OMP Declassified Part III (the Mikuni)
 

In the last OMP Declassified the Denso OMP oil flows were evaluated; in this post the Mikuni pump is evaluated. The Mikuni is much different than the Denso as it has two oil pumps (one large and one small), the stepper motor is rotary in contrast to the Denso's linear motor, and the position sensor is a rotary potentiometer rather than the Denso's linear sensor. Below are the two pumps side by side:
https://www.rx7club.com/picture.php?...ictureid=22829

Drive Side View:
https://www.rx7club.com/picture.php?...ictureid=22830

The following picture shows the Mikuni OMP disassembled. Note the two oil pumps (one large and one small)
https://www.rx7club.com/picture.php?...ictureid=22835

The next image shows the Mikuni pump stroke length means of adjustment and the position sensor:
https://www.rx7club.com/picture.php?...ictureid=22833

The two pumps can be switched from the drive side of the pump housing to the port side of the housing. The drive side rotates at one half the speed of the port side and the stroke length has a different profile and length between the port side and drive side. So one could swap the two pumps and obtain higher flow at low engine RPM's and lower flow at high engine RPM's or Lower flow at low engine RPM's and higher flow at high RPM's. The factory arrangement is the latter.

The following data provides the Port side pump flows by stroke length and position sensor output in Ohms. Note that I have given flows for both the small and large pump in the Port position.
https://www.rx7club.com/picture.php?...ictureid=22832

The next set of data provides the Drive side pump flows by stroke length and position sensor output in Ohms. Again I have given flows for both the small and large pump in the Drive position
https://www.rx7club.com/picture.php?...ictureid=22831

Now below shown is the data put together including output by E-Shaft revolution:
https://www.rx7club.com/picture.php?...ictureid=22834

The following chart puts the above data in graphical format. You can see the two different profiles that can be created by swaping the pumps. The Red and Brown lines are by pump drive revolution at different stroke length settings in Ohms and the Blue and Purple lines are by E-shaft revolution by stroke length setting in Ohms:
https://www.rx7club.com/picture.php?...ictureid=22838

For flows by pump stroke length and engine RPM the following maps are given (blue map has small pump on drive side and large pump on port side and red map has large pump on drive side and small pump on port side). The corresponding graphs are for each map:
https://www.rx7club.com/picture.php?...ictureid=22840
https://www.rx7club.com/picture.php?...ictureid=22837

https://www.rx7club.com/picture.php?...ictureid=22839
https://www.rx7club.com/picture.php?...ictureid=22836

Looking at the flow rates from the Denso and Mikuni pumps if the Mikuni pump is set up at the higher flow rate (factory setting) the Mikuni has about the same oil delivery at low engine load settings and significantly higher oil flow at the higher loads which makes the Mikuni probably a more desireable pump. There is a glitch in using the Mikuni in the 93 - 99 FD's; that is the position sensor will output different voltages from what the PCM is looking for. The Denso position sensor is 500-1800 Ohms and the Mikuni is 500-3400 Ohms. I have heard these can and have been used interchangeably but it appears for the Mikuni the PCM for the 93 - 99 FD will look to set it at 1800 Ohms at idle which will set a very high oil flow at the lower engine load settings; this is speculation on my behalf. Hopefully, someone who has been running one of the Mikuni's can chime in.

Great info, thanks for sharing. I suspect if you were really interested it may be possible to trick an ECU into using a Mikuni pump, but is that necessary for the PFC? I'm assuming that if you need the additional oil injection the stock ECU isn't going to be sufficient in terms of fuel and spark.


Have you also measured pump flow rate to back up your calcs?

how did you even get your hands on a Mikuni pump?

I assume they'll show up every now and again off used motors. Wouldn't mind having one myself, I'm sure they're CRAZY expensive new from Mazda Japan and I'm sure they'd be hard as hell to find used.

Regardless, EXCELLENT info, BIG THUMBS UP for taking the time to plot this out!

I'm not totally sure if the PFC will even care that there's a difference in the pump. Stock ECU might pitch a fit, but if you're at the point in modifying your car that you're thinking about swapping the OMP out, you most likely have a PFC at that point.

Dale

+1

reverse engineering factory stuff can be fun. it really forces you to think and do research.

