Almost put this in the interior section, but it's FD-specific.

I'm installing an AEM methanol kit, and as part of my fancy plan I'd like to put a blue LED into one of the eight lower cluster light spots (where the CEL, ABS, etc lights are) to use as the AEM module's status light. Clearly there are a million other places the status light can go, but I think it would be clean to have a frosted blue rectangle down there next to the other lights to indicate low fluid level or faults.

So I pulled the cluster today and was disappointed to find that every single slot is already in use by something from the factory. I figured there would be at least one blank in there. Bummer.

However, there are two slots I would be willing to replace in favor of the AI status light: ABS or airbag. I could just pull the bulb out of one of these and poke an LED in there, but that seems kinda janky. Ideally the silhouette cut-out in that slot would be removed or replaced. Has anybody done something like this before? I'm tempted to disassemble the cluster down to the point where the bulb silhouette covers are and try, but past experience has taught me that disassembling these types of things often ends poorly...

 

^Do you still have a working ABS system and airbag? If you do, it would be foolish to remove either of those lights - how would you know if you have a failure in either system?

FWIW, I'd advise finding another place for the water injection LED, and just avoid the risk of trashing the cluster.

 

i would tee it in with the Check Engine light

 

Its not on the cluster, but you could pretty easily use the exhaust overheat light in the center console that probably isn't super useful otherwise. Maybe give it a little buzzer as well.

You could also see the threads Dale has up about making the CEL work with the PFC. You could tap that same wire to toggle the CEL manually.

The airbag light is similar. It triggers when a certain circuit is cut off by the ABS module. You can see threads where people use jumper wires to disable the light completely when they remove the module. Just rig up a circuit to do the opposite. I wouldn't go this route unless your airbag system is not being used. Absolutely don't try modifying the airbag circuit with the controller still installed. I wouldn't even suggest doing it if your airbag(s) are still plugged in just in case you jump the wrong pins.

 

I thought about using the overheat exhaust light as well, but it seemed too easy to overlook. My main objective with the light is to avoid running the AI pump dry. Adding a buzzer would indeed solve that issue, albeit in a rather 1980's way.

As for the airbag, yeah, that thing scares the hell out of me. I pulled the OEM steering wheel off a couple nights ago and was scared to death while fiddling with the airbag connector to remove it. I had it positioned in such a way that it would hopefully not remove my head if it went off lol.

I really like @j9fd3s 's suggestion about using the CEL. I've seen Dale's post about using it with the PFC -- I also looked up the location & color for the wire at the back of the cluster and it's super easy to get to with the instrument hood out (it's the only orange base/black stripe wire in the several large connectors at the back of the cluster). I think what I will do is tee into the CEL wire and put it on a relay connecting it to the dimmed cigarette light, triggered by the AEM's "Boost Safe" (fault) output. That way the CEL will come on any time the fluid is low or there is something wrong with the system. I'm also setting up a second fog light switch to both disable the AI system & set the PFC to low boost mode when in the ON position. Maybe I'll do a write-up about my setup when it's done.

EDIT: I did tee into the CEL light but not relayed to a dimmed light source as I realized this was not a good idea for two reasons: 1) dimmed light sources are only active when the headlights are on and b) if the AI system triggered the relay, it would back-feed current from the dimmed source into the CEL circuit. Gremlins! I just relayed it into switched IGN power. Won't be testing it until thanksgiving time, however.

 

I have seen some OEM Tachs that actually have the Automatic Transmission position indicator markings there but not lit up (as the car is MT). I wonder if that location could be used.

 

there are studies about this, but i think humans can react to sound more quickly than light... a buzzer would work, and there are a bunch in the car

 

That's a good Idea to use the CEL.

I have a dual gauge pod that goes where the center speaker goes (I think it's an old Tripower pod) and I drilled a small hole below and between the 2 gauges in the pod and put a green LED there. It works well and is easy to see in your field of vision.

You may want to look at how your AI controller uses that light. I have the 1st gen AEM controller and it turns the light on any time the pump is running and will blink the light if the washer tank runs dry. Also I know it does blink codes if there's a problem with something.

That said, that would mean every time you're on boost your CEL will be lit up - that could be a bit much.

Dale

 

Thanks Dale.

