Mazda 6 Forums banner
1 - 5 of 5 Posts

04 Mazda 6 Sport Wagon 5 Speed Manual
10 Posts
Discussion Starter · #1 · (Edited)
Hello all,

I've been driving around in my 2004 Mazda that I bought a little over a year ago. 3.0L V6 MTX. The entire time since I've owned it I've had a check engine light relating to the EGR system. I ignored it for the 1 year since I've had it as I originally didn't know how long the car would last me. I also figured that if the EGR is kept closed then all the better for my performance and engine. However the car turned out great and now I have to pass emissions tests.

I am getting codes

P1487 - EGR Boost Sensor Solenoid Valve circuit malfunction
P2227 - EGR Boost Sensor Circuit performance problem

So from here begins my adventure.

I am still waiting to see if what I did has managed to fix my car but I have written this post in the style of a write-up so perhaps someone can benefit from my own untangling of the information available in regards to these two fault codes. I still would really appreciate some help and at the end of this post are a few questions.

Note 1: When you look up the P1487 fault code on google you find that the generic fault code relates to a stuck open EGR valve or EGR short. However for Mazdas the P1487 code is actually in relation to the EGR Boost Sensor Solenoid Valve. The EGR Boost Sensor Solenoid Valve is NOT to be confused with the EGR Valve itself. This confusion is what prompted me start by removing and cleaning out my EGR valve, which I'm still glad I did but no real problem was found there. The valve was cleaned even though the valve wasn't too dirty, it sealed fine, and the plunger moved freely before I put the valve back into the car. There is a vacuum nipple on the side of the EGR valve, this connects to the intake manifold/plenum side of the EGR valve and is open regardless of whether or not the EGR is open or closed.

The EGR valve in this car is actuated electrically and not via vacuum. I tested the electrical actuator of the EGR valve and found that all coils measured the specified 22 ohms as shown in the factory service manual. I believe for the 4 cylinder model this measurement is around 12-16 ohms but is performed in the same way. As expected this did not resolve the issue and I still got the same error codes in the end.

For the EGR valve electrical tests, see this page. Mazda 6 Service Manual - Egr valve inspection - Emission system


Once I discovered that the two fault codes relate closer to the EGR Boost Sensor and Boost Sensor Solenoid Valve i moved forward to the next step. I am using an OBDII scanner using Torque Lite. When I read the Barometric Pressure in Torque Lite I get a steady line that never moves no matter what. I'm not sure if this is normal. The line measures at about 55, though I have no idea what type of units this graph is supposed to measure in.

I made sure that all the vacuum hoses between the EGR valve and EGR Boost Sensor were not leaky or clogged. They were fine.

Next I tested the EGR Boost Sensor and the EGR Boost Sensor Solenoid valve. For short I'll just refer to them as Boost Sensor and Solenoid Valve respectively.

Note 2: A lot of people seem to have a hard time finding the numbers for these two parts and are buying them from parts store for very high OEM prices.

EGR Barometric Boost Sensor. Part number E1T10372. Available on eBay for less than $20 for USED OEM pulls or NEW Third Party Manufacturers. Don't get screwed by prices of $100 plus!!!


The second part is the EGR Boost Sensor Solenoid Valve Part Number K5T49091. Available on eBay and Amazon for around 15 dollars. Don't get screwed by OEM prices of $100 or more!! The part I aquired from Amazon seems to have the same quality housing, electrical resistance, and weight. Bought from seller/brand ZBM.


Code P1487 refers to this specific part and it's respective circuit. EGR Boost Sensor Solenoid Valve Circuit Malfunction.
If this valve is clogged or stays closed the Boost Sensor does not receive vacuum from the EGR valve and it cannot produce a reading. This in turn gives you code P2227 as well. Here is how I tested this part:

Method 1: Disconnect the harness from the Solenoid Valve. With the engine running apply 12V to the coil and the engine should fall flat on it's face and then recover. You should hear a click from the solenoid valve. On my Torque Lite app, doing this makes an actual varying reading appear on the BARO graph. However on one occasion doing this did not produce any change in engine noise, perhaps as the ECU learned to accommodate to the difference. Because of this I also recommend testing this part using the second method below. Another thing to note is that positive voltage should be applied to where the white wire from the harness normally connects to this valve in case there is a clamping diode inside.

Method 2:
Step 1: Blow into front nipple # 1 with no voltage applied. There should be no air flow.
Step 2: Blow into front nipple #1 with 12V DC applied, valve should open and air should come out of nipple #2 on the side of the part.
Step 3: Blow into nipple #2 with no voltage applied and air comes out of of the body of the part. This seems to be normal.
Step 4: Blow into nipple #2 with 12V DC applied and air should come out of nipple #1. Plug nipple #1 and with 12V still applied there should be no air flow.
Step 5: Measure across coil with a multimeter in the 200 ohm range and you should get a reading of 38 ohms.

