Author Topic: Rigol DS1052E self pickup of power supply noise (Read 9562 times)

Lightages · « **on:** January 16, 2012, 06:04:54 am »

I don't remember this being a problem earlier with my scope but something seems to be getting worse. My scope seems to picking up the internal power supply noise much worse than before. It now is making the scope almost unusable for many things.

I was testing a Chinese power supply and preparing for a review here while using my scope. Any noise or problems I might be able to observe is being swamped by this problem. The noise it is picking up is 125MHz ringing pulsing at 135Kz at around 40mV PP. I know other have complained of noise problems from the power supply but has anyone come up with a definite fix for this?

I am not sure I would recommend this scope anymore to anyone.

saturation · « **Reply #1 on:** January 16, 2012, 01:05:38 pm »

Do you have photos? There are posts of A Hellene in the archives regarding system noise but you suggest yours was better before and is worsening?

RJSC · « **Reply #2 on:** January 16, 2012, 05:33:48 pm »

Mine picks noise even when probing an unconnected power track on a breadboard (ground and tip on the same track), but if i short the ground clip to the tip of the probe its completely flat. If you short the probe ends does it pick noise?

I was checking a shunt regulator circuit with tl431 with no load connected and always got 40 to 80 mV noise.

I thought "but... shunt regulators are supposed to be super clean".

I disconnected the power to the circuit and the same noise was still there (but around 0V).
Then I removed all the components from the breadboard, probed the ground rail and the same noise was there.

I thought it must be external EMI, but I'm not sure what to think now...
Is it the scope, or is it external? Shorting the probe tips kills internal noise?

Lightages · « **Reply #3 on:** January 16, 2012, 08:58:07 pm »

Quote from: saturation on January 16, 2012, 01:05:38 pm

Do you have photos? There are posts of A Hellene in the archives regarding system noise but you suggest yours was better before and is worsening?

I do not remember this being a problem a few months ago when I last used this scope.

Here are the images. The first is with no probe attached. The second with the probe attached and grounded though the ground lead. The third is a closeup of the pulse. This makes it very difficult if not impossible to look at signals at high sensitivity settings on the scope.

RJSC · « **Reply #4 on:** January 16, 2012, 09:02:27 pm »

Maybe your probe is damaged. Since it doesn't pick up the spikes without the probe.

Lightages · « **Reply #5 on:** January 16, 2012, 09:07:27 pm »

It is the same with different probes, different ages, different types, different manufacturers.

tinhead · « **Reply #6 on:** January 16, 2012, 09:12:44 pm »

or maybe your neighbourhood is making cookies, or Asgard spaceship flying near your house ...

> probe attached and grounded though the ground lead <

this is the best antenna you can get!, i would worry if there would be nothing on the display with such "test setup".

Lightages · « **Reply #7 on:** January 16, 2012, 09:26:35 pm »

It would appear that you are correct. I shorted out the probe tip directly with a piece of aluminum foil and the signal goes away. Even so, this means that the power supply, in my scope at least, is extremely noisy. I don't believe that I should have to worry about the scope itself interfering with its own measurements. I really need to look at how to kill this EMI from the power supply.

Perhaps I didn't see this before because I had not tried to make such sensitive measurements before with this scope, but I can't remember.

amspire · « **Reply #8 on:** January 16, 2012, 10:44:41 pm »

The problem could be the scope power supply, of it could be you have another noisy inverter nearby.

Take the scope to a totally different place like a friends house and see if you have the same problem. That kind of noise is caused when a semiconductor in a switching power supply switches off causing ringing with the leakage inductance of a transformer. It could be a transistor of Mosfet on the primary, or a fast diode turning off on the secondary.

A failure in parts that are meant to damp out the ringing could cause it to get very bad, and it is the type of problem that can cause power supplies to fail.

What frequency does the Rigol power supply switch at? I waved a probe around the back of my Rigol looking for pickup, and it looks like it may be a flyback supply running at about 135KHz. If it is, I am not getting any excessive ringing, so if your noise is coming from the Rigol power supply, then there may be a fault that is probably fixable. I found that touching the shield of the BNC connector on the back (which looks like an isolated output) picked up a superb power supply waveform. If the same thing works for you, then you can see if your noise problem is in sync with the Rigol power supply or not.

If it is the Rigol power supply, I can give some suggestions towards tracking down the problem.

Richard

Lightages · « **Reply #9 on:** January 16, 2012, 10:59:57 pm »

It is most definitely the scope power supply. When I hold the probe nearer the scope the noise goes up, and when I connect the probe to the probe compensation ground it is also evident.

amspire · « **Reply #10 on:** January 16, 2012, 11:10:49 pm »

Quote from: Lightages on January 16, 2012, 10:59:57 pm

It is most definitely the scope power supply. When I hold the probe nearer the scope the noise goes up, and when I connect the probe to the probe compensation ground it is also evident.

