Spikes in Power Supply

[premkumar9]

Hello,
I am attaching the photograph of noise coming on my 12V supply line.How can I eliminate such noise?

 

[Colin]

Is the power supply connected to a project?

 

[MikeMl]

If it is a SMPS, live with it, or take the damn thing out and shoot it with a gun.

 

[MCU88]

If it is a SMPS, live with it, or take the damn thing out and shoot it with a gun.

I have an lifetime ban on AAC for sticking up for you. I won't lose any sleep though

 

[premkumar9]

If it is a SMPS, live with it, or take the damn thing out and shoot it with a gun.

Not SMPS. Series regulator supply.

 

[premkumar9]

Is the power supply connected to a project?

Without load and with load, the change is marginal only.

 

[audioguru]

A linear regulated supply will make the output noise if it uses a lousy old 741 opamp in its circuit. Please post the schematic.

 

[MikeMl]

Not SMPS. Series regulator supply.

Post the circuit. Likely missing capacitors or bad capacitor.

 

[Nigel Goodwin]

Looking at the scope settings it's only 20mV p-p anyway

 

[premkumar9]

Looking at the scope settings it's only 20mV p-p anyway

Do you mean that that much level is acceptable for a series regulator? Is there no way to eliminate that?

 

[crutschow]

Do you mean that that much level is acceptable for a series regulator? Is there no way to eliminate that?

Two choices are to modify/fix the circuit (if you have its schematic) or add an LC filter on the output.

Edit: That noise looks like it's coming from a switcher, not a linear supply.

 

[MCU88]

Do you mean that that much level is acceptable for a series regulator? Is there no way to eliminate that?

5mV peak-to-peak is a spec quote from many half decent lab power supplies. Trying 100nF capacitors on the rails. Try an LC line filter.

 

[premkumar9]

Theoretically what could be the reason for such spikes? Where do they originate from?

 

[Nigel Goodwin]

Do you mean that that much level is acceptable for a series regulator? Is there no way to eliminate that?

I would imagine it's perfectly acceptable, I'm also confused as to why you were scoping it in the first place?

 

[MikeMl]

What kind of regulator?

 

[Tony Stewart]

Voltage spikes are a simple result of Ohm's Law. Current Spike * ESR of Cap.

To improve this impedance ratio consider adding a series ferrite bead and lower ESR cap to ground measured in milliohms from spec.

But 10mV peak is fairly good. If you need much lower then specify or cascade more RLC or RC filtering, but series R affects load regulation by load impedance ratio as well.

 

[JimB]

First, find where the noise is coming from.
My first thought is that it does not come from a linear regulator.
Could be coming down the mains wiring from who knows what, who knows where.

It look like some of the crud which comes out of the control system for my CNC milling machine.

JimB

 

[MikeMl]

I ask: when you recorded the noise, where was the scope probe ground clip?

 

[audioguru]

18 replies and we still do not have a schematic. I think the power supply was purchased and is either welded together or it is cast in concrete so it cannot be opened to see what is inside.

 

[Tony Stewart]

I presume you can get a flat line on the closest ground pin.

If not then remove gnd clip and probe tip and use probe barrel and pin between 2 pins or wires added for low noise probing.

If still you have noise ingress, make an antenna with probe gnd grounded to tip and locate the signal or make one.