Multiple threads and articles discuss conventional power supply quality which makes sense because it is usually bad when you burn or blow up your computer. I do not see similar concerns raised regarding brick/wall wort power supplies. Do design features make it difficult to burn or blow up your computer with a brick?
Use of Brick to Blow up Computer?
- Thread starter dealcorn
- Start date
I would say that it isn't such an issue as there will be voltage regulators in the computer/laptop to create the voltages they actually use.
Modelworks
Lifer
- Feb 22, 2007
- 16,240
- 7
- 76
Wall warts destroy equipment all the time. It is just usually blamed on the device itself rather than the power supply. Getting a decent quality wall wart supply is actually becoming harder rather than easier due to them using switching supplies more and more because of the cost of copper. Companies are cutting corners to try to meet demand for low prices. The supplies that are really light weight are switching supplies and if care is not taken you can have the full voltage of the wall outlet supplied to the device. They also tend to have poorer isolation from the outlet voltage even when working normally.
The supplies that are heavy and bulky are transformer based. With a transformer based supply you get isolation from the outlet just because that is how transformers work. The other benefit is that the output voltage is limited by the secondary windings to a set voltage . Between a switching supply and a transformer supply the chances of damage are about 100x greater for switching because it requires multiple parts. With transformer supplies you rely on about 3 parts to remain healthy. Of course the downside is cost, transformers have to use copper and a lot of it.
The supplies that are heavy and bulky are transformer based. With a transformer based supply you get isolation from the outlet just because that is how transformers work. The other benefit is that the output voltage is limited by the secondary windings to a set voltage . Between a switching supply and a transformer supply the chances of damage are about 100x greater for switching because it requires multiple parts. With transformer supplies you rely on about 3 parts to remain healthy. Of course the downside is cost, transformers have to use copper and a lot of it.
bryanl
Golden Member
- Oct 15, 2006
- 1,157
- 8
- 81
They do vary in quality, and those not certified by UL or CSA can be very bad in regards to reliability and perhaps also safety. There was a Youtube video showing the insides of some units and demonstrating how the worst ones had smaller and fewer parts, and when tested, those supplies caused much more AM radio and low band VHF analog TV interference.
SparkyJJO
Lifer
- May 16, 2002
- 13,357
- 7
- 81
Another downside is efficiency. A classic transformer is not near as efficient as a switching model.
Switching wall warts aren't that bad when done correctly however. Remember every PC power supply - both good and bad - are switching power supplies. So just like standard PC power supplies, you can have good ones or crappy ones.
Generally speaking, however, even when a wall wart is the old-style, direct transformer type, even a dismal efficiency only results in a couple of watts lost at the most since they are all pretty much powering low wattage devices to begin with. The benefit of a switcher however is a smaller and lighter unit vs a classic wall wart transformer.
Last edited:
All power bricks use transformers of some kind - these reduce the voltage and provide an isolation barrier between the mains voltage and the output.
The difference is in how the transformer is connected. The traditional method has been to use an iron core transformer connected to 50-60 Hz mains voltage, then rectify and smooth the output, and then optionally regulate the output.
The more modern way (called a switching supply) is to use an inverter circuit to convert the mains voltage from 50-60 Hz to 20-50 kHz, usually the inverter has some form of control circuit in it to control the output. This voltage then goes to a transformer, and is then rectified and smoothed. The inverter takes care of regulation.
The advantage of the switching supply is that the bulky, expensive and heavy transformer can be replaced with a tiny, cheap and light transformer.
Efficiency of a power supply is highly dependent on design and quality of manufacture. Traditional mains frequency transformer power bricks often used the worst possible transformers, with every possible corner cut to minimize the cost of the expensive iron, copper and shipping. These transformers often had losses that dwarfed the actual energy usage of the load. A number of years ago I measured and catalogued the losses in dozens of home transformers that I had collected. The results suggested that "off-load" idle power consumption, more often than not, exceeded the actual rated load. In other words, a PSU capable of 5W output, would often take 7-8 W idle wastage when plugged into mains power, but disconnected from its load. The efficiency of the transformers was otherwise dire, and for small transformers it was unsual for the efficiency to be above 60%. The worst I saw was a "6W" transformer that consumed 8W at no-load, and 22 W at 6W load. For something like a cell phone charger that would be "on load" for maybe 2 hours per day - it was entirely typical for the amount of energy wasted by the transformer to be 15x as much as the energy used by the actual phone, if it was left plugged in when unused.
Despite the simplicity of some of these traditional power supplies, reliability was not as good as you might think - especially for regulated supplies. The incredible inefficiency resulted in extremely high temperatures, which severely degraded any electronic components in the PSU. I had some old computer external PSUs which I had to replace on an annual basis because they burned out, precisely because of this inefficiency. I eventually bit the bullet and bought a much bigger industrial supply and modified it - it outlasted the computer.