So mazda thought the oil injection should double?
Is that why they switched over to the muniki?
or was it because the increase in Horse Power in the latter FD that they increased the omp flow?

Also whats up with the difference in the reservoir? are the latter FD front cover different?


http://i705.photobucket.com/albums/w...beoner/omp.jpg

Didn't flow oil but the pump is positive displacement so the calcs should be close.

Japan2LA sent it to me. He is one of our club vendors; great to deal with and excellent parts resource.

Looks to be only a difference in machining; will align with the cover OK.


Looking again at the position sensor and the difference between the two, it looks like they may be interchangeable:
https://www.rx7club.com/picture.php?...ictureid=22889

Nice work, ttmott.

Here's a little tech info on some of the changes to the version VI (99-02) engine, taken from an SAE article:

Great work. If I had known about the Mikuni, it would have saved me all the headaches of custom made oil return line, fittings, adapters and the like in order for the RA OMP adapter to work.

As others have said, thanks for the information. I don't have the engineering background like some but it's still fun to try and follow.

Thanks for the great info.

So has anyone actually tried this? Do you need to run the newer/smaller injectors to run this pump? If I have a Power FC, it'll just plug right in and not throw any additional lights?

Not sure if you MUST change to the later injectors with the Mikuni. I have read reports of the the new(er), smaller diameter injectors plugging-up over time. Other than the smaller port, it wasn't clear why. It may have simply been from poor scheduled maintanence.
And yes, the PFC does not throw CELs and won't care what OMP you have. It won't even care if the OMP is working, or there at all. Because of that, even with the OMP some pre-mix a little to augment.

Shouldn't throw codes with a stock ECM either based upon it's VMOP output of about 1.1 to 4.1 VDC (See my previous post). The resistance measurements across the stepper motor is slightly out of family (~35 ohms vs. 16-31 ohms spec.) for the 93 to 99 but it shouldn't be a problem. I have a PFC so can't really test it in a stock configuration.

The injector port size was reduced (see Gomez's post above) to solve a technical issue regarding the engine vacuum non-boost modes from sucking the oil out of the injector cavities then during accelleration having no oil until the OMP could catch up. Most of us have the rubber filler under the check valve which should solve any potential issue in the oil refill lag with a large cavity. I don't believe we really had issues with the Denso pump and larger injectors so if you run the Mikuni pump you definately would not have any issues.

I have installed the late omp model(mikuni) together with the new injector grommets 3 years ago,with no problems whatsoever.Also,all the new mazda replacement housings come with the new grommets...

So I'm still not sure I understand the benefits of the newer pump as the signal values the pump is expecting (as you stated) has a different range for each pump style (new and old). Would you happen to have a flow curve for the denso pump to compare to the new style? And even if we CAN compare flow rates of the two pumps directly to one another, we still haven't factored in the influence of the updated nozzles, so we still don't know exactly how different the total flow per RPM is when you compare the entire package/setup (old vs new). I'm really interested in finding out however as Mazda seems to be heading down the trail of adding more and more oil (updated pump for 99+, rx8 gets two oil injectors, and supposedly the 16x will have 3). I have a hunch they know something we should know... but we still have to prove it.

Also, this kinda begs the question... what is on the 8? I know the 8 has two nozzles per rotor housing, so 4 lines total from the OMP. Now I'm curious :)

I think that is the omp unit he installed on my car. I was actually going to ask him what his thoughts are on this info.

As it ends up (see previous post from me) the VMOP is the same for both pumps so for the factory ECM's should not be an issue. For the PFC or other controllers the VMOP is ignored and the pump output curve can be changed in the software so again no problem.

I profiled the Denso pump in OMP Declassified Part II https://www.rx7club.com/3rd-generation-specific-1993-2002-16/omp-declassified-part-ii-895388/ so you could compare the two. If I have time I'll overlay the flow curves. The real benefit is the Mikuni can be set up to flow quite a bit more oil at high engine loads and pretty much match the Denso at low engine loads. I think they bring a lot to the table for those who do not want to premix and running chamber pressures higher than stock.

