wstrohm

Dave,

Thanks for your response. I will do some more checking; might have missed something.

1) Yes; I understand. Sometimes finding the switching threshold voltage and then comparing with others yields useful information. For this latest testing I used about 10.5 volts.

2) In general I just tried to find a single air path from the valve and after verifying there were open lines on other ports, seeing whether the single path would switch on and off. I'll use your input to retest.

3) Will do that, thanks. PITA, though, doing a select-in-test on those four valves.

4) Pizza test? Wifey will love that.

I take it you do not trust these valves to last the life of the car, LOL!

wstrohm

Update...

All solenoids are now OK. I had tested the Charge Relief solenoid incorrectly, and I fixed the Turbo Control solenoid with WD-40; it no longer sticks. Finished the overall FPD replacement/O-ring replacement job yesterday. Unfortunately, the engine runs extremely rich, rough, and nasty at all rpm. (But there is no fuel smell... yet.)

Checked the MAP sensor calibration per the factory manual, and it's fine.

Compared the photos I took of the solenoid rack hookup this time with those I took last time when I did the vac hose replacement, and I can't see any differences, although I suppose something could have come loose down behind the rack where the camera can't see.

I do know that the EGR connector on the wire harness side is shredded... the whole interior is broken up and the internal terminals are floating. However I tried to connect the thing properly even though the connector body is missing some pieces. And there is no EGR fault code (no fault codes at all). With "TEN" and "GND" connected in the Diagnostic Connector, I can adjust the idle speed with the AAS screw right through the computer-controlled idle speed, so I matched those. The Accelerated Warmup system seems to be working, i.e. the fast idle cam does disengage the roller as the car warms up, so it's not the choke itself, unless there is some problem internal to the valve, which I didn't disturb.

Any ideas what would cause this condition? The car is driveable, and I may have to give up & take it to Irvine Mazda if there is no help here... I can't see taking it apart again without having the slightest clue what might be wrong.

Attached a pic of the rats nest after re-installation, FWIW. Thanks for reading!

wstrohm

Just had a thought... what would happen if the FPR end of its vacuum line to the PRC solenoid were not connected?

dgeesaman

iTrader: (7)

Well, the fuel pressure would be too high at low load, and too low at high load. I believe it does need to be connected properly.

Dave

wstrohm

Yes, of course it should be connected. My problem is that I don't remember disconnecting it from the FPR when I changed its O-ring, but I also can't remember verifying it was connected when I reinstalled the fuel rail and primary fuel distributor. I have pics of the solenoid rack and the hose is connected correctly, but naturally I have no pic showing the other end. I suppose I could put a "tee" into the fuel line and measure the fuel pressure (I have a gauge, but no tee). Easier than tearing everything apart again. Off to Kragen and/or OSH tomorrow to buy a tee.

Can't think of anything else that would cause rich running... ??

FDeez

Your TPS may need to be recalibrated. Out of calibration, your car can display the symptoms you are experiencing. The TPS sensor is located in the area of the MAP sensor. It's a little black unit with two philips screws on it. Good luck!

sevensix

iTrader: (1)

hey wstrohm im from fv too, haven't seen your fd around before. i've seen a red, a silver, and a montego blue though, and also a black one.

dgeesaman

iTrader: (7)

I think the FPR hose is sorta reachable from around behind the manifold, or under the car if you reach over the tranny bellhousing.

Stuck open injector could too.

Use a 9V battery and jump across the injector - I think I used the pin at the ECU and I believe it grounds through the main relay. In any case, make sure each one is clicking. Sometimes, the exposure to air, etc causes them to stick open. If it's not clicking, try tapping on it until it unsticks - someone wrote in detail about it earlier this summer.

Dave

wstrohm

FDeez,

TPS should be OK. I didn't touch it on this job, and the car ran fine previously.

sevensix,

Our car is pictured to the left of each of my posts. What color is yours?

Dave,

Thanks for the insight into access to the FPR end of the PRC solenoid valve control hose. I'll have a look. I have checked the injectors, as I wrote previously: "I have replaced the O-rings of all four injectors, and electrically tested them. They all open at around 3.4 volts. I also put a little finger over each pintle and verified that it retracted when the injector switched "on.' "

Reviewing photos of my previous work on this car, I noted that the routing of the PRC line to the FPR is now different. Maybe the hose is connected, but either kinked or squeezed shut by the new routing. Looking at the pics below, the left one is factory, the middle one is after I replaced the vac hose with Viton (both had no problems) and the right one is as it is now (rough running). If the hose routing is the problem, obviously I have to tear this puppy down once again.

