casio

just as i wasn't sure what to name this thread, i wasn't quite sure what to search for. i'm trying to trace and solve a couple of check engine light issues, but along with those is my issue with my check engine light. with the key in the 'on' position, sometimes the light will work and sometimes it won't. because of this, sometimes i will drive with the light on and other times not. its off about 95% of the time which makes pulling codes a pain. and after trying to fix whatever problem, it would again be a pain to wait for the light to work again to see if the problem is fixed.

my question deals with the circuit. i don't know how the check engine light functions which makes my idea harder to deal with. the light itself may just need resoldering (so all hereafter deals with this assumption). what i would like to do is find an alternative way to check my codes (possibly using the same clip underneath the hood that is normally used [or some part of its circuit]). is there an easy way to do this? i dont have a schematic in front of me, so i'm not sure what i'm dealing with.
thanks for any help; if this has been covered, feel free to point me in the right direction (link me).

igottafc

iTrader: (6)

wellll i just grounded my engine code plug to my batter, there for its telling codes all the time right when the problem happens. This is on an S5 though, pulling the codes on an s4 is a little different.

J-Rat

iTrader: (2)

Its also causing you to have your ISC jumpered all the time (IIRC)

casio

right, but thats the ground side of the circuit. i guess that would still work if i put a light in that side of the circuit. i'm still very amateur in electrical and was hoping someone had done this and could tell me what they did or anyone who is good with electrical.
i guess i could try something tomorrow. if that side of the wire (ground side) is getting power (like i said, i dont know what the CEL [aka MIL] circuit looks like), than wiring in an alternative light would be the same as wiring it in series. i'm afraid that the CEL (wherever its problem may be) is breaking up the circuit and therefore the ground side may be useless.
so i guess that would lead to the question 'if indeed wiring an alternate light on the ground side (under hood single connector) would work, would the CEL's problem (not working) affect the circuit?'

hey jrat (or anyone), whats the ISC? i went and looked in the FAQs to see if it was there; obviously it wasn't.

casio

the LED idea didn't work. anyone have a quick remedy for pulling codes if their CEL isn't working??

casio

i tried it again and got the light to come on. i went from the single female clip on the driver's side for checking codes. i attached a male end into it with the alligator clip (which would normally attach to the battery). i attached the gator clip to the LED and grounded LED to battery. the light came on and stayed on. it was a shot in the dark since, as i mentioned, i have no idea how the CEL works.

i tried checking the power from the ecu, but wasn't getting anything. the only schematic i have is for the N353; i have an N351. i searched about their similarities, but didnt find anything on wiring. the wire on the diagram (N353) shows the MIL to be 1E; Y/Gr/Gr. on my ecu, that position is Y/Bl/G. i tried anyways, but wasnt getting any power.

question stands; how can i check ecu codes if MIL (CEL) isnt working ??


i would like to add that i reset codes (removed negative battery cable) yesterday while installing my safc. i know before it spit a 9 at me (thermosensor) which i havent touched. the LED got power, but never blinked. i assume when pulling codes, if there are none, the check engine light turns off and stays off. this further makes me believe that my initial idea using the LED was useless (since instead of blinking or not turning on it lit up and stayed that way.

Kenteth

ISC and initial set coupler. Its used for diagnostic functions. You shouldn't be driving with the ISC jumpered. It does several things, but I don't think you have variable timing, or closed loop with the ISC jumpered. I'm sure you lose a few other goodies too, but I'm not sure.

casio

well i'm not doing that. of course, as mentioned, my CEL isnt working (almost never) so it wouldnt do me any good.
man, i wouldnt think it would be this hard to check codes with the CEL not working. isnt there ANY way to rig up an LED just to check codes??

Kenteth

I've seen a write up that had someone making a CEL harness that installed the engine bay and ran a couple LEDs to the cabin, and would display engine codes *realtime*. I can't find it in the search though, might check the archives.

casio

see, i dont even need it in realtime, just a quick solution to pulling codes. realtime is cool and all, but i need somethin simple for now.
i did go from CEL clip >> LED >> ground, but the light just stayed on. i sat and stared at it for a few seconds, but it just stayed lit.
yea, i started to search before i made this thread, but was like "what the hell would i search for?"

Kenteth

Sounds stupid, but the CEL may be switching ground, not switching power, so try adding constant power to the LED, and sticking the other end to the other pin on the harness.

