Quick Verfication--> Floating power supply

Thread Starter

chimera

Joined Oct 21, 2010
122
Hello! I jut wanted a quick verification regarding a small concept I seem to have confused.

I am building a DC power supply and I have attached the basic schematic. Its the traditional power supply: AC to DC rectification, smoothing cap etc.

My question is:

Is this power supply, a floating power supply?

I seem to be unclear becaz according to the definition, I dont have a earth return (ground) for my power supply. That is partially becaz my transformer is not center-tapped. Therefore, all I have is a positive DC supply and with a common return to complete the circuit. There is no actual ground present in my design then, is there?
 

Attachments

chrisw1990

Joined Oct 22, 2011
551
its my understanding that what you have posted is a floating supply, unless, the ground for your supply (e.g.: +5V and GND) is tied to the earth of the mains, to give it a reference, make sense?
 

Thread Starter

chimera

Joined Oct 21, 2010
122
its my understanding that what you have posted is a floating supply, unless, the ground for your supply (e.g.: +5V and GND) is tied to the earth of the mains, to give it a reference, make sense?
Exactly! it is floating. In the picture that I posted, I have a ground symbol but that is the RETURN back to the rectifier-->not an actual ground reference (earth)

Basically, What I have is then a simple FLOATING power supply that gives out positive voltages.--> +ve and a common terminal (if I was to build it up)

Now, I want to have an actual ground becaz if I dont, I wont be able to use my oscilloscope to measure stuff becaz that scope is ground referenced to actual earth (0Volts).

However, on a different site, someone replied the following to similar question that I have regarding the o-scope. Please explain/elaborate what the bold statement means below:

When you connect the clip of the scope probe to the common of your floating power supply then you are indeed connecting common to chassis (Earth) ground. The power supply should also have a chassis terminal where the common of your supply can be connected. In both cases it does not matter since you are tying common and earth (two reference points) together.

One benefit of a floating power supply is that you can put them in series for a larger voltage.

What one cannot do is, for example, measure the voltage across a transformer that is connected to the mains with a grounded scope. Sparks will fly
 
Last edited:

chrisw1990

Joined Oct 22, 2011
551
Now, I want to have an actual ground becaz if I dont, I wont be able to use my oscilloscope to measure stuff becaz that scope is ground referenced to actual earth (0Volts).
When you connect your ground clip you will be referencing the oscilloscope to the ground of the local supply, i.e. your oscilloscope will measure voltages with reference to the ground where you are measureing the voltage, as long as you connect it correctly..

as for that bold comment.. im not sure there, it seems to imply that when you connect the ground clip, youre tying it to earth, thats wrong, as i have measured relative voltages before, and not just from the ground..

hope this is all making sense
 

Wendy

Joined Mar 24, 2008
23,415
You show a ground on your schematic, but in this context it is meaningless.



Ground is generally a common point in a schematic where all voltages and signals are referenced.

Since you are using a transformer the power supply is isolated. If you were to connect the AC ground then you would have problems. I would not recommend it in this case. You can always ground the output of the power supply after the DC voltage has left the case, this is the norm.
 

Attachments

Thread Starter

chimera

Joined Oct 21, 2010
122
When you connect your ground clip you will be referencing the oscilloscope to the ground of the local supply, i.e. your oscilloscope will measure voltages with reference to the ground where you are measureing the voltage, as long as you connect it correctly..

as for that bold comment.. im not sure there, it seems to imply that when you connect the ground clip, you're tying it to earth, thats wrong, as i have measured relative voltages before, and not just from the ground..

hope this is all making sense
So basically, it doesn't matter if I have an actual ground (earth) connection present in my supply, I will be able to measure correctly using the 0-scope since I am referencing the scope's ground to the common of DC-supply (as shown in my picture attached)


You show a ground on your schematic, but in this context it is meaningless.
Ground is generally a common point in a schematic where all voltages and signals are referenced.

Since you are using a transformer the power supply is isolated. If you were to connect the AC ground then you would have problems. I would not recommend it in this case. You can always ground the output of the power supply after the DC voltage has left the case, this is the norm.
The ground symbol in the picture is there becaz I needed to simulate the circuit. I agree the ground symbol is not needed and I mention that in my beginning post of this thread.

Can you please elaborate on, with a visual aid (if possible), what you mean when you say

" You can always ground the output of the power supply after the DC voltage has left the case, this is the norm".

SO BASICALLY, here is what I am getting at:

1- I do not need a proper earth-ground connection (the housing of this supply is plastic--so no shock hazard)

2- I can use o-scope with FLOATING power supply by placing the ground lead of the probe to the ground('common') and have the scope reference to that point (common).

