Would you do this to a 7812/7912

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
This is being sold as a replacement for the "WallWart" by a famous manufacturer:
 

Attachments

  • LM7812.gif
    LM7812.gif
    9.7 KB · Views: 418
No, I would not. In addition to the previous post, the common mode rejection transformer should be on the input side of the regulators followed by a capacitor, and the inductors to pos common and neg common are just (wrong) asking for oscillations, just tie them to ground.
 
Last edited:
:envy:
Lots of postings with readings of an AP...
Is this becoming a standard equipment for DIY?
...must start saving money....
:envy:

Erhm, back to topic:
1. Pretty strange, it also does not appear to be escpecially cheap.
2. Are you sure that the cap on the 79 output is 1uf?
 
Lots of postings with readings of an AP...
Is this becoming a standard equipment for DIY?
...must start saving money....
2. Are you sure that the cap on the 79 output is 1uf?

it's 1mF, my bad.

The Z-Out of an LM78xx regulator would be about 8 milli-Ohms @1kHz.

I bought the AP pretty in-expensively, but had it upgraded by selling my 1987 MB. :wink:

I like the line of Inspector Clousseau ("Pink Panther") when he was asked how his wife afforded such expensive clothes on a police inspector's salary: "She saves a little bit out of the household budget."
 
More problems -- the 7912 died! -- can't take the stress of all that unfiltered ripple. The choke values are specified as 22uH and 2x22uH, but the common mode choke measures 36uH.

Here's a simulation of what the output is supposed to look like on a 7812 without (green) and with the chokes:
 

Attachments

  • LM7812_Improved.gif
    LM7812_Improved.gif
    16.6 KB · Views: 257
There's not even a manufacturer listed on the device. It says, however: "Designed in the United States and Made in the PRC".

Does not give me a lot of confidence in the state of engineering education in the US.

...and as I am in the "coatings business" in the US, I can smell the organic solvent based coating which would probably be illegal in the US.
 
lots of products aren't 'designed'. They are just copies/clones/imitations done by people who don't know what they do. Obviously someone has mixed up the wiring, since it seems to contain all the necessary parts.

...you are guessing that both 1000uF were intended to be in the traditional postion in front of the 78xx and 79xx... - just accidentially mixed up ?
:rofl::rofl::rofl::rofl:
 
In case of really just being an accident, then it is a result of our simple mind setting: 'If it is cheap, we will buy it.'
Stinginess on one side and greedyness on the other side knocks down the full chain... to components, production sites (ref. also to the jumpers / I-slaves ) and also to the R&D guys.

Safety:
There is no evidence whether the design is save or not.
The key components for safety in this application are the housing and the mains transformer. Especially the transformer you cannot fully judge without unwinding it. High pot test alone can only prove missing compliance in case of failing the test. To ensure safety the construction has to fit the thermal limitations of the isolation system, creepages, clearances, min material thicknesses, high pot testing and more....
 
The key components for safety in this application are the housing and the mains transformer.

The "green wire" is connected to the PCB -- another "tit on a bull" -- as there is no connection to or from the case. It should, of course, be connected to the case.

As a sanity check, I scanned the bottom of the PCB and re-oriented it in Photoshop -- then traced out the top layer with a red sharpie:
 

Attachments

  • LM7812_3.jpg
    LM7812_3.jpg
    34.3 KB · Views: 117
Is it the green wire, which is directly coming from the PE (Protective Earth) at the mains plug?

In this case it would not be a tit on a bull.
When you have a proper connection of PE to the secondary of the transformer, then the isolation requirements inside the mains transformer are relaxed. In this case the transformer does not need to fulfill double or reinforced isolation.
The transformer itself would just need to fulfill basic isolation, which is much easier and cheaper to implement in a small transformer.

BTW, you say the green wire should be connected to the enclosure.
Does this mean you have metal enclosure?
 
Is there anything written like PC 1 (Protection Class 1) ?
This would fit to have the secondary connected to PE.

From view of safety there is nothing wrong with a correct implementation of PC1 (judging correct is complex...).
The only downside is that combining multiple devices of PC1 can cause issues with hum, due to ground loops.
I.e. Computer + Beamer + Play Station + Amp... in worst case running on different mains outlets...
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.