Turbo Upgrade for Mazda DiTD

Hi,

I drive a Mazda 323F 2.0 DITD turbo diesel 16v direct injection, manufactured in 1999. The car originally has 90 horsepower. I have installed an 'Alpin Powerbox' which increases the fuel injection. This significantly improves the car's performance, but it now produces too much smoke. If anyone has the time and inclination, I can also provide photos of the engine. There is a boost sensor mounted on the intercooler. I am grateful for any advice you can offer!

"Please measure the signal from the boost sensor, and then we can help you accordingly. It's especially important to..."
- Supply voltage (5V or 12V?)
- Output voltage at atmospheric pressure (engine off)
-Output voltage at maximum pressure (the motor must be operated at maximum load across the entire speed range, which can only be achieved while driving or on a test stand).

Regards,
Guste.

Hi,

'I finally got around to measuring things. The input voltage is 5V. The output voltage is 2.6V with the motor off, 2.6V with the motor running and idling, and 4.6V at full load. So, there's a 2V difference. According to Mazda, the boost pressure is 0.6 bar. If I want to increase the boost pressure to 0.75 bar, I need to reduce the output voltage to 4.1V. That would be 0.5V less, which is a quarter of the 2V difference, and 0.6 / 4 = 0.15 bar. 0.6 + 0.15 = 0.75 bar. Please correct me if I'm wrong. Do I need to solder a resistor in parallel with the output? And what value should that resistor have?'

If you want to reduce the voltage, you simply need to connect a diode in series with the output signal. This will reduce the output voltage by approximately 0.6-0.7V.

However, that won't help in your case, because the microcontroller (MCU) will try to regulate the input voltage to approximately 4.6V. If the MCU sees 4.6V, then there would need to be 0.6-0.7V more before the diode. However, since that is already above the operating voltage of your sensor, it won't work.

Conclusion: You need a sensor that doesn't reach its maximum output voltage until 4.6V. Therefore, you need a sensor with a wider input range. Since yours already outputs almost the full supply voltage at 0.6 bar, it can no longer measure higher boost pressures and is therefore unsuitable for your project.

I'm afraid I can't tell you which sensor would be suitable for your Mazda.

Regards,
Guste.

If you want to reduce the voltage, you simply need to connect a diode in series with the output signal. This would make the output voltage approximately 0.6-0.7V lower.

However, that won't help in your case, because the MSG will try to regulate the input voltage to approximately 4.6V. If the MSG sees 4.6V, then there would need to be 0.6-0.7V more before the diode. However, since that is already above the operating voltage of your sensor, it won't work.

Conclusion: You need a sensor that doesn't reach its maximum output voltage until 4.6V. Therefore, you need a sensor with a wider input range. Since yours already outputs almost the full supply voltage at 0.6 bar, it can no longer measure higher boost pressures and is therefore unsuitable for your project.

I'm afraid I can't tell you which sensor would be suitable for your Mazda.

Regards,
Guste.

Germanium diodes have a voltage drop of approximately 0.3 volts. Maybe that will help. If that doesn't work, you'll need a different sensor. What kind of charger does the car have? With or without VGT (Variable Geometry Turbo)?

schnappi wrote:

Germanium diodes have a voltage drop of approximately 0.3 volts. Perhaps that will help? If that doesn't work, you won't be able to avoid using a different sensor. What kind of charger does the car have? With or without VGT

?
Okay, I didn't mention it for two reasons:
Please provide the text you would like me to translate from German to English.
With a voltage drop of 0.3V, the sensor output voltage would already have to be 4.9V. I would definitely examine the sensor's datasheet carefully before experimenting with that. And obtaining that datasheet might be quite difficult (unless you know which specific sensor is being used internally).

2.
Konjoy desired an increase in the load pressure (Ld) of approximately 0.15 bar with a voltage of 0.6V. With a voltage drop of 0.3V, the increase in load pressure (based on a linear voltage curve) is only 0.075 bar. This makes it possible to minimize soot only to a very limited extent. Furthermore, I would not risk using a pressure sensor with such a small voltage difference between input and output, as this could, in the worst case, lead to an increase in LD until something gives way.

