How Thick is Your Baselne on Your Digital Oscilloscope?

[Paul Price]

How thick is your baseline?

One of the tings I've always liked about my old analog oscilloscopes is the low noise baseline display. I see a nice thin flat line. By positioning the free running trace at the bottom baseline and  just using the graticule and subdivisions for readout I can make DC voltages measurements quickly and quite accurately.

It was so convenient to use my scope as a voltmeter, with it's tiny 1 or 10x probe tip, to see all the important  voltages at a glance, and not upset the circuit as what happens when using the unshielded probes on my trusty Fluke DVM.

I have a RIGOL DS4024 and viewing the baseline, without any probe connected or with a 10x probe connected, always has a thick baseline, at 200mHz bandwidth, a fairly thick one(.2 major div), with 100mHz bandwidth(>1 minor div.) and with 20 MHz bandwith setting still about 1 minor div.  And there is no noise nearby.The noise source is within the scope itself. My 20MHz analog scope sitting directly below it has a nice flat baseline under the same conditions. My Tek 60-MHz 465 has also a beautiful flat baseline.

But at any bandwidth, and with the input set to Ground I finally see the beautiful flatline display that I am so used to seeing with my analog scope.

I am wondering if I bought the wrong scope. I really miss my baseline!

Does anyone have a Tek DS403x series baseline picture with no averaging and just a probe connected. I've heard that 9-bit vert. digitization give a better baseline.

If anyone has an approx. equiv. model of an Agilent or a  Tektronix DSO or DPO, how thick is your (Non-averaged) baseline?  Please post a picture!?

[jpb]

My scope is at home (a LeCroy WaveJet) and I'm at work but my measurements of the noise level were in line with those reported in Agilent's app note:

http://cp.literature.agilent.com/litweb/pdf/5989-3020EN.pdf

The main issue with a digital scope is it is only 8 bit (though more in high res mode). If you limit the bandwidth and turn on averaging the line is a lot smoother of course.

If the scale is set up properly, and you're not trying to measure small voltages below a few mV, then I would say the noise is around one or two bits so you should be able to measure to between a tenth and a fifth of a division by eye.

The advantage of a digital scope is it can also give you a mean value as a number which is probably more accurate.

What you need to ask is for people to post pictures of zero input on a scale that you're interested in, then we could do some sort of comparison.

To be fair the bandwidth needs to be limited to the same amount otherwise wider bandwidth scopes would be penalised. Also I presume this needs to be with probe attached?

[SLJ]

Are you using the same prob on both scopes to compare the two?

[Paul Price]

Yep. Same probe.

[nctnico]

You probably have peak-detect active. The analog scope doesn't show the noise. The digital does.

[AndyC_772]

It's not really a fair comparison, because the noise bandwidth is completely different between the old 20 MHz scope and your newer 200 MHz scope. What do you get if you engage the 20 MHz bandwidth filter on the Rigol?

If the two scopes have a similar noise figure in terms of V/sqrt(Hz), then a scope with 10x the bandwidth would be expected to show a line more than 3x thicker with no input signal for this reason alone. It's one reason scopes include bandwidth limiting filters that you can turn on. At a guess, when you select GND coupling on your Rigol, it's either a software feature that ignores the ADC output entirely, or it couples the ADC input directly to a fixed voltage and completely bypasses the front-end amplifiers.

Also, one of the major advantages of a digital scope over an analogue one works against it here. On the analogue scope, you see a line whose cross-section accurately reflects the temporal distribution in voltage, ie. it's brightest in the middle and rapidly decays to an invisibly low level as you look above and below. In other words, it has very high dynamic range in the 'Z axis'.

The digital scope's display system, however, tends to exaggerate the brightness of infrequent captures, and though this is normally a benefit as it makes infrequent events more clearly visible, it does make the baseline appear thicker than arguably it should be.

[Paul Price]

I don't have peak detect active making this test.
No special features turned on at all, but I compare the effects of bandwidth limitation on the Rigol DS4024.

[KedasProbe]

You may want to put High Resolution on, not sure where it is located in the Rigol.

[Harvs]

You may want to put High Resolution on, not sure where it is located in the Rigol.

DS2000 is the same (as is I imagine most DSOs with graded intensity displays).  It's not something that I've really ever thought about, as I just turn on Avg voltage measurement function (it's been about 5yrs since I've used a CRO though.)

Pretty much anything you turn on like Hi Res mode or averaging (which does significantly improve the line width) kills the update rate, so kind of a no win situation there.

[Paul Price]

KedesProbe:  Correction: Hi-Res setting can be tuned on, traces becomes very thin and dim, update rate of display about 1 sec/update at 1mS/div horiz. sweep. on the DS4000 series

[dfmischler]

Here's a trace from the 50 MHz Tek TDS310 I keep in my office.  Not as nice as the trace on my 250 MHz Tek 475A, but you need storage for this, and it was $125 with 2 Tek P6109B probes.  I get about the same amount of noise on my TDS380.

