Siglent SDS1102CML+ and capacitor discharge

[kiltro]

Hello everyone!

I'm the owner of a Silgent SDS1102CML+

I was measuring the ripple of a simple power supply (transformer, diodes, cap, load resistor) and what I'm noticing is that the capacitor discharges are curved instead of straight lines... 

What I've tried so far with no avail:

1. substitute the cap
2. trying the circuit on a friend's oscope (he sees straight lines indeed)
3. changing probe (I was using the included ones)
4. update firmware and do self calibration

Any suggestion would be appraciated

[StillTrying]

Are you using the channel AC Coupled, try it DC Coupled if there's enough Y offset to get the trace back on to the screen.

Or have the scope on DC coupled but probe the signal through a 22 or 47nF cap.

...To see if it's anything to do with this ?
https://www.eevblog.com/forum/chat/scope-acdc-coupling-at-60hz/

[kiltro]

I was using AC coupled

But its on the discharge
" alt="" class="bbc_img" />

[StillTrying]

I still think it's something to do with the scope's AC coupled LF frequency response, perhaps just not low enough - orange trace.
Try with the scope DC coupled but probed through a 100nF or 220nF for a much lower LF response - yellow trace

[kiltro]

Without any cap

" alt="" class="bbc_img" />

Yellow 100nF / Violet 200nF

" alt="" class="bbc_img" />

[tautech]

Probes in 1x or 10x ?

[kiltro]

Both switches on 1x

[tautech]

Both switches on 1x

You said:
2. trying the circuit on a friend's oscope (he sees straight lines indeed)

From your results with 200nF one might deduce your friends scope has far more input capacitance but nF instead of pF doesn't make sense. 
IMO the P-P measurement (ripple) accuracy is the greatest concern and less the transition between peaks but you seem to trying to prove something which IMO doesn't matter as not all cap discharge is linear.

[StillTrying]

"Yellow 100nF / Violet 200nF"

That difference between the channels in the 2nd pic is strange, swap the probes at the scope's BNCs or something to see what changes.

In a simulation it makes no difference if the 100nF is driving a X1 1M 115pF probe, or a X10 10M 17pF probe, they both give exactly the same result, same for a 200nF.

When the bridge has stopped conducting the discharge curve of the smoothing cap into the load looks like a straight line because you're only looking at about 1.4% of the curve, a ~140mV drop in ~10V.

[bdunham7]

If you try in with the probes (and scope) set to 10X, your LF response should be 10X lower, although the signal will be a bit noisier.

This is just from the LF response of the scope.  If you use DC coupling and a film cap as large as you can obtain (say 1uF or larger) or even an electrolytic cap with low enough leakage, that should eliminate any distortion due to LF cutoff.

[kcbrown]

Hello everyone!

I'm the owner of a Silgent SDS1102CML+

I was measuring the ripple of a simple power supply (transformer, diodes, cap, load resistor) and what I'm noticing is that the capacitor discharges are curved instead of straight lines... 

Isn't that what you'd expect?  The capacitor's discharging current into a load (likely a resistive load that has a fairly constant resistance), the discharge rate depends on the voltage and the load resistance, and so for it to not appear be curved, you'd need either a fairly high RC time constant relative to the time period you're measuring, or you'd need a load resistance that just happens to be proportional to the voltage applied to it. 

The replies here suggest that I must be missing something ...

[bdunham7]

The replies here suggest that I must be missing something ...

It's a matter of degree--the sawtooth decline on a low-ripple PSU is usually pretty straight simply because you are only seeing the very beginning of the decline.  If you turned the power off and watched the voltage ebb away, it would indeed have a curve.

[2N3055]

Voltage on a capacitor discharged with a resistor will not drop linearly  (in straight line) but exponentially.
To get linear ramp (straight line) you need to charge/discharge capacitor with constant current.

That being said, if you charge/discharge capacitor only partially (which is the case here) a short part or that exponential discharge curve might visually look like it is straight, because you're looking at a very short segment.

You didn't give any values here. But try to put larger filtering capacitor. Amplitude of ripple will be less and it will apear more linear.

You are obviously experimenting and learning here ( very good! ) if you allow me one advice.

Download free LTSpice and draw circuits you are learning about. Then connect them in a real world. And compare..
It is a great learning experience to do so..

[StillTrying]

If you turned the power off and watched the voltage ebb away, it would indeed have a curve.

Easily done in the simulation, it's ~8.5 VDC, 1000uF and a 300R load, bottom plot is a close up of the first.

I don't know if the OP's is wildly different but he has seen the straight line twice, on another scope and on his own CH2, going by my simulation his 100nF on CH1 only being about 10nF would just about explain it.

[kcbrown]

It's a matter of degree--the sawtooth decline on a low-ripple PSU is usually pretty straight simply because you are only seeing the very beginning of the decline.  If you turned the power off and watched the voltage ebb away, it would indeed have a curve.

Sure, but that's highly dependent upon the power supply, no?  In particular, the RC time constant versus the ripple frequency.

