SpdySylvrEX

Well, I am really at a loss here fellas. I have had nothing but problems since the turbo install and I can't seem to catch a break. I don't have any idea what to do next and I am SICK AND TIRED of dumping money into this thing. I think I need a bigger fuel pump but I am not totally sure. It has been sitting for about 6 weeks (started it a few times) and just last night it would not start.

I had what I thought was a relay problem and I had a thread on here where ezone and I went back and forth on what it could be but it is really hard to explain by typing words what exactly is happening.

It's not the battery, starter or anything in that area. I tried to jump the car and it sounds like a really weak fuel pump.
Plus I there is a clicking relay under the dash that comes and goes apparently when the key is in the "II" position. I know it's the relay clicking because I unplugged it and it didn't do it.

I know I need a bigger fuel pump but is the 190LPH strong enough for a turbo set-up and 410cc injectors? According to procivic.com it's not. Here is that Link

I have a 255 HP LPH but that supposedly needs a return system (no money for that). See link above

I also know that everyone is so against the AEM F/I C but the car was tuned by a great dyno guy, there was just a vac leak on one injector that was unknown to me. Plus the O2 sensor was all wired wrong (not sure if that matters). Anyhow, I feel that with only 6psi, the AEM is just fine for my tuning needs.

All issues have been fixed (except the fuel pump) and I am still having problems. New O2 sensor, rewired, vac leak gone...

Please, I don't want to quit on this thing but I don't have any idea what else to do.

SlimRisk

Splice into the fuel line right after the pump and make sure it's putting out the correct gph. Also check fuel pressure

MindBomber

To start, the factory fuel pump shouldn't need to be upgrade and a return system isn't necessary at 6psi.

Thinking about the starting issue, will read the thread with ezone and return with my thoughts.

ezone

Oh Gawd, what have I done this time?

With a fuel pressure gauge on it, if it maintains proper pressure at full boost and max RPM, then you have sufficient fuel flow.

ezone

My search-fu is weak. Linky?

MindBomber

The site search engine is not behaving, it's on the list of things that need to be fixed, damn Russians.

ezone

I thought it was just me. I can't find a darn thing I have replied to when I need it--if it isn't in that one subscribed threads place.

SpdySylvrEX

Convo between ezone and I
So I don't need a return system with the Walbro 255? OMG, I am so confused

Isn't this the same thing? If not, what is the difference? I am pretty mechanical but I don't see "splicing" into a pressurized fuel line as reasonable...Unless I disconnect at the rail and pump fuel into a 1 gal bucket....

I am back at the relay being and issue since it doesn't sound like my pump is even priming. All I get is a "clicking" noise coming from the relay, as if it isn't catching...but they are like $50...Don't really have that...

SpdySylvrEX

Just tested the fuel system and it starts with testing that Fuel relay. Needless to say it failed. Honda apparently is the only people who can get this part, unless I get it from a junkyard. Who wants to play ? The 4-pin and 5-pins relays do not interchange, so that idea is out the window...

ezone

You can plug a 5 pin relay into a 4 wired hole, as long as there is room for the 5th pin. Or you could cut the center pin off. The relay internals are the same except for the 5th pin.

This kind, not GMs or Bosch types


No idea if it would handle the amps that a fuel pump draws though.

SpdySylvrEX

Yeah, that's the one I need. I need the 5-pin, already called Honda, I guess the internals are what matters. I suppose to test these things you have to have power. This Link says how so I am going to give it a shot. With the clicking, I am thinking it is a "switch" that is jumping back and forth. Unless for some reason it is getting power losing power, but I figured that was covered in the testing I did.

p.s. The magical testing book I have is the nice computer program one from Honda, I think it's the one the Mech's use....

ezone

I do all tests live, in circuit, and loaded.
Ohm testing out of circuit anything with contacts is worthless.

Many other manufacturers use the same relays.


BINGO!!!

http://www.autozone.com/autozone/parts/Duralast-Relay-Daytime-Running-Light/_/N-9na14?itemIdentifier=206204_0_0_&newYmme=true

HTH


Edit: Why no linky there?? C&P if necessary.

SpdySylvrEX

Alright, here is the dealy-O. After doing Numerous tests, following the ones for PGM-FI relay 1, (not PGM-FI relay 2 which is the one we have been talking about) it all ended with the test saying, "replace with known good ECU and see if problems go away". I would attach the full test, but it is very large 6 or 7 pages, if not more. Unless someone wants to see what I am actually doing.

