Deleting the Oil Metering Pump System
01-06-2010 | 01:29 PM
#1
Guest
Posts: n/a
Deleting the Oil Metering Pump System
I would like to keep this thread on how to delete the oil metering pump, not throw a check engine light and prevent the power control module from going into a fault/limp mode, using the Access Port via the Access Tuner Race software.
My perspective on deleting the pump when I turbocharged a 2004 Renesis with a custom top mount:
Pros:
(1) Simplicity,
-If there's a supply of fuel, the engine is being lubricated with premix.
(2) Reliability,
-(see reason #1)
-No melted lines or sensors from the turbo manifold and/or downpipe.
-Metering pump is maxed at higher rpms/loads and does not deliver an adequate supply, necessitating the need for more lubrication by premixing.
(3) Valuable real estate,
-With a medium sized turbo (T4/GT3582R) & larger you start using every square inch.
(4) Alternative fuels,
-With E85 you just might want a different source of lubrication than what is in the oil pan.
Cons:
(1) Precision,
-Not as precise, as the oil metering pump can deliver smaller amounts of lubrication at lower rpms/loads and not waste your valuable premix.
(2) Expense,
-The cost will be more considering you are premixing for worse case scenario.
(8500 or 850 rpms, the engine is receiving the same ratio of gas to oil when premixing)
(3) Convenience,
-It takes more diligence, considering you have to mix with every fill up at the pump.
I had Richard Sohn’s adapter for quite a few years and premixed. It’s a great piece, I converted the windshield washer reservoir for the tank and it worked great. However, the adapter as some of you found out with a top mounted turbo system kicks out the OMP to the point of making the downpipe (even when modified) really close to the plastic sensor on the pump and those without semi solid motor mounts (ie: modified mazport mounts) just doesn’t work out in the long run.
Deleting the oil metering system outweighs the drawbacks in my situation, so this is what I have gotten myself into…
My perspective on deleting the pump when I turbocharged a 2004 Renesis with a custom top mount:
Pros:
(1) Simplicity,
-If there's a supply of fuel, the engine is being lubricated with premix.
(2) Reliability,
-(see reason #1)
-No melted lines or sensors from the turbo manifold and/or downpipe.
-Metering pump is maxed at higher rpms/loads and does not deliver an adequate supply, necessitating the need for more lubrication by premixing.
(3) Valuable real estate,
-With a medium sized turbo (T4/GT3582R) & larger you start using every square inch.
(4) Alternative fuels,
-With E85 you just might want a different source of lubrication than what is in the oil pan.
Cons:
(1) Precision,
-Not as precise, as the oil metering pump can deliver smaller amounts of lubrication at lower rpms/loads and not waste your valuable premix.
(2) Expense,
-The cost will be more considering you are premixing for worse case scenario.
(8500 or 850 rpms, the engine is receiving the same ratio of gas to oil when premixing)
(3) Convenience,
-It takes more diligence, considering you have to mix with every fill up at the pump.
I had Richard Sohn’s adapter for quite a few years and premixed. It’s a great piece, I converted the windshield washer reservoir for the tank and it worked great. However, the adapter as some of you found out with a top mounted turbo system kicks out the OMP to the point of making the downpipe (even when modified) really close to the plastic sensor on the pump and those without semi solid motor mounts (ie: modified mazport mounts) just doesn’t work out in the long run.
Deleting the oil metering system outweighs the drawbacks in my situation, so this is what I have gotten myself into…
01-06-2010 | 01:36 PM
#2
Guest
Posts: n/a
(1) OK, so I started with the ATR software and modified these three:
-Zeroed out the oil metering Load Based Map
-Zeroed out the oil metering Throttle Based Map
-Brought up the DTC’s Base and unchecked the P1686, P1687 and P1688 From Zoom44’s Tech info (thank you):
P1686 Metering oil pump control circuit low flow side problem ON 1
Other ´ (See DTC P1686 )
P1687 Metering oil pump control circuit high flow side problem ON 1
Other ´ (See DTC P1687 )
P1688 Metering oil pump control circuit initial check problem ON 1
Other ´ (See DTC P1688 )
(2) I then turned down a few M10 x 1.25 Allen head bolts on a lathe and reused the oil injector washers:
(they need to be 13mm long instead of the 20 that I had)
(3) I had an aluminum block-off plate made for the lack of oil metering pump:
(Drilled and tapped for an 1/8” npt fitting… a delrin bushing was also made to slide over the keyshaft and against this plate.)
