# Step-down transformer secondary center tap to chassis ground?

#### maw455

Joined Jul 20, 2016
30
Hello:

I'm an amateur trying to do my homework before attempting to build a small, regulated bench power supply and have a question about center-tapped step-down transformers, specifically what to do with the secondary center tap, i.e., leave it unconnected or connect it to chassis ground when using a full-wave bridge rectifier (4 diode). I'm in the US using standard 120V household voltage as a mains supply.

Say I have a 15V 300 mA step-down transformer 7.5-0-7.5V, and I want to get 15VDC using an off-the-shelf full-wave bridge rectifier (4 diode). As I understand it, there will be a voltage drop across the diodes and that with no load the output voltage will be higher than 15V, but I'm ignoring that for purposes of this question. To do this, I connect the two outside legs of the secondary winding to the transformer input legs of the bridge rectifier. The positive output of the bridge rectifier becomes the +DCV for the circuit, and the negative output becomes my 0V, correct?

My question is what to do with the un-used center tap from the secondary? As I understand it, leaving it unconnected will not hurt anything other than leaving the circuit floating (i.e., no earth ground). From what I can uncover by searching here and elsewhere, connection of the center tap to the chassis, which is ultimately connect to earth ground via the AC earth ground, protects against electrocution at 120VAC in the event of a short between the primary and secondary windings of the transformer. If correct, that makes sense except for one thing - I know the center tap is connected to the halfway point of the secondary winding, so there is continuity between each outside leg and the center tap. So, why doesn't grounding the center tap to earth via the chassis cause the chassis to be energized? Please help me understand.

Thanks

#### Papabravo

Joined Feb 24, 2006
21,264
Ignore the center tap unless you want to build a bipolar supply. You should also know that 15 VAC is RMS(Root Mean Square). The peak voltage of the AC waveform is 15VAC*SQRT(2) = 21.21 Vpk. Allow 1.4 volts for the bridge rectifier and you should get about 19.8 VDC out of the filter capacitor. You would then regulate this down to +15VDC.

#12

#### crutschow

Joined Mar 14, 2008
34,714
For safety, you connect the chassis to the main's safety ground.
Leave the center tap unconnected.
Typically a bench power supply has a separate chassis ground terminal at the front, along with the power terminals.
You can then connect the chassis to your circuit ground or not as desired.

#### maw455

Joined Jul 20, 2016
30
Ignore the center tap unless you want to build a bipolar supply. You should also know that 15 VAC is RMS(Root Mean Square). The peak voltage of the AC waveform is 15VAC*SQRT(2) = 21.21 Vpk. Allow 1.4 volts for the bridge rectifier and you should get about 19.8 VDC out of the filter capacitor. You would then regulate this down to +15VDC.
Thank you Papabravo. When you say "Ignore the center tap", do I leave it unconnected to anything, or connect it to earth/chassis for protecting against a short between the primary and secondary?

#### Papabravo

Joined Feb 24, 2006
21,264
"Ignore the center tap" means you leave the center tap unconnected for your application.

#### maw455

Joined Jul 20, 2016
30
"Ignore the center tap" means you leave the center tap unconnected for your application.
Thanks. Sorry to be dense. So, does that mean the suggestion to connect the center tap to earth ground to protect against shorting across the main and secondary windings of the transformers is not correct?

#### Papabravo

Joined Feb 24, 2006
21,264
Where did that advice come from and how does it offer any protection? Normally the steel laminations are bolted to the chassis and they are insulated from both the primary and the secondary windings. The windings are also insulated from each other. So explain to me how chassis grounding the center tap will do anything except prevent you from making a 15VDC supply

#### maw455

Joined Jul 20, 2016
30
It did not make any sense to me either, which is why I asked the question in the first place.

#### #12

Joined Nov 30, 2010
18,224
The mistake waiting for you is that the negative output of the bridge rectifier is going to become your zero voltage point for the regulated output. If you ground the center tap, and then ground your zero voltage point, you will short out the rectifier bridge, then the smoke will escape.

#### hp1729

Joined Nov 23, 2015
2,304
Hello:

I'm an amateur trying to do my homework before attempting to build a small, regulated bench power supply and have a question about center-tapped step-down transformers, specifically what to do with the secondary center tap, i.e., leave it unconnected or connect it to chassis ground when using a full-wave bridge rectifier (4 diode). I'm in the US using standard 120V household voltage as a mains supply.