Regarding the PRC system, Mazda says on page F-111 of my 1994 workshop manual: "This system cancels the vacuum applied to the pressure regulator and increases the fuel pressure during hot engine start-up and for a period immediately following engine start-up. This improves hot starting as well as providing smooth idle. - To prevent vapor-lock during hot restart idle, vacuum to the pressure regulator is momentarily cut, and fuel injection pressure is increased."

Comments:

1) To clear a vapor lock condition, the fuel flow through the rail should be a maximum to flush out the vapor and replace it with liquid fuel. In order to do that, the FPR should be fully open to allow maximum flow of fuel back to the tank. And to do that, the FPR needs maximum vacuum, not minimum. I don't understand the logic.

2) A hot engine does not require more fuel during restart... a cold engine does. I don't understand this either.

The 1990 - 1993 Miata connects its FPR control hose directly to the intake manifold. My '90 has always started instantly for any condition of the engine. What if I ran the FPR control line directly from the vacuum tap to the FPR, bypassing the PRC solenoid altogether? Has anyone tried this? That way the hose would be contained totally below the fuel rail and behind the wiring harness, and would be much less likely to foul or be pinched.

dgeesaman

iTrader: (7)

Bill, I think you're onto something with the pinched hose. I remember doing the exact same thing the first time I did the hose job - there is some big widget (ISC? I don't feel like looking it up) under the UIM that gets VERY close to the rats nest rack. I pinched off this hose, and I now use a zip-tie to keep the hose pulled about 1" toward the front of the car and away from this pinch point.

Check that out.

Dave

wstrohm

Dave,

I did make that mistake also, but with the secondary air valve solenoid hose from the solenoid 3rd from the right, in the right-hand pic in my last post. Actually that pic was taken with that hose in the interference position... I had to rotate the connection at the top of the ACV clockwise to reposition the hose to the front of the AWS solenoid valve. It's the one on the far right... not pinched by the AWS, it's too far to the rear, but I'm worried about routing it next to the circular portion of the ACV, and what is happening behind the wire harness to its right as the hose turns to get to the FPR.

wstrohm

Post-Thanksgiving re-teardown of the car showed that I was barking up the wrong tree in chasing a pinched FPR control line. It was tight on the ends and loose in the middle, just like all the other lines.

I have the extension manifold off, but the coil assembly and solenoid rack, alternator, and all the intake air tubes (on passenger side) are still in place. I checked the FPR diaphragm using my MityVac, and it is solid for both pressure and vacuum. I also looked at the PRC solenoid, and it is in the right mode (through path from engine tap to FPR is open with ignition on).

Have looked at as many solenoid vacuum paths as possible and cannot find anything incorrect or not working. Since the problem is excessively rich/rough running at both cold and warm coolant temperatures, it seems to me that I still need to put a tee in the fuel line and check the fuel pressures for both atmospheric and idle vacuum conditions on the FPR control line. It is really hard for me to believe that an injector is leaking, since all I did was remove them, replace their O-rings, and replace them.

Is there any way at all that a new FPD could cause high fuel pressure? I guess I'll wait until I test that...

One of the things I'm planning is a new set of plugs and plug leads. The plug leads are original, and although they test in spec, as long as this thing is apart, I will go ahead and change them out. The plugs have 17 K miles on them... copper, not platinum... couldn't find Pt NGK plugs anywhere last time.

Any other suggestions or comments would be helpful! 90 K miles on an almost stock engine... ran fine except for fuel smell before I replaced the FPD and put new O-rings on the injectors and FPR. Everything went so smoothly... have no idea what the problem is.

DaleClark

iTrader: (55)

Few things I could think of here.

First, the most dumb thing - make sure the MAP sensor is hooked up properly. Vacuum line should go from the MAP sensor to the long nipple on the back of the upper intake manifold. In the vacuum line to the MAP sensor there should be the black factory filter for the sensor. There SHOULD NOT be a check valve in this line - the MAP sensor needs to see both vacuum and boost.

Next is the water temp sensor - it's at the back of the water pump housing, it's a green square 2-wire connector. If the wiring is bad, shorted, or broken, the car will run like crap when it's cold, but OK when warm.

Next is the fuel injector O-rings. The one drawback with side feed injectors is the bottom O-ring - if it leaks, it will drip pressurized fuel straight into the engine. Had a buddy with a 240sx fight this problem. I wish I would have seen this post earlier - I ALWAYS recommend doing a pressure test on the fuel system once you re-assemble it and you're still in there. Jumpering the FP (fuel pump) connector to ground at the diagnostic terminal will run the fuel pump with the key in the On position, making it very easy to spot leaks. Whenever I get into the fuel system on a car, I run the fuel pump for about a minute, carefully looking and smelling for leaks.