(4) Relocated and modified a used oil metering pump to fit next to the passenger side shock tower along where the air pump/fuel pump resistor used to reside. I also lengthened the wiring harness a few inches with an engine harness I got off of e-bay, since I ended up melting the original connectors.
(The plan is to eventually pull the omp out completely)
??(5) Figure out if a relay can be used to supply the sensor switch side with the required battery voltage at startup for the pcm's learning mode. (only after memory has been cleared?)
O.K. there is an initial check by the pcm @ start up:
(Thanks Dannobre and Rotary Enthusiast)
(This is the Manual that I used)
So, I found out this is called a can stack permanent stepper motor. Now, since it has 8 fingers on both north & south sides on each coil totaling 32, I suppose it moves 22.5 degrees in full steps and with two coils offset it can do 11.25 degree half steps in both forward and reverse.
With 16 teeth on the pinion and turning the shaft it takes two full rotations to go from min. to max. If this vibes, then it takes 32 full or 64 half steps to hit maximum on the omp motor and activate the switch.
(info derived from www.ibiblio.org)
These oscilloscope measurements below are from the shop manual posted above.
Ok, from what I understand, this motor is a unipolar motor that feeds a positive supply through the windings and is completed by grounding the signal through the pcm driver. (looking for an activation signal for the relay)
With the pulsing from the pcm wires (2W & 2V) I wonder if I can use one of these for the activation and set the relay "on" time to a whatever amount of time to mimick the motor hitting max on the omp and activating the switch or possibly using one to activate the "on" and the other to trigger the "off" signal?
Considering I have all the omp tables on the ATR software at a value of 0, the omp motor (other than at learning mode) shouldn't move.
What I would really like to do is use the omp motor harness to wire in a relay as it's right next to the omp switch, making it a clean & easy install with something like the DEI 611T relay.
My idea so far with the DEI 611T relay:
(there are (5) wires on this relay)
-Black: to chassis ground = (several to choose from)
-Green: to + input for activation = (mop control 1 (2W/E) or mop control 3 (2V/A) ???
-Red: to +12v = (C or D / red or orange wire on omp motor harness or white/red on pcm side)
-Yellow: for +12v output = (A/2N or brown wire on omp switch harness)
-Brown: to +12v input = (C/1U or black/yellow wire on omp switch harness)
As if you didn’t already figure out, I’m not all that familiar with the electrical side and I'm slowly learning. If anyone can shed some light, I appreciate it.
Those that don’t want to help or think that this is below them please don’t bother posting. Thank you.
-Zeroed out the oil metering Load Based Map
-Zeroed out the oil metering Throttle Based Map
-Brought up the DTC’s Base and unchecked the P1686, P1687 and P1688 From Zoom44’s Tech info (thank you):
P1686 Metering oil pump control circuit low flow side problem ON 1
Other ´ (See DTC P1686 )
P1687 Metering oil pump control circuit high flow side problem ON 1
Other ´ (See DTC P1687 )
P1688 Metering oil pump control circuit initial check problem ON 1
Other ´ (See DTC P1688 )
(2) I then turned down a few M10 x 1.25 Allen head bolts on a lathe and reused the oil injector washers:
(they need to be 13mm long instead of the 20 that I had)
(3) I had an aluminum block-off plate made for the lack of oil metering pump:
(Drilled and tapped for an 1/8” npt fitting… a delrin bushing was also made to slide over the keyshaft and against this plate.)
(4) Relocated and modified a used oil metering pump to fit next to the passenger side shock tower along where the air pump/fuel pump resistor used to reside. I also lengthened the wiring harness a few inches with an engine harness I got off of e-bay, since I ended up melting the original connectors.
