Mazda 626 DITD - Starting Problems

Hello!

My name is Stefan, and I'm new to this forum because I have a problem that I'm simply unable to solve.

Here's the translation:

It concerns the following vehicle (I hope you can help me despite the VAG-centric nature of this forum ).

Mazda 626 2.0 DITD.
2-liter turbocharged diesel engine.
Denso V5 Distributor-ESP.
Direct injectors.
VJ27 Turbocharger.
Year of construction: 1999.
90 horsepower.
Combination.

It started in the summer, when the engine sometimes wouldn't start when it was warm - turning the ignition off and on again would then solve the problem. There were no problems when the system was in a cold state.

In mid-August, I was on vacation in Croatia, and my car broke down on the highway while I was stuck in traffic (my fault!). In the midday heat, he then tried to restart the engine only after turning off all consumers and pumping the diesel filter a few times with the hand pump (possibly the hand pump had no effect, I just did everything at the same time out of panic, as I could hear horns honking behind me).

So, the initial assumption was a dead battery, especially since the accompanying vehicle was able to jump-start the car at the next rest stop using jumper cables.

However, the battery is only about 3 years old, has a capacity of 80 Ah, and a nominal current of 660 A. Furthermore, I was able to rule it out later on, as it also started having problems when jump-started.

A friend of mine who is very knowledgeable about older diesel engines was then able to pinpoint the problem.

The ESP (Electronic Stability Program) has a mass airflow control valve that is controlled by the control unit via an 'injector driver module'. This valve only has two states:
-) When open, diesel fuel is pumped back from the high-pressure side of the pump to the low-pressure side.
-) If the engine stalls, the diesel fuel is forced into the injection system.
-) The longer it lasts, the longer the injection duration, and the higher the injection volume.

When cold water was poured into the valve while it was in a warm state, the engine started very reluctantly. However, even the starting behavior deteriorated over the course of the autumn.

So, I bought a used, but functional MRV (I was able to install and test it myself in the seller's car, a Premacy with the same engine).

Valve replaced, result: It doesn't start at all; it cranks over nicely, but doesn't start.

A friend of mine, who is a very good mechanic (although I have a vehicle technology education, I'm more of a theorist and a hobbyist ), took a look at it and suggested that I try starting it with silicone spray.

Yes, it's not good for the engine – I know. But it helped... in a way. The engine started, ran for about 2 minutes at idle, and then just turned off. Subsequent attempts to start the engine resulted in it starting, but then immediately turning off while idling. However, if you kept your foot on the accelerator, you could operate it normally.

Meanwhile, I have consulted various Mazda workshop manuals available online for that engine, studied numerous forums, and ruled out all sorts of other potential problems.

-) Override nuts on the fuel injection lines opened during starting - fuel is reaching the injectors.
-) 'Fuel Shut Off - Valve': A dedicated valve that cuts off the diesel supply when the engine is turned off, in case the quantity control valve fails. The relay for this valve, located in the passenger footwell next to the control unit, was tested - both are functioning correctly.
-) The fuel injection pump speed sensor is working correctly.
-) Fuel injector pump temperature sensor disconnected - problem still persists.
-) AGR system disconnected (hub sensor and control line (pneumatic) alternately and together) - problem still occurs.
-) 'Injection Driver Module' tested on an identical engine - functioning correctly.
-) I can rule out glow plugs, as the problem persists even after the engine has warmed up, and the engine is a direct injection engine.

The only thing I noticed was that the hand pump offered very little resistance to the diesel filter, especially immediately after an attempted start.

Then, about two weeks ago, something big happened. A friend came over and wanted to see what I was doing with the car, just out of curiosity (he's not really a car person), and we tried starting it with silicone spray, and suddenly it could idle normally and could be driven like a regular car (without me changing anything!). However, it still doesn't start on its own.

Since it was a bit warmer that day, the next assumption was that a diesel hose might be porous in the cold, causing it to draw in air.

So, I replaced the hose between the diesel filter and the ESP (Exhaust Gas Recirculation system), and I bled the diesel system again using a hand pump and by draining diesel from the filter. And lo and behold, the hand pump now feels really hard after a few strokes, just like it did before.

However, it still doesn't start automatically.

So, this is my current situation: It can only be started with starting spray, but once it's running, it runs without any problems.

I hope you can help me, because I'm completely at my wit's end. I thank you in advance .

Best regards, Stefan.

P.S.: The error reading also didn't reveal anything!

Hello,

Please note:

Quote:

We cordially invite you to participate. The requirement for posting in diagnosis-related topics (everything for which a vehicle diagnostic system is needed) is... a diagnostic tool from a diesel mechanic's shop, or... our software, KDataScope, KPower, or the Diesel Mechanic license, or... a vehicle diagnostic support key. Diesel mechanic customers receive extended support (expert assistance) and full access.

Has it already been checked, during a startup attempt, whether the necessary starting speed is actually reached?

Best regards, Rainer.

Hello Rainer!

I assumed that, since I wasn't receiving any error codes, it wasn't a diagnostic issue - I apologize.

No, it has not been reviewed. I simply assumed it would fit, as it seems to be the right size acoustically, and he also experiences the same problems when starting the car with an external jump start from a car with a similarly powerful battery.

However, I suspect that the problem doesn't lie here, as there have also been occasional issues with the engine idling, where the starter motor and the starting state in the ECU shouldn't play a role.

Best regards, Stefan.

Have the temperature sensors for coolant and diesel temperature ever been temporarily replaced with a resistor that simulates a "cold" reading? This could help determine whether the problem is related to the actual temperature or only to the temperature that the engine control unit (ECU) believes to be present.

Background: In at least some VW vehicles, different starting conditions and parameters are required at high temperatures compared to low temperatures. For VAG vehicles, simply disconnecting the coolant temperature sensor can be used to check this (the control unit then assumes a temperature of -40°C). In VAG vehicles, the minimum engine speed that must be reached before sufficient fuel is injected is often a problem at higher temperatures (typically >300 RPM when warm, less is sufficient when cold). In such cases, a weak starter motor or battery might start the engine more easily in cold conditions than in warm conditions. Of course, Mazda vehicles may have different parameters that are relevant to starting. It is certainly important to be able to distinguish between the actual temperature (problems with ESP, etc.) and the measured temperature (sensor or control unit problem).

Regards,
Guste.

Hello Guste!

As far as I know, the amount of fuel injected during startup definitely depends on the coolant temperature. However, I need to read the workshop manual more carefully or find out whether the engine has two temperature sensors. (Because the instrument cluster inside the car provides a plausible reading.)

Regarding the fuel temperature sensor on the ESP, I measured it with an ohmmeter when it was cold, and according to the Mazda internal workshop manual, it's within the correct resistance range. However, this doesn't tell us anything about its performance when it's warm.

I will read into that more carefully and then report back. Thank you very much in advance for any answer .

Best regards, Stefan.

Quote:

whether the engine has two temperature sensors. (Because a plausible reading is displayed on the instrument cluster inside the car)

At VAG, 2 sensors are built into 1 "sensor," so the display in the instrument cluster doesn't really tell you much.

If the engine suddenly shuts off, I would also check how the power supply to the engine management system is implemented. Have there ever been any issues with faulty relays (109) in VAG vehicles?

Hi, there are two more relays that affect the fuel injection – one supplies power to the control unit, and the other to the quantity control valve. I've checked both, and they are both okay.

Regarding the issue with the temperature sensors:
-) There is a coolant temperature sensor located at the front of the engine head.
-) Two intake air temperature sensors (one at the air filter and one after the intercooler).
-) A fuel temperature sensor on the ESP (Electronic Stability Program).

According to the information in the workshop manual, the following factors are involved:
-) The coolant temperature sensor influences both the fuel injection quantity and the fuel injection timing.
-) The intake air temperature sensor located on the air filter affects the timing of when the intercooler activates and adjusts the fuel injection amount.
-) The fuel temperature sensor only affects the fuel injection amount.
-) The starter signal also affects the amount of fuel injected.

And indeed, the only part that also affects the idle speed is the coolant temperature sensor. Furthermore, it's located not far from the fuel quantity control valve, which, when exposed to water, led to slightly improved starting behavior. I think that's really a good tip, I will take a look at it and get back to you as soon as I know more .

I will also follow up on the issue with the starter speed, but for me, it's easier to start by checking the sensor first.

Best regards, Stefan.

Hello!

First of all, I would like to apologize for the delay.

Okay, I've now managed to do the following with the help of a friend/colleague:

He is skilled at programming Arduino microcontrollers and has written two programs that independently control both the mass flow control valve and the timing control valve on the injection pump, regardless of the engine control unit.

We connected an oscilloscope to each of the connections to monitor the signals being sent by the control unit, and we read out these signals. He then simulated them and programmed the control frequencies.

'These also worked; that is, during the startup process, we were actually able to extract a high volume of fuel from the open diesel lines, even with an extremely high injection setting. However, when the control unit was disconnected, nothing came out.' We managed to get the timing control valve working even when the vehicle was otherwise turned off, and we could clearly hear it operating (both are floating and not grounded).

We read the starting speed using the fuel injection pump speed sensor – it was just under 300 RPM, which is definitely sufficient.

However, despite completely and independently modifying both the fuel injection amount and the injection timing, the engine didn't even make a sound. It starts up cleanly, but it doesn't actually turn on.

So, we tried using starting spray again, but because of the excess diesel in the combustion chamber, it first backfired, then revved up quickly, and a lot of black smoke came out the back - so we quickly turned it off again.
Ouch...

I don't know if the engine has improved mechanically thanks to the starting spray. It still spins very smoothly, but it's still possible for it to fail.

I'm completely at my wit's end. In summary:

-) Injection amount varied from 0 to full, with no success.
-) Injection timing varied from extremely early to extremely late - no success.

But what else could it be due to?

I mean, the engine is actually completely fine, except for this unpleasant issue with the starting spray... But the problem was already there before. By the way, I understand that the car is no longer usable, but I would just like to find out what the problem is.

Oh yeah, I also bypassed the coolant temperature sensor, but that didn't work either.

Best regards, Stefan.

There are automotive diagnostic systems specifically designed for this purpose. Anything else, including DIY projects with Arduino, will only waste your time. Either get a suitable system for your car or dispose of the device. At least, the Arduino tinkering had some educational value.

He's going to be scrapped anyway, I just want to find the error.

The error reading process, as described above, has already been performed. No results .

Best regards, Stefan.

You shouldn't just blindly read the memory; you should look at the measured values and draw logical conclusions.

Furthermore, you haven't revealed how the engine control unit's fault memory was accessed, because you're unlikely to get very far using OBD2 for that.

I'm sorry for my blunt words: Your actions are a real pain and only make sense to people who are extremely bored right now.

There are people who file a nail from a thick wire, although whether it's a sensible thing to do is another matter.