Eric Michel

I received a PM asking for more information on a modification I have done to the OMP, since I took the time to write this up I figured I’d post it for everyone’s amusement. I begin with a brief and somewhat simplified explanation on how the OMP works, hopefully this makes sense.
The Oil Metering pump (OMP) is supplied oil from the high-pressure side of your engine oil pump, it is driven by a gear assembly in the front engine cover that is directly connected to the drive shaft. Within the OMP is a shaft that rotates with the engine RPM, this shaft has a hole (approx 1 mm) drilled through it, on one side of the shaft is the high pressure engine oil passage on the opposite side is a passage going to the oil lines that drip oil into your engine. So as the shaft holes align with the passages a small amount of oil is sent into the engine. The rotating OMP shaft can also slide in and out and is spring loaded, the oil flow is adjusted by moving the shaft in or out causing the shaft hole to align more or less with the oil passages. On the FC & FD RX7’s the shaft position is set by the ECU using a stepper motor with a spiral gear, as the stepper motor is turned its output shaft will move in or out. Using the stepper motor the ECU can precisely control the amount of oil flow by sliding the shaft back and forth. The other electronic component of the OMP is a potentiometer that tells the ECU exactly where the shaft is positioned which it translates into oil flow. When the OMP goes bad it is usually the stepper motor or the potentiometer that malfunctions, unfortunately Mazda does not sell those parts separately and they are not exactly off the shelf parts so you are forced to replace the entire assembly when this happens.
As you can see the OMP is more like an electronically controlled valve than a pump, when your car shuts down the ECU adjusts the OMP for 50% flow for starting oil lubrication. If you are using an aftermarket ECU (i.e. Haltech) and you simply disconnected the OMP electrical connections (most are like this), your engine is getting 50% OMP oil flow all the time. Other than speculation about how synthetic oils might leave harmful deposits in the engine I have not seen any definitive proof that this happens or have even heard of even single engine that has failed because of this. From what I have learned, the amount of oil being injected by the OMP has been reduced significantly in the later models RX7’s. The speculation for why this was done is to help with emissions and because oil in the gas/air mixture reduces the octane rating of the fuel increasing the risk of pre-ignition. To reduce the need for oil injection Mazda started impregnating the housings with graphite. I have seen a number of rotary engines torn down and rebuilt, some used synthetic and others were using petroleum based oils, IMO the only side affect from using synthetic oil is extended engine and turbo life. The biggest problem with carbon deposits comes from running rich, after a while your plugs foul and then you really start building carbon deposits.
I am running a Haltech so the stepper motor is not used to adjust oil injection, rather than leaving the OMP at some arbitrary flow rate I made it adjustable so I could determine the optimum fixed rate oil injection. I removed the front cover plate of the stepper motor (see attached pic) and simply tapped the hole with a 6mm tap and installed a bolt with a locking nut, (see pic attached on next message). With this set up you can adjust your oil injection to whatever amount the OMP will allow.
When you see oil on your plugs don’t be to alarmed, when you cut the ignition the engine kept turning long enough for the injected oil to spray the plugs, when the ignition is on this oil is vaporized. I am still in the process of determining the optimum setting, when I first installed this mod I had way to much oil injection (lots of blue smoke out the exhaust) I reduced the oil injection by bringing the bolt out counter clockwise until the exhaust was clean. I have a wide band permanently mounted in the car, with my A/F ratios optimized I am attributing excessive plug fouling to the oil injection. I keep checking and cleaning the plugs, each time reducing the injection a quarter turn. Once I have it set so the plugs look good I may experiment with a little premix but I don’t really know if that will be necessary. My engines always seem to fail due to some catastrophic malfunction long before they have a chance to wear out due to low oil lubrication on the apex seals.
I can understand why people just remove the OMP and pre-mix, this allows you to precisely control the amount and type of oil. The ratios some people say they use 100:1, 80:1 require almost a quart of premix with every tank full of gas, what a PITA. I am not sure if these ratios are really necessary and I would be very concerned about what that much oil would do to your fuel octane rating, especially if your running a lot of boost. If you set the OMP for a constant low level at least you have some lubrication if you run out or simply forget to add premix.

Eric Michel

Another pic

Bacon

iTrader: (12)

nice write up....

how much bolt do you have pussing the innner slide in courently?

Eric Michel

Well I'm not exactly sure, I started with the bolt sticking out 8mm, that caused a lot of oil smoke out the exhaust. While it was idling on jacks I started backing it out until it ran smooth and the exahaust smoke subsided, about one and a half turns. I have done two 1/4 turns CC since then , I would guess around 5mm.

DamonB

Bumped. Good stuff.

Madmax670

what exactly does premix reffer to, and what does it do?

