Richter12x2

iTrader: (1)

It sounds like you're misunderstanding what I'm saying. There are three plugs that go into the ECU and 2 more plugs that go to other boxes within the car. There are a great many wires that go from one plug on the ECU to the other, and to the other plugs that go to other processors WITHIN the car. The only thing the AFM needs supplied to it is voltage, that's correct - what I'm saying is that, without looking at the diagram, it's possible that the input from the AFM is converted on a sub-level board within the ECU from a sawtooth or square wave signal into a constant voltage that's sent into the main program segment of the board. I understand if you don't believe ME, but some pretty smart people have already confirmed that the output of the AFM is pulsed, not constant DC, and that the information is listed in Mazda's service manual.

This was relevant because it was showing how throttle position information is useful to the tuning process, where AFM information isn't. Or rather, useful to the tuner/driver during the tuning process.

RotaMan99

I haven't seen it.

BTW, you must not be understand what im say either, the OUTPUT of the AFM which goes to the ECU which is what the ECU uses to control fuel amount is linear. Disconnect the wire from the ECU which comes FROM the AFM and hook it up to a multimeter and you will see the voltage is linear, by no way is the ECU messing with the signal now nor has it before.

Better yet, check the output voltage AT the AFM and watch how linear it is and how its NOT pulsed. Also you can measure the resistance and watch how when the flapper is at ONE position, there is only ONE resistance value given which will give you ONE voltage valve. Gez.

There is no "other" processors in the car, there is a CPU which is basically a relay box, PS control box, Alarm system cpu and possibly a few others for some other features such as cruise control or AAS. The ECU is the only electronic computer that deals with the ONE wire from the AFM.

Please look in the FSM for the wiring diagrams and training manual to lean how the AFM works and how the ECU uses it if you don't believe me, follow the wires in the harness too and also use a multimeter and probe the wire.

Ask me how I was able to get the engine to idle and to rev up to about 2800rpm without the AFM using the stock ECU and SAFC.

Your right, but we are not talking about tuning with the AFM are we? The SAFC, whether or not its usful to the driver or tuner, DOES show the AFM voltage which go down is more air flows through the AFM.

RotaMan99

Here is the training manual for the EFI system
http://wombat.sevarg.net/RX7/RX-7_Tr...%20Systems.pdf

Go to page 23 and it tells you how it works. In the training manual it says the voltage goes up, but I see the voltage going down which makes more sense since the testing I did, the lower the voltage at the plug with my own variable resistor, the more fuel was injected.

Here is the FSM for a S4 engine. Tells you how to test the AFM. We wouldn't be able to do this with a meter if it acted that way you are saying. We would have to test it the way you mentioned.
http://www.teamfc3s.org/main/factory...SYSTEMS_NA.pdf

Page 55

I have yet to see in the FSM or Training manual where it says the signal is pulsed.

RotaMan99

Also page 31 in the FSM in the link above is the start of the ECU pin description table and 33 is the last page of that table with the locations of the wires in the connector ON the ECU.

If you click here
http://www.teamfc3s.org/main/factory...NG_DIAGRAM.pdf

And go to the fuel and emmisions wiring diagrams, scroll for either turbo or non turbo and you will see where the wire goes from the AFM directly to the input on the ECU on pin 2E for S4 turbo and non turbo ECUs.

NZConvertible

Where?

Sorry, but there's some fundamental misunderstanding here. You're talking about waveforms, which have nothing at all to do with the AFM's signal. At a constant AFM flap position, the signal voltage is also constant. It's not pulsed or oscillating. You can put a DMM on the AFM output wire (either at the AFM or ECU) and see this clearly. The input voltage is constant (5V) and the output voltage is roughly proportional to flap position, highest when the flap is closed and lowest when it's fully open.

What you're described is a stepped output, i.e. one that changes in discrete steps rather than perfectly smoothly. The output voltage still drops as the flap is opened, and rises as it opens. It never goes up and down when the flap is moving in one direction. This is completely different to a sine wave, sawtooth wave, or square wave. You certainly don't need an oscilloscope to check an AFM.

As a sensor check it'll let you if the sensor is working properly, which helps when troubleshooting. Not silly at all, very useful in fact.

There is no wave. A DC voltmeter works fine for check the AFM signal as a said above.

You can display "Airflow" as a percentage or raw AFM voltage.

The number in the Sensor Check display are not scaled, they are the raw sensor output voltages.

If you have a bad AFM or TPS (or bad wiring to either) that causes poor running, and the Sensor Check display shows you something is wrong, is that not of use to a driver or tuner?

There is no pulsating voltage from the AFM. You can measure this directly at the AFM.

Richter12x2

iTrader: (1)

I apologize, I guess I'm mistaken - are their multiple inputs of different voltages to the AFM? Because I know it reads the same resistance at several positions along the vane's travel. And while yes, checking the voltage of the AFM (which is accomplished through a subscreen of the SAFC during configuration, not through the normal monitoring window, which is what I was saying) would be useful IF there were some doubt as to whether it was operational or not, but, as I said before, seeing the input voltage from the AFM as you operate the vehicle is not useful to the tuning process, because it doesn't correlate to anything useful to you as the driver.

NZConvertible

No, there's only one constant 5V input voltage from the ECU. There's one output voltage varying with flap position and one output voltage varying with air temp.

I assume what you mean is that the resistance changes in discrete steps as you move the flap. This is just because of the way the AFM is constructed and can be seen on any variable resistor. The wiper attached to the flap is sliding across a resistive coil, and at each turn in the coil there's a step in resistance. The resistance should still only increase as the flap opens (causing the output voltage to drop) and decrease as the flap closes (causing the output voltage to rise). If the resistance moves up and down while the flap is moving in one direction then the AFM is fucked...

