SuperMID installation - help needed!

[GasSavers_Yoshi]

Hi Matt,

Would you please try following?
- connect the INJ wire to your speed sensor
- connect the SPEED wire to your injector

Yes, it is reversed.
My bet is that you'll be able to see injector pulse and some fuel usage as you drive (pulse from speed sensor) and able to see some km number as your engine runs.

Regards,
Yoshi

[Gary Palmer]

Matt: I may have missed something, but it sounds to me like you may have wired something funny, for your system. I believe that the injector's run off of
+12, which is switched to the all of the injector's, when the car is started. The individual signal's to each injector are switched to ground, when the ecu enables the injector. It's really a sort of inverse logic, with a low voltage at the injector signal, turning on the injector.

In your inital post, you said you hooked the MID to ground, the same as the injector's. If your using the injector common wire, then your not hooked to the chassis ground, your hooked to the voltage supplied for the injector's to run. This would give you some very squirlly results, which is what you seem to be seeing.


I believe you should be able to check to see if your getting the voltage you expect, at both your injector and your speed input, with just a voltmeter. If you set it to measure 15 volt's, DC, you should be able to get some measurement and the voltage will probably be bouncing around, since your not using the DVM in a traditional manner. However, you should not get any voltage from the speed sensor, until you start to move the vehicle.

Hope this helps.

From what you've described, you should only have

[krousdb]

Originally Posted by Yoshi

Hi Matt,

Would you please try following?
- connect the INJ wire to your speed sensor
- connect the SPEED wire to your injector

Yes, it is reversed.
My bet is that you'll be able to see injector pulse and some fuel usage as you drive (pulse from speed sensor) and able to see some km number as your engine runs.

Regards,
Yoshi

Exactly what I was thinking!

[landspeed]

My initial post was wrong!. It is wired in correctly (I think), I will post a scan of the ECU wiring diagram, and where I have connected the SuperMID. If I had wired the power backwards the SuperMID would probably be broken!. (but it has always powered up).

My current tests so far:

(1) It runs the same if powered from the body ground, or the engine ground
(2) All the injectors work the same way (i.e. give the odd results), ruling out a faulty injector
(3) Tested the voltage of the injectors, according to the service manual diagnostic section, and they are within normal limits.
(4) Noticed that
: when injectors are on, the fuel usage increments at a certain rate (which is the same under light load, or full turbo boost) but pulse width =0
: When throttle is off, and coasting in gear at above the idle speed, then
the the fuel usage increments much more slowly, and pulse width =0.5
(which is paradoxical).


What I'm going to do.

(1) Lunchtime : Connect the speed sensor wire to the injector wire
(2) See if I can find my spare ECU and try that (to rule out faulty ECU although I'm sure that isn't the problem).
(3) Buy an oscilloscope off ebay (expensive ), and see what the difference
is in the signal between the LPG injector emulator, and the actual injector
being connected.

(4) A question for Yoshi, if it is OK:

(a)

If this is the injector voltage trace:

On Off On Off
,,,,I^I,,,,,,,,,,,,,,I^I,,,,,,,,,,,,,,,,,,,,13V
--+ I,,,,,,,,,,,,,,+,,+-----,,,,,,,,,,,,12V
,,,,,,,I,,,,,,,,,,,,,,I
,,,,,,,I,,,,,,,,,,,,,,I,,,,,,,,,,,,,,,,,,,,,,,,6v
,,,,,,,I,,,,,,,,,,,,,,I
,,,,,,,I,+------+,I,,,,,,,,,,,,,,,,,,,,,,,,0V
,,,,,,,I_I,,,,,,,,,I_I

I have drawn transient spikes in, to show what it probably actually looks
like.

Does the SuperMID measure the changeover at e.g. 6V (which would mean it
will catch all the 12V-0V transitions), or does it measure the changeover at 12V (meaning that, if the voltage only changes between 0V and 11.9V, with tiny spikes up to 12.5V, would cause the problems seen in my case.

(b)

Is it possible to have Pulse Width 0, but have the fuel usage increasing? (i.e.
does your code use two algorithms, one for pulse width, and one for fuel usage, or, does it calculate the pulse width, *then* use this width to calculate fuel usage.

(c) I am going to have to re-wire my speed sensor (since it is disconnected on my car). It is just a simple on/off switch that goes on twice per revolution.

