My Variable DC Power Supply (30V/1A)

[birrbert]

Hi!

It occurred to me just now that I didn't specify what the input is for this power supply. So, where it says input there it's already clean DC voltage entering the circuit. Rectifying and smoothing takes place elsewhere. I'll wrap everything up hopefully in a video and post it in the giveaway topic... sooner rather than later. It's gonna be cool, count on that.

Cheers!

[birrbert]

Hello!

Just to make sure nobody missed the video, here's the link again: http://www.youtube.com/watch?v=T2hxFi8Rmno
Thanks for all who supported me!

Now that I have my own oscilloscope let me "bombard" you with questions. Here's the first one: after checking the attached screenshot what is your thought? It's showing a signal while the power supply is turned on, output is turned off and of course oscilloscope probe is connected to the output cables. If I touch the metal parts where the probe and positive output cable meet then the Voltage rises; if I take away my finger then the signal goes back to this size. I feel that something is not right here, but I have no idea what. When I turn the output on then this sine wave signal disappears and the oscilloscope is showing the output DC signal (i.e. a straight line).

edit:
I just checked because I was curious... I see the same signal when the probe is connected to nothing... It's just that the amplitude is low, but when I "zoom in" all the way to 20mV per division I can see it clearly. I hope I'm not making a big thing out of this, but is this normal or what?

[SeanB]

Mains pick up on the open circuit output. Quite normal. add a 1k resistor across the output and it will mostly disappear. This is mains voltage being capacitively coupled into the 1M or 10M scope input impedance, most likely from inside the power supply.

[birrbert]

I'm trying, but I don't understand this. If I completely remove the probe from my oscilloscope, even then there's this signal shown; it's barely visible but it's there. If the probe is connected to oscilloscope then the amplitude is higher, if the probe is connected to my power supply then the amplitude goes even higher, if I touch the metal parts with my finger then the amplitude is even higher. I watched Martin's episodes about using an oscilloscope, but he didn't have this; neither on the Agilent nor the ISO-Tech. This has nothing to do with my power supply project, but still, what gives? Is there something wrong or it's just how the oscilloscope is?

I'm attaching two more screenshots. The first one I made simply with my finger on the tip of the probe; the signal was happily climbing up, then going down and so on randomly. The second I made in Single Shot mode; you can see the sine wave and when I turn the output of the power supply on then you see the ramp up followed by a rough but straight line.

PS: I'll go to bed now cause it's late and I hope I won't be dreaming of waveforms.

[birrbert]

Again, a bit off topic, but is it possible that I have a grounding issue? I should fix this somehow because in some cases it's affecting the Single Shot mode, i.e. it's triggering even when I'm not expecting it to.

[SeanB]

Ground the power supply output and you will not have this any more, as the power supply and the scope will have a common ground point. A floating power supply is good, but only if you need a floating power source if you want to have a ground at some point like at a half rail fed from a function generator or a scope probe. Otherwise it is a way to get mains noise into the powered device.

[birrbert]

OK! After reading some more, I started to understand it. The probe is so sensitive that it picks up mains hum, as they call it, and that's why I see the 50 Hz sine wave. It's not always 50 Hz and not always a sine wave... sometimes it's just gibberish, of course. The source of mains hum can be, in my case at least: the power cable of the oscilloscope, the power cable of the power supply, the power supply itself or the power strip. Since my bench is very small, the oscilloscope can pick the signal up easily from any of these sources, even if the probe is not connected to the oscilloscope at all. Also, I checked by moving the probe around.

As far as I understood, this doesn't interfere when taking an actual measurement, but please correct me if I'm wrong. The other thing worth mentioning is that Martin didn't have this mains hum picked up by any of his oscilloscopes (Xytron, Agilent, ISO-Tech) and I don't seem to recall explanation in the videos.

I'll be back on topic soon with measurements about the output noise of my power supply.

[birrbert]

Last night's results with storyline below. It was interesting to see for example how the image changes as I vary the trigger level (i.e. first six images).

1. No load, trigger level set low



2. No load, trigger level a bit higher



3. No load, trigger level more higher



4. No load, trigger level in the middle



5. No load, trigger level on the positive side



6. No load, trigger level higher on the positive side



7. Rise time of the output turning on a 100 W incandescent light bulb



8. Output with light bulb running on full power



9. Output with light bulb and reduced Voltage to 21 V



10. Output with light bulb and reduced Voltage to 10 V

[birrbert]

What would be an acceptable noise level in the case of a power supply? How should I interpret my oscilloscope screenshots?

edit:
Another thing worth looking into is the following: when I power this 100 W light bulb the power supply can't push more Voltage through it than the bulb wants... or at least this is what seems to be happening. Let's say I turn the Voltage all the way up to 30 V on the power supply, I turn on the output, then bulb lights up, but the Voltage shown on the LCD drops to around 24-25 V. Normally if I power a lower Voltage bulb or an LED or any type of load then as I crank up the Voltage I'm pushing more current through it; this is why at a certain point the load gets damaged, but not with this bulb.

I should make a new video showing cases which I would like to understand and solve, because there are quite a few similar ones to the one I just described...

[birrbert]

Hello again!

I bought a couple of latching push buttons similar to this one:
http://www.ebay.com.au/itm/16mm-12V-Car-Silver-Aluminum-LED-Power-Push-Button-Metal-Switch-Latching-B-/390731993896?pt=AU_B_I_Electrical_Test_Equipment&hash=item5af9711328

But I can't figure out how to wire it so that the LED lights up when the power supply is on and turns off when the power supply is off. Because now if I apply 12 V through the positive and negative pins the illumination turn on and stay that way no matter what. Can you please help?

[SeanB]

The switch and the LED are separate items. You need to use the switch to light the LED, or at least connect the LED to something controlled by the switch. As you have discovered you can light the LED at any time by powering it. This can be used to your advantage by having a low current ( like 1mA) through the LED at all times ( 10k resistor to 12V) and simply apply 12V to the led to have it dim when not switched and bright when on.

[ax2013]

About oscillating: how strong was it? I have very similar circuit to this one:
http://electronics-lab.com/projects/power/001/schem.gif

and indeed the output of U2 is oscillating 20Vpp. It is filtered by C7 so no visible noise on output. But why this type of opamp circuit is oscillating in the first place and how to prevent from it?

Ax