(The plan is to eventually pull the omp out completely)
??(5) Figure out if a relay can be used to supply the sensor switch side with the required battery voltage at startup for the pcm's learning mode. (only after memory has been cleared?)
O.K. there is an initial check by the pcm @ start up:
(Thanks Dannobre and Rotary Enthusiast)
(This is the Manual that I used)
So, I found out this is called a can stack permanent stepper motor. Now, since it has 8 fingers on both north & south sides on each coil totaling 32, I suppose it moves 22.5 degrees in full steps and with two coils offset it can do 11.25 degree half steps in both forward and reverse.
With 16 teeth on the pinion and turning the shaft it takes two full rotations to go from min. to max. If this vibes, then it takes 32 full or 64 half steps to hit maximum on the omp motor and activate the switch.
(info derived from www.ibiblio.org)
These oscilloscope measurements below are from the shop manual posted above.
Ok, from what I understand, this motor is a unipolar motor that feeds a positive supply through the windings and is completed by grounding the signal through the pcm driver. (looking for an activation signal for the relay)
With the pulsing from the pcm wires (2W & 2V) I wonder if I can use one of these for the activation and set the relay "on" time to a whatever amount of time to mimick the motor hitting max on the omp and activating the switch or possibly using one to activate the "on" and the other to trigger the "off" signal?
Considering I have all the omp tables on the ATR software at a value of 0, the omp motor (other than at learning mode) shouldn't move.
What I would really like to do is use the omp motor harness to wire in a relay as it's right next to the omp switch, making it a clean & easy install with something like the DEI 611T relay.
My idea so far with the DEI 611T relay:
(there are (5) wires on this relay)
-Black: to chassis ground = (several to choose from)
-Green: to + input for activation = (mop control 1 (2W/E) or mop control 3 (2V/A) ???
-Red: to +12v = (C or D / red or orange wire on omp motor harness or white/red on pcm side)
-Yellow: for +12v output = (A/2N or brown wire on omp switch harness)
-Brown: to +12v input = (C/1U or black/yellow wire on omp switch harness)
As if you didn’t already figure out, I’m not all that familiar with the electrical side and I'm slowly learning. If anyone can shed some light, I appreciate it.
Those that don’t want to help or think that this is below them please don’t bother posting. Thank you.
Last edited by GTAW; 01-17-2010 at 02:27 PM.
01-06-2010 | 02:03 PM
#3
One note though, and you touched on this, you will be swapping your oil metering from a 2 dimensional rpm/load map to a singular scalar value X% of fuel injected. Oil metering at 8500 rpms might have its highest flow rate, but I think it's actually at its highest value per unit fuel at low RPM/load. I've never gone beyond knee deep in oil metering science, but as you lean out your mixture at cruise running near stoich, the burnoff rate of the oil may increase as a function of each revolution. Certainly you can run a rotary purely on premix as it's been done for years, but did Mazda implement the variable OMP solely for convenience of the owner or is the two dimensional variable injection a more important aspect of it than you credit it?
01-06-2010 | 02:24 PM
#4
Guest
Posts: n/a
I would surely think that OE design is more than the convience factor. I'm viewing this as a balance of what I'm wanting out of turbocharging this engine and an overall "safety" net, hence the simplicity comment.
Dealing with the give and take of FI on a high compression rotary, I decided (not saying it is the best idea overall) to attempt an old school practice that has, as you said, worked in the past and implement it into this case.
I have changed the way I drive this car. When one moves away from an OEM platform, sacrifices are to be made.
Dealing with the give and take of FI on a high compression rotary, I decided (not saying it is the best idea overall) to attempt an old school practice that has, as you said, worked in the past and implement it into this case.
I have changed the way I drive this car. When one moves away from an OEM platform, sacrifices are to be made.
01-06-2010 | 02:29 PM
#6
1) Low mount? Why take the pump out at all if you have the room for it?
2) I'm not sure why you would get the "limp mode" since the pump is attached, but I wouldn't "zero-out" the OMP tables since the OMP isn't doing anything, anyway. Just leave the values alone. Perhaps there is a hidden "check" for pump movement at all.