My EGR Boost Sensor Solenoid Valve passed all tests. Just to be sure I changed it anyway. The problem persists.

Next I tested my Boost Sensor. Note 3: At least in this particular Mazda, EGR Boost Sensor and Barometric Pressure Sensor are two terms which appear to be used to refer to the same thing.

During my testing of this gizmo I made another dopey discovery. Note 4: Resistance measurements do very little here as this part isn't a simple potentiometer. I made the mistake of going by resistance measurement and since the resistance doesn't vary when applying pressure, I thought my problem was a frozen Boost Sensor. I replaced the part and my BARO reading in Torque Light went up to around 100 but still does not vary at all despite acceleration or load, whether or not that's normal, the error codes appeared again.

Next I needed to see if both my new EGR Boost Sensor and it's related circuitry was functioning properly and that it was reporting something to the ECU. This is the best way of testing this sensor and it's circuitry and it should have been what I did to begin with. Note 5: To test Boost Sensor, connect it directly to the EGR valve's vacuum nipple, bypassing the Solenoid Valve. I did this with the engine running and doing so produced an engine stumble that eventually evened out. It also gave me a reading on my Torque Lite BARO graph that varied when I revved the engine. This also made my Vacuum/Boost gauge in Torque Lite go to 0. The fact that the BARO graph actually gives me a reading like this tells me that the Sensor itself is good as well as the wiring to the ECU. It also means the vacuum line all the way to the intake manifold is clear or mostly clear. If you deleted your EGR valve you could use a syringe on your Boost Sensor to produce a change.


Here is a diagram of the EGR system. The yellow line is the bypass you use to test the Boost Sensor. The line that goes from the EGR valve to the Solenoid Valve is made of two lengths with a plastic coupler in the middle. You can use this coupler to splice the vacuum line from the Boost Sensor directly into the Line going into the EGR valve.

Up until this point we know that.

1. EGR Valve is not clogged and at least electrically tests good. Mechanically moving it by hand produces a good seal. It also provides vacuum at the nipple on the side.
2. EGR Boost Sensor and it's wiring to the ECU is good.
3. EGR Boost Sensor Solenoid Valve is good and not clogged.
4. Vacuum lines between EGR and Sensor are good.

The only thing I could think of after this point is that the Solenoid Valve was not being told to open by the ECU and this did not allow the Boost Sensor to take a reading. I took my multi meter to the control wire of the Solenoid Valve which is Green with a Yellow Stripe. This wire connects directly to the ECU and should measure 0V when the Solenoid Valve is told to open. When the Solenoid Valve is NOT being told to open this wire should measure 12V. The white wire on the Solenoid Valve is it's 12V source and should always measure 12V as long as the ignition is on. If there is no voltage on the white wire you will always measure 0V on the control wire as well.

No matter what I did I never saw the Green and Yellow control wire for the Solenoid Valve go down to 0V. For some reason the Solenoid Valve was never being told to open. With the connector removed from the Solenoid Valve I ran the following electrical tests. First a diode test. Multimeter red probe to ground and black probe to the control wire. Multimeter set to diode mode. This should read around 0.5V to 0.6V. A resistance measurement in the highest scale should read O.L, otherwise there may be a short or current leak. Next I did an End to End continuity check of the control wire. One probe of the multimeter to the end that connects to the Solenoid Valve and the other probe to the end that connects to the ECU. For this you'll have to remove the ECU connector. You'll also have to remove the cover that goes over the ECU connector in order to find the Green with Yellow strip wire. Roughly speaking it should be the 6th pin from the right and the second row facing you. This should be measured on the lowest resistance scale of the multimeter and should be as close to 0 as possible. I got something like 0.160 which is perfect. If you have high resistance verify that you are using good quality test leads for your multimeter and any alligator clips you may be using for extensions are also of good quality. I use test leads from PartsCandy .com.

All my electrical tests passed. This lead me to believe that my ECU was bad and for some reason it refused to ever ground out the Solenoid Valve. Out of curiosity I replaced the Solenoid Valve with a 200 ohm resistor and by doing this I did see the Green and Yellow control wire ground out a couple of times very briefly near engine start up but never saw this happen with the acual solenoid valve in place. This is actually what prompted me to replace the solenoid valve more than anything incase the solenoid valve was partly shorted. Upon arrival of the new part I discovered that the new part also measured 38 ohms and upon instalation the problem persisted.

So now I've changed out my ECU and am driving around to see if the code is gonna come back. So far only 6 miles on the trip meter but no code and am hoping to God they don't come back or else I'm gonna pull my hair out. I do remember the Pending code taking less than that to return.