Nobody ever said anything about using the raw AFM voltage during the actual tuning process, that was just your misinterpretation. It was only mentioned as a way of seeing that the voltage is not an oscillating signal.

RotaMan99

I rechecked my AFM voltage a few days ago and it measures down to 1000th of a volt. .001 volt.

Don't know how much linear you can get.


I mentioned the "sensor check" screen many times and you still doubted me.

So it IS usfull information.

If you know what voltage the AFM should be at, at a certain RPM , at a certain throttle range, and its not getting to that voltage, then yes it can be very usfull. At WOT on my S4 N/A. The voltage drops a tenth of a volt for every 1000rpm. I will have to check my notes again to see exactly what voltage it is. Would take quite the memory to remember but it could be usfull if you were some how able to put it to good use. Mainly its good for troubleshooting TO the DRIVER and TUNER. No one ever said anything about tuning with the AFM voltage.

NZConvertible

The AFM doesn't "measure" voltage, your voltmeter does, and it can display to three decimal places (0.001V). The resolution of your voltmeter has nothing to do with the AFM's linearity...

RotaMan99

This was from the SAFC but I would still get the same reading on my MM anyways.

I know the AFM doesn't measure voltage, how could it? I am talking about how the resistance change in the AFM is small enough to allow it to change the voltage by a 1000th of a volt.

I would not see a 1000th of a volt increase or decrease if the resistance of the variable resistor was not capable of it. If the resistance was making large enough steps, then I would see large jumps in voltage. Large i mean by 10ths of a volt and not down to 1000ths of a volt

Richter12x2

iTrader: (1)

I know how a potentiometer works, what I'm saying is that if you open up your AFM, you'll see on the wiper there are like 5 seperate fields that vary (iirc) from 500 ohms to 1.5 ohms from one end to the other. With the ohm meter on the same pins of the AFM, it reads like 500 ohms at the low end of each section and 1.5k ohms at the high end of each section, repeatably. I've already admitted I was apparently wrong, I'm just trying to reconcile what I have witnessed with my own eyes with an apparently linear dc voltage output, since voltage varies proportionally to resistance of the circuit.

Richter12x2

iTrader: (1)

How many bad ones have you had? If the answer is zero, then by definition it's NOT useful information. I haven't seen "AFM volt meter" appear on any aftermarket gauge pods, either.

RotaMan99

It is rare they go bad. I have only heard of a few, I actually have a bad one in my shop. Once my engine stalled out and wouldn't even start, I went into the SAFC only to see the voltage on the AFM was WAAYY off. I checked the afm only to find that some gunk was holding it open causing the engine to run very rich.

I don't know why you insist on arguing weather its usfull or not. Having a "AFM VOLT METER" in your gauge pod is rediculous and I know you are being sarcastic. Again, has anyone but your self ever said it was usfull for tuning?

If you think that there are ZERO bad AFMs then you are by definition, wrong.

The information is provided by APEXi in the SAFC, whether its usfull to you or not, its there and CAN help diagnose issues. Now go in the FSM and buy a spare AFM, play around with it and stop bitching.

Richter12x2

iTrader: (1)

I may be mistaken, but that seems to be what you're indicating here

Which was in response to me saying that MONITORING the AFM is NOT USEFUL to a DRIVER or a TUNER.

What I'm saying is that they don't go bad often enough that you need a gauge connected to it all the time. There are better devices for measuring it, like a multimeter/voltmeter.

I've already admitted that I was wrong about the linear output of the AFM, now it just seems like you're arguing just for the sake of doing it.

RotaMan99

Did I or anyone else besides yourself, ever say you should have a gauge connected to the AFM? I don't think so. You are the one implying this.

Answer me this, using a SAFC which is for the main purpose, to adjust fuel, which also happends to have a sensor chk feature built in so you can easily, without digging up wires, or getting down to the ECU, check the voltage. You say its easier to use a multimeter and dig through the wires to meter one wire out?

Your the one that brought up the "AFM volt meter".

Again, whether its usfull to YOU or not, its a feature that IS BUILT IN to the SAFC. I already told you how its usefull to me and can be to many others. If its not usfull to you, fine, leave it at that, stop trying to say that no one else will ever use it.

I supose you could also argue about the voltage for the TPS or knock sensor, or 02 sensor if you have it hooked up or the boost sensor if you have the safc hooked up to that instead of the TPS. Volts for all of these items are usless to you im sure. Thats only because you don't know how to utilize the readings.

Im sure you are one of those that think a narrow band o2 sensor is crap right?

NZConvertible

Yes you most certainly would. Just because a voltage display shows thousandths of a volt doesn't mean the device being measured is that precise.

It's not hard to see how it's not possible for the AFM to make discrete 1mV changes. There's ~3.5V difference between fully open and fully closed, which means 3500 steps of 1mV. To do that you'd need at least that many turns on the coil of the potentiometer. I've never inspected it that closely myself but I'd be surprised if the coil actually has more than about 100 turns, in which case the smallest voltage step the AFM can make would be ~35mV.

RotaMan99

really? I always thought that if my meter shows a .001 VDC increase or decrease, then the VR in the AFM must be fairly precise. Im not talking about the actually see a number in the .00# spot. Just the fact that the voltage can increase by .001vdc. I guess my thinking is slightly off. Although reading below, I can understand how it wouldn't be precise compairing voltage to opening.

Understandable.