Switch mode supplies tend to be more efficient because the manufacturers are less likely to cut costs in the transformer, given that the costs of a switching transformer are much smaller than those a traditional transformer. And even if the manufacturer does cheap out on the transformer, most switching control chips, will automatically sense no-load conditions and go into a deep-sleep mode, regardless of how cheap the overall design.
There's nothing that inherently makes a switcher more efficient that a simple transformer - just that making a high efficiency transformer requires lots of high-quality materials (or expensive manufacturing processes - e.g. winding copper onto a toroidal core, which is very labor intensive and requires very expensive machinery compared to winding copper wire on a two-part iron core).
If someone really, really wants high efficiency (e.g. very powerful transformers for power plants which are extremely difficult to cool because of their size - very high volume for their surface area - extreme efficiency, in the region of 99.7%, is necessary to prevent the transformer from getting incinerated) then they can have it. It's just very expensive.
The difference is in how the transformer is connected. The traditional method has been to use an iron core transformer connected to 50-60 Hz mains voltage, then rectify and smooth the output, and then optionally regulate the output.
The more modern way (called a switching supply) is to use an inverter circuit to convert the mains voltage from 50-60 Hz to 20-50 kHz, usually the inverter has some form of control circuit in it to control the output. This voltage then goes to a transformer, and is then rectified and smoothed. The inverter takes care of regulation.
The advantage of the switching supply is that the bulky, expensive and heavy transformer can be replaced with a tiny, cheap and light transformer.
Efficiency of a power supply is highly dependent on design and quality of manufacture. Traditional mains frequency transformer power bricks often used the worst possible transformers, with every possible corner cut to minimize the cost of the expensive iron, copper and shipping. These transformers often had losses that dwarfed the actual energy usage of the load. A number of years ago I measured and catalogued the losses in dozens of home transformers that I had collected. The results suggested that "off-load" idle power consumption, more often than not, exceeded the actual rated load. In other words, a PSU capable of 5W output, would often take 7-8 W idle wastage when plugged into mains power, but disconnected from its load. The efficiency of the transformers was otherwise dire, and for small transformers it was unsual for the efficiency to be above 60%. The worst I saw was a "6W" transformer that consumed 8W at no-load, and 22 W at 6W load. For something like a cell phone charger that would be "on load" for maybe 2 hours per day - it was entirely typical for the amount of energy wasted by the transformer to be 15x as much as the energy used by the actual phone, if it was left plugged in when unused.
Despite the simplicity of some of these traditional power supplies, reliability was not as good as you might think - especially for regulated supplies. The incredible inefficiency resulted in extremely high temperatures, which severely degraded any electronic components in the PSU. I had some old computer external PSUs which I had to replace on an annual basis because they burned out, precisely because of this inefficiency. I eventually bit the bullet and bought a much bigger industrial supply and modified it - it outlasted the computer.
Switch mode supplies tend to be more efficient because the manufacturers are less likely to cut costs in the transformer, given that the costs of a switching transformer are much smaller than those a traditional transformer. And even if the manufacturer does cheap out on the transformer, most switching control chips, will automatically sense no-load conditions and go into a deep-sleep mode, regardless of how cheap the overall design.
There's nothing that inherently makes a switcher more efficient that a simple transformer - just that making a high efficiency transformer requires lots of high-quality materials (or expensive manufacturing processes - e.g. winding copper onto a toroidal core, which is very labor intensive and requires very expensive machinery compared to winding copper wire on a two-part iron core).
If someone really, really wants high efficiency (e.g. very powerful transformers for power plants which are extremely difficult to cool because of their size - very high volume for their surface area - extreme efficiency, in the region of 99.7%, is necessary to prevent the transformer from getting incinerated) then they can have it. It's just very expensive.
Last edited:
Modelworks
Lifer
- Feb 22, 2007
- 16,240
- 7
- 76
I have been finding some really scary switching designs lately in wall warts. These contain no transformer and instead are using a current limiting resistor connected to a bridge that is then switched to get the output voltage. These have zero isolation from the mains as they take the output directly off the FET. These adapters were shipped with routers and when I checked to see how they got UL certification, it seems this is allowed as long as there is no exposed metal contact point on the outside of the product. I hope nobody goes poking around inside one of these routers with the power on 
Who makes good quality, efficient brick that might, for example, power a Pico PSU? It looks like Seasonic has some but they are not yet available in the marketplace.
I understand bricks are generally switching power supplies just like conventional power supplies. However, efficiency of conventional power supplies deteriorate rapidly at loads less than 20% of their rated capacity. My understanding is that bricks may maintain good efficiency at low load levels. Why is that?
I understand bricks are generally switching power supplies just like conventional power supplies. However, efficiency of conventional power supplies deteriorate rapidly at loads less than 20% of their rated capacity. My understanding is that bricks may maintain good efficiency at low load levels. Why is that?
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 23K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter