FHS

iTrader: (2)

I had my 4 wire O2 sensor tapped into the utility plug on the driver side but the flush headlight kit is also tapping 12v there and overloads that circuit when I have the lights on.

There is a pigtail that comes off the igniter harness with a 12v source and a ground. It's plugged into something but the signal seems pretty stable. What the heck does it do and would there be a problem tapping my 4 wire 12v from there?

Thanks!

The_Pipefather

iTrader: (5)

I think that's the purge solenoid if I'm not mistaken. Logically it should handle the current draw from the sensor heater, however, the duty cycle of the purge solenoid is controlled by the ECU so its not a good source of power if you're running the stock ECU.

Oscar

iTrader: (2)

I used the stock O2 sensor heater and ground sources, near the cas sensor

FHS

iTrader: (2)

Oscar, the single wire O2 sensor is unheated as far as I know. What color is the stock heater wire on the cas if it is present? How about the igniter 12v or anything that will provide a nice stable signal?



Thanks guys!

Oscar

iTrader: (2)

my bad, I now see you have '91

FHS

iTrader: (2)

No worries, thanks anyway!

Joe Perez

iTrader: (8)

This is a good spot for the regular 4-wire NB sensors- it's where I had mine tapped in when I had one.

On the downside, this connector is fed from the "Heater" fuse, which is not one of the Main Relay circuits, rather it comes directly from the "IGN2" position on the keyswitch, which is not hot during cranking. Un-noticable for a regular NB sensor, but it might cause problems with a wideband.

Ideally, I'd think you'd want to take your wideband power off of the white/red wire which comes from the Main Relay supplies the injectors, the ECU, the CAS, the IAC valve, etc. I spliced into the white/red wire at position 1B of the old ECU harness, and that's where I take power for both my MS and my WBO2. Fortunately, the current draw of the wideband is relatively small.

Another option would be the blue wire that comes off the Main Fuse via the combined IG1 terminal of the keyswitch and supplies the ignitor and coils, although this one is not easy to access inside the passenger compartment.

FHS

iTrader: (2)

That's what I ended up tapping into for the 4 wire. I also have my current WB and will have my PLX SM wideband and DM display pulling power from the same source. My car will run stupid lean at idle if the 4 wire 12v is tapped into any and blue wires in the engine bay with the lights, heater, and fan on, as it turns out.

My most stable electronic devices are grounded off my ECU. Is it possible to have too many devices wired into the ECU grounds?

Joe Perez

iTrader: (8)

Assuming you mean the blue wire which hits the aux power connector, that's somewhat surprising. This is exactly where I used to have my 4-wire connected, and there were no issues. That wire comes from the 30A "Heater" fuse, which actually doesn't power all that much. The headlights have their own fuse (only one for all four, shockingly). The Heater fuse, so far as I can tell, supplies only the interior blower motor and the controller for the rear defroster.

A while back, I took a current profile of the heater circuit on my 4-wire O2 sensor. Here it is:




This graph is scaled to 500ma and 4 seconds per major division. So at first startup, you can see the big >3A inrush of current as the cold filament first comes on. The current quickly tapers down as the filament comes up to temperature, and it quickly stabilizes at around 1100ma (1.1A).


I'd make sure not to ground the 500 watt subwoofer amp through there.

The only problem with the ECU grounds is that one or more of them I suspect are being used as injector and solenoid grounds, and are thus going to be much noisier than the others. I should have current profiled the grounds back when my ECU was still installed so we could answer that damn question once and for all, but that time is long past...

In a perfect world, if you need to run a new device and make absolutely certain that it's at the same ground potential as the ECU, yet not on the same wire, just run a new ground wire out through the firewall hole that the A/C pipes pass through and connected it to the ground point at the back of the head (cold side) where the rest of the critical ECU grounds are.

FHS

iTrader: (2)

Thanks Joe, that's damn helpful.

Yeah, in my OP I mention that the reason why I can't tap into the aux power connector is because I have a set of AAC flush mounted headlights that are drawing power from that plug. They have their own harness (they don't use the stock Miata pigtails) with their own fuses and when they come on, there is a noticeable voltage drop.