[Harvs]

KedesProbe:  No Hi-Res setting available on the DS4000 series

According to the manual it does.  It's the same as the DS2000, under the Acquire menu.

[KedasProbe]

KedesProbe:  No Hi-Res setting available on the DS4000 series

According to the manual it does.  It's the same as the DS2000, under the Acquire menu.

Yes, I just checked it the lab, it's there.

[Paul Price]

Thanks dfmischler for the picture of the analog scope trace.

KedasProbes and Harvs: Thanks for reminding me where the Hi-Res is located.  Not too easy to setup if I suddenly decide to make some fast DC measurements.


Yes, Hi-Res mode gives a sharp but dim display but it takes 1-Sec to update the display in Auto Sweep even at 1mS/div Horiz. Sweep.

It is even worse at lower time/div settings. Too long for me to wait to jump about a circuit and having to wait a second for the trace to reset to the baseline.

Turning on Averaging if even worse, take at least a second to display the result and two sweeps at say, 100mS/div that is way to long to wait for the purpose of fast voltage measurements. Doesn't change the noise level at the flat top of a sq. wave nor at the baseline of it.

I would like to see what a trace looks like on an Agilent (8-bit) or Tek (9-bit) scope.

[dfmischler]

Thanks dfmischler for the picture of the analog scope trace.

That trace was from a Tektronix TDS310 50MHz 500MS/s 1Kpt/channel digital storage scope built around 1993.

If anybody cares it is part of a Dallas 1-wire search sequence that I was having trouble with.  The view is from about 110 meters cable distance from the DS2480B master.  Note the nice slew rate controlled edges and the horrible glitch (I think this is when the DS2480B's active pullup tries to take over from the passive pullup).

[Salas]

That Tektronix TDS line has nice waveform depiction due it uses the phosphor's natural persistence. Its bulky using a CRT but it was a good, reliable line. Still many of those at work out there.

[jpb]

Here are screen grabs at 2mV division (the most sensitive) and 1V a division with 20MHz bandwidth. They are flat enough for me but perhaps too fat for your liking.
(I've just used normal mode without averaging or other schemes to flatten them further.)

Note 8V/256 = 31.25 mV and 16mV/256 = 62.5 microV so at 1V/div the peak-to-peak is only 1bit whilst at the most sensitive range (2mV/div) it is 3 bits
which I think is as good as you might expect.

[dfmischler]

Here are screen grabs at 2mV division (the most sensitive) and 1V a division with 20MHz bandwidth.

Do they look that nice when viewed on the screen?  How about a photograph (or a bad cell phone picture like mine  )?

[jpb]

Here are screen grabs at 2mV division (the most sensitive) and 1V a division with 20MHz bandwidth.

Do they look that nice when viewed on the screen?  How about a photograph (or a bad cell phone picture like mine  )?

I'll try and get a photo but because these were screen saves and the screen is a VGA lcd there isn't much difference between the two.

edit: here is the photo - unfortunately there is a bit of reflection but I think you can see it well enough.

[AndyC_772]

I would like to see what a trace looks like on an Agilent (8-bit) or Tek (9-bit) scope.

Here's a few captures from my Agilent MSO-X3054A and Tek TDS754D 

Both scopes are 500 MHz bandwidth, all images are taken with no probe attached.

[Paul Price]

Thanks for the pinup's JBP. This looks twice as flat as my Rigol which shoes about 1-minor div noise bar regardless of the volts/div setting when viewing a top or baseline bottom of a clean square wave.
It is a little hard to see your picture result since you don't display a graticule, so I must visually guess.

AndyC_772, thanks much for taking the time to send me these screenshots.

The Agilent Normal Mode display,(top of your set of pictures displayed) looks close to what I see on my DS4024 in the Normal mode.
The TDS754 scope noise floor looks no better then my DS4024.

In the Hi-Res mode on the DS4024, the baseline is minimally thick and quite dim altogether just slow to respond to a voltage change.

I must try to take some display shots myself to show what I am talking about with my cellphone camera and post them here.

[ftransform]

if you are measuring DC voltages, did you consider turning on a low frequency low pass filter?

they tend to clear stuff up on oscopes

[Paul Price]

Ftransform:  Thanks for the tip, but in general, I use a scope to see what the electrons are doing out there, not hide their play from my eyes.The noise I see  is generated in the scope, not external to the scope, so no Lo-Pass Filter would work here.

[ftransform]

Ftransform:  Thanks for the tip, but in general, I use a scope to see what the electrons are doing out there, not hide their play from my eyes.The noise I see  is generated in the scope, not external to the scope, so no Lo-Pass Filter would work here.

it totally cleans up my 1052e display to a flat line

[JuanPC]

Noise coming from the wall is tremendous,.

get a PS Audio Power Plant Premier,
use shielded power cables.
eliminate all common noise. see Dave video: