NEW PSU Electrical CRACKLING

[akio123008]

36 minutes ago, jonnyGURU said:

know measurement on DC.

I assumed the displays on his test rig are DC voltmeters? I'm not sure if he also measured the output current somehow. As for AC input, well I guess you could get something more fancy than a meter like that.

 

36 minutes ago, jonnyGURU said:

Using light bulbs as a load?

Why not? 

 

36 minutes ago, jonnyGURU said:

Are the probes properly coupled for reading ripple? 

He did mess around with the scope on video and give commentary, I assume he talked about how he set it up and what exactly he measured every time. Good chance he mentioned how it was wired up.

 

Edit: Actually the scope is AC coupled as far as I could see on the video so he got that right. The information on the amount of ripple is probably in Italian so can't judge on that.

 

36 minutes ago, jonnyGURU said:

What bridge rectifier is used?  Primary switchers?  Synchronous rectification uses, what?  How were the protections tested? 

Same goes for this stuff; he talked quite a bit during the part where the psu is opened up, quite possibly he was discussing the internals.

 

I mean sure that's not a very well produced video and it looks like one heck of a dodgy setup so I can see what you mean but assuming you don't speak Italian, how can you be so sure so much is missing?

 

Edit: I guess what's missing regardless is some tables and figures with actual test data, but that's not really something for in a video anyway, I'd publish that in a separately available document. 

 

[Guest]

2 hours ago, akio123008 said:

I assumed the displays on his test rig are DC voltmeters? I'm not sure if he also measured the output current somehow. As for AC input, well I guess you could get something more fancy than a meter like that.[/quote]

Those are volts.  Which means he is not measuring current.

2 hours ago, akio123008 said:

 

Why not?

Because incandescent light bulbs are not a linear resistance with respect to voltage.

2 hours ago, akio123008 said:

 

He did mess around with the scope on video and give commentary, I assume he talked about how he set it up and what exactly he measured every time. Good chance he mentioned how it was wired up.

 

Edit: Actually the scope is AC coupled as far as I could see on the video so he got that right. The information on the amount of ripple is probably in Italian so can't judge on that.

As per the Intel spec, Outputs should be bypassed at the connector with a 0.1μF ceramic disk capacitor and a 10 μF electrolytic capacitor to simulate system loading.

2 hours ago, akio123008 said:

 

Same goes for this stuff; he talked quite a bit during the part where the psu is opened up, quite possibly he was discussing the internals.

 

I mean sure that's not a very well produced video and it looks like one heck of a dodgy setup so I can see what you mean but assuming you don't speak Italian, how can you be so sure so much is missing?

 

Edit: I guess what's missing regardless is some tables and figures with actual test data, but that's not really something for in a video anyway, I'd publish that in a separately available document. 

 

HE opened it up.  That's it.  He didn't desolder any of the components to take a closer look as to what actual components were used.  That's like opening the hood of a car and saying "yep.  It's a V6!"

[akio123008]

6 hours ago, jonnyGURU said:

Those are volts.  Which means he is not measuring current.

No it means there's no reading of the current in the video.

 

6 hours ago, jonnyGURU said:

Because incandescent light bulbs are not a linear resistance with respect to voltage.

Doesn't matter, if you monitor the voltage and current, you can work out the power just fine. For that you shouldn't be relying on the resistance of the load anyway. A light bulb isn't as good as say a power resistor, but it's pretty good for load testing IMO; it's a purely resistive load with rather consistent behavior.

 

7 hours ago, jonnyGURU said:

As per the Intel spec, Outputs should be bypassed at the connector with a 0.1μF ceramic disk capacitor and a 10 μF electrolytic capacitor to simulate system loading.

Not sure what spec you're talking about there, or what exactly that method is supposed to achieve, but what he did is:

 

-he attached a load

-he attached a scope

-he set the scope to AC coupled

-he monitored the ripple for the given load on the screen.

One perfectly valid way to measure ripple.

 

Then he didn't desolder any parts indeed, but it's not impossible to know what components are used without desoldering them. He was talking for quite a while during the part where he films the opened up unit, it's quite possible he inspected the components before filming and discusses them there.