The part about lighting every time it’s spraying definitely is a consideration. I imagine you couldn’t hook that “status LED” lead straight up to the CEL and expect it to work because it’s probably a really low current circuit coming from the AEM, but who knows. I ended up using the “Boost Safe” (ground output) on a relay to turn the CEL on (comes on any time there is a “fault”’ including low fluid). I also added a second fog switch which when left in the “off” position while IGN power is on arms the AI system, and when clicked to the “on” position uses a crossover relay to disable the system. I replaced the light inside the switch with a blue LED hooked it up to the status LED output from the AEM. This way, when there is a problem or the fluid is low the CEL comes on, you can then look down at the switch and see the blink code (1 flash/second for low fluid), and decide if it’s time to disable the system by clicking the switch to the “on” position. I also used a label maker to put an "AI OFF” label on the switch. I’m pretty happy with this setup!

 

There are ways to invert a signal so that it is always off, and blinks on for a problem. Just depends on how clever someone wants to get.

What's posted above seems like a good way to work around it though.

 

Some pics of teeing the CEL cluster wire & installing the switch LED. The LED fits pretty much perfectly after very slightly enlarging the existing hole. I chose to wire mine into the existing pigtail with a 4-pin Deutsch connector. The LED is the 5mm version from Amazon here:
I ordered another fog switch from Atkins to experiment with. When I was working with the WI one I noticed that the top "rocker" portion of the switch is actually a translucent white plastic that is painted over on the top with a stencil for the fog light image. I want to try removing the paint and re-painting with a new stencil. Maybe I can mask a small circle into the "WI OFF" switch and just paint the antenna one all black. We'll see.

 

So I ran into a snag regarding the CEL. Page Z-46 of the FSM shows the orange/black wire I spliced into (in the post above) as the CEL wire -- and it.. sort of.. is. Here is what I've observed:

- Turn ACC power on. Connect one end of a a multimeter to the orange/black wire, and the other end to ground. Meter reads 0V. CEL does not light.
- Turn IGN power on. Connect one end of a a multimeter to the orange/black wire, and the other end to ground. Meter reads 12V. CEL still does NOT light.
- Remove key. Connect orange/black wire directly to +12V. CEL lights.
- Turn ACC power on. Connect orange/black wire directly to +12V. CEL lights.
- Turn IGN power on. Connect orange/black wire directly to +12V. CEL does NOT light.

This is with a PFC, mind you.

So, with the orange/black wire left unmolested, even when it reads +12V, the CEL never lights. Ever. However, if you APPLY +12V to that line, the CEL will light if a) the key is removed or b) the key is in ACC state. BUT, the CEL turns off when the key goes to IGN.

I suppose I could pull the cluster out and solder directly onto the CEL light itself, but I worry about damaging things, and I want to understand what is going on here. Any ideas?

 

Not totally sure. You may need to use a diode in the circuit.

Dale

 

Talking out my ***, but I think we are all assuming the ECU does a simple 12v signal to trigger the CEL. That may not be correct? Someone would have to monitor that line when the ECU triggers the CEL while the car is running to verify.

I believe under normal conditions the CEL can be triggered as solid on, or blinking by the ECU. It is possible that it expects some sort of simple PWM signal.

If that's the case you may not be able to observe it with a multimeter (it may just read 12v when the ECU illuminates the bulb) and in that case someone would need an oscilloscope.

And again full disclaimer I have no idea how the ECU physically controls the CEL in a FD. I could be overthinking it and the solution could be much simpler. I am not sure if that level of detail would be included in the FSM. It would not seem like it.

 

I think the CEL output actually comes from the ELD box or something. Been a while since I looked at that wiring. I know that's what I had to tap into to do the PFC check engine light blink.

Found my old write-up on getting the CEL to work -

http://www.freelancemotorsports.com/...10/pfcwarn.pdf

Dale

 

The thought of that being some sort of data line crossed my mind as well, however you'd expect if the car was using some sort of data bus there would be less wires/connectors on the back of the cluster...

I actually have access to an oscilloscope, but my stock ECU is 200 miles away at the moment. I will have to remember to grab it over Thanksgiving. I suspect what is going on is there is some other control wire that must see +12 or a ground to allow the CEL to work. What is very curious though, is Dale seems to be tapping the same orange/black wire at the ELD box. I wonder what kind of signal the ECU is putting out at that new pin in his guide.

Maybe I will troll some Miata forums and see if any of the NA guys have figured this out.

 

Yeah he taps that line to the correct CEL pinout on the PFC. I have done Dale's modification on my FD as well.

I guess my question was more of what the PFC/ECU actually sends out on that line. I wouldn't go as far to say it is a data line in the sense of what modern cars have, but it might use different 12v pulses to indicate different CEL states or something.

