FYI I installed an RX-7 FD alternator and dealt with everything once and for all. As an added bonus, its a 100A alt compared to the 80A of the stock 99-05 Miata or the 65A 90-97.

Jim

So I've got my little printed boards working, and the cars run and all that, but the alternator section doesn't seem to work at all.

Issues:
1)The battery light doesn't come on when the battery is low. In fact, it JUST starts to glow as I hit 14.0V, but dim, and gets slowly brighter.

2)Q8 gets hot - like ~60-65* C external, the ZTX690 I have there. It's a small little thing, maybe too little for the job. But that doesn't mean the light should behave differently.

3)The alternator control line never gets any real voltage on it (~0.025 volts or something....), and there's no current to ground or anywhere else useful.

4)I blew a fuse. This might not be related, and it took a long time to do it, running for 20 minutes (though only the last few at full voltage - I finally ran the car off the Mopar part).


Anyone got ANY ideas where I should start?

AbeFM : photos : Schematics and Issues schematics.

Here's another idea. Did I get my pin assignments wrong?

http://abefm.smugmug.com/photos/611410954_jgj4t-O.png

I wonder how it matches up with

http://abefm.smugmug.com/photos/611372777_chgid-O.png
R16,18 -> 3.3k
R17 -> 33k
Otherwise all values check out. And I think that's ok, you get the same ratios.

and
http://abefm.smugmug.com/photos/611372778_XxrHv-O.png

Answer: Yes. I used a generic footprint, and I should have checked it first. Turns out I had the pin assignments wrong on the TL431. I'll check the rest and edit this post to reflect it.

That seemed the only chip with any real issues - replaced them with new, rotated parts... and still nothing works.

edit: Note to self - swap anode/cathode. Also, sleep before I do anything else, the second time I tripped just walking to the garage should have been a sign. At least this is all making a self-consistent story. :-)

go figure, I put all the parts in how they are supposed to go, and it works. Neat.

Also, touching on what I saw posted earlier, as I connect the alternator (and go from 11.5 to 14.6 volts) RPM jumps about 200, every time and reversibly.

Thanks Jason!!

Is the 1k resistor you guys are using to limit the current to the 12v field coil circuit really necessary?
I ask because the adaptronic apparently now has a voltage sensing ability meaning I could potentially wire an output directly to the field coil...

Even if not, put a 1k in the wire. :-)

Jason, your PM's are full!
 

A bunch of people are using this circuit now, and it seems to work flawlessly. The complaint I keep getting is the light doesn't shut off all the day. Is there a way to make the light more digital?

Just for the sake of completeness:

I tried this and it works.

The 1k resistor protects the output of the circuit from accidental shorts by the installer.

It *may* also be necessary for the alternator - I don't rememeber, it's been so long.

Re: controlling it from the adaptronic. It is unlikely that the Adaptronic aux output will have a fast enough sampling/looping/polling to properly control the alternator. That was the problem with the Hydra which was what I originally designed this circuit for. This circuit is designed to do it at 200-300 Hz IIRC.

If you insist on trying to make the Adap do it, do a PWM 2D table with duty cycle that goes from 100% at 14.1V to 0% at 14.5V.

Can a mod edit this original post and say "a 0.1uF or 0.22 uF cap between top and middle terminals of TL431 is necessary to make the warning light act digitally"?

https://www.miataturbo.net/showpost....6&postcount=44

By your command.

http://farm4.static.flickr.com/3612/...eb9becd844.jpg

Heh, that only took like 6 months. Glad to hear it works. More and more I'm getting to hate my set up, but I don't want to rebuild my MS to work with this board. :-)

7 month necro!

Is there a specific voltage I'm looking for on the field coil control output? I'm not EE-smart enough to determine what the magic circuit puts out. If I had to guess, it'd be 12V across a 1k.

I'm looking at having the megasquirt directly control the alternator with a generic output and some hysteresis. loop speed should be sufficiently fast except in extreme cases of high rpm and lots of processing going on--where it drops down to 125Hz or something.

I don't think it's looking for a voltage but rather for a current. The closer the output gets to 14.4V, the lower the current it sends through the field wire.

well if I send 12V out through a 1k resistor... V/R = 12mA

And when the voltage reaches the setpoint + hysteresis, V becomes 0 and V/R also becomes 0.

Ok bringing this thread back from the depths. I installed the control box in the first post but its not working for me. At first I thought it was something I did wrong but after some testing today I'm pretty sure I've got it all hooked up right. The blue wire is getting 12v from the fuse box, case is grounded and the green wire is connected to the grey/red wire. Ok so I did some testing, with the stock wiring the ecu sends around 1v to the alternator through the grey/red wire and the system charges. While the car is running the green wire puts out almost exactly the same, slightly more maybe. But if I cut the grey/red wire the voltage on the green wire output drops to .58 volts. Stays at .58 when I connect it to the grey/red on the alternator and doesn't charge. If I disconnect the green and touch the grey/red wires back together the idle stumbles but recovers and returns to 1v and charging. Any ideas?

EDIT:
Oh forgot one thing, while the grey/red wire is cut, the dangling wire on the ecu side is putting out like 12v.:confused:

Matt:
Shit, I've been gone so long, I don't remember how many T's you get. I remember reading the voltage out of that thing, and yeah, it was very low. I think I posted it, but it was ~0.6-1.5V or something like that.

Luke,
Cutting the wire and seeing 12V is not odd - it's the car trying to get current to flow through something which isn't taking it.

As far as ideas.... I wonder if cutting the other wire helps? I know, since I was confused on mine, I ended up cutting lots of wires. :-) And some people have this issue, some don't. Wire "1-T" of the alternator. I know several people have had problems, perhaps it's worth doing - that and telling grey-red-striped from grey-red-dotted.

Oh hai.

So I have run into what may be an issue with the alternator control circuit: SERIOUS BELT SQUEAL.

At startup.

I suspect the circuit is saying "Scotty, Gimme that power!" since voltage drops and the circuit engages the alternator thus increasing the load seen by the belt/engine during cranking.

So I want to add a transitor inside the megasquirt to address this.

will switching the ground of the circuit through a TIP120 cause problems? Would I be better off switching out the output wire or something else?

I have a TIP120 already wired in the megasquirt that switches to ground and I was just going to use that.

Thanks nerdy bros.

i dont see any reason the circuit ground couldn't flow through the TIP120. Heck i suggested a pn2222 last night.


but im also no EE.