3) Why not adapt an old-style metering system that injects at the intake port?
2) I'm not sure why you would get the "limp mode" since the pump is attached, but I wouldn't "zero-out" the OMP tables since the OMP isn't doing anything, anyway. Just leave the values alone. Perhaps there is a hidden "check" for pump movement at all.
3) Why not adapt an old-style metering system that injects at the intake port?
01-06-2010 | 03:02 PM
#7
Guest
Posts: n/a
1) Low mount? Why take the pump out at all if you have the room for it?
That's just it, I didn't have the room with a top mount turbo setup and won't (if I reinstalled the pump) for an intake on a T04S style housing.
2) I'm not sure why you would get the "limp mode" since the pump is attached, but I wouldn't "zero-out" the OMP tables since the OMP isn't doing anything, anyway. Just leave the values alone. Perhaps there is a hidden "check" for pump movement at all.
Either do I, however it does it regardless of the inputed values in the ATR software. I've actually had the oil system deleted for a few years now and the issue has become more common recently...I'm stumped.
I have noticed when logging with the AP with the maps set with a value of zero, the motor will still move. If this is a check system of some kind or not, I'm unsure..
3) Why not adapt an old-style metering system that injects at the intake port?
I'm not sure how beneficial this would be considering the fabrication over premixing?
That's just it, I didn't have the room with a top mount turbo setup and won't (if I reinstalled the pump) for an intake on a T04S style housing.
2) I'm not sure why you would get the "limp mode" since the pump is attached, but I wouldn't "zero-out" the OMP tables since the OMP isn't doing anything, anyway. Just leave the values alone. Perhaps there is a hidden "check" for pump movement at all.
Either do I, however it does it regardless of the inputed values in the ATR software. I've actually had the oil system deleted for a few years now and the issue has become more common recently...I'm stumped.
I have noticed when logging with the AP with the maps set with a value of zero, the motor will still move. If this is a check system of some kind or not, I'm unsure..
3) Why not adapt an old-style metering system that injects at the intake port?
I'm not sure how beneficial this would be considering the fabrication over premixing?
01-06-2010 | 03:06 PM
#8
Maybe not a big deal, but that is how I would do it.
As you noted, Mazda decided to have a precise oil proportioning system for some reason. Whether or not its purely emissions related is up for discussion.
01-08-2010 | 11:02 AM
#9
Guest
Posts: n/a
Definitely a valid point if I was going to keep an oil metering system. However that's not the case, I appreciate the advise.
Preferring the more basic layout in the engine bay is still the priority, so the goal for now is to see if I can figure out the feedback system with pcm > omp motor > sensor > pcm. I've never used a scopemeter, so this will get interesting.
Preferring the more basic layout in the engine bay is still the priority, so the goal for now is to see if I can figure out the feedback system with pcm > omp motor > sensor > pcm. I've never used a scopemeter, so this will get interesting.
01-08-2010 | 01:14 PM
#10
The Local Idiot
Joined: May 2008
Posts: 406
Likes: 0
From: Big D, Texas
I've "deleted" the OMP on my car by zeroing out the same oil metering tables you did with ATR. I however left the OMP where it was since I don't need the space yet. I didn't bother suppressing the DTC's associated with the OMP, and haven't had any come on.
Review the service manual to see if there is any information about how the PCM monitors the OMP functionality. You say you "modified" the OMP... in what way? I would also check your wiring extension... it's possible that the added impedance of the harness extension itself is the problem even if the connections are all good (although this seems unlikely).
Review the service manual to see if there is any information about how the PCM monitors the OMP functionality. You say you "modified" the OMP... in what way? I would also check your wiring extension... it's possible that the added impedance of the harness extension itself is the problem even if the connections are all good (although this seems unlikely).
01-08-2010 | 01:27 PM
#11
The PCM does a check pump routine in the first 6-8 sec after startup. If it doesn't like something about the pump it goes into limp mode and the throttle is limited.
I ran into this problem doing my Haltech install. There is no code set...but it does have to do with the position sensor and pump wiring/operation. I have three pumps that scope out all the same...same resistance values..same position switch settings as near as I can tell...two of them will not work....and the third does...