Say I have a 15V 300 mA step-down transformer 7.5-0-7.5V, and I want to get 15VDC using an off-the-shelf full-wave bridge rectifier (4 diode). As I understand it, there will be a voltage drop across the diodes and that with no load the output voltage will be higher than 15V, but I'm ignoring that for purposes of this question. To do this, I connect the two outside legs of the secondary winding to the transformer input legs of the bridge rectifier. The positive output of the bridge rectifier becomes the +DCV for the circuit, and the negative output becomes my 0V, correct?

My question is what to do with the un-used center tap from the secondary? As I understand it, leaving it unconnected will not hurt anything other than leaving the circuit floating (i.e., no earth ground). From what I can uncover by searching here and elsewhere, connection of the center tap to the chassis, which is ultimately connect to earth ground via the AC earth ground, protects against electrocution at 120VAC in the event of a short between the primary and secondary windings of the transformer. If correct, that makes sense except for one thing - I know the center tap is connected to the halfway point of the secondary winding, so there is continuity between each outside leg and the center tap. So, why doesn't grounding the center tap to earth via the chassis cause the chassis to be energized? Please help me understand.

Thanks
Center tapped secondary options.
You want the top one?

#### Attachments

• 40 KB Views: 34

#### Papabravo

Joined Feb 24, 2006
21,264
It did not make any sense to me either, which is why I asked the question in the first place.
I'd still like to know where the advice that confused you came from.

#### Tonyr1084

Joined Sep 24, 2015
7,965
The "Dumbest" question is the one nobody bothered to ask. It's always better to ASK a dumb question and learn something than to neglect to ask - then learn the hard way.

ANY question you have is valid. And it's also good to get more than one answer. If a debate erupts then more people have a chance to learn something.

I've given dumb answers. Actually, WRONG answers. And others have kindly (usually) corrected me. So don't feel dumb for asking. It's certainly better than LOOKING dumb for blowing up your project.

Funny story: I built a 12 volt supply for a car radio. It had a horrible hum. No capacitor filtration on it. Someone told me to put a cap on it. So I tried in several places. Nothing I did helped (cap was too small). So I tried putting it on the mains (120 volt side of the transformer). As you read this - know that lots of others who've read this are laughing. 120 volts AC across a 16 volt electrolytic capacitor? BOOM! No, not smoke leaking out - all out BOOM. And the can the capacitor once lived in evacuated the location in a BIG hurry. I guess I should have asked more questions before building my supply.

Last edited:

#### maw455

Joined Jul 20, 2016
30
Center tapped secondary options.
You want the top one?
I believe so, but in that schematic what is shown as earth ground (vs. chassis ground on the AC side of the transformer) is actually common 0v for the entire circuit, connected to the negative output of the rectifier, correct?

Something like this schematic:

Different mains voltage and regulated voltage and related values for caps, resistors, etc., but the same basic design as what I'm contemplating. Is my understanding correct?

#### Papabravo

Joined Feb 24, 2006
21,264

I readily admit that I may be misinterpreting the information in this thread, which is why I wanted to clarify things.

Thanks
I don't see that there was a definitive answer to the original question. The evidence suggests that bad things happened when the CT was grounded which is consistent with what others have said. The only reason to do it would be to make a bipolar supply. See the following page for details. NOTE carefully the difference between chassis ground and DC ground.

http://www.bristolwatch.com/ele/power_supplies.htm

#### maw455

Joined Jul 20, 2016
30
I don't see that there was a definitive answer to the original question. The evidence suggests that bad things happened when the CT was grounded which is consistent with what others have said. The only reason to do it would be to make a bipolar supply. See the following page for details. NOTE carefully the difference between chassis ground and DC ground.

http://www.bristolwatch.com/ele/power_supplies.htm
Thanks. I believe I now understand things much more clearly. I definitely do not want a bipolar supply. So, the answer to my original question would be to leave the CT unconnected, protecting it from shorting out with shrink tube or some insulating material over the end, and use the negative output of the bridge rectifier as the 0V or common connection for the circuit's negative connections, with nothing in the circuit connected to earth ground, correct?

Thanks.