So could I be looking at a bad ECU? If so, then this will have to be a FULL PART OUT!!

ezone

This is the relay that is buzzing, right?
Can you tell which wire is the one with the pulsating (voltage or ground) CAUSING the buzz? (It will be on one of the 2 silver colored terminals in the pic above.) What wire colors?


Looking up a diagram, brb.

sdaidoji

iTrader: (1)

i guess i cannot be of much help (never had to do that deep a electric check), but good luck, spdy...

ezone

Main #1, coil side:

wht/blk comes from the fuse (6?) Hot at all times....,

red/yel comes from the PCM, (relay control) should be a ground when turned on........

Relay supplies power to the injectors and other stuff....


Before you condemn the computer:

Ground for the relay comes through the PCM. The PCM has to use one of its chassis grounds to operate this circuit.
You need to load test (PCM out of circuit) or volt drop test (live testing) all of the external PCM grounds, and you probably should do it with the key on, everything powered up, and hopefully with the problem happening.

At the PCM, connector A, 31 pin, gray, on the engine harness:

A23, A24, brn/yel
A4, A5 blk
Measure with a voltmeter, each of these should be under 1v at all times under all conditions!

All end up at Ground G101, The 10mm bolt right next to the upper radiator hose. Edit: on the engine.

You don't have a simple loose ground bolt, do you? Or a ground cable to the engine bad?

SpdySylvrEX

Passed

Passed

How is this done?

With Ignition OFF, connector unplugged, PASSED. (all good grounds)

Then for whatever reason, I turned the Ignition on and tested again. OHMS readings were high. (OHMS>1v) The display on my voltmeter is sorta busted so I couldn't tell exact number. I will try to get a better voltmeter and try again later.

The battery voltage is dropping and I am going to try to get it load tested when I get home from work.

ezone

NOPE.


"Believe it or not, it's possible to check a power supply circuit for voltage and show a good reading of 12 volts, but still have a faulty circuit. You could check that same line for resistance and show an acceptable reading of .2 ohms, and again still have a faulty circuit."

STOP.
HOOK EVERYTHING UP (the computer, as if the car will run)
DO YOUR TESTING ON LIVE CIRCUITS
OHM TESTS ARE WORTHLESS (for this).

Connect your meter to battery ground post (or the frame of the dash is usually ok too).
Turn the key ON so everything is powered up like normal.
Probe each ground circuit I mentioned earlier with the other meter lead.
Measure VOLTS. Not ohms.

I did NOT say ohms here. I said "under 1v". That's VOLT. Measure VOLTS.
"All conditions" means to make that sucker do some work while you measure! Crank the engine if you have to!
Measure while everything is powered up and trying to operate. (Hopefully with the actual problem happening!)
Ideally, there would be zero volts measured on each ground wire at the PCM.

-------------------------------------

You are supposed to be doing a "voltage drop test". (Google it, here's one: http://www.engine-light-help.com/voltage-drop.html )
Measuring for a difference in ground potential while loaded.
The battery ground and the PCM grounds are all supposed to be grounds connected by ground straps and the engine block, right?
There should be zero difference in voltage (when measured under a normal load) since they are all grounds connected together, right?
If there is any voltage showing between any two ground points while under a load, there is a weak point between the battery ground and the point you are measuring at.

An ohm test will not work for finding these types of things, the meter alone cannot load the circuit enough.
High resistance rears its ugly head under a load to cause problems. It won't always show up when only using a wimpy voltmeter with a 9v battery.

-------------------------------------

Another way to check it with the PCM disconnected would be to see if each ground can support the operation of a 55w Halogen headlight bulb.
A 55w headlight will need about 4.6 initial amps to shine brightly.
(This might not prove anything unless you can load all 4 of the ground wires at the same time with a separate 55watt bulb for each one, since they all connect to the same place.)

See, all these ground wires need to be able to support a few AMPS of current flow.
You could measure voltage just fine on a single strand of wire (the thickness of a human hair) if it is still making a decent connection, but that single strand of thin wire wouldn't support much of a load once you turn something on. (Does this make any sense yet?) A corroded or loose connection can have the same effect.
A voltmeter can't put that kind of load on any circuit by itself. But it can measure it while the circuit is operating.

------------------------------------

Have you checked that the ground wire bolt on the engine is tight, like I said in the other post?