4Fun

iTrader: (8)

Good info, but i would like to see pics. They seem to be down right now. Thanks for sharing your info.

spurvo

Premix reffers to additives you can place into your fuel that will lubricate the apex seals from fuel injection cycle. There are pros and cons, including the PITA factor and injector shutoff during deceleration reducing the lubrication at that time. For those of us running stock, Pettit provides a premix additive called Protec-R that they say (and others seem to agree) can work in conjunction with the OMP to improve performance (increased chamber sealing, maybe?). Protec-R is also reccomended for OMP deletion, described below. Note that there has been some reasonably strong evidence that the Protec-R is really just repackaged Marvel Mystery Oil (MMO for short), and many foks have reported perfectly fine results using this material instead. Another argument for premixing on stock OMP is the possibility of OMP failure taking out the engine, where the premix would provide a measure of protection. Towards this last point, I reccomend you do a search to find how many engines have failed due to OMP failure, versus, oh, say, increasing boost with no commensurate increase in fuel and compensation in spark timing.

That stated (written?), for folks who have removed their stock engine control unit in favor of an aftermarket programmable unit, there is typically no provision to progresiively control the OMP function, leading to over or under injection of oil, with the problems expected from this (and nicely compensated for by the discussion generating this post). One very effective way of dealing with this is to block the OMP mounting point altogether, and place the lubrication in the fuel. With standalone ECU's, you can always program a small amount of injection on high vacuum deceleration to ensure lubrication. Or go the route described above.

Or just not bother with it except for changing the oil injection lines when you replace the engine at 60K miles from, oh, say, O-ring failure due to overheating?

InsaneGideon

Good find. Been doing some thread digging lately, Damon? OMP on the fritz?

I see this was posted back when the "3rd Generation Specific (1993-Current) " forum was considered an "Advanced Tech" forum.

I didn't realize this was such an issue. I wonder if there's a market for an external rpm/load-dependant OMP controller...

DamonB

Yeah, I just replaced my OMP a few weeks ago and was digging around here. I've had a front cover leak for a while (fixed yesterday. Cross your fingers!) and hot oil has been dripping on my OMP for sometime. The position sensor on the OMP finally started tripping the CEL intermittently. The OMP itself was functioning just fine but the sensor was fouled as it was full of oil. Apparently it doesn't like that I'm going to see if I can't clean it up and keep it as a spare.

z8cw

Just to clarify this even further, more questions. I have an AEM after market ECU which allows to control the OMP via some base and target tables and a basic flow rate (which is set at 50%). There is a provision in the AEM menu that instructs you to add to the target table the proportional increase of extra fuel you running. Basically, with larger injectors you need more oil. I am having a hard time conceptualising this. Does more fuel really affect the viscosity of the mixture that much. I can see that the amount of oil should be proportional to the RPMs or even exponetional since the apex seals will be forced against the rotor housing that much more at high RPMs.

Thanks
CW

DamonB

I'm going to disassemble my old pump. If it is engine oil pressure that supplies the force to push the oil through the OMP tubes then the Rotary Aviation adaptor would not work as the lube it sends to the OMP is merely gravity fed. The tang on the OMP that engages the drive on the front cover must have a small rotary type pump behind it so that the OMP in fact does make pressure all by itself when driven by the engine. The Rotary Aviation piece would thus seal the oil supply from the block and instead provide a gravity feed of premix to the OMP which is then pressurized inside the OMP itself.

I have to open up my OMP to verify but if it doesn't work this way the Rotary Aviation adaptor wouldn't work.

DamonB

Opened up my old OMP today. It appears to work exactly as Eric Michel explained but with one exception: the OMP itself provides the pressure to drive the oil through the OMP tubes, not the engine oil pump. The OMP in fact has a single piston pump located underneath the rectangular cover near the tube connections that is driven by the tang which locks into the drive gear on the front cover. The tang drives this pump to create pressure and the stepper motor adjusts a needle type valve to regulate the volume of oil admitted through the OMP. The sensor is just a linear potentiometer that provides feedback to the ecu so it can be certain the valve is positioned correctly.

Engine oil is delivered to the OMP from the front cover and then the OMP pressurizes this oil while also regulating oil volume. The Rotary Aviation adaptor is essentially a block off plate that blocks engine oil from the front cover but still allows the tang drive on the OMP to connect to the drive gear inside the front cover. The new source of OMP lubricant is then plumbed to the adaptor.

the_glass_man

iTrader: (2)

So you would just hook up another oil source with better oil?

z8cw

Exactly,
I did some research and are giong this route. Synthetic in the engine and 2 stroke through the OMP. Best of both worlds.

CW

the_glass_man

iTrader: (2)

Where and how much are the RA adapters? I can't seem to find them anywhere on their site.

afterburn27

iTrader: (5)

This thread was a great read

rotarygod

iTrader: (1)

You guys are all wrong with how the omp works. First of all the omp is not on the high pressure side. It is above the oil pan in the front cover receiving drainage. Second, the omp is not a pump at all. It does no pumping. It is just a valve which opens to vary the amount of oil passing through it whether by electronic control as in '89 and up or mechanically before this point. The oil is actually pulled through the lines by the engine. As the engine turns, it puts a vacuum on the lines. The reason you still have vacuum lines on the oil injectors is because if you didn't, oil flow would be much faster through the lines. The oil injector vacuum lines actually tie in before the throttlebody. They only run to the manifold to get filtered air. Technically you could leave them unplugged but you risk them getting dirt inside them. The air passing through the vacuum lines from in front of the throttlebody allows some pressure leakage. The air moving past the oil lines however still exerts a small pull on the oil. If you were to block off these vacuum lines, the engine would just suck the oil through these like a straw at a much quicker pace. The reason that the rotary aviation reservoir adapter works without a pump is due to this suction. If you really want to get crazy, run the oil injector lines from the engine to a container of oil. It will still suck it through. This, incidentally enough, is how Racing Beat uses oil metering on the engine dynos. The engine is the pump.