After replacing the ECU I measured the control wire to the Solenoid Valve and found that a few seconds after start up the line does ground out for a very brief moment. However this doesn't seem long enough to register any type of reading in my BARO graph in Torque Lite and this graph stays the same flat line at all times. At the same time that's more than I ever saw before changing the ECU.

After this whole ordeal my questions to you guys are.

1. IS the EGR Boost Sensor Solenoid Valve supposed to be open at all times as long as the engine is running or does it only ever open during specific conditions? I've not yet gotten up to highway speeds but have tried hard launches, soft launches, with no reading on my graph.

2. Is there a MAP sensor on the V6 3.0L version of this car? According to this post here P2227 is derived from comparing MAP sensor readings and BARO sensor readings. When I set up a MAP sensor graph or dial in Torque Lite I get no data. Simptoms connected to p2227 - Mazda Forum - Mazda Enthusiast Forums

3. What does a typical BARO sensor reading look like? According to some information I read online this is only taken as a quick sample before start up to detect the Ambient Barometric Pressure. However in this car this seems to double as a MAP sensor? Why is it tied to the EGR valve?

4. What is the normal amount of vacuum I should be reading from this car? Mine stays at around 20in/Hg at idle. I noticed that when I bypassed the Solenoid Valve this would go down to 0 and teeter-totter between boost and vacuum.

5. How long does it take for the EGR system to come up as ready after having had the battery disconnected?

I appreciate you guys reading this and I would highly appreciate your help. I really hope these engine codes don't appear again but am already preparing for the worst.

104 Posts
I'm curently getting intermittent P1478 and P2227 but mostly P1478 but never at the same time and they are mostly pending codes. I did also notice that I'm getting P0138 with P1478.

One thing I'm noticing is that these errors tend to show up during rainy days when I'm driving through lots of watter, which leads me to think moisture is getting somwhere it's not suposed to. So far I have not been able to find the culprit.

Sent from my SM-G973U1 using Tapatalk

04 Mazda 6 Sport Wagon 5 Speed Manual
10 Posts
Discussion Starter · #4 ·
I'm glad to announce that so far no engine light has come back after just under 100 miles and I passed my emissions test. The CEL used to come back after just idling sometimes. I plan on rewriting the post as more of a write up with step by step instructions on how I diagnosed my EGR system. As I mentioned I'm not an expert on the subject but the information out there regarding the EGR system is very spread out and confusing.

The issue indeed turned out to be the ECU in this case. In fact I tried to open and repair the ECU itself. There is a series of motorola driver ICs along one side of the board that interface all the input and output pins of the ECU. One of these, closest to the connector, seems to have been the problem. The ICs themselves are not available for purchase and although I tried resoldering and adding a wire bridge between the interface IC and the external control pin, I did not stick with the old ECU long enough to confirm that this fixed it. My plates were way overdue and I did not want to wait around to see if that fix would work in the long run.

Shout out to Flagship One Inc for the ECU. No I'm not being paid to say that. They had the exact same ECU with lifetime warranty and 2 pre-programmed keys. Beats digging one out of the junkyard and figuring out key programming and all that jazz. I even ended up being able to program my old key by doing a little ON/OFF sequence with all 3 keys at first start up after installing the new ECU. Perfect solution and they were a life saver, especially since I was in a hurry.


The EGR BOOST SENSOR is actually just a barometric pressure sensor. Most of the time.
A barometric pressure sensor just measures the outside atmospheric pressure of where ever you are in the world. This is just like the barometer on a weather station. This is why the second nipple on the EGR Solenoid Valve when not activated just leaks air out through the body. This diversion lets the boost sensor sample the regular outside air when the EGR Solenoid Valve isn't activated, which is apparently most of the time. Only on certain occasions does the EGR Solenoid Valve activate which diverts the EGR Boost Sensor to instead measure the intake manifold pressure through the EGR Valve vacuum nipple. This measurement sequence occurs when the EGR Valve is told to open and the resulting measurements confirm whether or not the EGR valve has opened and if it has enough airflow going through it.

Thus seeing a non changing reading at the BARO graph like I was seeing in torque light is completely normal. Atmospheric pressure does not change in any way that would be instantly noticeable. This infact may be normal even during EGR Solenoid Valve activations as the data is probably sent to a different data group and is not shown in the BARO graph, I'm not really sure.

@Dr Feel Good thank you very much for those manuals! I actually had already found the V6 service manual through one of your old posts and it was like finding gold! I also went ahead and downloaded all the other manuals you posted. These will be very very handy. Huge thanks.

104 Posts
Do you have any or can you take some high res shots of the old ECU internals? If it comes down to it I may have to take my ECU to a shop since here in Poand i doubt I can find a replacement ECU, and it's just not worth it for me to import a replacement one.

If I have some high res shots I can send them to the shop before I take out my ECU so the shop can see what they will have to work with.



Sent from my SM-G973U1 using Tapatalk
1 - 5 of 5 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.