If I connect one side of the switch to the battery + terminal, and connect the
other side straight to the SuperMID sensor, will this be OK. e.g. If the SuperMID speed sensor wire has a resistance of 0 ohms, then the SuperMID will explode, but if the speed sensor wire has a very high resistance, it will be OK. Is it safe to do this?

(A simpler way is to say, if I touch the speed sensor wire to the battery +12V
terminal, will that damage the SuperMID due to high current flowing?

Thanks for the help!

Matt

Edit : the ,,,,,s are there for spacing purposes or the board software messes up
the formatiing.

[landspeed]

OK. I have connected one injector wire to the speed sensor, and one to the injector SuperMID wire.

The distance travelled goes up, and the fuel used goes up, and it shows an average KM/L reading.

However, the speed is 0.00KM/H, and the Injector Pulse is 0.0ms.

Can anyone explain how this could be?

[landspeed]

Edit : The above is slightly incorrect (after I reset all the values it changed).

I have connected the speed sensor to one of the cars injectors
I have connected the injector sensor to another of the cars injectors.

Results:

The Instantaneous km/l always stays at 17.6. When I go downhill with
engine braking, it changes to 99.9.

The speed sensor wire is actually reading the injector pulses correctly, and
when I go 1 mile (at any speed), the display goes up 1km (as long as I am in 5th gear).

The Injector pulse width (on the top right of the display) is 0.0 all the time, except when it is 0.5,

when going downhil with engine braking.

This means that

- The speed sensor wire is able to 'read' the injector pulses perfectly
- The injector sensor wire is unable to read the injector pulses perfectly
- The injector sensor wire is detecting every firing of the injectors, but
it is unable to detect the length that the injectors are opening. This is
why, even when cruising, or accelerating hard, the instantaneous
economy is staying at 17.6km/l.
- When the car is doing 50km/h (on the SuperMID display), it says that the economy is 17.6km/l, *BUT

THE INJECTOR PULSE = 0.0* still (which makes no
sense).



The above information should be useful to someone who knows how the
SuperMID works, in order to detect what might be going wrong. Is there
any chance the fault is in the SuperMID conversion box, or the SuperMID
itself?

Attached - is a diagram of how I have wired up the SuperMID...

[landspeed]

Here is a youtube video of the above (note it goes to 0.5 when decelerating, and always shows 17.6 km/l.

http://www.youtube.com/watch?v=KNaqoXxJItg

Any help would be appreciated

Matt

[GasSavers_Yoshi]

Originally Posted by landspeed

OK. I have connected one injector wire to the speed sensor, and one to the injector SuperMID wire.

The distance travelled goes up, and the fuel used goes up, and it shows an average KM/L reading.

Thank you for your testing.
Now, I think I can solve your problems.

However, the speed is 0.00KM/H, and the Injector Pulse is 0.0ms.

Can anyone explain how this could be?

If the Injector Pulse is measured as more than 328 msec (no realistic value, something wrong), it shows 0.0ms.
The average speed KM/H calculated by traveled distance/time.
If the time is big than the distance, it shows 0.00KM/H.

Anyway...
The INJ wire checks the negative current to the ground.
The SPEED wire checks the positive current to the interface circuit.
On Prius, Civic and Metro, everything worked fine.
However, on your Nissan, everything worked in reverse, I believe.

I'll try to build another interface box to meet the behavior above and send it to you.

Regards,
Yoshi

[GasSavers_Yoshi]

Originally Posted by landspeed

Edit : The above is slightly incorrect (after I reset all the values it changed).

I have connected the speed sensor to one of the cars injectors
I have connected the injector sensor to another of the cars injectors.

I see this now.

And, thank you for your video.
I could see it was displaying 45-5x KM/H, so it measured something.

Again, would you please connect the INJ wire to your speed sensor signal and test how it works?

Yoshi

[GasSavers_Yoshi]

Originally Posted by landspeed

Does the SuperMID measure the changeover at e.g. 6V (which would mean it
will catch all the 12V-0V transitions), or does it measure the changeover at 12V (meaning that, if the voltage only changes between 0V and 11.9V, with tiny spikes up to 12.5V, would cause the problems seen in my case.

The threshold of INJ wire voltage is about 0.7V.
The SuperMID detects logical "1" above it, 6V and 12V.
It detects logical "0" below it, 0V.

Edit: But, the current direction is improtant. The interface detects the INJ signal the current flow from the interface to the ground. The SPEED signal is opposite.

Yoshi