With SMPS all over the place, there are tons of options for AC/DC conversion.

Anyone in the biz have any insight into exactly how stable some of the over-the-counter supplies are?

First, you see a lot of wall wart style SMPS supplies. For this particular project, I am looking for something in the 12V range, but there are a slew for the 5V regime. Do these, as a general rule, provide clean power? Realizing they generally do nothing but charge a battery, I'm wondering if a filter may be in order to actively supply say a sensor array or mcu.

Next, you see the laptop style supply. These seem to be often less noisy, and provide more power that their wall-wart counterparts. Do these, as a general rule, provide more stable power as well?

I've used all of these for various projects with good luck. I suppose now I"m curious how well they may work for applications that require a little more precision.

ETA: Yes, I just realized the title is ridiculous. If a mod wants to change to "Consumer SMPS Stability" I would be appreciative.

 

You forgot to mention the numerous varieties of unregulated wallwarts. The output voltage is as specified, ONLY when the rated current is being drawn. this has been a topic of some interest on this and other forums. Without being able to examine the internals it may be difficult to determine weather a given wallwart is an SMPS or a linear regulator or something else.

 

I use CUI 3A-181WP12 12V,1.5A (and the 9V version) SMPS wall-warts for powering many projects during development. Most of these have various combinations of PIC processors with internal and external ADC/DAC and other analog which runs fine with the power supply. There are stepdown regulators on board of course, but everything seems happy with the SMPS as the primary power source. In all the scoping around during programming and debug, I've never seen any issue that could be attributed SMPS power issues with the CUI stuff.

 

define stable/precision?
But in general regulated ones are fine for most stuff and the accuracy is always in the datasheet for a particular power supply.

For example
http://www.meanwell.com/search/gs05u-usb/default.htm

 

Papabravo,

I've played the unregulated transformer supply game. They are, however, becoming much less common. Generally weight and listed input voltages makes culling these an easy task.

John,

I too have had good luck as long as I had a dc/dc regulator in the mix. For this latest project, I plan to use the supply to directly excite a wheatstone bridge for a sensing array. Just wondering how much filtering I should do to expect good results.

 

John,
I too have had good luck as long as I had a dc/dc regulator in the mix. For this latest project, I plan to use the supply to directly excite a wheatstone bridge for a sensing array. Just wondering how much filtering I should do to expect good results.

Based on my experience (and I have used the CUIs on load-cell (w-bridge) development) the SMPS is a non-issue. The production scale comes with one. Your filtering problems are more likely to be in the RF range - assuming a load-cell... You didn't say.

But for a few bucks you can buy one and start from there.

And.. wow! I googled the P/N to link a reference and the first hit was this thread. Spooky. Its deja-vu all over again.

 

Even cheap SMPS supplies are usually pretty good. I've tested a lot of cheap SMPS chargers and SMPS wall-warts etc, and they regulate to within a couple of mV, generally as good or better than a good old 7805 regulator.

The SMPS supplies use modern SMPS driver chips, which have modern technology internal voltage references which are temperature compensated etc.