Regards,
Guste.

Wouldn't it be possible to install some kind of electrical component that would cause the voltage to reach 4.6V sooner? At full throttle, the output voltage is 4.5V, but for a brief moment, it jumps to 4.6V. If I install a resistor that drops 0.4V, wouldn't the boost pressure increase until the output voltage reaches 5V, and the control unit receives a value of 4.6V? Does the sensor perhaps have an internal resistance that prevents it from producing 5V at the output? There must be some good electrical engineers out there who can help me with this.

You misunderstood.
The sensor is powered by 5V and can only output a maximum of 5V. At that point, it reaches its measurement limit. If you now connect a diode in series (-0.5V), the control unit will continue to try to regulate to 4.6V. However, the sensor would need to output 5.1V to achieve 4.6V after the diode. Since it cannot do that, the wastegate/VTG will be fully actuated, and your turbocharger will almost certainly fail immediately. Even a slight increase is dangerous because it can lead to insufficient counter-regulation during overshoots, and it will hardly suppress the smoke.
Chip tuning cannot solve this problem either.
The only possible solution would be to install a different boost pressure sensor.

The 4.6V reading is with overshoot. At full throttle, it reads 4.48V. But perhaps I could try using a germanium diode that only draws 0.3V. Will the engine control unit trigger an error message at startup? Is that even checked? And in idle, I wouldn't have 2.6V anymore, but 2.3V – will it somehow detect that as an error?
'Another sensor would be good. One that operates with a 12V power supply would be ideal. But I'm sure I'll have to search everywhere before I can find something like that for such an unusual car. I wonder if anyone knows what other sensors might be compatible?'

Well, according to my calculations, you'll get 100 mbar more with 0.3V. But please be aware that this is extremely dangerous.
1. Overbrakes will no longer be recognized.
2. You cannot assume that the sensor is still operating linearly at the measurement limit.
It could go well, but it doesn't have to.

Apparently, there are no error entries in this case.

A sensor with 12V does not fit!
Ideally, you'd want a sensor from the 1Z engine produced up to the end of 1993. It has an external pressure sensor with a 5V output at 200kPa (which is 1 bar relative). Yours likely can handle a maximum of around 175-180kPa. You would need to adjust this using an operational amplifier circuit (a multiplier) and that would be it.

wolfi_b wrote:

The sensor is powered by 5V and can also only output a maximum of 5V.

I don't know the details of this sensor, but what you're describing is an ideal sensor. With a real sensor, the output voltage will always be lower than the input voltage. I assume that the sensor is already close to its limit at 4.6V output.

Regards,
Guste.

Yes, that's correct, which is why I wrote that.

wolfi_b wrote:

2. You cannot assume that the sensor is still operating linearly at the measurement limit.

I know that mine outputs at least 4.96V during overshoots.

@wolfi_b

Your suggestion sounds good. If the external boost pressure sensor from the 1Z model has a 5V output at 1 bar, then it would have a 3V output at 0.6 bar. Using an operational amplifier as a multiplier with a gain of 4/3, I could then have a 4V output at 0.6 bar. The control unit would then increase the boost pressure until the output voltage reaches 4.6V. This would correspond to 4.6V / (4/3) = 3.45V at the sensor, and I would achieve my desired boost pressure without significant risk!

konjoy wrote:

@wolfi_b Your suggestion sounds good. If the external boost pressure sensor has a 1 bar pressure corresponding to a 5V output voltage, then at 0.6 bar, it would have a 3V output. Using an operational amplifier as a multiplier with a gain of 4/3, I could then have a 4V output at 0.6 bar. The control unit would then increase the boost pressure until the output voltage reaches 4.6V, which would correspond to 4.6V : 4/3 = 3.45V at the sensor, and I would achieve my desired boost pressure without significant risk!

Hi konjoy,

Could you please provide the exact model name on the sensor, including everything that's written on it? Also, please provide the hole spacing for the mounting screws and the diameter of the opening in the coolant reservoir. It's possible that I have something suitable, or at least I can find the specifications for it.

'Couldn't I just install a steam turbine? I already have a Westgate turbocharger. Surely that would work, right?'