Jeff: Do you have any info on the pump check routine? I know it exists...cause after a reset with the non-useful pumps the throttle works fine till the car has been started and runs for about 10 sec...then it goes into limp mode. It can be repeated after a reset
PS: you might want to fiddle with the OMP position sensor a bit...that might help.....I got that to work for about a year on one pump
I ran into this problem doing my Haltech install. There is no code set...but it does have to do with the position sensor and pump wiring/operation. I have three pumps that scope out all the same...same resistance values..same position switch settings as near as I can tell...two of them will not work....and the third does...
Jeff: Do you have any info on the pump check routine? I know it exists...cause after a reset with the non-useful pumps the throttle works fine till the car has been started and runs for about 10 sec...then it goes into limp mode. It can be repeated after a reset
PS: you might want to fiddle with the OMP position sensor a bit...that might help.....I got that to work for about a year on one pump
Last edited by dannobre; 01-08-2010 at 02:31 PM.
01-08-2010 | 09:49 PM
#12
I would definitely reset the OMP values back to OE rather than zero since it makes no sense to zero them out with it disconnected from actually pumping oil into the engine. Can't say for sure, but I'm with Jeff that this may be the cause. Otherwise it makes no sense for it to throw a limp mode since it can't actually detect oil flow directly. 
the only other thing that might make sense is that the new OMP connector wasn't wired as well as it should have been.
.
the only other thing that might make sense is that the new OMP connector wasn't wired as well as it should have been.
.
Last edited by TeamRX8; 01-08-2010 at 09:54 PM.
01-08-2010 | 09:58 PM
#13
3-wheeler
Joined: Dec 2005
Posts: 2,734
Likes: 0
From: Phoenix
01-09-2010 | 10:48 PM
#17
Guest
Posts: n/a
I've "deleted" the OMP on my car by zeroing out the same oil metering tables you did with ATR. I however left the OMP where it was since I don't need the space yet. I didn't bother suppressing the DTC's associated with the OMP, and haven't had any come on.
Good deal, as suspected the position switch was the culprit.
Review the service manual to see if there is any information about how the PCM monitors the OMP functionality. You say you "modified" the OMP... in what way? I would also check your wiring extension... it's possible that the added impedance of the harness extension itself is the problem even if the connections are all good (although this seems unlikely).
Thanks, I searched and found this Manual via a google search . I just cut it down to fit in the alloted space. Wiring checked out good. Thanks for the suggestions.
Good deal, as suspected the position switch was the culprit.
Review the service manual to see if there is any information about how the PCM monitors the OMP functionality. You say you "modified" the OMP... in what way? I would also check your wiring extension... it's possible that the added impedance of the harness extension itself is the problem even if the connections are all good (although this seems unlikely).
Thanks, I searched and found this Manual via a google search . I just cut it down to fit in the alloted space. Wiring checked out good. Thanks for the suggestions.
01-09-2010 | 10:58 PM
#18
Guest
Posts: n/a
The PCM does a check pump routine in the first 6-8 sec after startup. If it doesn't like something about the pump it goes into limp mode and the throttle is limited.
I ran into this problem doing my Haltech install. There is no code set...but it does have to do with the position sensor and pump wiring/operation. I have three pumps that scope out all the same...same resistance values..same position switch settings as near as I can tell...two of them will not work....and the third does...
Jeff: Do you have any info on the pump check routine? I know it exists...cause after a reset with the non-useful pumps the throttle works fine till the car has been started and runs for about 10 sec...then it goes into limp mode. It can be repeated after a reset
PS: you might want to fiddle with the OMP position sensor a bit...that might help.....I got that to work for about a year on one pump.
Thanks, I totally overlooked one of the DTC's (P1688) that I posted here that was deleted using the ATR software
You hit that one on the head, it was was the position sensor. I'm glad that was an easy fix.
I ran into this problem doing my Haltech install. There is no code set...but it does have to do with the position sensor and pump wiring/operation. I have three pumps that scope out all the same...same resistance values..same position switch settings as near as I can tell...two of them will not work....and the third does...
