Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Why does (LT)spice require a ground?

Status
Not open for further replies.

carbonzit

Active Member
Question basically says it all. Why is this a requirement?

A lot of small projects have no real ground, but only positive and negative supply rails. Nothing is actually connected to the case (which may be plastic anyhow.) So why do we need to add a superfluous connection point that doesn't even really exist in reality?

It seems the simulator could just as easily use 0 volts as a reference.

Now I know there are many circuits where ground is not only necessary, but the particular configuration (e.g., star) may be crucial, particular RF circuits. But in low-frequency or DC circuits that run on batteries, it seems totally unnecessary.

It isn't a safety issue, is it? What would a simulator care about that?

Color me curious.
 
Question basically says it all. Why is this a requirement?

A lot of small projects have no real ground, but only positive and negative supply rails. Nothing is actually connected to the case (which may be plastic anyhow.) So why do we need to add a superfluous connection point that doesn't even really exist in reality?

It seems the simulator could just as easily use 0 volts as a reference.

Now I know there are many circuits where ground is not only necessary, but the particular configuration (e.g., star) may be crucial, particular RF circuits. But in low-frequency or DC circuits that run on batteries, it seems totally unnecessary.

It isn't a safety issue, is it? What would a simulator care about that?

Color me curious.

hi cz,
It not really a 'ground' its a 'reference' point or node that LTS uses for its calculations

If you think about it, without that reference node, how could it take measurements.??? bit like you with a DVM or scope hooking onto the 0V[gnd] reference....OK.?
 
It not really a 'ground' its a 'reference' point or node that LTS uses for its calculations

If you think about it, without that reference node, how could it take measurements.??? bit like you with a DVM or scope hooking onto the 0V[gnd] reference....OK.?

Well, that's why I asked why one couldn't just designate the negative rail (0 V) of the power supply, for instance, as the reference point.

I suppose it accomplishes the same thing in practice, but it just seems strange to have to add a "ground" to a circuit that really doesn't have one.
 
Because it's a computer program. If you've doing just a little bit of programming, you would have known that it require you to feed it with any relevant information.

Say you have a 12V battery in your circuit. All humans does understand that there is 12V and that voltage, and that voltage in most cases is based on the negative pole on the battery, that is zeero volts.

A computer on the other hand isn't just capable (or in this case not alowed) to assume anything. It doesn't know where to put zeero volts. You have to tell it.

Let's take another practic example: If there was no gravity, why should direcion down always be toward center of the earth? Why couldn't "down" just be a reference to your house (that floats around in the air)?

Hope this explanation helps :)
 
Last edited:
Well, not really, as I already understood all that.

What Spice is asking for is really a zero-voltage reference, not a ground. However, I'm sure it's called "ground" for historical reasons, and since nobody else seems to have a problem with it, I guess there's little point in speculating why.

If it were me designing the software, I'd just ask for a zero-voltage reference point, with "ground" as an option for those circuits that really do need a ground (audio amps, RF circuits, etc.).
 
Well, not really, as I already understood all that.

What Spice is asking for is really a zero-voltage reference, not a ground. However, I'm sure it's called "ground" for historical reasons, and since nobody else seems to have a problem with it, I guess there's little point in speculating why.

If it were me designing the software, I'd just ask for a zero-voltage reference point, with "ground" as an option for those circuits that really do need a ground (audio amps, RF circuits, etc.).


hi,
Get a simple circuit running in simulation. note the plot values you have selected to see.

Now move the 'reference' symbol, which looks like a ground/0V symbol to different part of your circuit. Re-run the sim and note the changes in the plot values.

As I said earlier, its like you using a voltmeter to check a circuit, you can connect the meter 0V anywhere onto the circuit and then that becomes the reference point for your measurements... have you got that.?
 
As I said earlier, its like you using a voltmeter to check a circuit, you can connect the meter 0V anywhere onto the circuit and then that becomes the reference point for your measurements... have you got that.?

Why do you seem to assume I don't understand this? I do.

I just object to the placement of a "ground" in a circuit which really has no ground. It's a matter of semantics, I suppose: I'd rather see it called "0 V reference".

Have you got that?
 
Why do you seem to assume I don't understand this? I do.

I just object to the placement of a "ground" in a circuit which really has no ground. It's a matter of semantics, I suppose: I'd rather see it called "0 V reference".

Have you got that?

hi cz,
You are reading something into that post thats not intended.

The point I am trying to get across is that someone at Linear choose a symbol that is normally used as a 0V [common] signal symbol.
I guess if the had created a 'special' symbol for the reference point, we wouldn't be having this discussion.

I would also agree that the name of the ref point isnt the best.

BTW: for future ref, when in LTS click F4 and you will see another symbol called 'COM' sometime you may need to use this version

While you have F4 pressed, note the 'Gnd' is also called 'Global Node 0'.

EDIT:
I would suggest that that you use the terminology as shown in this link, its one of those points that is often raised.
https://en.wikipedia.org/wiki/Ground_(electricity)
 
Last edited:
I believe part of the confusion is your understanding of the word "ground" as used in SPICE simulators. It refers specifically to the 0V or circuit "common" point. It has no further meaning beyond that. It has nothing to do with earth or chassis ground that may or may not be connected to this common point. Thus the "Ground" symbol is used (and required) by spice to designate the 0V point in the circuit since it has no other way of knowing that.

And "ground" is commonly (and perhaps inaccurately) used interchangeably with "common" in many circuit descriptions and schematics without meaning earth or chassis ground.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top