You know, I've been thinking about this all day while reading RX7MPGUY's thread(another mystery, lol), and it's not clear on the schematics- just exactly WHY does the boost sensor need that 12v source, and HOW does it use it? Could there maybe be a little vibrator in there to keep the diaphragm's output from sticking (like an aircraft altimeter), or is it for the positive boost pressure voltage output on the turbos? And if that's the case, why do NA's have the input?

It uses the 5vref input source for the output we see at the ECU, why does it need the 12v?

The guy with the correct answer gets to be my new hero

 

What's stranger is that the S5 MAP sensors don't have a 12V input.

 

We're doomed. Wayne is asking US questions.

 

/\ lol /\ ya we are doomed... damn wayne i remember when i was having such harsh trouble with my car in feb 2004 you helped alot and you had about the same post count as me 300 or so, now your way up there... guess you just like to help huh... btw we should meet up sometime... im just south of you not to far away in nacogdoches

christopher

 

An interesting question, that desers an answer..


So.. Im taking mine appart..

I would appear to me, the 12v goes into an small PCB that is seperate from the PCB that the actual sensor and other circuitry is on...

This small board is all surface mount stuff, so its a bit hard to see what is actually going on. But it would appear to me that there is a device on it that looks suspiciously like a photo cell, no idea why there would be one there because theres no light in there..

So it's either that, or a temperature sensor.

Yeah.. thats what it is..

So it must be like a self linearizing (is that even a word?) circuit..
ie, compensation for temperature.. for what reason? i dunno, but someone must've thought it was a good idea..

..sorry guys just talking out aloud here while im looking at it..

I'll post again later after i've had a better look, hehe.

 

It's also a mitsubishi part, with the big old diamond logo there..

hehe this is fun...

Still trying to fully understand what is happening in this circuit though...


Does anyone have a S5 sensor they want to donate to the cause?
Because that would be really interesting.

 

I've got the guru's stumped???

Lol, I was going to take my spare apart too, for a little "reverse engineering" project, if no one came up with an answer that satisfied my curiosity...

Are you sure it's not a mini-vibrator, WhiteFC? I don't know if I buy the temp sensor guess...For one thing, you said the two boards were seperate...

What about a little mini-heater? To keep the diaphragm from freezing, or to help keep the internals moisture-free (since a vacuum tends to turn any humidity into condensation drops of water...)

We're still guessing- got any pics of your little "mystery object" on that circuit board?

I figured Hailers or NZ would have nailed it by now, lol...

Edit: cwwtsu- you're north of me, maybe 50 miles by raod...The wife says she likes your car

 

How about chemistry. I'll buy the temperature sensor for only one reason:

Gas law constant.

If you know pressure, you know nothing but pressure.
If you know temp, you know nothing but temp.
If you know the amount of air, you know only the amount of air.

BUT, if you know any of the two, you can determine the third.

So by knowing the pressure and temperature, the ECU therefore can determine how much actual air is part of the suction.

For example, 1.5 bar vc at 30*c is more of theoretical vac (how much mass of air would be reqiored to refill the vac at atm) than 1.5 bar at 40*c. Its not by much.. in theory hardly any-- running w/o the temp sensor is so trivial it would probably not be noticed if the ECU determined 27*c for it.

 

Damn good guess, Kenteth, and you very well could be right, but there are a couple of holes in the theory:

1) the resolution of the sensor output is huge, on a units scale- the ECU sees less than a 4 volt range, tops, (from about .2v on decel during cruise, to 3.3v WOT on my NA). Using a guess that the 8-bit chips can't resolve anything finer that .1v increments for mapping (it would amaze me if it could), that leaves us with about 32 different possible inputs, over a pressure scale of what, maybe 25" vac to atmospheric, or, about .1v per inch of mercury pressure. Now, an inch of mercury difference in pressure is HUGE (consider the earth's atmosphere, which varies by only about 3" at sea level, from the eye of a hurricane, to a high pressure weather system). So in conclusion, I doubt the resolutions seen at the ECU would require a temp correction. Besides, there's an AFM intake temp sensor (also with poor resolution) siiting within 6" of that boost sensor...

2) If we are to believe WhiteFC (and I almost always do), the two inputs (12v and 5v) are totally seperate from each other in the sensor, and if the 12v was the source for some form of "correction", the two boards would be electrically connected, so that the output signal could be "corrected" before inputed to the ECU. It doesn't appear to be doing this.

3) I'm fairly certain this 12v is not inputed in any way to the ECU, as far as fuel scheduling is concerned. It wouldn't make sense, considering all of the scheduling inputs are between 0 and 4.8 volts or so- a 12v input wouldn't fit on the "map"...

4) Some components that I know well (aircraft altimeters, air data computers) are the closest analogies I know of to our boost sensors. They take absolute pressures using a diaphragm, convert the diaphragm movements to an electrical signal, and use this signal to display altitude (electronic instrumentation- not the "old" mechanical needle guys)...They all use small vibrators to prevent the diaphragm from "sticking" at any one point, for a smoother output signal (even the mechanical gauges use a vibrator to keep the needle from "sticking"). That's why I threw out the vibrator hypothesis earlier (just wanted to clear that up so you guys wouldn't think I'm any more insane than I really am, lol).