Tom

Thanks for the information! Yeah more oil at higher RPM's is a must... I'm just curious if it actually works out to more flow when the smaller nozzles enter the mix, but I have a feeling it does.

Also, looking at my 20b pump I believe its the same denso unit as the FD (but with a 3rd output). Now I'm interested in opening it up to see if anything is substantially different or just a 3rd port.

I am sure there is more to it than the 3rd port. The needle that fits inside sector valve / pump is probably a larger diameter for the additional oil volume required. Quite possibly also the worm gear ratio and / or the drive under the engine cover could be different. It would be interesting to add it to this data collection.

If it wont be destroyed by taking it apart I'm not opposed LOL Would give me a chance to clean out the internals

I have a question here... I took my spare OMP apart to mess around with it and i see that When the SECTOR VALVE NEEDLE is pushed all the way IN it cuts flow to the secondary hole in the SECTOR VALVE. Is it safe to assume that this secondary hole is opened up at higher rpms? If so what if i bore out the opening, will it increase oil to the housing injectors?

any thoughts?

sorry for the bad pics,camera phone...

Attachment 711925

Attachment 711926

I thought that also but a couple of things:
The small hole is drilled at an angle downward to as to not be in the area where the needle is.
The needle does not insert fully into the bore of the sector valve and it's maximum stroke is rather short; look at the height of the cams on the sector valve drive gears.
As the sector valve rotates the holes align either with the oil supply or the discharge ports.
Also as the sector valve rotates it moves up and down. The length of the up and down movement is dependent on the position of the linear (Denso in your case) stepper motor.
As the valve rotates and moves down the oil supply is open to the lower sector hole and the needle is retracted. This draws oil into the sector valve internal area.
The valve continues to rotate and then moves up in concert with the small upper hole aligning with the oil discharge port holes. As the valve moves up the needle is depressed into the sector valve forcing the oil out of the OMP and to the injectors.

you should try disassembling a series 5 FC omp. You can get them cheap. Those are the first electric OMP's to be fitted to an Rx-7. I wonder what the differences are between that and say these late model Mikuni pumps.

If someone will contribute one to the cause I'll set up and profile it's performance and add to this data. Same for the three rotor.... Taking them apart does no damage as long as they have been maintained well and the O-rings are still pliable.

same as the denso,, but 4 output ports

so is the pumps intercheangeable?
i disasembled a JDN engine from a 97 car ind i found a mikuni pump installed.
now i am rebuilding that engine , i have a pfc and wonder if i can keep the newer mikuni .
is the wirring the same? i notised some color diferences in the wires.

it is stated that mazda used these from 2000 but how it was installed on a 97 car?

Hey guys,

I got the CEL on and it gives me error #26 MOP Stepper Motor, and my 7 doesn't boost up properly/power loss! "Limp Mode" others told me that it's my vacuum lines, but since the ECU flashes 26 then it's definitely not the vacuum lines, because I replaced the vacuum lines with new silicon ones.

So what should i do exactly? Replace the Metering Oil Pump? which costs $1700 from Atkins Rotary.

The ECU also flashes :

23 Fuel Thermosensor
25 PRC Solenoid (3Way Valve)
44 Turbo Control Solenoid

But No 27 which is the whole OMP.

Please help guys.

Thanks

Crap I put this in the wrong OMP Declassified thread, had all three open in tabs and picked the wrong one - this being about the Mikuni and not Denso. If a mod sees this can you please relocate if possible as I cannot delete the post.
===
Working on restoring my FD and wanted to see the condition of my OMP, plus I wanted to see for myself how it worked. The info in the three OMP Declassified threads was awesome, and though I might be able to add.

I am fairly certain I have the following two pictures correctly demonstrating how the OMP uses the rotation from the e-shaft and a cam profile to pump/meter the oil. The operation of the OMP's pumping action is similar to one piston in an axial piston pump.

High flow rate. This is when the servo motor forces the flow rate control shaft downward. This allows the longest travel of the piston, increasing cylinder displacement and increasing the pumps output.
Attachment 675452

Low flow rate. This is when the servo motor allows the flow rate control shaft to rise. This reduces the travel of the piston, decreasing cylinder displacement and decreasing the pumps output.
Attachment 675453

Few assumptions:
-even if the servo motor dies, the OMP should still function. However, I would assume the cam profile would remain on the setting when the servo died. Either injecting to much or too little oil depending on RPM.

