RhinoPower Ltd

Members Login
Username 
 
Password 
    Remember Me  
Post Info TOPIC: Trying to run a spare 1995 Tracker/Suzuki ECU on my kitchen table
rab


Member

Status: Offline
Posts: 9
Date:
RE: Trying to run a spare 1995 Tracker/Suzuki ECU on my kitchen table
Permalink  
 


Back on my project now with success at last! Applied Guru's advice of connecting the CAM signal not only to pin B3 but to the ignition signal pin A8 as well. But as Guru says, the ECU wants to see a 50 volt pulse here, so my zero volt to 5V cam signal cant go directly to A8 but has to connect inside the ECU, after a voltage dropping resistor. Pin A8 goes to R206 (47K ohms) then to a diode (D110) then to R25, a 4.7K resistor. From there it goes on into the twilight zone of complexity. So I made my connection between R206 and D110.

I can now get my injector to run at 5% up to 62% duty cycle by playing around with the MAF and Throttle pots. I get 5% at 800 crank RPM and 62% at 1500 RPM. Cant go higher than 1500 due to signal generator limitation. But I'm happy.

Didnt simulate the O2 sensor but the ECU never threw a fault code. Dont have a vehicle speed sensor either but it did throw a code for that. But who cares.

Thanks everyone !

 



__________________


Guru

Status: Offline
Posts: 1194
Date:
Permalink  
 

rab wrote:

 It's a 16 valve.   PSB #E331B641B


 Yes its the same as a 59B series ECU, I have one of those. I'll have a look at it at the weekend.



__________________

1984 Suzuki SJ413K pick up, 1.6 16V Baleno engine
2000 Suzuki Vitara 1.6 8V, many mods
2004 Suzuki Ignis 1.5VVT 4Grip
2006 Suzuki Jimny 1.3VVT JLX+
and many more.

rab


Member

Status: Offline
Posts: 9
Date:
Permalink  
 

 It's a 16 valve.   PSB #E331B641B



__________________


Guru

Status: Offline
Posts: 1194
Date:
Permalink  
 

The ECU expects a signal of around 50V on the IFS pin, at the front end of the circuitry is a resistor divider to reduce that to a sensible voltage. You need to link from the CAS input to the centre of the resistor divider. Is this for a 16V or an 8V? from memory the 1995 16V is a standard 'world wide' ECU with US mapping, what is the PCB part number?



__________________

1984 Suzuki SJ413K pick up, 1.6 16V Baleno engine
2000 Suzuki Vitara 1.6 8V, many mods
2004 Suzuki Ignis 1.5VVT 4Grip
2006 Suzuki Jimny 1.3VVT JLX+
and many more.

rab


Member

Status: Offline
Posts: 9
Date:
Permalink  
 

Are you saying run a wire from the cam signal pin of the ECU over to the IFS pin? In effect, fooling the ECU into thinking the cam pulses are ign pulses?



__________________


Guru

Status: Offline
Posts: 1194
Date:
Permalink  
 

rab wrote:

  What do you mean about ECU noise? How is that problem handled inside the car? They shield the cam signal wire, so maybe I should be concerned about it too. Should I keep the coil and cam signal generator far appart? Run shielded cables? Other tricks? What can substitute for a real coil?

 


When the ECU switches the ignition the high voltage spikes (400V+) causes bursts of high current that can cause ground shifts in other circuits. This is usually dealt with by good grounding, particularly through the RF ground which is through the ECU case to the vehicle chassis - that is difficult to replicate on a normal bench. As I stated above I bypass the IFS circuitry for this sort of test.



__________________

1984 Suzuki SJ413K pick up, 1.6 16V Baleno engine
2000 Suzuki Vitara 1.6 8V, many mods
2004 Suzuki Ignis 1.5VVT 4Grip
2006 Suzuki Jimny 1.3VVT JLX+
and many more.

rab


Member

Status: Offline
Posts: 9
Date:
Permalink  
 

 You're right. As rpm goes up there's more power strokes for the same hp.  So naturally each stroke doesnt have to have as much fuel, so D.C. decreases.

  I got around 2% duty cycle at 6 Hz cam signal (which would equate to only 210 engine rpm! ). I see now that to be realistic I need to increase my cam signal by 4 times! Anyway, I remember seeing it go up to 22% at 20 Hz at one point. Guess that would be a pulse width of 1 divided by 20= 50 ms  x .22 = 11ms pulse width. Did I do that right?  But something's not working right. My generator probably did get spiked by the coil like you said. I'll employ the noise filter my next try.

  What do you mean about ECU noise? How is that problem handled inside the car? They shield the cam signal wire, so maybe I should be concerned about it too. Should I keep the coil and cam signal generator far appart? Run shielded cables? Other tricks? What can substitute for a real coil?

 



__________________


Guru

Status: Offline
Posts: 1194
Date:
Permalink  
 

VE = volumetric efficiency, for a constant air flow this will drop as the RPM increases because you have less air per stroke (Mass Air Flow is measured in grammes/second).
If you assume you have 1000 gms/s at 1000 RPM then you have 1000 grammes shared over 2000 injection events so you have 0.5 grammes of air per stroke (injection event). At 6000 rpm you have 1000/12000 grammes of air per stroke so for a constant AFR you need to inject 1/6 of the fuel so the PW should drop. What pulse widths are you seeing? there is a relatively long period between injection events at low RPM, to get a higher duty cycle you may need to increase the rpm though you can reduce coolant temperature to add enrichment.
The injectors are always grounded in a consistent manner.

