SMPS supplies are rubbish

[restorer-john]

8x24" LG LED monitors used in our offices. 4 of them on a single computer (quad screens) used 4-5 hours a day, the other 2 in 2 pairs on other PCs that see 1-2 hours per day.

12 months old approx. They started failing 6 months in and now 5 out of 8 PSUs have died. We are only talking 23-32 Watt supplies (max).

Here are 4 of them (dead):

Let me open them up and we'll do some autopsies...

(top bet: bad caps)

 

[restorer-john]

reserved for pics of the guts

 

[HammerSandwich]

Bad batch sounds reasonable, but (all) SMPS are rubbish might be a stretch.

 

[restorer-john]

Bad batch sounds reasonable, but (all) SMPS are rubbish might be a stretch.

If you'd seen as many prematurely dead ones across all brands as I have, you'd change your statement.

Honestly, I'm sick to death of SMPS supplies. They are all crap.

 

[Don Hills]

Different labels, so different batches at least if not different manufacturers (subcontractors). I'd be a little surprised if it was the classic bad caps, I would have thought there's been enough time for the bad ones to have been flushed through the system. I've recapped my share of monitors with built-in power supplies, but never had an external brick type fail. Your mileage clearly varies...

Hint for anyone who isn't familiar with disassembling such things:
Sometimes they're screwed together with screws under the feet / caps at the corners, but they're usually welded. Stand or clamp it so one of the seams is facing up. Line up a wood chisel along the seam and tap with a hammer, gently at first and then increasing. At some point there'll be a cracking sound as the weld separates. Move the chisel along and repeat until you've been all the way around. Done properly, there's little danger of the chisel penetrating and damaging the internals. I've been opening battery packs this way for many years without incident. I reassemble with ABS solvent from a hobby shop specialising in plastic model kits.

 

[maxxevv]

Might be worth investigating your mains supply too if so many of them has died on you over such a period of time.

I have never had one die on me all these years.

 

[solderdude]

I have seen my share of SMPS fail as well.

Many small wallwarts, many in cheap DVD players and some in network equipment.
In most cases the start-up cap.
These generally work untill they are switched off from mains and then won't start up again.

In most cases it is a small cap (10uF to 47uF) in the primary side that is mounted (deliberately) very close to a heat generating part (the switcher).
Replacing that one only with a 135oC cap and bent away from the heat source is often enough to extend its life.

In my case its not mains related (230V and stable).

Have to admit that I too have many bricks and wall warts working trouble free for years.

 

[restorer-john]

Hint for anyone who isn't familiar with disassembling such things:

Easiest way on RF/Microwave melt sealed ones is to place and hold the unit flat on a cement floor and 'crack' the joint by using a hammer on the end edges. Doesn't hurt a thing and they can be glued back together with little or no evidence if you so wish.

Same basic procedure as yours except no clamps are required, so no damage is done to the casework.

 

[garbulky]

Give me a nice torroid. Everytime I encounter switching power supplies, I know I'm not guaranteed anything having encountered many failures. A torroid, so far, I've never had failures. Though, to be fair, there are gear with switching supplies that have never failed on me and have given close to a decade of good service.

 

[solderdude]

Had 2 small toroids (7W encapsulated) fail on me.
Not due to drawing too much power.

A linear PS is reliable but bulkier. heavier and more expensive and generally draws more idle power.

Have some very old switchers at work that still work flawlessly.

 

[sergeauckland]

On balance, I prefer SMPS to linear. I agree that SMPS seem to fail more than linear supplies, but when they work, they are mechanically quieter, physically smaller and lighter, no transformer hum, draw much less idle power, and, the ones I've measured, also have less 50 and 100Hz ripple. This may perhaps be at the expense of HF noise, but audibly that's less troublesome.

In confined installations such as 19" racks, mechanical hum and heat are an issue, so anything that helps is welcome.
S

 

[KSTR]

SMPS design and construction is not a job for junior engineers. That is the underlying problem, IMHO, SMPS can be made to last for decades (except maybe a change of e-caps). I have several SMPS-powered lab equipment from the early 90'ies that has been running heavy duty, never had any issue.

