Continuity test doubt

[Ram__46]

Hi everyone. I have a doubt. If I have to test a ground pin of a 5V connector (suspect bad ground), to do a continuity test is it necessary to touch with the multimeter probe any point of exposed metal of the chassis? Or do I have to touch exactly and directly the negative pole of the battery?
Is it the same thing?
Thank you for help

 

[DPG]

What application is this?
Is the chassis definitely supposed to be connected to 0V?

 

[Ram__46]

Hi and thanks for your reply, it's a 5V (2 pin) connector of an exhaust gas temp sensor in a car. I suppose I have a bad ground wire, but I can't reach the ground collecting the wires of that sensor (and many other sensors) because it's located under the dashboard (hours of labor to reach it). I read 4.963V between the pins, but the ground pin is at 0.037V with respect to a metallic part in the engine.

Instead, from the other pin of the connector to the same metallic part of the engine I have 5.00V. So the ecu sends perfectly the 5V, but the ground it's not perfectly at zero.

The question is, is the ground wire a bad ground? I'm a novice in electrical repairs
For comparing, I checked another EGT sensor connector, and tension at the pins is 4.995V, quite perfectly 5V (I suppose also within the multimeter error sensitivity (it's a cheap multimeter)).

 

[DPG]

Yes, those very slight differences are to be expected. It may be worth you try to connect a temporary wire from ground to the pin which you know should be grounded.

 

[Ram__46]

That's what I wanted to do. Create a new ground wire and get it to another ground in the engine bay. If the sensor starts working properly again, I know the answer. But could a bad ground wire give that 0.037V value? I really don't get it thank you

The other sensor working properly had 4.995V at the connector. So I suppose that at the pins there must be quite exactly 5.00V and not 4.96 or so.

 

[DPG]

Rather than test voltage, what you need to do is a resistance measurement between the ground of the sensor connector, and a known ground. Set your meter to resistance/Ohms.

 

[Ram__46]

I did it but I don't remember the value. I remember I read around 3.6 kOhm (probably) but it seemed too low to me and strange, I'll repeat the measure. What range of values should I expect?

 

[DPG]

I did it but I don't remember the value. I remember I read around 3.6 kOhm (probably) but it seemed too low to me and strange, I'll repeat the measure. What range of values should I expect?

If the wire should definitely be grounded, then 3.6k is far too high. Depending on meter quality, lead resistance, length of cable etc I would expect around 1 Ohm. Again, don't worry about slight variations.

 

[Ram__46]

I'll let you know, thank you

 

[Ram__46]

Minutes ago I did the attempt. Cold engine. Diagnosis told 130 °C for that faulty sensor. I grounded it to another ground in the engine bay and I immediately got plausible results: 31°C while other sensors were at 29 °C. I think I found the problem. Now I have to detach wires from the connector but I don't know how to do it properly. Thanks again