But that does not exclude noise that is entering the scope through the mains power cable.

If the noise is not synchronised to the power supply switching waveform, then it is probably not from the scope. Put an ungrounded scope lead on the rear BNC and see if you see a noise picuk from the power supply on the 20mV range. If you do, is it sync'ed to the problem noise?

You can capture the waveform if you like and I can compare it to my waveform.

Lightages · « **Reply #11 on:** January 16, 2012, 11:36:32 pm »

Yes it is sync'd with the noise on the BNC on the back. Most definitely the scope's power supply is causing the noise.

amspire · « **Reply #12 on:** January 17, 2012, 12:36:38 am »

That is clear.

I get the same waveform on my scope without the spike that you are getting, so I think something has failed in your power supply.

If you feel like fixing it yourself, you obviously need to open the scope up.

Don't try and probe anything directly. If the ringing is on the primary, the amplitude of the ringing at over 100MHz will probably exceed your probe's ratings. What you want to do is to work out if the ringing is on the primary or the secondary. to do this,add some tape or something to a scope probe so it cannot make contact with anything and just use it as a proximity probe. In particular, you are looking at the switching transistor(s) on the primary, or the rectifiers on the secondary. The ringing should be much stronger on one side of the transformer to the other side.

From the waveform, it looks almost certain that the ringing happens when the switching transistor turns off. The other lower frequency oscillation at the end of the plateau is totally fine - that is standard in a flyback switching power supply where the core current is allowed to decay to zero.

If it turns out to be the primary, then as a start, it is not the switching transistors/mosfets or the transformer with the problem unless the transformer has an extra winding that is used only for a snubber diode circuit, and that winding has gone open circuit. See what other components are connected to the primary windings. If there is a snubber circuit, it could consist of resistors, capacitors or diodes.

Richard.

Lightages · « **Reply #13 on:** January 17, 2012, 12:59:33 am »

Actually I have some 100:1 250MHz probes. I will start digging when I have time. Thanks for the pointers.

amspire · « **Reply #14 on:** January 17, 2012, 01:32:40 am »

I just pulled out my scope again, and repeated the test. My waveform on the rear BNC is actually almost identical to yours, but I do not have it getting into the scope channel at all.

I had my scope set to dots modes before, and the ringing wasn't as obvious. The ringing on the switching waveform is 25MHz, and your channel noise was 125MHz.

Can you zoom in and see what frequency the switching noise is in the rear BNC connector waveform?

Richard

amspire · « **Reply #15 on:** January 17, 2012, 01:49:07 am »

Quote from: amspire on January 17, 2012, 01:32:40 am

I just pulled out my scope again, and repeated the test. My waveform on the rear BNC is actually almost identical to yours, but I do not have it getting into the scope channel at all.

I had my scope set to dots modes before, and the ringing wasn't as obvious. The ringing on the switching waveform is 25MHz, and your channel noise was 125MHz. If you are getting 125MHz ringing on your power supply and I am getting 25MHz, that would imply that I have some extra capacitance present that you do not have.

Can you zoom in and see what frequency the switching noise is in the rear BNC connector waveform?

Richard

Lightages · « **Reply #16 on:** January 17, 2012, 04:43:14 am »

The frequency in the narrow positive glitch is 25MHz.

amspire · « **Reply #17 on:** January 17, 2012, 05:02:09 am »

Your power supply is working exactly the same as mine - possibly does not need fixing. As I said, I have no channel noise spikes at all - just a few pixels of random noise.

So the 125MHz channel glitch is not coming directly from the power supply. and if you directly short the input socket, the noise goes away.

Is it being radiated from the power cable? Does the amplitude increase if you move the probe close to the power cord? If it does, you can get the clip on ferrite RF suppressors.

Also, could it be a ground failure in the probes. I wonder if your 100:1 probe will have the same problem?

saturation · « **Reply #18 on:** January 25, 2012, 11:55:43 am »

I did a check on my scope and find my PSU noise is for the most part undetectable even with probes on and maximum vertical gain. When I use the probe as an aerial I do pick up more noise only if I put the setup right ontop of the scope. If you have more metallic structures around it could be serving as unintended aerials. If you add more shielding to the scope it might suppress the noise.

big t · « **Reply #19 on:** January 31, 2012, 06:53:51 pm »

Hello,

My advice, try to change PSU (and in PSU ways on PCB) electrolytic capacitors to low ESR for example Panasonic FC or FM series.

Lightages · « **Reply #20 on:** January 31, 2012, 07:07:20 pm »

Now that might be worth trying. Or alternatively paralleling some ceramics with the electrolytics.


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: Rigol DS1052E self pickup of power supply noise (Read 9562 times)

Share me