Alternatively, we should be able to track the ECU pin number for the CEL in the fsm someplace...

casio

i have a diagram for an N353 (i have an N351; non-convertible). the MIL is 1E (top right ecu connector, 3rd to left from right side). its color shows to be Y/Gr/G. on my ECU, that wire is Y/Bl/G. i turned the key to on position, not pulling codes, so i figured there'd be a constant voltage to the MIL. i got NO voltage from that wire. i was trying with the LED because i thought it would work, then i used a voltmeter. nothin.

btw, nice penrose triangle.

casio

due to the curiosity of wayne... and a lot of my whining... rise again.

WAYNE88N/A

All right, you got one of those weird S5's, so you've confused me already. I have no idea how electrically similar the S4 and S5 CEL circuits are, but I can tell you how the S4 works, and maybe that will spark some insight into what you're trying to accomplish...

Typical of Mazda's ***-backwards circuit design, the ECU provides the ground, not the power, for this circuit. In effect, with the circuit completed (by an LED in the proper connector pins, say) there will be 0 volts at the connector when the LED is ON, and 12v when it is OFF. The ECU's circuitry handles the ground switching to produce a certain code (flashes) from a certain sensed failure, probably using simple AND & OR circuits for the switching functions.

Now, IF the S5's CEL assy is wired similar to what is accomplished by installing the LEDS in the S4, you really shouldn't have a problem. This connector you S5 guys jumper, is it to ground? If so, will keeping this ground in place result in real-time code instigations? Will the ECU still perform normally with the car running in this case? As you can tell, I can only speculate until I have an actual schematic in front of my face...

Electricians without schematics for guidance is akin to a lawyer without his law books, lol...

casio

i must say, the ecu/ground thing is new to me. which makes me ask this question: while trying to check the MIL from the ecu, i was sticking the LED into the ecu and the other side to ground. was i going ground to ground? man, i must say, this is an anomaly.

as for the S5s function, someone mentioned driving with the CEL function always on (yes, the clip under the hood goes straight to ground). it was mentioned that, though this can be done, closed loop and something else (i believe) are said to not function. otherwise, yes, normal real-time codes.

WAYNE88N/A

Yes, you were going ground-to-ground. Amazing, eh? A lot of guys get confused trying to understand the way Mazda wired this thing...

Hmm, closed loop and something else...What the hell does the O2 sensor/fuel scheduling circuit have to do with the CEL circuit???

casio

haha, buddy, i was confused when i read that as i was when you said the ecu was the ground (though i learned in electrical that electricity flows ground to power). if i find that thread, i'll let you know. false statement? its possible. if i remember right, it was a name i trusted. or maybe the words just sounded so damn confident.

so at the ECU, how would i check the MIL wire? would i have to connect the LED between the MIL circuit (socket; area; place thingy) on the ECU to power on the ecu?? i'm not running a power wire all the way to the battery unless absolutely needed.

WAYNE88N/A

Yes, the actual little electrons "flow" neg to pos, but do yourself a favor and think of it the other way around for troubleshooting a chassis grounded vehicle. Also remember the ECU is not the ground, but CONTROLS the internal transistor switches to ground the circuit to make it complete. It does this on several components: the BAC, the injectors, the emissions solenoids, the fuel scheduling sensors, etc...

As far as checking that LED, unless you want to provide your own ground (which wouldn't accomplish much), you're going to have to get the ECU to provide it for you, by simulating a defect that would trigger the light normally (disconnect the TPS, or the O2 sensor, something easy like that). Since I'm thinking along the lines of the S4, I'd put it out on the check connector, but I have a feeling it's not going to work that way for the S5's.

Still need a schematic

casio

wow, this is getting tough. i'm used to working with basic electrical which is all i know.
i've thought about going into the auto class (shop) at my school and trying to check out a book (from the lab). even if i got in and got the book, i'd still need quick access to a copier which, obviously, isnt going to be found in the auto shop. they have the series of Chiltons and, my favorite (thus far), Mitchells. the Mitchell had all the schematics and diagrams and i grew to love it.
as of now, i dont have a schematic. my CEL has been working, though. last i pulled codes, i was getting thermosensor and pressure sensor. i dont know whats up with my pressure sensor. i guess i'm gonna have to pull voltage at the ecu for it. but now with the whole "ecu is ground" thing, i'm completely lost.