 

[Killingbeans]

Makes me wonder what happens if you ditch the 'small' part of 'small and efficient' and build a SMPS with no electrolytics but with huge MKP film caps instead. Probably opens a whole new can of EMI-worms, but the thought is fun to me.

 

[svart-hvitt]

I have never had an smps fail in office. Having worked where anywhere from 200 on a floor to 20 on a floor, this theme - failing screens or pcs due to smps - has never been an issue.

Maybe clean energy where I live?

 

[Soniclife]

High ambient temperature?

I've not found SMPS often, I can only think of one that failed in a PC after a few years use, and one in a Samsung TV, that failed a few months past it's 5 year warranty. In the case of the Samsung TV they sent an engineer to the house who replaced the problem caps with better versions, and didn't charge as they admitted the original parts were too optimistic for longevity, it's still works perfectly several years later.

 

[Veri]

On balance, I prefer SMPS to linear. I agree that SMPS seem to fail more than linear supplies, but when they work, they are mechanically quieter, physically smaller and lighter, no transformer hum

About transformer hum, never had this problem before but with current DAC the toroid is humming (at 230V EU voltage). It's no biggie but I wonder this means there is a bigger chance of premature failure over the years? Or is it relatively benign to have a transformer hum..

 

[solderdude]

In case of power bricks and wallwarts the tempertaure inside can be higher. This reduces life of wet electrolytics (capacitors).
Often using physically bigger/higher voltage rated or higher temp rated caps are not used to save on costs.
When manufacturing 1000's of SMPS every cent counts and cheap poorer caps are used.
Not so in lab equipment or higher end stuff.

So ... it's a financial decision mostly. Cost reducing by using parts that last just long enough.

 

[solderdude]

About transformer hum, never had this problem before but with current DAC the toroid is humming (at 230V EU voltage). It's no biggie but I wonder this means there is a bigger chance of premature failure over the years? Or is it relatively benign to have a transformer hum..

Hum can become audible because of direct coupling to the enclosure (which makes it louder) it can be a manufacturing thing (some transformers hum more than others) or it can be that too much current is drawn which is unlikely in a DAC. In that case the transformer would get quite hot and could have a shorter life.

 

[SIY]

I use SMPS bricks regularly in diy projects. Exactly one failure ever.

There's a lot of variation in output noise between different models/brands, so I always check that before pressing a brick into audio service. The 19V Lenovo bricks I have are quite good in that respect, so they're my go-to for supplying tube heaters and op-amp rails. I snagged a pile of them when my old company tossed a lot of older IT equipment...

 

[sergeauckland]

About transformer hum, never had this problem before but with current DAC the toroid is humming (at 230V EU voltage). It's no biggie but I wonder this means there is a bigger chance of premature failure over the years? Or is it relatively benign to have a transformer hum..

I've never found a mains transformer that didn't have at least some small amount of mechanical hum. Toroids tend to be quieter than laminated transformers, but they still hum. Toroids are particularly sensitive to any DC on the mains, as their primary DC resistance is low, so any DC can saturate the core. There's always some small amount of DC on the mains, as any half-wave rectified equipment, like hair dryers on 'half power', battery chargers, welders etc don't balance out perfectly. It doesn't have much effect on small toroids, but large toroids as one gets in power amps can hum audibly. That's why I use a DC blocker on my HiFi as it helps keep the transformers a bit quieter, although there's still some hum.

I don't know of any studies that look at transformer lifetime against hum, but clearly, a humming transformer will be losing energy, and consequently heating up, and heat tends to reduce life. I suspect, however, that this is more of a theoretical than practical concern.

As to SMPS, I have a number of SMPS working 24/7 and have done now for 10+ years without problems, but last month I had to replace two small SMPS in our mixing consoles as the large input capacitors had both failed, causing hum. This was after 88000 hours of continuous use, so shouldn't complain too much.

S.