Golf 3 TD AAZ engine problems: preheating and tachometer

Hello, I have two annoying problems with my Golf 3 TD Variant (manufactured in 1998, AAZ 75 hp) which is still almost rust-free and runs excellently. Perhaps someone with experience in automotive electrical systems might know the possible causes.
"For a few days now, the glow plug indicator has only been lighting up occasionally. Most of the time, it takes about half a minute of cranking before the engine starts. The control unit is the 166 unit located above the relay panel (controls injection timing, exhaust gas recirculation, and glow plug control). The corresponding relay in position 12 is working. Sometimes the glow plugs do light up, but for too short a time. Occasionally, simply opening the door triggers the automatic glow plug system, and the engine starts immediately. Otherwise, when the engine is cold, nothing happens. If I disconnect the four-pin connector at the coolant outlet on the engine head (which connects the coolant and engine temperature sensors), the glow plugs light up properly. When the engine is cold, the relevant engine temperature sensor reads approximately 800 ohms. I suspect the temperature sensor (which may be reporting a warm engine temperature when it's actually cold)." All the connections have been checked, the contacts have been cleaned, but it didn't solve the problem. Does anyone know about something like this?
The ESP is prepared for Pöl; in the older predecessor model (manufactured in 1994), I had a control relay in slot 12 with an extended afterglow time. Can the control unit 166 be reconfigured to have a longer afterglow?
"Since replacing the alternator with a refurbished unit (Valeo 70A), the D+ signal seems to be malfunctioning. The tachometer almost always stays at zero after starting the engine and only becomes active at higher RPMs, then it functions normally." After a while, it turns off again, and the oil pressure warning light blinks when it's off.
The wiring to the alternator has been checked and is OK. An additional ground connection to the alternator has also been installed. Another speedometer (a VDO, while the one installed is a Motorola) shows the same problem. Does anyone have any ideas?

Hi,

Regarding preheating:
If you state that the dual-temperature sensor has a resistance of only 800 ohms when the engine is cold (around 20°C), then it is defective.
These sensors contain two identical NTC thermistors, also known as "hot-element" resistors. The warmer it "feels," the lower its resistance becomes. One person is reporting an issue with the control unit, while another is reporting an issue with the instrument cluster.

"For a healthy antenna, the nonlinear curve indicates:"

0°C = 5000 Ohms
10°C = 3000 Ohms
20°C = 2000 Ohms
30°C = 1500 Ohms
40°C = 1000 Ohms
50°C = 750 Ohms
60°C = 550 Ohms
70°C = 400 Ohms
80°C = 280 Ohms
90°C = 210 Ohms
95°C = 180 Ohms
100°C = 165 Ohms

=> So, your sensor is reporting that it's too warm, and therefore => it's not preheating, which your test confirms by unplugging it.

"Replace it and get a good one, buy an original part. The imitations differ significantly in their resistance curve and don't last very long."

Regards, Mpire.

Thank you for the quick and professional response. Unfortunately, I've already ordered a Hella replacement part, but it should be okay. Hopefully, that will fix the problem.
I'm actually considering retrofitting the old glow plug relay in slot 12 and disconnecting the engine control unit for the glow plug function (which would also allow for a longer afterglow duration for diesel engines). Control unit 166 also regulates the exhaust gas recirculation (which has been removed, resulting in a fuel saving of one liter per 100 km, and keeps the intake manifold clean), processes the injector signal, and controls the injection timing advance on the old VP system, as well as the idle speed increase during cold starts. He probably needs the temperature signal, which I will extract as a low-impedance voltage (I suspect it's simply a voltage divider using the NTC resistor and a fixed resistor to ground within the control unit).

I made a mistake in my thinking about the problem with the tachometer; the W signal is the correct one. I'm going to re-check the entire cable run and wiggle it around, maybe it's just a loose connection or corrosion.

I still don't quite understand your explanation.
Does the vehicle with the AAZ engine have the control features as described (e.g., sensors in the sealing plug), or were they added later? In its standard configuration, the engine temperature would likely be monitored.
Therefore, I don't understand your statement about separating things. What is it good for?
hg
Herbert.

The fuel pump is standard equipment, as is the electric fuel injection start adjuster (produced from the end of 1998). Therefore, as you mentioned, the engine control unit will likely need the temperature value for fuel injection timing and idle speed adjustment, which will also be used for the preheating control.
If I connect the wires in parallel to slot 12 and plug in the standard glow plug control there instead of the current relay (the battery voltage and glow plug connections, as well as the relay control, are definitely connected), and I need to test whether the temperature sensor, door contact, starter, and glow plug indicator on the instrument cluster are connected to slot 12.
If I connect two control units to the temperature sensor, the value will likely be completely distorted. I suspect that in both control units, a resistor to ground forms a voltage divider with the NTC, and the resulting voltage value (which is inversely proportional to the temperature) is evaluated. If I connect two load resistors in parallel to the NTC, one in each control unit, this voltage value will decrease significantly and be interpreted as a "too cold" engine. Therefore, only one control unit is allowed to be directly connected to the temperature sensor (using the original voltage divider), and the resulting voltage must be buffered with a low-impedance output (using an emitter follower) so that the second control unit receives the correct voltage value, just like the first, without the second load resistor distorting the value. The emitter follower thus simulates the temperature sensor by providing a voltage that follows the original value (minus 0.5V, but this would not be a problem for the glow plug control). At least, that's what I think, but I'll have to try to see if it works.

In its original state, the preheating time is likely determined by the engine control unit. The preheating control unit is connected downstream of the control unit.
What you want (as I understand it) is a separate relay that evaluates the temperature itself. The term "regular preheating control" is misleading in this context.
The temperature sensors are connected to ground on one side only.
Alternatively, check what kind of signal the engine control unit sends to the glow plug control system.
You'll have to replace the temperature sensor anyway.
hg
Herbert.

PS: Or simply install a switch for manual pre-glow/post-glow. An AAZ driver can definitely handle that. There's definitely a slot available in the dashboard, and the relay with fuse will be mounted on an additional bracket.

I'm fine with the preheating button, but my partner wouldn't be as thrilled.
You're right, I want to disconnect the preheating element in the control unit (terminal 166 above the relay panel), but I want to keep the rest active, which is why the temperature sensor needs to stay connected. I'll remove the simple relay on pin 12 and replace it with the preheating control. This requires the temperature sensor, so I need to tap into its signal in a way that doesn't affect the function for the control unit.
The temperature sensor is new, but the engine wasn't cold yet.

It was the temperature sensor (approximately 1.3 kOhms at 15 degrees Celsius, which is clearly too high). Switched and it's good.
That just leaves the issue of the tachometer.

LaurenzBo wrote:

It was the temperature sensor (approximately 1.3 kOhms at 15 degrees, which is clearly too high.

Too low, therefore insufficient preheating.

Herbert.