------------------------------------

Sorry if this sounds snippy, man. I've been an electrical whiz for many years (at my first wrenching job the boss called me "fusebox"), but I'm a lousy teacher. I can see it in my head far easier than I can explain how to do it and why.

HTH anyway.

SlimRisk

Yeah one time I tried to start my car without a ground wire and it would crank but not start. Hooked it up and it started fine

SpdySylvrEX

Sorry if I misinterpreted or used the wrong lingo. The Multi-Meter I had only has an on/off switch. It reads volts/OHMS automatically. I guess it read as continuity on those 4 wires. I will retest with a different Meter.

I have a basic and I mean basic military teaching on electrical circuits. The Coast Guard taught me a little about 8 years ago and I have been teaching myself ever since. I know the difference between OHMS and Volts...

Well with all this said, I went on a whim and decided to have the battery load tested and charged since it was going low from all this testing. Load test passed and I put the battery back in and the car started right up. It ran EXTREMELY ROUGH, surging idle, and threw a CEL but it ran. With that said, I do not think I have fixed the problem, whatever it may be. Reason is, I tried to jump it the other day, with no joy....

These problems have been hit or miss...I swear, it's there one minute, gone the next....

I checked the tightness of that cable and it seemed fine. I will clean it up, and I have some extra 3/0 cable and a few leads i can make a nice new one...
Nah man you're good...You are helping me, that is enough.

Also, another thing I noticed is when the car was first started (tonight, after the battery was charged) my AFR Gauge was right in the 12:1 range...I let it run for a bit and after a few minutes the CEL came on and it started surging again. I am not sure when this happened (warm or not) because I walked away from the car. I do know it took a few minutes after the initial start. I will reset the ECU and try again.

I think this covers everything...

ezone

12:1 is normal for a cold engine, and it can get a bit lower than that depending on the temps.
12:1 is somewhere close to a mixture you would use when tuning for max power only, without any regard for efficiency or emissions.

What mods does this have on it? I thought I saw turbo somewhere. Are you using the stock PCM? A piggyback controller? What? What?

That's scary to me. This isn't a time to have to guess.

Ok. Your meter must be damn confusing then. I would shoot it, then go spend a 20 at Sears. Surely they have a cheapo meter that would be better than that.

Ok, it's tight. Great.
Sooooo....what if one of the many ground wires is broken, down inside the sheathing? How are you gonna find out? Volt drop test is how.

I'm trying. I can't do the work though, and all I have to go on is your info to work with. I'm also probably firing far too many ideas in a single post.

Shoulda started with verifying that the PCM is the cause of the relay buzzing though, because I just guessed. It could be the power side causing the problem instead. Test test test. Test, don't guess.

Here's another thought (groan)....
Would it run right if you ran a separate ground to the buzzing relay (on a toggle switch) to take the place of the PCM's control input?

SpdySylvrEX

Yes, turbod with the AEM F/I C as my piggyback. I know, I know, they SUCK!! The full kit is on procivic.com...
Bigger rad fans (1100 CFM if I can recall correctly)
2 core radiator..

I just replaced the whole thing with a big 3/0 gauge wire.

Nope your good man, I would rather have more tests than have to wait a while after doing one....


hmmm....That is a good thought, but how do you think I should go about doing it?

I don't know if you caught it but I did say the car started but ran extremely rough and threw a CEL right? I don't own a scanner, but my neighbor does...also, the RPMS were surging from all different points, never reached above 4000 though....it was just really erratic. Although it has started, I don't think the problem is fixed.

So what's next?

ezone

I meant the small ground wires for the PCM that bolt to the stat housing.
Just to test with, I always have an assortment of jumper leads with alligator clips. Right now that's all you would need, temporary for diagnosis.

I don't usually go looking for a silver bullet fix first.
I try to figure out what exactly is going on first, then work through to find the root cause.
Example: If the hot side of the injectors was being turned off and on, I go back to the relay that controls that power, and check the control side of the relay (not bloody likely that the relay can cause it) for which side is the cause of the relay turning on and off. Then work up the system from there.
It's a method. Methods vary depending on the situation. I have to be able to adapt on the fly.

I deal with OEM, no aftermarket.
I would eliminate all the aftermarket stuff first. Reconnect the factory stuff as designed. (I mean the piggyback)
IDK if you have big injectors or anything, so fuel control might be out the window for the moment if you do.
All you need right now is to see if eliminating that aftermarket stuff will let it run normally.


The OEM PCM is pretty darned reliable. That's why I think it is something else.

