Need help understanding how to determine GND and VCC pins from unlabelled pins

Thread Starter

Steven_Mark

Joined May 28, 2020
1
Hi All,

Recently I started to learn electronics, and saw a measurement method to determine GND and VCC pin.

So basically there is a few unknown pins on a PCB circuit. The guy first used a multimeter to probe the pin against GND and measured a resistance value of 0 ohm. Next, he probe the same pin against a VCC point on the board from a IC chip. The measurement is 100 ohms. Lastly, he switched on the device, and measured the voltage of the pin to be 0V. He went on to derived that the pin is actually a GND pin.

Similarly, the guy used a multimeter to probe another pin against GND and measured a resistance value of 100 ohms. Next, he probed the same pin against a VCC point on the board from a IC chip. The measurement is 0 ohm. Lastly, he switched on the device, and measured the voltage of the pin to be 3.3V. He went on to derived that the pin is actually a VCC pin.

I thought about it but still could not wrap my head around how this method works.Would you guide me to understand the logic behind it? Thank You.
 
Last edited:

MaxHeadRoom

Joined Jul 18, 2013
20,698
Does not sound a very good definitive method of determining VCC/VSS pin connections etc.
I would be hesitant at using it, personally.
Max.
 

MrSoftware

Joined Oct 29, 2013
1,780
If you have a reasonable idea for the purpose of a chip, and a good idea for how the board works, you can probe some pins and make a somewhat educated guess as to what the pins are, but the datasheet is the only 100% accurate way to go. For example if you have a voltage regulator chip that you know has a 5V in, 3.3v out and a ground, but you don't know which is which; then in this case you can probe around, or power it up, make some measurements and reasonably determine which is which. BUT if you've got a chip with a number of pins, you have no idea what they do, then you can probe some pins and perhaps determine which pin if any is connected to the 3.3v supply net, but you cannot say with 100% certainty that it's the Vcc pin. It might an enable pin, reset pin, or one of 100 other purposes, probing around will only tell you where it's connected, not what it's purpose is.
 

dl324

Joined Mar 30, 2015
10,745
Welcome to AAC!
I thought about it but still could not wrap my head around how this method works.Would you guide me to understand the logic behind it?
Sounds like some typical YouTube snake oil to me.

If it's an unknown/unmarked chip, there are some standard places for ground on many IC's. On non-military logic chips in dual in-line packages, power is usually the highest number pin and ground will be the pin kitty-corner to it. There are exceptions, but that will cover a majority of the chips in dual in-line packages.
 

jpanhalt

Joined Jan 18, 2008
9,419
I would rely mostly on the PCB for an unknown and unmarked chip. Of there is a marked chip on the same board, you can use its datasheet to find the PCB traces that are GND and VCC. If the board is battery powered, you should be able to determine the same assignments. Then, anything with "zero" ohms to ground may be the ground pin. And similarly for VCC. But that is no guarantee. Dome chips (probably many today) have multiple VSS/ground and VCC connections.

Once you establish which trace is ground and which is VCC, you can usually confirm those connections to pins on the chip. The fact that a certain pin on a chip is connected to ground or VCC does not mean it is a power ground or power VCC. If there is only one such connect pin on an IC,it might be a safe bet, but on rare occasions someone might use a small value resistor on VCC to limit current. I have never seen that done on ground.
 

BobaMosfet

Joined Jul 1, 2009
1,108
Normally, you can use the board layout to help you determine VCC, and ground, usually in relationship to other components (regulators, bypass caps, etc). Trace thickness, and which plane the pin is attached to.

Beyond that, you need to figure out what the board is doing enough to be able to discern what the chip is doing, it's type, and pin count, orientation, and then start comparing what you believe it is to datasheets and see if you find something that seems to match everything.
 

Bernard

Joined Aug 7, 2008
5,524
I look for polarized caps - sign. If several - are common, pretty sure it is power supply common.
Almost always there are exceptions.
 

jpanhalt

Joined Jan 18, 2008
9,419
Beyond that, you need to figure out what the board is doing enough to be able to discern what the chip is doing, it's type, and pin count, orientation, and then start comparing what you believe it is to datasheets and see if you find something that seems to match everything.
Are you serious? If you have the datasheet, why do any of the preceding?
 

andrewmm

Joined Feb 25, 2011
323
From what I read of the OP,

The test was measure the resistance from a known ground and Vcc to an unknown pin on a chip.

If it reads "zero" from ground to that pin and a higher resistance from Vcc to that pin, its liable to be a copper connection between the gnd and that pin, its a ground,

conversely,

If it reads "zero" from Vcc to that pin and a higher resistance from Ground to that pin, its liable to be a copper connection between the Vcc and that pin, its a Vcc,

Now how the original video "knew" a Vcc and a ground to test against, is open to debate, but it sounds a fairly simple and reliable test to me.

You could improve it by reversing the meter and test again, to detect a diode, but most "all" meters would not read a forward biased diode as zero.
 

jpanhalt

Joined Jan 18, 2008
9,419
From what I read of the OP,

The test was measure the resistance from a known ground and Vcc to an unknown pin on a chip.

If it reads "zero" from ground to that pin and a higher resistance from Vcc to that pin, its liable to be a copper connection between the gnd and that pin, its a ground,

conversely,

If it reads "zero" from Vcc to that pin and a higher resistance from Ground to that pin, its liable to be a copper connection between the Vcc and that pin, its a Vcc,

Now how the original video "knew" a Vcc and a ground to test against, is open to debate, but it sounds a fairly simple and reliable test to me.

You could improve it by reversing the meter and test again, to detect a diode, but most "all" meters would not read a forward biased diode as zero.
Yes, it is likely the first instance is a "grounded" pin. But that does not extend to it being power ground, i.e., "GND pin," which is how I and most here seem to read the OP. Similarly, pins can be tied high and not be VCC.
 
Top