Also, make 100% sure the fuel feed and return lines going into the rail are hooked up properly. If they're backwards, you'll get all sorts of weirdness. It looks like you did a good job keeping things labeled and straight, but just double-check.

On the fuel pressure thing - fuel pressure would have to be sky-high to really make the car run poorly. But, it might be worth teeing in a fuel pressure gauge to the feed line and running the fuel pump to make sure your pressure is right - if it is, that's one less thing you'll have to chase.

Another dumb thing to check - make sure the spark plug wires are hooked up properly. If they're hooked up weird, the car will run VERY strangely.

Big thing is to focus on the problem - either the engine is getting too much fuel, mechanically (fuel pressure too high, fuel leaking around injector into engine), electronically (ECU is confused by a bad sensor input), or your spark isn't lighting the fuel it's getting off properly or at the right time. It will definitely be in some part of the system you touched - a component on the car most likely won't go bad just sitting in your garage, if it worked when you started, it's likely still OK.

Keep up the good work, I know you'll find your answer, then you'll be a Certified FD Bad-***.

Dale

wstrohm

Dale,

Thank you very much for taking the time to answer this thread.

MAP sensor... not only checked the hookup and the gas filter flow both ways, but also probed the rear of the connections with MityVac as a driver. Voltages are correct for all vacuum/pressures.

Water thermosensor... had looked at its voltages about a year ago; it was fine. However, I did have its connector off & back on during this job, when I had to get the wiring harness out of the way of the fuel rail etc. No problems disconnecting/reconnecting the connector, but I guess something could have happened. However, the roughness and black smoke occur at all temps.

Fuel injectors...as you read, I replaced the O-rings with new ones. They went on easily after coating them with dielectric silicone grease. I was careful to wipe excess grease off of all the new O-rings... injectors, new FPD, and FPR. I also connected F/P to GND in the Diagnostic Connector, turned on the ignition, and listened to fuel gurgling through the (new) fuel lines for about 2 minutes. No indication of a leak anywhere. The fuel hoses are connected as originally... input to FPD and primary fuel distributor (lower hose), output from FPR.

Fuel pressure... today I will try and find a brass tee to do that test. (I have a gauge.) I plan to use the F/P jumper and run the pump while varying the FPR control hose vacuum with mty MityVac, to see if I get the 28 - 32 psi specified for idle, and to see what the fuel pressure goes up to as the control pressure goes up.

Plugs/wires... they were re-hooked up correctly, but I had to struggle with the coil tower connection at the #1 rotor trailing plug wire. I might have damaged it there; the oil filler pipe is in the way. I removed the two rear hold down screws of the filler pipe, but the front fastener traps it, and I can't get to it to remove the pipe. So I guess after I buy new wires, I will dismount the coil assembly and lift it high enough to get that tower connection clear of the pipe. Wifey wants me to replace the plugs even though they have only 17 K miles on them. I'll try to find platinum (again). When I removed the old plug wire connections on the #1 rotor housing, both leading and trailing plug wires were "glued" to the plug ceramic. This stuff is leakage from the knock sensor, and it really sticks!

It seems to me the most likely possibility is some kind of contamination in one or more of the injectors holding the pintle open, but I was as careful as I know how to be in handling them. They tested fine on the bench, i.e. open/close voltage was around 3.4 volts on each, and I could feel the pintles pull back when opening, and "seal up" when closing.

Anything else to look at?

Thanks again, Dale!

DaleClark

iTrader: (55)

Bill -

When you pressure tested the fuel system, did you have the fuel rails bolted to the engine? I just want to make SURE fuel isn't dripping around the injector body itself. My buddy went through a few O-rings on his 240sx that one time - there may have been a small sharp spot that nicked the O-ring or something like that.

New plugs is always a good idea. I wouldn't kill myself to get the platinums - they just have SLIGHTLY longer life than the regular coppers, that's about it. A fresh set of plugs is always a smart idea.

You should be able to make a tee for your fuel pressure tester from the hardware store. You'll need either 5/16" or 3/8" brass nipple fittings (might want to take a spare piece of fuel line for a fit test at the store) and you can get 1/4" NPT brass tees. Screw the nipples into each side, screw a 1/4" NPT to 1/8" NPT adapter into the top, and attach your fuel pressure gauge. Make sure to use Teflon tape on the fittings.