-To increase ouput, bore out cylinder sleeve, and replace the piston with a matching increased diameter piston. Or, alter the profile of the flow control rate shaft.

Please let me know if any of my statements or assumptions are incorrect so I don't mislead anyone. I just though having pictures might make the OMP's pumping action easier to understand.

Anyone have clarification on this? I am installing the Mikuni pump with the new small diameter injectors. I had Ray at Malloy send me 2x N326-14-633 which are two small rubber grommets. Where do these even go?

What's weird is that N326-14-633 comes up as the injector plug itself in the housings: 84-95 13B Rx7 Oil Metering Jet In Rotor Housing...

Perhaps the small rubber grommets are incorrect and I need the "newer" plugs that sit under the injectors?

On a side note the bottom bolt is also now too long thus will need to be replaced with a shorter 10mm bolt.

Have the grommets been repackaged? Too lazy to look up the part number again to verify, but this is from back when....https://www.rx7club.com/3rd-generati...estion-660786/

The grommets sit under the injector. The oil jet itself, I'm surprised they'd even offer them and can't recall seeing them as a separate item to the housings - late ones at least.

Can't understand the bit about the bolt.

Thanks, from reading that thread it appears the newer jets are pressed into the housings? Thus for us with old housings the rubber grommets are essentially useless?

Probably depend if the jet size changed - a set of .1mm increment drills probably required to test that on old and new housings. Here's a couple of pics I'd forgotten I'd taken from years back comparing them...

Attachment 622643

Attachment 622644

doesn't seem to be a huge difference to the eye at least!

Thank you for the pictures, they say 1000 words!

I can't see a point of using the grommets on the older jets when there is a ridge all the way across. They will not seal anything like the newer jets. At this point I am debating just using the old larger injectors with the newer mikuni pump for more oil to flow into the housings.

No worries.

Going by the failure rate of oil injectors, I'm surprised there's any good ones left in the larger size. New type is all that has been available here for a very long time.

Dropping this here, if anyone has any intel on it.

I've been running the Mikuni pump, with a PFC and the newer style injectors with 20/50 oil for a while, and now am suddenly having issues with blowing injectors. Like it blew one on the dyno, and then blew two brand new ones after a day on track. In addition to the more oil that the Mikuni pumps already, the settings in the PFC are pumped up a bit as carryover from previous tune. This is on a new motor.

The theory is, it's pumping so much oil it's overwhelming the check valve (though not sure why it didn't happen before). Going to try the OE/93 style injectors (bigger holes), but wondering if anyone else has had experience with this.

I'm under the impression that brand new you can only get the newer style injectors, so I guess you would have to use a set of used older style ones.

So when this happened to you how did you tell that you blew an oil injector? Did you knew right away or later on?

I have the vent lines running to a little filter, which started puking oil.

Now I am worried about my combination of new oil injectors and mikuni pump.

Do you have the ability to see the OMP setting on your (assume) PFC.

One other thought is, is there some sort of failure mode in the pump that would cause it to run wide open or something?

" There is a glitch in using the Mikuni in the 93 - 99 FD's; that is the position sensor will output different voltages from what the PCM is looking for. The Denso position sensor is 500-1800 Ohms and the Mikuni is 500-3400 Ohms. "

How does this work out for modern stand alone ECU's? Is there an option to input the sensor data so the ecu knows full open or closed?

Does this mean the mikuni is going to be pumping more oil at same settings?

Has anyone ever removed the little brass oil jet from the rotor housing (especially while it's still in the car?)

It looks like it has a flathead screwdriver head, and would just thread out, but I have a replacement set sitting in a baggie, and there are no threads on it at all. How would one even get ahold of the thing to pull it out?

Good question. This is one of the reasons I want new housings. The old style brass oil jet with the slot will not create a seal like the new style does with the rubber grommet.

Come to think of it, I don't know if my current ones have the slot or not. I can see the o-ring in there, but I want to order a couple new ones before I attempt to pull them out, because it looks like they'd be easy to lose/damage.

