12 Volt USB Adapters

[Superscope]

Not sure if this is posted in the correct Area or not.



Has anybody else actually checked out the performance of those
12 Volt Car USB Adapters?

I have a Project I'm working on where I need to supply 5 Volts, and ideally
between 500 - 600 mA from a 16 - 18 Volt supply.

I tried the Linear Regulator Route, but the Heat generated is simply to much!

Then I thought, I know, I'll use one of those USB Adaptors, their cheap and can supply 1 A (at least that's what they say)
and it saves me trying to build a Switched Mode Circuit which I probably would never get working anyway.

However,

All the Adaptors I've tested so far, at best, can supply 250mA before the Voltage starts to fall well below 5 Volts.
It doesn't really matter what they claim on the Label, they are all pretty much the same.

I tested one of those 3 Port (supposedly 5 Amp Total) Adaptors, and found the exact same
250mA maximum current before the Voltage drops significantly.
Thats Total current.

I found the 3 Ports connected together internally, making a complete mockery
of the 2.1 A, 2 A and 1 A Port Labelling.

The one good thing about Ebay, is you get your Money back pretty quickly if
an item is not as described.
The Sellers live in denial when you tell them their Products are Fake!

Anybody else noticed these ubiquitous Fake Devices.


Ian

[G6Tanuki]

See:

https://www.digitaltrends.com/mobile...charging-work/

there are several different DC-supply 'modes' available depending on how the charger recognises the attached device. In the absence of you providing the necessary signalling to the charger (by links/resistors between some of the pins), it should current-limit at the lowest of the USB-specifications.

[peterpiper]

Hi Ian, have you looked at the switched mode variable voltage modules using the LM2596S chip on eBay? Cheap and very easy to use. Output is a bit noisy for some uses but OK for many.

Peter

[Superscope]

The USB Adaptors I'm talking about are not chargers.
They are just universal Adaptors.

I am not trying to charge anything, only power up.

I had a read through the Link, and that seems to concentrate
on the Charging Technology of the Battery's and supplied Chargers.

Intelligent Charging of course is a very different matter.





Ian

[Oldcodger]

I've always gone down the 7805 ,1-1.5Aroute with a heatshunt on the IC. But I have two 3.15 A 12v adaptors in the car. One is in the front, the other is in the rear cup compartment ( for GS with an apple iPAD.These are stat4d to be car/marine use, and I've had no problems.

[broadgage]

A 16 volt to 18 volt input is almost certainly too much for a unit intended for in car use.
A nominal 12 volt car electrical system should not exceed about 14 volts to 14.4 volts. Anything more than that will rapidly destroy the battery.

I would purchase from ebay a DC/DC converter module with a regulated 5 volts or 5.2 volts output.
Choose one with a stated output capability of at least twice your actual current demand.
Choose one that is stated to accept at least 24 volts on the input.

These cheap modules work well IME provided that they are not pushed to anything near the limit of current output or voltage input.

[Superscope]

Thanks Peter,

I will order a couple of them to try out.
They are clearly a more substantial, dare I say professional device.

Although, if the USB Adaptors did what they say on the Tin they would have been fine.
I have one open and I'm trying to identify the Chip.

LC51 and A341 are writen on it, but I haven't found any info on it.
Bottom line is, they don't do as they claim.



Yes, the Linear Route is much simpler, but my Project involves squeezing
everything into a very tight space with little ventilation.
I have tried it, but I'm not happy with the Heat produced.



Ian

[Superscope]

Hi Broadgage,

Yes, you are correct,
but I chose the Adaptors because they claim 12/24Volt input.
Presumably for 24 Volt Truck use.

I have tested them, and they do work at 24 Volt, but they also limit the current
a bit more (to around 200mA) I guess as a result of the extra Power dissipation.

They seem quite safe, as in they clearly limit the Current, but they simply don't
deliver the Current they say they do.
Because nobody ever tests them, they get away with selling Fake Items.


Ian

[GMB]

Strictly speaking, a USB host (which is what these devices would be expected to be, given the socket) is not expected to be able to supply more than 100mA unless explicitly asked for more by the device, and that could be as low as 4.4V

Now in real life I do not think this is usually how things work, so I would expect these crude power devices to just be a PSU and nothing else.

But before claiming that they are fake I would check whether they have any circuits that are actually talking to the device because they might be working as per the spec.

When we view this from the device's point of view - if a device just plugged in and immediately consumed more than 100mA then it is taking a huge liberty and is way out of spec.

So if these adaptors are claiming to be able to supply a genuine USB device up to 2A then they would have to engage in actual USB protocol to let a well-behaved device know that it is OK for it to draw that sort of current.

[AnalogueMan]

Originally Posted by broadgage

I would purchase from ebay a DC/DC converter module with a regulated 5 volts or 5.2 volts output.
Choose one with a stated output capability of at least twice your actual current demand.
Choose one that is stated to accept at least 24 volts on the input.
These cheap modules work well IME provided that they are not pushed to anything near the limit of current output or voltage input.

I'll second this. I recently converted a couple of indoor-outdoor Christmas lights to go over an arch in the garden. They needed 4.5v DC and my garden ring-main is 12v DC. Tried a 7805 first, too much voltage to drop and it got ridiculously hot. So I went for a DC-to-DC buck converter module off the aforementioned online market. It takes up to 24 V input and gives 5 V out at up to 5 A. Running at 2 A for me with no external heatsink and it's as cool as a cucumber. For a little board about an inch square that cost me six quid (you can get them for three quid if you're prepared to wait a month for delivery from China) that's pretty impressive

[Superscope]

Hi GMB,

I have the device open.

The Data Pins (2 and 3) are connected together on the PCB.
There are no other connections to the Circuit.

So, one assumes this could send some sort of message back to
anything plugged into it.

But it is difficult to see how anything outside, could communicate with it.

So, at the moment, unless somebody knows how to get something
more than 250mA out of them (And I'm willing to try anything suggested)
these Devices appear to be Fakes, in that they cannot deliver the Power
they say they can.




Ian

[PJL]

I have ordered a similar one but this time the application is to power a low power USB device from 8-9V input. If it won't work I will use a 7805 and patch it to the USB socket.

There is a USB charging spec whereby the capability of the supply is "sent" to the device by having various resistors on the data pins connected to 0V and 5V. Having designed a few USB hosts I can tell you that if it really obeys the rules lots of things don't work.

[Scimitar]

The device plugged in has various resistances between the four pins. That is how the PSU decides what current it should provide. If you are only wiring the two power pins, it WILL fold back.

[paulsherwin]

Going from 12V to 5V using a 7805 type linear regulator will indeed dissipate quite a bit of heat at significant currents. This isn't necessarily a problem - they are designed to run hot and will be fine with a suitable heatsink.

Some of the heat can be dumped by a suitable resistor in the 7805 input.

[duncanlowe]

Linear regulation will have to dissipate power. P=IxV so the bigger the volt drop and current the higher the power. But with switched mode the power dissipation can be much lower. So look for a 12v to 5v switched mode supply. I have seen devices with a claimed 10A output so I'm confident they could supply your needs.

EDIT: in case its not clear, in the calculation V is the difference between 12v and 5v (IE 7v) and I is the current you need to draw.

EDIT: something like this: https://uk.rs-online.com/web/p/non-i...rters/1834382/

[robin0577]

My "go to" device for this kind of thing is the LM2576 series of switching regulators. Readily available, good for up to 3A, and not much external circuitry required. Availiable in 12V, 5V and variable flavours, and good for up to 40V input, more on some versions.

https://www.hobby-hour.com/electroni...-regulator.php

Also available as a kit with a PCB if you want to make things really easy..
https://www.horter-shop.de/en/i2c-di...404260912.html

[Terry_VK5TM]

It is interesting to do a search for tear downs on various products and here is one for the centre (black) one pictured above https://www.ifixit.com/Teardown/Gene...Teardown/10108

Schematic included, which shows it is not even capable of supplying the supposed sticker current.

The only thing it didn't show was the D+/D- interconnection (shown in the attached pic) from here https://www.youtube.com/watch?v=txvqJRyYZBY

[Keith956]

Ah iFixit, the company that pontificates about the 'Right to Repair' yet won't even sell you a spare bit for the soldering irons they sell.

[Superscope]

Thanks for the Links Terry.

Just confirms what I've been saying really.

The Teardown was dated 2012.
My Adaptor is different inside (Photos attached), so probably
more cost cutting has been going on since then, but at least
it does look slightly better built inside.


Once again, if anybody can tell me how you can get more out of one of these,
I'm more than happy to try, otherwise this is just 250mA max.
So significantly less than the one torn down!

So I stand by comments that these do not perform anything like as stated.




Ian