Why 99% of Power Supply Reviews Are Wrong

[Olle P]

It's an interesting article all right, but I still feel that Mr Torres is too harsh on voltage measurements.

Sure it doesn't say anything about the power capabilities, and switched supplies keep the effective value of the output voltage levels steady (up to the point where it can't reach the target level at all).
But! The voltage is balanced relative to some reference, and if that reference isn't well calibrated the voltage will be wrong, all the time!

Therefore it is valuable to measure the voltages (using a calibrated precision instrument) as the first step of a technical examination.
If the voltages are outside the allowed range then there's something very wrong with the PSU.

Cheers
Olle

[empire23]

Originally Posted by Olle P

It's an interesting article all right, but I still feel that Mr Torres is too harsh on voltage measurements.

Sure it doesn't say anything about the power capabilities, and switched supplies keep the effective value of the output voltage levels steady (up to the point where it can't reach the target level at all).
But! The voltage is balanced relative to some reference, and if that reference isn't well calibrated the voltage will be wrong, all the time!

Therefore it is valuable to measure the voltages (using a calibrated precision instrument) as the first step of a technical examination.
If the voltages are outside the allowed range then there's something very wrong with the PSU.

Cheers
Olle

.
It's true since most PSUs have some offset, but we still must remember that proper calibration or trimpotting is the responsibility of the manufacturer and that any such offset shouldn't be there in the first place if they did they're job well. Most high quality PSUs i've seen follow reference very closely and are hand calibrated, PCPC silencers for one are calibrated, although the pots are glued stuck to prevent user fiddling.

Secondly, most SMPS PSUs need somekind of load to get them working. Question is, what percentage of load would be acceptable to get a basic offset reference value? 10? 20? 30? Very hard to define. The effects of loading on the circuit is never known to a reviewer, so a base level should always be defined. (although the definition is still alien to me) The question of whether the variance is caused by the load or offset requires some standards. Although if all PSUs came with trimpots, we'd all live easier lives.



Anyways guys, i'm still 1.5k ringgit short for my scope. Mind pitching in?

[Olle P]

Originally Posted by empire23

... proper calibration or trimpotting is the responsibility of the manufacturer and that any such offset shouldn't be there in the first place if they did they're job well.

Which is why it's a good idea to see how well they've done their job, isn't it?

Originally Posted by empire23

Most high quality PSUs i've seen follow reference very closely...

Many (but far from all) low quality PSUs don't, which makes it also a fairly good quality indicator.
"Bad" voltages => Low quality!

Originally Posted by empire23

... most SMPS PSUs need somekind of load to get them working. Question is, what percentage of load would be acceptable to get a basic offset reference value?

Anywhere in the range 20-50 W should do well for all PSUs. Well designed PSUs only need a few watts output to work well, whereas others need closer to 10 W before they run nice.

Originally Posted by empire23

Although if all PSUs came with trimpots, we'd all live easier lives.

No we wouldn't! A proper calibrating of the voltages require tools that very few individuals have access to.

Originally Posted by empire23

Anyways guys, i'm still 1.5k ringgit short for my scope. Mind pitching in?

I got my oscilloscope (used, but fully functional) for free, with cables, probes and all!
It was either me taking care of it or sending it for scrapping, since it wasn't used anymore where I work and we needed the space.

Cheers
Olle