SpdySylvrEX

Injectors are 410cc...
ok, I thought you meant the ground wire that goes from the motor to the radiator support bracket, my bad...

ok, I did the test again on the four wires you said; a4, a5, a23 and a24. all read .3 volts with ignition on.

Is this a voltage drop test:
I had everything hooked up and touched each individual lead and got .3 volts from each one with the ignition on. Now, do I need to do this same thing with the ignition off? then car running? (since it can) So I have a comparison.

Do i need to do the same test ^^on those ground wires that go to the stat housing? or was that done when I did a4, a5, a23, a24?

I am only sort-of comprehending what I am doing and what you are telling me. I am sorry. I really appreciate your help.

ezone

Should I guess those injectors are bigger than stock? Yeah. Duh. But the numbers are somewhat meaningless to me without reference material.

I don't get paid to dink around and make unnecessary wild stabs. I get paid for results. I document and justify the work I do (most of the time).
/rant

Sounds like you got it, provided the meter was connected to a good ground the entire time.
Where was the other meter lead at?


Remember, the spec said "under ALL conditions"
Only checking with the key on is not ALL conditions, is it?

Was the buzzing problem happening when you did the check? No.
I want to know the results of the test while the problem is occurring. If it isn't acting up, the problem isn't there, and your test won't show a damned thing wrong.

*facepalm*
That is what you are already doing.

Here is a problem I want to see checked and results posted, because I haven't seen that you did this yet:
You said this all started with a main relay clicking and buzzing, right?
What is causing the relay to buzz?
TEST each of the 4 relay wires WHILE IT IS ACTING UP, because I only took a guess at the PCM as the cause. Look back to where I told you which 2 pins I thought you would get a result on. https://www.civicforums.com/forums/1...ml#post4584779

YOU have to check, I can't do it from here....
If something besides the PCM is causing the relay to go nuts, then I have been sending you down the wrong track.
Get it?


Also, testing for something broken is worthless while everything works perfectly. It must be broken.

Sorry, I can't give you a crash course in electrical theory, and I can't easily explain how I think.


Break is over, gotta go.

SpdySylvrEX

What kind of references are you looking for? They are out of the Acura MDX (the turbod SUV)

Grounded to the vehicle on some metal inside the dash. I checked for voltage on a known power source.

I guess I have to wait for it to act up again.

ezone

I wasn't really looking for references, I meant that injector flow numbers are not something I ever have to deal with, and that's info is not something I would find in a normal shop manual.
Remember, I deal with bone stock cars. It's rare that someone would bring something like this into a dealer.

I was figuring a turbo'd car would have bigger injectors or at least some super high fuel pressure, and operating fat injectors with the stock unmodified PCM means the engine will run very rich. Probably set a P0172 within a few minutes of the O2 sensor coming on line if it doesn't flood itself out first.

I think the stock PCM is quite reliable, so one step I may take is to eliminate all the questionable aftermarket electrical/electronic stuff. Then resume testing, if necessary.


Real life example of eliminating non-factory modifications as a cause of a problem:
If you ever read about the light duty diesel stuff, some of the chipped PCMs will set P06xx codes for processor faults.
What do you do if you are a tech, you have never seen the truck before, nobody told you that it is modified, and you find a code like this?
You could throw parts at it all day long and never fix it.
Return the chipped PCM to stock configuration, the fault codes go away.
/CSB, really doesn't have much to do with your problem....

Good enough (usually).
Yup. That's your best bet. I'd leave the test equipment hooked up and ready for when the problem happens.

HTH

SpdySylvrEX

Thanks again man, sorry if I seem like a moron. I am just sick and tired of chasing problem after problem. I wonder if whatever "was" happening has anything to do with the erratic idle I have now. I am just so afraid to drive it to advanced with the way the RPM's were all over the place.

sdaidoji

iTrader: (1)

nope, keep going, buddy. yer keeping up with ezone (which is rare 'round here )
g'luck again

ezone

[/quote]nope, keep going, buddy. yer keeping up with ezone (which is rare 'round here )
[/quote]
He's doing better than most. I know I lose a lot of people right off the bat, deer in the headlights looks and all.

Hey, I'm happy he has a voltmeter and can use it. I'm trying to work with and expand on that.
Most people that have a voltmeter can't even check a flashlight battery with it.


I wonder what happened to "all the idiot lights flashing like a Christmas tree in Times Square" that made me think ignition switch??

Or was that a different car and thread?