Weird signal on oscilloscope

[viktor18]

Hello everyone! I'm fairly new to the oscilloscope, and when troubleshooting, I poked my oscilloscope at both terminals of an inductor 220, as shown in the picture:


The bottom terminal of the inductor is connected via U2 at pin 6 and 5, which is SW (Switch Pin. Connect external inductor/diode here) and continues on to the D1 diode, and going downwards C2, CT3, CT4 as so on.

With my multimeter, both terminals of the inductor shows 5volts. However, when using my oscilloscope, I'm getting very different results.

First picture is from the upper terminal in the photo, and the last picture is from the bottom terminal connected via the U2 SW pins. It looks very messy and I don't know why.




This device is supposed to be attached to a hard drive so you can use it as an external hard drive. However, it fails to start the hard drive properly. What happens is that the hard drive starts spinning (and the amp usage goes up) and after a second, it stops spinning and goes down to a lower amp usage. So basically, you can hear the hard drive start spinning (but not as much as when I plug it into another adapter).. It seems to fail before it gets up in working conditions, so it's not completely dead, it just fails to drive the hard drive.

[bktemp]

The waveform looks normal.
What you see is a stepdown (buck converter) operating in discontinous conduction mode because of the low output load.
It is ringing because of the inductance of the inductor and parasitic capacitance of the switch in the stepdown ic. That is the normal behaviour.

Look at the output voltage (the stable 5V signal), set the trigger level slightly below the 5V line and trigger on the falling edge.
Then power on the drive. If the voltage drops because of a high current draw, the scope will trigger ans you will see how much the voltage has dropped.
If it does not trigger, the voltage is stable.
Do the same with the 12V rail, because the motor on many hard drives is powered from the 12V rail.

[viktor18]

5v rail seems stable. However, I set the trigger to falling edge and level at 10.5volts and measured at a capacitor on the 12volt rail, and when plugging in the hard drive, it triggers at 10.5 volts. (The waveform gets stuck there, I suppose that's how it works?)

The 12volt line was stable until the hard drive gets plugged in. What does it mean that I'm triggering at falling edge? And what does this measurement tell me? Apparently the rail is not steady.

It even triggers at 9.40volts

[bktemp]

A short dip is expected, because when you connect the hard drive, all its internal capacitors charge up.
If the voltage does not recover from <11V after a couple of 100ms, then your 12V power supply is probably the problem and can't supply enough current for the hard drive to start up.

[viktor18]

It fails to boot up the hard drive, and then it just keeps jumping from 12 to 9.5v all the time.

[bktemp]

Then either the hard drive is faulty (drawing too much current) or the 12V power supply can not supply enough current.

[viktor18]

The hard drive is not faulty, and the power supply is my lab PSU, so neither of those are the cause of the problem.. Must be something along the 12volt line on the "adapter" between the hard drive and the PSU.. Hmm, but it seems like the 12 volt line is just going straight through with some capacitors? I can't be 100% certain but it looks like it.. So maybe any of the surface mounted capacitors are faulty?

[bktemp]

If the 12V go straight  through, then start probing the 12V rail from the power supply to the hard drive.
At some point there might be a bad connection or the current limit of your PSU is set too low.

[viktor18]

Perhaps, but what happened to this device was that lightning striked, and it blew the wall adapter (ac to dc converter), and luckily the hard drive survived but had to switch out the TVS diode on the 12volt rail to make it work.. So now the hard drive works, but this external adapter is simply not powering up the hard drive like it should.. It feels weird that it would be a connection problem after the event, but you never know.. I'll look at it but feels like the issue would be somewhere else

[PA0PBZ]

How do you know the hard drive is ok, did you test it in a pc? Are you aware that a PC PSU can deliver in excess of 20A on the 12V rail?
Did you try to power up the hard drive using 12 and 5V from another source? Can you measure the current(s)?
Maybe the logic on the board is broken and it keeps powering up and down the drive?

Enough for now 

[viktor18]

Yes, the hard drive is functional when I plug it into a PC, and I have also tried to power another hard drive with the adapter, and the same problem.. And the PSU should be able to provide the current.

I hear the HDD start spinning, the PSU amp goes up to 1.14 on the display, and then the hard drive stops spinning, and starts beeping and the usage goes down to about 0.20amps and jumps around 0.20-0.60 and doing some beeping sound

[bktemp]

Around 1A sounds ok for a starting hard drive.
As I said: Follow the 12V from the power supply to the hard drive (or the other way round). A voltage drop down to 9.5V can cause the hard disk to stop (undervoltage lockout protecting the motor driver).
Now you need to find where the 2.5V drop happens.

[viktor18]

Around 1A sounds ok for a starting hard drive.
As I said: Follow the 12V from the power supply to the hard drive (or the other way round). A voltage drop down to 9.5V can cause the hard disk to stop (undervoltage lockout protecting the motor driver).
Now you need to find where the 2.5V drop happens.

To find the drop, can I just measure the voltage at the positive side of the capacitors along the rail? Doesn't the voltage drop affect the whole rail though?

[bktemp]

Since your scope has at least 2 channels, I would connect the first channel to the 12V rail near the hard drive because there you have a large voltage drop, giving a clean trigger signal whenever the voltage drops.
Using the second channel, I would follow the 12V rail to the power supply. Then you can use the math function for showing the difference between both channels.

[viktor18]

Since your scope has at least 2 channels, I would connect the first channel to the 12V rail near the hard drive because there you have a large voltage drop, giving a clean trigger signal whenever the voltage drops.
Using the second channel, I would follow the 12V rail to the power supply. Then you can use the math function for showing the difference between both channels.

I used your method, and the voltage drop is the same at the input as the output. This is the waveform I'm getting when it's trying to spin up the hard drive

So, basically I'm not getting any difference on the 12volt line, but i don't know what causes the voltage drop.

[bktemp]

So, even directly at the output of the power supply, the voltage drops to <9V?
Try increasing the current limit of the PSU or add a large capacitor (around 1000uF).
The waveform looks like the commutation signal of the motor.