Another dodgy USB charger.

[bd139]

Got this with a digital camera kit. Found some interesting design aspects.

Firstly it appears there is no feedback mechanism between the primary and secondary at all that I can see. Or any voltage regulation on the secondary. I suspect it’s using the ratio of the transformer and doing feedback on primary. Hmm. It does output 5.8v out of the box which is more than a little high  . I don’t have a load handy to check it.

Secondly, clearly there is a problem in the factory with them blowing up from poor soldering and manufacturing tolerances because someone nearly scored around pads of the board around the transformer primary and the IC with a scalpel. First time I’ve seen that one 

Saving grace is there is actually quite a lot of clearance between primary and secondary. No slots cut though.

Gore attached.

[magic]

Firstly it appears there is no feedback mechanism between the primary and secondary at all that I can see. Or any voltage regulation on the secondary. I suspect it’s using the ratio of the transformer and doing feedback on primary.

It's common in miniature USB chargers. I suppose it's good enough given a transformer with high enough coupling.

Another annoying feature found in USB chargers is "cable resistance compensation", i.e. the chip knows nothing about the cable but it raises its output voltage proportionally to load according to some designer's guesstimate.

It does output 5.8v out of the box which is more than a little high 

Less common

[Dennis Frie]

I guess the 5.8V is measured unloaded? It's not unusual to see significant voltage-variations depending on load, if the output is not regulated directly. Flyback converters are great for making multiple outputs (in this case probably just an output to drive the primary side and the 5V secondary output). However, voltage tracking between outputs depends a lot on the transformer design and it will never be perfect.
Primary side regulation is a cheap way to design, but far from ideal.

The "cuts" in the PCB however, seems like a really bad idea. Those USB-chargers from China is pretty much a lottery.

[David Hess]

Primary side feedback is possible with a flyback converter:

https://www.analog.com/en/technical-articles/primary-side-sensing-takes-complexity-out-of-isolated-flyback-converter-design.html

[SmokedComponent]

Wow, those cuts I sometimes do on my protoboards. Never seen it in a commercial product.
They have an extra assembly step and line workers to do those cuts...must be an expensive, premium product