#### KeepItSimpleStupid

Joined Mar 4, 2014
5,088
So, the answer to my original question would be to leave the CT unconnected, protecting it from shorting out with shrink tube or some insulating material over the end,
So, your the guy that has plastic outlet protectors all over the house to keep the electricity from leaking out?

The low voltage stuff is usually left alone especially when the xformer is in the case.

FWIW, some transformers have an electrostatic shield that is sometimes grounded. The transformer metal chassis being grounded and the metal case being connected to ground can protect against the 120 VAC shorting to ground.

I fixed an amp where the primary was shorted to ground, but since it was a 2 prong outlet it just made the receiver live. Not good.

and use the negative output of the bridge rectifier as the 0V or common connection for the circuit's negative connections, with nothing in the circuit connected to earth ground, correct?
Ground is a really weird concept. The big point is that there should be only one place where your circuit gets connected to ground if it gets connected at all.

We can talk about ground being a REFERENCE or as PROTECTIVE GROUND or a COMMON terminal.

A house in the US is not wired like a piece of electronic equipment. If it was, then there would be no outlets that would be in parallel. Every one would terminate at the breaker panel. Every one would also have a protecttive ground and a reference ground.

If your lucky, that ground is a reference because it doesn't carry any current.

When a fault develops, it does carry a current and hopefully the breaker pops.

When you have a bunch of parallel outlets and the last one develops a fault. all of the other outlets frame of reference moves because there is current in the ground wire. If that was a single outlet wired to the breaker panel, the reference to the house doesn't move relative to the reference.

#### maw455

Joined Jul 20, 2016
30
So, your the guy that has plastic outlet protectors all over the house to keep the electricity from leaking out?

The low voltage stuff is usually left alone especially when the xformer is in the case.

FWIW, some transformers have an electrostatic shield that is sometimes grounded. The transformer metal chassis being grounded and the metal case being connected to ground can protect against the 120 VAC shorting to ground.

I fixed an amp where the primary was shorted to ground, but since it was a 2 prong outlet it just made the receiver live. Not good.

Ground is a really weird concept. The big point is that there should be only one place where your circuit gets connected to ground if it gets connected at all.

We can talk about ground being a REFERENCE or as PROTECTIVE GROUND or a COMMON terminal.

A house in the US is not wired like a piece of electronic equipment. If it was, then there would be no outlets that would be in parallel. Every one would terminate at the breaker panel. Every one would also have a protecttive ground and a reference ground.

If your lucky, that ground is a reference because it doesn't carry any current.

When a fault develops, it does carry a current and hopefully the breaker pops.

When you have a bunch of parallel outlets and the last one develops a fault. all of the other outlets frame of reference moves because there is current in the ground wire. If that was a single outlet wired to the breaker panel, the reference to the house doesn't move relative to the reference.
I'm not that AR - ;-) No, my concern stemmed from what Papabravo said above: "The evidence suggests that bad things happened when the CT was grounded which is consistent with what others have said." So, it follows that if the CT does have voltage on it, that if it were to be left unconnected and at some point get in contact with the metal case because of movement, etc., it could cause some issues with the circuit. Plus, that's AC current, not DC current, at the CT.

My knowledge of DC circuits is primarily from automotive work, where there is never an "earth ground", and chassis ground and common are the same grounding point back to the negative terminal of the battery. I assume the same concept applies in DC circuits for DC power supplies. However, the circuit designs I have seen vary in the way they identify common or negative. For example, some have the circuit negative connection points all connected together, back to the negative voltage source, and labeled "-" or "0V" or "Common". This my feeble brain understands. Others have all negative points terminating individually at what I typically identify as "earth ground" by using the symbol for that. This is what is confusing for me because, in automotive DC circuits that symbol means common or 0 volts because it is going back to the negative post of the battery and there can never be an earth ground unless your car is running on its rims. Plus, it must be connected back to the negative post of the battery or the circuit will be broken at that connection point.

In power supplies and other DC circuits with AC mains, there IS an earth ground on the AC Mains side if using a 3-conductor supply. So, when I see that someone has used the symbol for earth ground on the DC side of the circuit, it makes me wonder which "ground" they really mean.

#### BobTPH

Joined Jun 5, 2013
9,186
The symbols for earth, chassis and common grounds are abused all the time. Consider them interchangeable.

Bob