressure sensor next to the spool valve is at the end of the oil passage before oil falls back into the pan. External oil pressure gauge is definitely more accurate than dash gauge because the reading is not an electronic signal. It is a safe testing precedure as explained on the factory service manual. Everyone claims VTEC will affect oil pressure drop, any real data to back up that claim ? My reading at 7K. dash gauge reads 1-2 bar, oil pressure sensor external gauge reads 6 bar. At 8K, dask reads 0-1 bar, oil pressure sensor external gauge reads 7 bar. After replacing oil pressure sensor unit, the dash gauge reads normal as it should. So first thing first, is your dash gauge reading accurate ? Or is it just a faulty electronic signal ?