RX-Racer

Let me just pull out this part! "The oil is actually pulled through the lines by the engine. As the engine turns, it puts a vacuum on the lines." Now you say there is vacumn in th oil injectors, but why is it that you can have them un-plugged with no engine problems (usually a vacumn leak tends to make the engine run a little rough, also you can put your skin up to it and feel no "suction". Please explain.

But I will have to see if that straw with a bottle oil works and then appligize, but until then it don't make since.

JP

rotarygod

iTrader: (1)

It sounds weird but definitely try the bottle trick. It works. I can't explain why the engine doesn't act like a vacuum leak. That part doesn't make sense to me either but nevertheless it works.

SomeGuy_sg

mm....very conflicting reads on how the OMP works ..mm...

DamonB

True.

Absolutely wrong, at least for my FD OMP that I disassembled. Want proof? How do you suppose turbo engines when under positive manifold pressure (boost) still have a vacuum present inside the motor to pull oil up through the tubes as you describe? Easy; they don't. When under boost the entire path from turbos to rotor face itself is under positive pressure. If the OMP did not drive the oil through the tubes by pumping it you would never get any oil into the motor because afterall there is NO vacuum present when under boost.

This describes why an OMP is an Oil Metering Pump. It both Meters the oil through the valve assembly and Pumps the oil.

If the OMP worked as you describe there would be absolutely no need for the drive tang to the OMP from the front cover; you wouldn't need a driven source. All you'd need is the valve assembly and the stepper motor. The fact that the tang drive is what drives the pump inside the OMP is why the Rotary Aviation adaptor still has this tang present. If there is no drive to the OMP it cannot pump oil and the engine absolutely does not suck it up on its own accord. As has already been shown that would be absolutely impossible to do when under boost anyway.

The oil injectors have one way valves (see the FSM). The OMP has to overpower the valve and the manifold pressure to deliver oil into the chamber. Without the one way valve boost would pressurize the tubes and therefore the sump (or the oil tank if using the Rotary Aviation adaptor).

The original post in this thread is exactly right except for the statement about engine oil pressure being the motive force for OMP oil up the tubes. The OMP in fact has a small pump inside that is driven by a tang on the front cover to provide the force to move the oil through the tubes and into the motor.

rotarygod

iTrader: (1)

There is no motor in the mechanical '87-'88 T-II omp's. How do you explain how oil gets to the engine under boost to them since there is no pump? Stop looking at it in terms of intake air pressure and start looking at it from an airflow perspective. The air still moves in the same fashion in a bosted engine. There is just more of it in the same spot.

DamonB

I'm sure there is a pump inside them. The stepper motor on the FD is NOT a pump; it controls a valve. The pump portion is driven by the tang on the front cover and I'm sure that's how the OMP's on all the other gens work as well. Oil enters the OMP and is pressurized by the pump inside the OMP which is driven by the gear in the front cover. The oil then must make it through the valve that is controlled by the stepper motor before it can be pushed into the engine. The stepper motor does not run full time and has nothing to do with pushing the oil into the motor, it merely controls the valve which regulates oil volume. The stepper motor only runs when the ecu adjusts oil volume and then the motor "steps" in short angular rotations to adjust how wide the valve is open. The stepper motor does not run at anything resembling high speed.

Let's make certain you understand that. The stepper motor on the FD OMP has absolutely nothing to do with pushing oil into the motor. It is merely an electronic valve that is fully proportional and can be held open, nearly closed or anywhere in between to regulate the volume of oil that flows out the tubes. The pressure to drive the oil through the tubes and into the motor is from the pump inside the OMP that is driven off the front cover. The mechanical OMPs from the other gens I'm sure still have the mechanical pump inside, they just don't have the electronic valving the FD does. The FD uses less oil and can more fully adjust it's OMP output volume to all engine parameters based on ecu programming. The pumps on the FB and FC are just "dumber".

You have no idea how airflow and pressure work then. Flow is always from high pressure to low pressure. When the engine is under boost there is absolutely no way for there to be a vacuum present inside the OMP tubes to pull oil in as the flow would in fact be the opposite direction.

If you still don't believe me remove the oil injectors from the motor and start it up. With the tubes not connected to the engine but only at the OMP you will in fact get oil from the ends of the tubes because the OMP is still being driven and pushing the oil out. Remember to hold the tubes high up in the air when you do this so you can't say the oil is mrerely leaking out from the cover.

The tang on the front cover drives a pump inside the OMP. Open one up and see for yourself.

rotarygod

iTrader: (1)

Apparently not.