3- I AM using a transformer and hence the power supply IS isolated. Of course I would not connect my 0-scope to mains.

4- If I am able to understand/employ Bills idea of "ground the output of the power supply after the DC voltage has left the case" (provided he explains it :) ), what advantage would that serve?

Is there any point I am missing out? Thanks!!!!
 

chrisw1990

Joined Oct 22, 2011
551
basically, if your power supply were being used locally (powering one item) it wouldn't matter if its floating.. if you're using the power externally.. you might need to reference it.. i guess for communications, unless you used a differential signal like RS485..
 

Thread Starter

chimera

Joined Oct 21, 2010
122
basically, if your power supply were being used locally (powering one item) it wouldn't matter if its floating.. if you're using the power externally.. you might need to reference it.. i guess for communications, unless you used a differential signal like RS485..
its general purpose bench power supply... how about then? I needed an extra one. So I figured I build one my self and get the whole 'every electronics design engineer must, at least once, design his/her own power supply'
 

chrisw1990

Joined Oct 22, 2011
551
i wouldnt worry so much about it, because its still being used locally, if that circuit was then communicating with another circuit on another supply, say through SPI or I2C, then a common ground would be needed.
 

Thread Starter

chimera

Joined Oct 21, 2010
122
i wouldnt worry so much about it, because its still being used locally, if that circuit was then communicating with another circuit on another supply, say through SPI or I2C, then a common ground would be needed.
I see.. so end result is:

Power circuits(multiples, if needed) up from the same power source. Don't mix up two different power sources or circuits powered by different power supplies till I am sure that both power supplies are floating and have a common ground.

You know, this stuff is basic and I never thought over it as i used my current power supply for 3 years, built tonnes of circuits using it.. but now that I want to design one (and yes, its NOT gonna be based on an LM317)... I am realizing that not a lot of ppl take the basics into account when designing a bench power supply.

Do you know of a good reading source where I can look this stuff up? I kind of googled it..but sometimes there is just no substitution for a solid reference/reading material

Thanks for all your help!
 

jtrent

Joined Mar 11, 2012
26
You can read anything with your scope if you connect the scope's power cord to an AC power outlet using a 3 prong to 2 prong adapter. This removes earth ground from the scope. So now the scope is floating and it's reference/common and the power supply or any circuit with floating common can be connected together to take readings. I've done it many times.
 

Thread Starter

chimera

Joined Oct 21, 2010
122
You can read anything with your scope if you connect the scope's power cord to an AC power outlet using a 3 prong to 2 prong adapter. This removes earth ground from the scope. So now the scope is floating and it's reference/common and the power supply or any circuit with floating common can be connected together to take readings. I've done it many times.
Sounds sketched to me.. but I suppose the logic makes sense in what you say.. I will consider that.. but for alternative, I want to verify something:

I have researched and often ppl have mentioned that they simply connected the common terminal of the (DC rectified voltage) circuit to the EARTH connection. In this way, there is a true 0V reference.

I really dont need a negative voltage (i prefer to work with positive voltages).. so I figured I can connect the common(return) of my circuit to a EARTH connection.

The transformer I am using is not a Center-tapped one; this means that the EARTH connection I am going to using, will come from a 3 pronged power switch which I will be using to deliver the power. So, this EARTH connection is what I am going to be using to connect the common(return) terminal to true 0V.

Is that safe? I have never built a power supply before, so I hope everyone can see and understand my concerns.
 

Thread Starter

chimera

Joined Oct 21, 2010
122
Sounds sketched to me.. but I suppose the logic makes sense in what you say.. I will consider that.. but for alternative, I want to verify something:

I have researched and often ppl have mentioned that they simply connected the common terminal of the (DC rectified voltage) circuit to the EARTH connection. In this way, there is a true 0V reference.

I really dont need a negative voltage (i prefer to work with positive voltages).. so I figured I can connect the common(return) of my circuit to a EARTH connection.

The transformer I am using is not a Center-tapped one; this means that the EARTH connection I am going to usi
 

Wendy

Joined Mar 24, 2008
23,415
You really do not need to worry about it. Leave the power supply floating. Connect the ground of the oscope to the negative lead, the probe to the positive. Simple.
 

Thread Starter

chimera

Joined Oct 21, 2010
122
You really do not need to worry about it. Leave the power supply floating. Connect the ground of the oscope to the negative lead, the probe to the positive. Simple.
Bill, I appreciate the fact you are giving me advice and I know you are talking out of experience, but can you please explain to me why I should for go my attempt to connect the common of my power supply to EARTH. In this way, I can have true 0V reference.

I would like to know the reason behind why you are saying when you said I should not worry about it; it will help me understand fundamentally what I dealing with :)
 
Top