Here, the ripple frequency is only 100 Hz, for a 10 ms period.  That seems to me to be a pretty long period for ripple, no?  The peak to peak ripple is about 120 mV.  This translates to a slew rate of 24V/sec, which seems rather slow.

In light of that, my sense is that the curvature seen is unsurprising.

[bdunham7]

It's a matter of degree--the sawtooth decline on a low-ripple PSU is usually pretty straight simply because you are only seeing the very beginning of the decline.  If you turned the power off and watched the voltage ebb away, it would indeed have a curve.

Sure, but that's highly dependent upon the power supply, no?  In particular, the RC time constant versus the ripple frequency.

Here, the ripple frequency is only 100 Hz, for a 10 ms period.  That seems to me to be a pretty long period for ripple, no?  The peak to peak ripple is about 120 mV.  This translates to a slew rate of 24V/sec, which seems rather slow.

In light of that, my sense is that the curvature seen is unsurprising.

So it is pretty obviously a 50Hz full-bridge cap-filtered supply.  Let's assume the load is resistive.  The ripple appears to be 134mV, but the OP hasn't given the other details, so let's assume that this is a 13.4V supply, and thus the ripple represents 1% of the total voltage.  This means that the peak current will be 13.4V/R_LOAD and the minimum current will be 0.99 x that.  OK?

So the dV/dt will be proportional to the current during the capacitor discharge (the downslope) and is represented by the angle of the tangential to the trace at any one point, since the vertical difference is the dV and the horizontal dt.  The actual angle will depend on the selection of V/div and s/div, but let's say we adjust those so that whatever the peak dV/dt is, it is represented by a 45 degree angle on the scope.  The tangent of this angle is 1, representing dV/dt * some constant, and thus the tangent of the angle at the bottom of the slope, dV/dt_min will be 0.99*dV/dt_peak.  Arctan (0.99) = ~44.7 degrees, so a 0.3 degree difference in the slope from the peak to the bottom is what you'd expect.  That's pretty straight.  I hope I didn't screw up the math and that it wasn't too unclear.

[StillTrying]

Years ago when I used to work out (guestimate ) pk-pk ripple by hand, I used to consider that the diodes where conducting for 1/5 of the time which left 8ms of straight line where the cap is discharging into the load, unless you're doing something very unusual, in practice that's close enough.

[kiltro]

I was using AC coupled

But its on the discharge
(Attachment Link) " alt="" class="bbc_img" />

First picture above is for 43V, 10000uF cap and 270ohm 10W resistor.

Without any cap

(Attachment Link) " alt="" class="bbc_img" />

Yellow 100nF / Violet 200nF

(Attachment Link) " alt="" class="bbc_img" />

Last pictures I posted are for 43V, 100uF and 36k resistor (1st AC coupled without any probe cap, 2nd DC couplet with 100nF, yellow, and 200nF, violet)

In both cases if I simulate with lt spice I see a straight line

[2N3055]

I was using AC coupled

But its on the discharge
(Attachment Link) " alt="" class="bbc_img" />

First picture above is for 43V, 10000uF cap and 270ohm 10W resistor.

Without any cap

(Attachment Link) " alt="" class="bbc_img" />

Yellow 100nF / Violet 200nF

(Attachment Link) " alt="" class="bbc_img" />

Last pictures I posted are for 43V, 100uF and 36k resistor (1st AC coupled without any probe cap, 2nd DC couplet with 100nF, yellow, and 200nF, violet)

In both cases if I simulate with lt spice I see a straight line

With 43V, 100uF and 36K resistor you shouldn't see much curve because discharging is not deep enough, like we said.

You are very vague with info.

What exactly probe you used? 10X or 1X..?

[kiltro]

You are very vague with info.

What exactly probe you used? 10X or 1X..?

Both switches on 1x

I can look if I find the probes specs when I come back home

[tautech]

Both switches on 1x

I can look if I find the probes specs when I come back home

OEM = PP510.
https://siglentna.com/wp-content/uploads/dlm_uploads/2017/10/Probe_List_05282021.pdf

[2N3055]

Try this.

You have to include all the schematics. Including probe and scope front end connected to the circuit.

In short, AC coupling capacitor can interact with 1X probe setting.
If you include that in spice it shows the curve.

If you change it to 10X (in sim and real life it will be more straight)

[kiltro]

Try this.

You have to include all the schematics. Including probe and scope front end connected to the circuit.

In short, AC coupling capacitor can interact with 1X probe setting.
If you include that in spice it shows the curve.

If you change it to 10X (in sim and real life it will be more straight)

I see
So I guess this is normal functioning and not an oscope fault that needs maintenance?

[2N3055]

I would say scope is not broken.

People seem to forget that 1X probes not only lowpass limit high frequencies to much less than in 10X (10x 150Mhz, 1x 12 MHz for instance) but also present different source impedance to input of scope and AC coupling circuit.
AC circuit acts as a highpass filter and corner frequency will be different for 10X and 1X..
It is basically RC highpass filter where R changes...

[StillTrying]

A 3A bridge rectifier, 3300uF and 220R gave 170mVpp of ripple on 16V DC.

The scope's AC Coupling gave a discharge line straight enough, so I didn't bother with a through a 100n version, probes X1.