a3dcadman

iTrader: (1)

I sprayed starter fluid around UIM gasket area, underneath the UIM, down in the turbo area, behind the UIM, and all around the area where the ch ch ch noise is coming from. NO DIFFERENCE!
chuck

tcb100

I am fixated on the running rich issue. All the unburned fuel is causing your cat to fire up. So why is there unburned gas when you are in closed loop? Your O2 sensor ought to be correcting an overly rich mixture. Yours is pretty new so it is probably working but it would not hurt to check.

Maybe you have a leak in the exhaust system introducing false air to the O2 sensor & it keeps richening you up? I saw that on a BMW once: hole in the header fooled the the O2 sensor and it ran rich as hell. Maybe throw in some ATF, make lots of smoke and see if there is a leak ahead of your o2 sensor anywhere?

Just for the hell of it, park your car in a dark place and spray water around the engine compartment to see if you have any kind of a spark leak too. That only takes a second.

a3dcadman

iTrader: (1)

Im leaning towards a problem with the egr / lim / exhaust area. On startup the fumes and exhaust smell is so bad, it makes my eyes burn and water when Im looking down into that area of the engine.

I also have checked for engine error codes - none came up.

I did the voltage checks for the MAP as per the shop manual and they were way off spec. This one Im not sure of my results and need to find out more about the test procedure.

Parts arriving tomorrow and I will be diving into this project. From whats Ive read, indications are that pulling the turbos should be a final confirmation on condition of the apex seals.

Also saw a post, think it was Jason (rx7 store) that stated there is a discrepancy in readings generated from a piston type compression tester as compared to a rotary compression tester. 70 -90 lbs for the piston tester is comparable to 80 - 100lbs on the rotary tester. If that is indeed the case, then my compression numbers are right in there.

chuck

tcb100

Chuck, just in case, here is the procedure for testing the map sensor.

There are three wires going into the connector. Left to right, a brown and white reference voltage, a black and green ground and a green and yellow signal. Either rig up three jumper wires or carefully stick pins into the back of the connector.

1) Engine running, warm and at idle, voltmeter + lead to reference (BW), neg lead to ground(BG) - test to make sure you are getting 5v power.

2) Positive lead to sensor ground(BG), negative lead to battery ground, test for voltage drop, should be very low. I get .06 volts.

3) Positive lead to sensor signal (gy) and negative lead to to ground (BG)- at idle 1.3 to 1.6 volts.

4) Disconnect the vacuum hose. At zero vacuum you should get 2.4 to 2.8v. I got a 2.6v

5) Use the pump to create vacuum. Vacuum "sucks the voltage out". The service manual shows idle, 0, and 19.7 inches so pump the vacuum up to 20. The manual says 1.25 to 1.55v. I got 1.14 at 20 inches.

Key on – 3.6v, Idle – 1.4-1.7v

a3dcadman

iTrader: (1)

Got up at the crack of dawn and started the disassembly process. I will be posting pics and updates on my progress, what Ive found, and hopefully a definite resolution to these perplexing problems.

I do definitely have an exhaust leak. I am amazed at how much **** is packed under the hood of this car. I am just about ready to pull the turbos.

Here is a list of parts I ordered for this project. I may not need everything, but with Turkey day next week, wanted to make sure I had everything necessary to complete the job (I hope)!
1. silicone hose kit (blue)
2. fuel pulsation damper
3. UIM and LIM gaskets
4. EGR valve and gasket
5. O2 sensor
6. MAP sensor
7. coolant hose kit excluding radiator hoses
8. built a heavy duty 2 tier rolling work bench
9. will be getting the injectors serviced and spec'd

Tomorrow, pics and discovery. Thank you to all of you that have provided data and input for this thread.

chuck

Speed of light

iTrader: (8)

With respect to the O2 sensor and running rich: Any misfire that occurs in the engine, for whatever reason, will cause the O2 sensor to think the engine is lean. I know that this may seem conterintuitive, but if you think about it for a minute, you can understand why: a misfire introduces uncombined oxygen into the exhaust, which is read as lean by the sensor. (It doesn't see the hydrocarbons.) The result, of course, is for the ecu to attempt to richen the mixture. And this, of course compounds any other problems...

DaveW

So, obviously, an ignition problem (fouled plugs, bad wires, bad coils, etc.) could cause a3dcadman's problem. Interesting...

Mahjik

iTrader: (1)

However, that won't cause the uneven compression between the front and rear housings; and his readings are different enough to be outside of Mazda's spec (no greater than 21 PSI difference).

a3dcadman

iTrader: (1)

The problem with the engine idle is still unresolved. Im starting the hose replacement this week and will be replacing the aforementioned parts. Here are some pics of the exhaust leak area underneath the LIM. Looks like its been there for a while. Originally I thought it was the downpipe which I had fixed. The leak almost looks like its coming from the bottom of the LIM. I will have to pull the exhaust manifold tonite and check the gasket. Also looks like there was a little water leak as well.

Look at the pics and give me your thoughts.

RotorJoe

So you have your turbos off now? Check those for sticky exhaust wheels. It’s a good check while they are off, spin them a few times and check for binding. I had my front turbo exhaust blade lock up on occasion and that would make my motor run like the exhaust was clogged or the motor only running on one rotor. I couldn’t figure out what was going on at the time. Since I have non cat, I thought my newly rebuilt seals were sticking. I didn't know the wheel was sticking occasionally until my motor let go (under a forth gear full boost run) and when we pulled the turbos off we noticed the wheel was stuck. It would free up for a bit then lock again. Just something to check.