I'm pretty sure my engine grounds are the big problem. My car is about to turnover 300k miles and I don't know the conditions of the grounds. I had a new engine put in but I don't think the shop cleaned or replaced the ground connections. I lost my Enthusiast Manual in my latest move so I'm having a hard time finding the locations to check myself.

Joe Perez

iTrader: (8)

Ah yes. You did say that, and I missed it. That I can definitely see being a problem. The wire coming into that blue connector is pretty tiny- it's intended for plugging a timing light into, not 10A worth of headlights.

Well, they're easy enough to check. Unbolt 'em, clean 'em, smear a little silicone grease on, reinstall 'em.

There are three ground points that are of importance to the ECU. The first goes to the chassis on the driver's side of the engine compartment. It's a 5-wire plastic connector with a metal tab labeled JC-01 in the picture below.

Grounds #2 and #3 are ring terminals- one is located at the back of the intake manifold, the other is located at the back of the head very near the intake manifold. They're not hard to find. They're labeled 2 and 3 in the picture below.


If I am interpreting the schematic correctly (and I find Mazda's ground distribution schematics a bit hard to read) then both of these grounds actually pass through a junction connector (the black plastic thing) in the passenger compartment on their way to the engine, which is where they get consolidated from many wires down to two. I think they are the ones labeled JC-02 and JC-03 below:


If you really need to make sure you've got a good, serious ground for something, locate the two ring terminals at the head, and run a dedicated wire from whatever needs grounding to one of those two points.

FHS

iTrader: (2)

Thanks a ton for the diagrams and info Joe.

I did a quick once over to locate the connections as pictured. I actually found a thick braided steel wire with two ring connections that I can only assume was supposed to be a ground connection to the engine block on the drivers side near the fire wall. I'm not sure where it was supposed to go but I cleaned up the connections and bolted it to the dipstick bracket. That alone made a huge difference and practically eliminated the lean idling conditions with the lights, fan, and heater on. I'm definitely going to dig in and clean up as many connections as I can get to.

Thanks again!

Joe Perez

iTrader: (8)

That would be the "main" ground interconnect between the engine and the chassis. There's another one near the back of the car that goes from the PPF to the chassis as well.

Curious that it had such a stunning effect on your mixture issue- I can see it improving the lighting situation perhaps since that would be a primary return path for the headlights back to the alternator (remember, the battery is more or less irrelevant when the engine is running- the alternator is the primary current source) but it puzzles me that it affected your engine management that much...

Here's a picture of the two ground points I was talking about at the back of the engine. One is on the hoist bracket attached to the head, the other is on the back of the intake manifold:

FHS

iTrader: (2)

Ah, I use those physical connections for my fuel pressure gauge sender and one of my 4 wire O2 sensor grounds. I keep those nice and clean, but I didn't know the o connectors already attached were ECU grounds.

The only thing I can think of is the fact that I do have the grounds on my 4 wire connected to the engine block. If the block itself was incompletely or badly grounded, wouldn't that cause problems for something as sensitive to voltage fluctuations as an NB O2 sensor grounded to it? The voltage draw from the lights etc only caused problems (as far as I know) during idling which would be when the output of the alternator is at it's lowest.

Joe Perez

iTrader: (8)

Well, ground is a relative thing...

When the engine is running, the alternator is the primary source of power for the car. Because of this, the alternator's chassis is the definitive ground for the car. The alternator does not have a ground wire as such, it grounds through its own case directly to the engine block. So from a practical standpoint, the engine block is the "reference" ground.

That's why those wires from the ECU ground (ground points 2 and 3) go directly to the engine- it's the best, most direct ground path. The big strap on the exhaust side that you cleaned up is one of two connections (the other being on the PPF in the back) that provides the primary ground return for the whole rest of the chassis- all the things in the car that ground to chassis use these straps as their return path. (During cranking it's exactly the opposite of course, these straps provide the return path for the starter to get back to the battery.)

That's why the concept of the block itself being "incompletely or badly grounded" is slightly puzzling- the block is ground.

Still, can't argue with success.

FHS

iTrader: (2)

Well, learning as I go and I've learned a lot about electronics ever since the gremlims started messing with my car. Sometimes, though, it is nice to get stupid lucky.

Thanks again!