Jeff: Do you have any info on the pump check routine? I know it exists...cause after a reset with the non-useful pumps the throttle works fine till the car has been started and runs for about 10 sec...then it goes into limp mode. It can be repeated after a reset
PS: you might want to fiddle with the OMP position sensor a bit...that might help.....I got that to work for about a year on one pump.
Thanks, I totally overlooked one of the DTC's (P1688) that I posted here that was deleted using the ATR software
You hit that one on the head, it was was the position sensor. I'm glad that was an easy fix.
01-17-2010 | 02:41 PM
#20
Anything is possible if you throw enough time and money at it....BUT for me it isn't worth the work....I found it easier to just supplement the OMP with pre-mix and leave it at that. Then you don't have to fight with the PCM logic...and have less headaches
01-17-2010 | 07:34 PM
#21
one thing i dont like about not having the omp----is during high load/high rpm--you let off the gas and very little lubrication is getting to the seals.
This is like on track use or mountain driving---idle amount of premix going in but engine is at a much higher rpm.
I agree with Dan--let pre mix be a supplement.
OD
This is like on track use or mountain driving---idle amount of premix going in but engine is at a much higher rpm.
I agree with Dan--let pre mix be a supplement.
OD
01-18-2010 | 09:37 AM
#22
08-30-2010 | 09:10 AM
#23
Ok so its not simple to turn off or un-check the OEM ECU's inspection of OMP using the COBB or anything else?
I wish anyone of our reputable performance companies could figure out a delete or relocate.
I wish anyone of our reputable performance companies could figure out a delete or relocate.
Last edited by ssspeedfreak; 08-30-2010 at 09:14 AM. Reason: add more
08-30-2010 | 08:20 PM
#24
Registered
Joined: Apr 2005
Posts: 654
Likes: 4
Somebody will have to find the limp mode system in the PCM code. Unfortunately, Cobb has abandoned the Rx-8 and opensource tuning hasn't really picked up steam.
As for OMP vs premix, well the OMP was originally developed to supply oil right to the rotor housing instead of only from the manifold as it was done on the old carbureted engines. The whole point is to reduce consumption. You can get the same lubrication effect with less oil consumption by injecting oil right on to the housing. So the first system was based on throttle position only because it had a mechanical rod connected from the throttlebody to a simple mechanical OMP.
Then in '89 Mazda implemented the stepper motor system that carried on until the Series 2 Rx-8. They figured out that at a given level of engine load, after a certain point supplying more oil accomplishes nothing and just increases consumption. Therefore they calculated oil requirements based on the airflow meter signal rather than just having mechical system controlled only by throttle position.
At a certain level of oil injection the apex seal temperature no longer decreases and thus there is no point of using more oil than necessary.
As for the question of letting off the gas on a premix only car, the channels in the chrome surface are designed to retain an oil film. I wouldn't worry too much about it. Still, you would think that after enough coasting eventually the residual oil film wwould break down and apex seal temperature would begin to rise.
As for OMP vs premix, well the OMP was originally developed to supply oil right to the rotor housing instead of only from the manifold as it was done on the old carbureted engines. The whole point is to reduce consumption. You can get the same lubrication effect with less oil consumption by injecting oil right on to the housing. So the first system was based on throttle position only because it had a mechanical rod connected from the throttlebody to a simple mechanical OMP.
Then in '89 Mazda implemented the stepper motor system that carried on until the Series 2 Rx-8. They figured out that at a given level of engine load, after a certain point supplying more oil accomplishes nothing and just increases consumption. Therefore they calculated oil requirements based on the airflow meter signal rather than just having mechical system controlled only by throttle position.
At a certain level of oil injection the apex seal temperature no longer decreases and thus there is no point of using more oil than necessary.
As for the question of letting off the gas on a premix only car, the channels in the chrome surface are designed to retain an oil film. I wouldn't worry too much about it. Still, you would think that after enough coasting eventually the residual oil film wwould break down and apex seal temperature would begin to rise.
Last edited by arghx7; 08-30-2010 at 08:22 PM.