You might be able to search for people who have done something with the CEL in cars from a similar production date.


Again I am just guessing based on the fact you stated that providing a steady 12v signal only operated the light before the engine was running.

It could also just be timing thing. ie maybe the 12v needs to stop for some amount of time before coming back on when the car starts.

 

So, I know that rusEFI has support for the NA Miata (and I think maybe the RX8?). They are open source so I went and took a peak at what looks to be a code snippet that controls the CEL.

I don't see anything fancy they are doing with it. I do notice that if I understand the logic correctly, they might be grounding the ECU pin to get the light on, but I could be mistaken.

If you are putting +12v on the wire tap, you might try grounding it instead.

I also found this in reference to the NB miata which seems to support that idea: https://rusefi.com/forum/viewtopic.p...c688a9d#p34351

 

Hmm. I'll give that a try. Remember, though, that when IGN is on that wire is driven +12.

 

I also cross-posted this to the miata.net forums, but it's pending moderator approval since it's my first post

 

Shouldn't hurt to try. I think a lot of other things the ECU controls are also ground driven if I remember right.

As for why it responds to you connecting +12v before ignition I couldn't say. Sometimes the electronics in this car can do odd things. If you read that thread I posted, you'll see one guy was hitting his light with +5v and +12v and getting some (dim) activation as well, but using the ground seemed to light it up correctly.

 

Yeah, I'm not going to connect a hot +12 to ground.

One of the dudes on the miata forum suggested simulating a fault elsewhere in the vehicle to get the CEL to come on. I like that idea.

With Dale's overheat exhaust/knock mod, does that light the CEL for other problems, or does the PFC just totally not support that?

Or perhaps there is something that can be jumped at the diagnostic plug to get the CEL to come on while driving.

 

Keep in mind the ECU activates the AC and I believe all of the solenoids by pulling their pins to ground, not by pumping out +12v.

If you de-pinned one of those connections and poked it with a multi meter that is connected to the chassis you'd probably read 12v, but the circuit wouldn't complete until the ECU pulled the pin to ground.

The code for that open source ECU project sets their CEL pinout to 0, which is ground, to activate the CEL, and in that thread the NB Miata required pulling the CEL pin to ground to illuminate it properly (worth noting that when he pumped +5v or +12v into it instead as you are trying, that he did get partial illumination sometimes)

Also when you activate the CEL code output on the stock ECU, you do it by grounding one of the circuits via the diagnostic port.

Anyways, that is why I suggested thinking about it. I'm not trying to trick you into sticking a fork between your battery terminals, and there is some logic behind the suggestion.

If I didn't have to pull my PFC to easily get to that wire, I'd try it on mine and let you know what happens. I can understand if you don't want to consider it though. My advice is typically worth exactly what I charge for it.

 

and I missed your other question.

The PFC with dales mod will illuminate as the stock one does on acc power for at least a little bit.

With the engine running, I believe there are some knock and other tolerances that will cause the PFC to blink the CEL. If you turned them to something stupid low I am sure you could get it to go off in normal operation. I believe Dale actually did this when he was originally testing his solution. I do vaguely recall that detail in one of his threads about it.

Diagnostic module can output CEL codes as I mentioned in my first response, but that is with the OEM ECU. I have never read that it will work that way with the PFC though since I don't think it stores fault codes.

 

Gotcha -- I was trying to understand if the PFC would monitor for the same faults the stock ECU does and light the CEL for them (like a bad O2 sensor heater, for example). That way you could just fake some sensor fault somewhere to get the CEL to turn on. Dale's PFC CEL light mod is actually something I've been wanting to do for a while just for the knock warning. I actually just bought a lopped off piece of Miata loom connector on ebay today for the extra pin

I think you are probably right about the ground, though -- I just instinctively try to avoid touching hot wires to cold But if that wire is connected to a filament, which is connected to hot on the other side, it all makes perfect sense. I will run outside and try it shortly (with a 1A fuse )

Unrelated, but I'm curious about this open source ECU. I've been wondering if you could script a Haltech for example, but never looked into it. I've wanted to do things like trigger an output to a relay if the intake air temp crossed a certain threshold for example. How mature is that project?

Edit: a few of the Miata dudes agree a ground is probably what it wants, too. Now if the moderator would just approve my second post because they keep asking me what I'm trying to do anyway (they think I'm trying to cheat the emissions CEL check lol): https://forum.miata.net/vb/showthread.php?t=728217