Technical Question regarding negative wire disconnect switch and voltage (AI answer)

Here's something relatively interesting (at least to me, maybe to some of you). I have this "situation" relative to a new wiring harness installation (some of you may have seen my other posts regarding new harness in '37 Plymouth). So, I posed this to "Grok" Artificial Intelligence (AI):
"I am putting a new wiring harness in an old Hot Rod truck ('37 Plymouth). It has a battery cut-off switch wired to the negative side of the battery. When the cut-off switch is off (ground cut-off), I can put one end of my multi-meter on the battery Ground terminal and the other on the chassis/body and it will read 12 volts (+ or -) [neg battery post to chassis ground]. When I turn the disconnect back to on (ground restored) and re-connect one end of the multi-meter to the battery neg post and the other to the chassis/body it shows no voltage. Will a negative side cut-off switch cause something like this? I am in the process of checking wires and do not see any issues (frayed, worn, misconnected)".
Here's its response:
A negative-side battery cut-off switch can indeed cause the behavior you're observing with your multimeter, and it’s expected based on how the circuit is configured. Let’s break it down:
Why You’re Seeing 12 Volts When the Cut-Off Switch is Off

• Setup Context: Your battery’s negative terminal is connected to a cut-off switch, which then connects to the chassis (ground). When the cut-off switch is off, the battery’s negative terminal is disconnected from the chassis, breaking the ground connection for the entire electrical system.
• Multimeter Reading: When you measure from the battery’s negative terminal to the chassis with the switch off, you’re seeing approximately 12 volts. This happens because:
  • The battery’s positive terminal is still connected to the vehicle’s electrical system (e.g., wiring harness, components, etc.), which is ultimately tied to the chassis through various ground points.
  • With the negative side disconnected (switch off), the chassis is "floating" relative to the battery’s negative terminal. The multimeter is detecting the potential difference between the battery’s negative terminal (0V relative to itself) and the chassis, which is indirectly connected to the battery’s positive terminal (+12V) through the vehicle’s wiring and components.
  • Essentially, the multimeter is completing a circuit from the battery’s negative terminal, through the multimeter, to the chassis, and back to the positive terminal via the vehicle’s wiring. This creates a voltage reading of approximately 12 volts.

Why You See 0 Volts When the Cut-Off Switch is On

• When the cut-off switch is on, the battery’s negative terminal is connected to the chassis, establishing a proper ground.
• Now, the battery’s negative terminal and the chassis are at the same electrical potential (0V, ground). When you measure between them with your multimeter, there’s no potential difference, so you read 0 volts.
• This is normal behavior for a properly grounded system.

Is This Behavior Normal for a Negative-Side Cut-Off Switch?
Yes, this is entirely consistent with a negative-side battery cut-off switch. The switch interrupts the ground path, isolating the battery’s negative terminal from the chassis. When the switch is off, the chassis is no longer directly grounded to the battery, which can lead to the voltage reading you’re seeing due to the multimeter creating a path through the vehicle’s wiring.
So, what is you guys' opinion of the response and the AI's accuracy?
Thanks.

 

It is accurate, if a bit verbose. Voltage = potential. With the switch in the Off position the battery is disconnected from the chassis, so there is potential (voltage) present between the chassis and the battery. With the switch in the ON position the battery is now connected, so there is no longer any potential between them.

 

Is AI traditional? A man (prophet? Savior?) once warned of the use of such things. He was the son of a single mother, his initials are J.C., perhaps you 've heard of him. He was born to deliver man from the evil of the sentient machines. John Conner, where are you when we need you?

 

recently retired so I can play with my toys after 50 years as an EE. Can't say I was ever a fan of auto-generated documentation. I will grudgingly admit its getting better. From your post, I didnt like the wording in this section:

• With the negative side disconnected (switch off), the chassis is "floating" relative to the battery’s negative terminal. The multimeter is detecting the potential difference between the battery’s negative terminal (0V relative to itself) and the chassis, which is indirectly connected to the battery’s positive terminal (+12V) through the vehicle’s wiring and components.

There is some truth in that, but the AI assumes newer cars full of electronic gizmos, they have lots of paths for tiny currents to flow back to ground keeping the clock and other junk alive even when ignition is off. May not be true with old cars...

I just went out and tried my Fluke meter on the 66 Corvair, its the closest one and the only WIP car here that still has a battery in it. The alternator is still connected, but nothing in the dash, no heater or radio, no light bulbs. Disconnected the ground cable from the battery. Reading across battery terminals shows more than 12V as expected, good battery. Reading from disconnected cable (chassis ground, meter neg lead) to battery neg post (meter pos lead) causes the voltmeter to jump, starts reading something negative but then drifts to lower voltage values. Got tired of watching that, might have got down to a few millivolts if I was patient. Then popped the hood on the '09 Crown Vic PI, pulled the ground cable, same test got constant -11.94V on the meter. Held it for a minute, no change. Maybe if I was patient...nah. Anyway, same meter, same test, different car, different current paths, different readings.

 

Guys, thanks for being good sports about this. Its definitely not easy for those of us who are a little older than this century to "embrace" AI. I don't want to get too far off of my original topic related to the readings I got on the multi-meter when testing from the battery neg post to the chassis/body with the disconnect in the "off" position. I just wondered what the "Grok" AI would have to provide. And to see if experts like many of you could critique and reject or verify its assessment. Seems like it did a pretty good job. I use it with a skeptical and critical perspective on all sorts of things. Its fun and simply (except not so simple) a different research resource (Grok AI app on Samsung or Apple phone).

Bandit Billy, as a fellow Christian, I understand your concern brother. We ought to be real careful about the nefarious use of AI. However, AI "in and of itself" isn't evil. (not saying it won't get there).

 

On your Corvair, what kind of meter are you using? Is it digital or analog? Some old analog meters will conduct some current which could account for your reading. It really should read battery voltage with no other connection to the battery negative post just like your Crown Vic.

 

John R, thanks for your perspective as a retired EE. Being just an amateur hot rodder I am always learning. I had not tested other vehicles to see if I could replicate it. Interesting.