The sense resistor for the MAF is 1k so 5mA will give 5V - the ECU can't read any higher than that.

The stock system has a noise suppressor between the coil and the ECU input to reduce the voltage seen at the input and to extend the pulse width, I would be wary about running without it. You may have spiked the signal generator, the ECU can generate a lot of noise on the supply which is why I won't use a coil on the bench.



__________________

1984 Suzuki SJ413K pick up, 1.6 16V Baleno engine
2000 Suzuki Vitara 1.6 8V, many mods
2004 Suzuki Ignis 1.5VVT 4Grip
2006 Suzuki Jimny 1.3VVT JLX+
and many more.

rab


Member

Status: Offline
Posts: 9
Date:
Permalink  
 

  If VE means injector voltage on the wire that connects it to the ECU, for a fixed flow of air (constant MAF value) fuel flow, and consequently duty cycle will remain constant because the horse power is staying constant. We also have to assume the injector always fires only once per cam shaft revolution. Do you agree?

  When I set the MAF to 3v and throttle to 4v I get a  D.C. of 13% which is the highest I can get right now. At those settings the MAF current measures 3.6 ma. Service manual says shouldnt be over 5 ma max.  All I want to do is pulse the injector at around 10% to 30% D.C. at a fixed frequency of around 20 Hz.

  Question: Does the ECU always ground the injectors in a consistant manner regardless of all other conditions? I mean, the transistor that grounds the injector, it always instantly turns on and instantly turns off, right?

  The ECU I'm using is a 33920-70EA.

  I took the igniter off the car and tried using it to more realistic operate the coil my signal generator quit working so I put the igniter & coil back on the car for now.

 

 



__________________


Guru

Status: Offline
Posts: 1194
Date:
Permalink  
 

Duty cycle will change with RPM because the amount of air per stroke will change, for a fixed value of MAF you should see VE drop as RPM goes up. The throttle pot will only change duty cycle on a change of throttle rather than a constant throttle, MAF can be a bit tricky with a pot because the MAF output is current and not voltage, the current is converted to a voltage within the ECU by a 1k current sense resistor, you need to ensure that you are actually getting 0 to 5V at the input to the ECU. I use a voltage calibrator for this because I found it quite difficult to accurately adjust a pot.



__________________

1984 Suzuki SJ413K pick up, 1.6 16V Baleno engine
2000 Suzuki Vitara 1.6 8V, many mods
2004 Suzuki Ignis 1.5VVT 4Grip
2006 Suzuki Jimny 1.3VVT JLX+
and many more.

rab


Member

Status: Offline
Posts: 9
Date:
Permalink  
 

  Adding a real ign coil got it going! But the MAF and throttle pots arent changing the injector duty cycle.  Duty cycle increases as I increase the cam signal frequency which doesnt seem right. The coil tickles my finger but not enough to jump a real spark. Transistor cant handle it? Maybe I need  to use the actual coil igniter. I left it on the car. But the CEL goes out as it should and no trouble codes, only blinks #12 which is good.

 



__________________
rab


Member

Status: Offline
Posts: 9
Date:
Permalink  
 

 Thanks for responding!! I have connected the IFS signal but substituted a 100 ohm resistor for the ignition coil so ofcource dont have a high kick back voltage, just 12 volts shorting to ground (via a transistor I rigged up) every time cam signal pulses.  I'll employ a real coil and let you know what happens.



__________________


Guru

Status: Offline
Posts: 1194
Date:
Permalink  
 

It sounds like you don't have a proper Ignition Fail Safe (IFS) signal - usually it taps off the negative side of the coil and takes the back-EMF pulse from when the engine fires. The pulse is a high voltage pulse and is divided down and shaped within the ECU. For a test like this I generally trace through the cam sensor signal and the IFS from the connector to a point where they are both 5V levels and then I put a wire link across from the cam sensor circuit to the IFS circuit.

__________________

1984 Suzuki SJ413K pick up, 1.6 16V Baleno engine
2000 Suzuki Vitara 1.6 8V, many mods
2004 Suzuki Ignis 1.5VVT 4Grip
2006 Suzuki Jimny 1.3VVT JLX+
and many more.

rab


Member

Status: Offline
Posts: 9
Date:
Permalink  
 

 I want to use it to fire up a fuel injector for flow tests. It clicks the injector once only and shuts it off. CEL never goes out and error code 41 flashes (no crank signal). Got ECT, AIT, throttle position potentiometer, MAF potentiometer, 13V battery power, square wave generator for Cam sensor set at 6 Hz (hot idle), ignition output pin going to a transistor switch acting as the igniter, ignition signal coming from this igniter going back to the ECU. Voltages look ok. What am I doing wrong?

Using resistors as engine sensors. Havent simulated the O2 sensor because I just want to run in open loop.



-- Edited by rab on Monday 7th of May 2018 07:40:13 PM



-- Edited by rab on Monday 7th of May 2018 07:54:23 PM

__________________
Page 1 of 1  sorted by
 
Quick Reply

Please log in to post quick replies.



Create your own FREE Forum
Report Abuse
Powered by ActiveBoard