Another thought - there have been problems in the past with the wiring harness going to the coil packs failing, basically the wire is run too tight and can break internally. Mazda updated the part. Anyhow, if that harness was goofed up from removing/reinstalling the coils packs, and you weren't firing on all plugs, that could be a problem as well.

Dale

wstrohm

Dale,

Yes, the pressure test was done with the primary fuel distributor and fuel rail mounted to the engine. Basically the whole stock setup. Didn't see any leaks... that was the original point of this project... to eliminate the fuel smell problem. I'm sure that if that odor shows up again (once this thing is running right), at least I will know there is no danger of a fire. I did get a private message from "tphan" of this forum, telling me he had fuel odor which turned out to be coming from the overflow tube on the charcoal canister, and he solved that by running a longer piece of tubing to the rear of the car.

I did buy platinum plugs and new wires today; $200.38 including $14.41 tax. I have the plugs; wires should be in Thursday.

I now have a brass female tee with 1/4" NPT female connectors. I also have two male hose nipples, but need to adapt my gauge to the tee. (Teflon tape is on the two hose connections.)

Good thought about the ignition wiring. I did check the resistances of the primary coils (a little less than 1 ohm on each) and the tower-to-tower resistance of the secondary winding on the L1/L2 tower; it was 12.8 K ohms. I will see if I get consistent results doing resistance tests on the wire harness connectors. Presumbly they should all look about the same, whatever the numbers are. You are right; that harness is quite tight, although nothing as bad as the TPS connector, whose wires fold back along its body, and right over the locking tab!

I have been ignoring the knock sensor altogether; not a possible cause of the rough running problem, but I tested it and it does not work per the manual; no voltage at all when engine is tapped with ignition on. Not surprising, since it has leaked out all that "tar." Do you think I should make an effort to replace it? Car is stock as far as boost, and we run CA 91 octane. I have never heard pinging, but don't know what pinging in a rotary sounds like.

Also, do you have any comment on my question about the logic behind the PRC solenoid's purpose (above)?

Mahjik

iTrader: (1)

If the tar has been leaked, the sensor should be replaced. Not replacing it can throw the car into limp mode (if you are running a stock or rechipped ECU).

Its to help with vapor lock on hot starts.

gracer7-rx7

iTrader: (16)

Hmmm. Friend of mine gcthree had a similar issue as yourself after doing a bunch of work in the rats nest. Might want to search for his posts to see if it helps. I think he had posted in damian's TPS adjustment thread.

wstrohm

mahjik,

I thought one of the purposes of a continuously flowing fuel system was to prevent vapor lock, by keeping the fuel from stopping and heating up in the engine space. Seems to me that the design should take the opposite tack and go to a minimum pressure, maximum flow rate to prevent vapor lock. Is that wrong?

Thanks for the input on the knock sensor. The output voltage is always at the "not pinging" state of zero volts, so I thought there would be no effect on normal operation, and I only had to worry about the PCME failing to retard spark during pinging. Is that also wrong?

gracer7-rx7,

Thanks for the info. I will search on gcthree's posts.

Mahjik

iTrader: (1)

The solenoid is basically only used for hot starts.

As long as the ECU is getting "allowable" readings, it should be fine. I believe the ECU needs first a proper reading, second it determines if that reading is too high. If the ECU is not getting an allowable reading, it will throw the car into limp mode.

However, if the "goo" is gone, it can make the sensor too sensitive. Thus causing the car to to retard when it shouldn't.

wstrohm

mahjik,

OK, thank you, I understand.

According to his posts in this thread, "gcthree" had an identical problem to my present one, and did the identical troubleshooting. Unfortunately, probably since the thread was hijacked once again, he never returned to say how he fixed his problem. Do you (or anyone) happen to know his resolution? I will send him a PM and hopefully he will respond.

wstrohm

Yuch... knock sensor lists at $240. Is this possibly available as an aftermarket item?

Ehron

iTrader: (1)

I also have a very strong fuel smell. It doesn't seem like the car is running rich; driveability is fine, no problems starting hot or cold.

The other day when I got home my girlfriend said I smelled like gas. I don't have a cat or airpump. Could that alone be enough for such a strong smell?

I had the injectors rebuilt, and the FPR replaced in February (and I haven't driven the car since then) so it isn't the FPR.

After reading this thread it looks like I have a ton of stuff to check this weekend. Lots of good information...

Mahjik

iTrader: (1)

Nope. I can't seem to find my receipt, but I thought I remember paying about $170 from Ray for mine (I put in a new one with my project).

Ehron

iTrader: (1)

For some reason I can't seem to edit my previous post. I meant to type that I had the fuel pulsation dampener replaced, not FPR...