tach not working ('97 HDFT 24V)

[Firewout]

it's not a drama, it didn't work when I bought the truck, but I would like to get it going again.
So here's the story :

the tach gets its signal from the IP.
the signal goes to the EGR ECU and from there to the dashboard.
first I checked the ground connection. this was OK
Next I measured the signal from the IP to the ECU with a scope.

that seems to be ok, it reacts nicely when I rev the engine.

next I opened the ECU and measured the Input from the pump and the output to the dashboard.

input ok, but the output is a steady 12 volt which doesn't change when engine is revved. so maybe the ECU is toast.?

Does anyone know how the signal to the meter in the dashboard should be ? I was thinking to bypass the ECU (EGR is diconnected anyway) and route the NE+ signal directly to the meter.
Will this work or will it end in a small puff of black smoke?
all input appreciated!

 

[StarCruiser]

it's not a drama, it didn't work when I bought the truck, but I would like to get it going again.
So here's the story :

the tach gets its signal from the IP.
the signal goes to the EGR ECU and from there to the dashboard.
first I checked the ground connection. this was OK
Next I measured the signal from the IP to the ECU with a scope.
View attachment 101519
that seems to be ok, it reacts nicely when I rev the engine.

next I opened the ECU and measured the Input from the pump and the output to the dashboard.
View attachment 101520

input ok, but the output is a steady 12 volt which doesn't change when engine is revved. so maybe the ECU is toast.?
View attachment 101521

Does anyone know how the signal to the meter in the dashboard should be ? I was thinking to bypass the ECU (EGR is diconnected anyway) and route the NE+ signal directly to the meter.
Will this work or will it end in a small puff of black smoke?
all input appreciated!

View attachment 101522

Sounds to me like you possibly have an output transistor blown. If it were me, I would look at that pin 3 to the tachometer, and trace the tracks on the PCB to then check where the transistor is picking it's signal up from and change the transistor.

 

[Firewout]

thanks SC, I could do that.
no SMD on the board yet, so it should be doable....

 

[StarCruiser]

thanks SC, I could do that.
no SMD on the board yet, so it should be doable....

SMD is a real pain. So are hair thin signal tracks!! Spent the best part of a day working under a microscope desoldering and reattaching a connector such as the one in your pic to a ribbon pigtail. Amazing how the dexterity improves to do the fine work when it is magnified like that.

 

[Firewout]

weekend again, so next step in fault finding.
Following SC's advice, I started looking for a dodgy transistor by following the traces. when I opened the back of the ecu, my enthousiasm faded a little because as opposed to the front, the back was fully loaded with miniature smd components...so no chance for my thick fingers to operate on there.
and then, almost by accident, I noticed a capacitor with a familiar black colour :

a quick trip to the local electronics shop to get some different capacitors. an investment for the stunning amount of 2.5 euro's.
soldered one in, attached the scope, plugged everything back into the cruiser et voilà : a nice signal on the tach output.

and the needle of the tachometer woke up again after 3 years :

up to next step : how accurate is the tach ?
at 120 kph (75mph) it indicates 2600 rpm with 285/75/16 tires ('97 24v HDFT manual)
is this realistic ?

nevertheless, I'm already very satisfied with the result I got.

costs so far : 2.5 euro for 4 capacitors (of which I needed just one) and 50 euro's for a pc oscilloscope from dealextreme (which came with 2 decent probes and works flawlessly).