I'd like to replace the jets just to eliminate clogging as a cause for blown check valve/nozzles.

Given the testing results provided by the OP and as I have come to understand it, the Mikuni (when installed in an older generation FD) will pump ~more oil at lower engine speed than the Denso; as it will not register a resistance above a 1800 ohms.

So...(and I'm making up #s here)
• if the Mikuni is registering 3400 ohms, say that equates to a flow of .05
• if the Mikuni is registering 1800 ohms, say that equates to a flow of .08

On an older gen FD, the lowest flow rate the Mikuni will provide is .08

Someone more certain in their knowledge than me is welcome to eviscerate my explanation.

I'm currently sitting on a Mikuni and have been following your build thread while also trying to learn more about this topic, hoping the right (or better) way to utilize the Mikuni (or OMP in general) with a simplified or single setup is determined.

One component I always come back to when I consider the OMP system, is the line/s routing from the injector nozzle to the pre-compressor portion of intake in the factory setup. The general argument seems to be that this is/was only done to provide a filtered air source to the OMP injection nozzle so that when there is negative pressure in the engine, filtered air is pulled through the nozzle instead of all the available oil in the OMP lines.

Ok, yes, a filtered air source is important and maybe that really is the only reason. But if so, why are there so many stories of people (such as yourself) blowing check valves so easily/frequently. It may be true, but it's hard to swallow that the Mazda check valves in the nozzles are simply that shitty. A lot of time and money went into the design of the system and it's components. And a lot of stock(ish) cars racked up a lot of miles on them(admittedly there is no immediate indicator if a check valve has gone bad in a factory setup). I'm also pretty sure race teams retained the OMP.

My point (which is more of a question) is this - is there truly no benefit to having a vacuum source attached to the nozzle? Is it not in some way beneficial for the whole thing to be a connected system(as opposed to having a piece vent to atmosphere)? Or beneficial for the check valve to be held closed via vacuum(pull from the intake side) before being slammed shut(hyperbole) by positive engine pressure(push from the engine side) after getting into boost?

I could be grasping at straws. But as mentioned, it's tough to swallow that the nozzle and check valve are simply junk.. so I keep trying to learn/think of something else.

I'm actually wondering, though, how much this is really a "single turbo" problem, or just a problem masked by the way it's plumbed in a typical TT system. If I recall, the lines from the nozzles just go to the primary turbo inlet, so there's no vacuum or pressure being applied to the nozzles, and if they were to fail and puke oil, it just get's ingested or puked into the intake so you don't really notice.

I only noticed because I had a dedicated filter to the lines. After it failed, I plumbed the lines to a nipple on the oil filler neck, and then you don't notice anything. It's also worth noting, this doesn't happen on the street. It happens after prolonged boost, like on the track or on the dyno.

One thing I wonder about is the whole relationship with the OMP nozzles and boost/vacuum. Non-turbo FC's have the same OMP system as turbo cars and I think the FD has the basic same design.

Big problem with the OMP system is it's not obvious how it works, there's just a lot of guessing and "seems like it works this way" involved. It would be nice to find a way to scientifically test how the system works and what external factors like high boost or changing the vacuum line to the injector will actually do to the system.

Dale

Dale, can you confirm my recollection of where the lines from the nozzles are plumbed to on a stock car? If I understand it, there's no vacuum applied to them, they pull vacuum by way of the motor, and are supposed to block boost being expelled. I do know that when they blow, they affect idle, just like adjusting the idle bleed screw.

I'd be surprised though, if they were being blown purely by "high boost" since I only run 13 psi at the track, and I don't understand how 13 psi from TTs is any different than 13 psi from a single turbo.

Attachment 735807I agree with this. To me, this currently indicates that there is something else at play here.

I've looked around for a cross sectional image of injector nozzle because I realized I don't actually know what type of check valve is in there or have any idea what it looks like inside. I've never found any specific literature on this but I did find this image.

https://cimg1.ibsrv.net/gimg/www.rx7...44d33a8299.png

This is from a 1999 Mazda press report, it seems to show that the only difference in the nozzle design is the inner diameter and the addition of the seal. Check valve and design appear to be the same (assuming the drawing is accurately representing that part).

Does anyone recognize the type of check valve depicted?