Take a look at your cat while you have some of your exhaust off. Make sure noting rattles around or sounds loose in there.

Also, now check compression. It may have changed if the exhaust was restrictive.

The exhaust leak could have been caused by a restriction somewhere.

I don’t know why I’m soo stuck on it being and exhaust problem.

Good Luck and Happy Trouble Shooting.
Joe

P.S. The map sensor was a good point to test also. If that fails the car runs like its blown almost or it will constantly flood itself out, depending on the failure.

Mahjik

iTrader: (1)

I'm curious. If compression is a measurement of how well the engine combustion chambers are sealing, how does the exhaust play into that?

RotorJoe

I saw on my first gen once when the exhaust was clogged (now mind you this was really clogged) it causes lower then normal compression. I saw 75psi (piston engine compression tester) and I usually saw around 90psi.

I thought the motor was going south at that point. But when I had a buddy fire the car up, I noticed my 6 port actuators (GSL-SE) slap open immediately. Knowing that those work off of back pressure I knew that there was a lot of back pressure in there at idle. So that pointed towards the exhaust.

Once I pulled the cat off all went back to normal.

So that is why I think compression numbers can be affected by a clogged or stuck exhaust turbo blade.

Too much back pressure could case the seals to not seal completely? Thats they only thing that I can think of that would cause what I saw.

Mahjik

iTrader: (1)

The exhaust port is on the other side from the spark plug hole (i.e. compression tester). The compression tester is reading a different face of the rotor than the one exposed to the exhaust at that time (unless I'm just missing something here).

I suppose they may be exposed to each other at that time but the comression is just showing how much air is being held by that chamber, not necessarily how much is being expelled out of the exhaust (which you would assume is the same, 120 PSI coming, in 120 going out..... ).

RotorJoe

No, you’re exactly correct. I was and still am pretty confused by what I saw. It could have been just a coincidence what I saw also.

Although I would be curious to see his compression numbers now. It won’t be hard to retest. At that point all admit defeat.

Another thing, I almost forgot. When I was 300-500 miles into my break in on my previous FD motor (the one that failed due to the turbos going out) would show 100 psi on all faces both rotors but when it started acting up I would see 100 on all faces of the rear and only 70psi ish on the front rotor all faces. This is another reason why I thought exhaust problems can play a part in compression numbers. Although at the time I though my seals were sticking and hell they probably were.

a3dcadman

iTrader: (1)

I will run the test again tonight. Any input on the pics?

chuck

RotorJoe

/\
Replace the turbo manifold exhaust gasket while you're in there.

Mahjik

iTrader: (1)

If I'm misunderstanding something, correct me as I want to learn too!

I would also like to see a new compression test.

RotorJoe

So do I, So do I (want to learn).

I don't think you misunderstood anything and I agree with your opinion/thought process. I think I'm just seeing weird results and am attributing them to other symptoms/problems that occurred at the same time. But that’s from my past experience so I have to take that into account and the results I got when I ran the compression test at the time. I too am confused by my results. Anything that goes on the exhaust side shouldn’t affect compression but I’m not so sure about that anymore. I guess I want to see what goes on in other parts of the world because I am stuck in my own.

tcb100

I agree that the sound is that of a leaking exhaust. As I said above it sounds like a BMW I saw a couple of weeks ago with a hole rusted around the header bolt.

That false air might be seen by the O2 sensor as a lean condition & causing the ECU to enrich the mixture, leading to a red-hot cat. I'm still stuck on the hot cat. Assuming no ignition issues, why is there more fuel than there should be?

a3dcadman

iTrader: (1)

Here is a pic of the new LIM gasket. What is the function of the 2 rectangular holes and 2 oval shaped holes in the middle of the gasket. Are any of these exhaust related? I havent pulled the LIM yet so I am not able to visually determine where these holes go. Looking at the carbon buildup on my engine, it really looks like the leak is coming out of the center of the LIM contact area.

chuck

RotorJoe

No, none of those holes are exhaust related. The two in question are the two center/primary iron intake holes. I forget what the rectangular holes do/are for.

I think your leak is at the top turbo exhaust manifold gasket and thats why you see black soot all over the bottom of the LIM. It was blowing hot exhaust gas up on the bottom of the LIM. Or it could have been where the twins mount to the turbo manifold.

I don't think you would see that much soot if you had a leak at the LIM gasket, plus if this was the case you may have a boost leak/vacuum leak.

a3dcadman

iTrader: (1)

Are any of the middle holes channels for coolant? There was an indication of a coolant leak as near as I can tell.
cp

Mahjik

iTrader: (1)

There is no coolant running through the LIM. The only liquid that is in the LIM is fuel.

CantGoStraight

When the compression test done is it not done with the TB wide open ??? Since this is a pump the compression can be affected by how much air it takes in and compresses it as well as being able to get rid of it. Perhaps a restricked exhaust would act the same as doing the check with the TB plates closed ? ? ? Just a thought

Mahjik

iTrader: (1)

Take a look at this animation:

http://www.rotaryengineillustrated.c...tions/IGN4.php

Now, imagine that the compression tester is one of the spark plugs. It's measuring the amount of air it's holding in that section. Yes, the amount of air coming in will change the results (i.e. warm vs cold engine), however it should still represent an "idea" of the overall compression (i.e. if you show a 25 PSI difference between the two housings, the compression per housing might go up or down, but the variance should still be 25 PSI).