Although at this point, that theory is kind of "holy", also, being that the S5's don't have the 12v input...

The mystery continues

 

Ok, here's you starter kit for figuring out the mystery: https://www.rx7club.com/forum/showth...l+cut+defender

Sorry, I have no idea what/how the series five is constructed. YEARSOFDECAY makes these. He might know.

Wayne gets a kewpie doll for the vibrator idea. Sorry, it's for a IC in the url above. No idea what it's for on the original sensor. Never took on apart. Just made the fcd itself. Works. Works better than RB.

 

How many wires go to the 12v board. If its only two, I'll go with the vibrator thing. If its three or more, its sending something back to the ECU.

 

Years of Decays first s5 setup just used an external +12v supply. My guess is that the circuit just uses the +12v as it's own internal power supply, instead of drawing off the 5vref, simply to power the whole circuit. But, I dunno, depends on how it's designed, and I don't feel like tearing apart my only s5 one to figure it out.

 

Maybe my first post was a little misleading, the smaller board is connected to the other board via metal tabs soldered into the other board, but they are physically seperate.

Hmm... I don't have a digital camera, but i'll see what I can do

I could nearly bet my left ******** that it is a temp sensor though....

 

Sorry guys, I broke the stupid thing trying to get a better angle from it for a picture.. hehe all because I couldn't be bother heating up the soldering iron.
Ah well, gotta be glad my microtech doesn't need this sensor hey? :p

But, never the less you can still make out this sensor thingy..



Now on this one, you can't really see it, but there is a square pad of some type of metal, or semi conduction material. Very shiny, although barely visible in this poor quality pic. This is the sensor in question

 

that makes perfect sense when working with a constant variable voltage. you would want a deidicate power supply to it to not interfer with proper opertion of the ICs.

does anyone have the acutal schematic for it?

 

Man, I can't get a good pic with my "hi speed" dialup, lol, but is that a piezo-electric buzzer? Are there any numbers or markings on it? Sorry I made you tear up your unit, WhiteFC

Hailers, what was I supposed to glean from that link you provided? I'm only on my first beer, but I think I missed something there, lol...

Projectredsuns- If I had an internal schematic, I wouldn't have ever asked the question, would have been able to figure it out from the schizmo

 

Stress less young grasshopper, I do not need it anymore as my microtech has an internal one.......


And nahh its definatly nothing that vibrates or makes sound at all..
It's a sold plate of very shiny stuff about ~ 1.5x1.5x0.25mm

Looks to me to be some sort of metal.. but.. could really be just about any conductor/semi conductor.

From looking at all the other devices on the small PCD, the voltage goes through lots of resistors/caps and possibly a diode or two by the look of things (friggen hard to tell with that surface mount crap).

So the voltage would be somewhat less than 12v once it got there.. like, alot less..

 

i dont have my copy of haynes repair book with me but im pretty sure their are schematics in it. i dont know if the boost sensor itself is in their.

 

well it certainly would be nice if it did, would you mind having a look for us?

I'm getting lost everytime I try...

but if we could get someone with a S5 to tear theirs open, then if that still has the sensor I can see, then thats not the cause for the 12v

either that or they made a refined PCB that ran on 5v..
grr.. anything is possible with mazda...

 

whats the brown stuff next to the circle

 

i wish i could but i live in florida and we are about to get hit with a hurricane im over a friends house and my copy is back at my apt.

 

WhiteFC- is there any material in the little mystery object that could possibly be a crystal (like quartz)? It's probably going to be inside of the shiny metal thingy...Can you "break in" to it?

If it's a temp sensor, it should have some form of an element or bi-metallic "fork" in it. If it's a piezo, it'll have that crystal I was talking about.

Did you put 12v DC on the "right" pins to see if anything happens? (besides fire, lol)

I've already searched the FSM, I'll do the Haynes for you, suns...Good luck with the hurricane, man

 

everytime ive seen a quartz clock on a circuit board it is usually in a metal tube

 

If it's what I think it is, it won't be a clock, it'll be a piezo buzzer, which could be ultra- sonic depending on the material...

I'm tempted to hook up a 12v source to my spare to see (or hear) what happens, if FC doesn't get anywhere with his "dissection"...

 

projectredsuns, the 'brown stuff' is a capacticor...

Interestingly enough rated at 100v!!?
bizare..


anyway, the black square is actually a 16pin surface mount IC, with some sort of sensor on it, so, my guess still stands as a temp sensor, some all integrated surface mount type though with the sensor all in one.

Thats my guess, it's definatly not a buzzer of any sort.

And yeah sorrt Wayne, you can't 'open' it.. unfoirtunatly, it's just a (very very small..) IC.

It was also surrounded by an insulating type of plastic/rubber, you can see the circular outline around it in the pic I posted.

So all things seem to be leading to a type of temp sensor at the moment..