OpenWrt Forum Archive

Topic: MR3020 U5 Chip Burned.

The content of this topic has been archived on 30 Mar 2018. There are no obvious gaps in this topic, but there may still be some posts missing at the end.

Hi everyone.Recently i've managed to burn my mr3020.The u5 chip is done,and i've searched only the part to replace it myself.I think is a SN74AHC1GU04DBVR(datasheet:http://www.ti.com/general/docs/lit/getliterature.tsp?literatureNumber=scls343q) or a TC1185(datasheet:http://ww1.microchip.com/downloads/en/devicedoc/21335c.pdf) but i'm not sure.I've connected 2 pins from the u5 with a pice of wire and it's working , but it gets hot.I've tried to power it up from 3.7v and 5.5v and the problem with hot chips is the same.The usb port gets about 4 volts and it's decreasing slowly to 0v.The IC is like the one found on wr703n on U6.So i hope anyone can help me.Sorry for my english,but you get the point..

(Last edited by eightysixphp on 18 Jun 2013, 11:42)

If the purpose of U5 chip on MR3020 is the same as U6 on 703n, then you can safely desolder it and short pins 1 and 5 on the board. You will lose the ability to turn off the usb device from software but as a advantage you will get a free GPIO port on pin 4.

PS: Look at http://squonk42.github.io/TL-WR703N/

Thanks for the reply.I've already done this.But the board is getting to hot on the flash mem and the soc.The safest thing to do in my opinion is to replace u5 chip with a identical IC.I don't know..And the thing is that the router power itself up ,but the wireless it's not working...

Hi all,
may be it's too late... Anyway, I've just burned my TL-MR3020 too. The U5 chip popped up making smoke, I think because I supplied the router with 5.26V through P1 pins (it may be too hi).

So, I found this post and read the docs linked by strby. Thank you very much for the link and thanks to the author of that documentation.

Now, looking on my board and comparing it with the circuit schemas, it seems to me that the U5 of the TL-MR3020 corresponds to the U5 of the TL-WR703N and not to the U6.
eightysixphp, U6 may be just a bypassable switch, but U5 is part of the 3.3V DC/DC converter circuit!

Please tell me if I'm wrong.
Otherwise, this may be a solution:

  1. find and buy an external 3.3V DC/DC converter that takes in 5V (check max current limits)

  2. remove U5 (the burned DC/DC) and L14 (inductor);

  3. solder a wire in the place of L14, on the pad directly connected to the 3V3 line;

  4. connect the wire to the external 3.3V DC/DC (also connect the GND to any free pad, or to the pin in P1)

Yes, the chips referenced as U5 and U6 on the TL-MR3020 and TL-WR703N are the same (U6 is on top side on the TL-MR3020 PCB, bottom side on TL-WR703N):

  • U5 (marking: BWMM7AG) is an AME5258 step-down 3.3 V switching voltage regulator , along with the L14 ferrite-core self, C112 input filter ceramic capacitor, C109 output filter ceramic capacitor and feedback circuit R99 / R100 / C106

  • U6 (marking: 16GBC) is an USB power distribution switch TPS2051B clone, with its related C119 capacitor

So if you blew up U5, you just blew up the main 5V => 3.3V DC/DC buck converter. Is your router still working when you apply 5V to its output?!?

You may well have sent your TL-MR3020 to the Paradise of blown routers hmm

Anyway, the AME5258 is a Chinese chip that you won't find on the retail market, except in China.

(Last edited by Squonk on 1 Oct 2013, 15:26)

Supposing that U5 is the only blown part, what do you think about connecting an external 3.3V power supply, as I said above?
I can find a lot of suitable 5V => 3.3V DC/DC on electronics online stores.
Or do you thing it maybe easier to replace the U5 chip with an equivalent one, instead of connecting a complete (more expensive) external circuit?

It's not important, but note that the U5 chip may differ from one board to another. My U5 shows the code "IC2BG", but the same pin layout.

(After the accident, the LEDs of my router didn't light up anymore--in the first 1 o 2 seconds, at least--, so I immediately supposed that the burned part was about the power supply. I haven't done any other experiment because I'm afraid of burning the rest of the circuits.
I don't understand what eightysixphp says about "the router power itself up".
Maybe he simply read 5V on a test point?)

DC/DC converters are not interchangeable.

Feeding +5V to the delicate AR9331 with 2.5V-tolerant GPIOs and +3.3V normal VDD usually have catastrophic effects on the whole board sad

My guess is that it is already too late, as this voltage level for just a fraction of a second is enough to kill most chips onboard.

I suggest you to find a new (used?) router, and not hook anything above +3.3V to it, or even +2.5V for the GPIOs.

Thank you for the reply.

Today I succesfully tested my burned TL-MR3020 using an external 3.3V power supply.
I did what I previously written... The U5 chip was the only blown part and the rest of the circuit was miraculously saved.
The test: I connected my PC to the TL-MR3020 used as WiFi AP. I was able to access the Luci administration page.

My problem is solved.
In the next days I'll provide further details and photos.

You are lucky wink

And you are even more lucky, as I just figured out that the U5 chip (AME5258) is just a Chinese clone of the Linear Technology LTC1877 step-down 600 mA synchronous converter!

I just googled for a singular text extract "The comparator OVDET guards against transient" in the detailed description on page 8 of the AME5258 datasheet and bingo!

Squonk wrote:

You are lucky wink

And you are even more lucky, as I just figured out that the U5 chip (AME5258) is just a Chinese clone of the Linear Technology LTC1877 step-down 600 mA synchronous converter!

I just googled for a singular text extract "The comparator OVDET guards against transient" in the detailed description on page 8 of the AME5258 datasheet and bingo!

That's actually wrong.  But you're not too far off.  To help those in the future who google and find this thread, I will explain.

The part in these portable wireless routers is a 5 pin SOT-23 package, which, according to the datasheet, the LTC1877 is definitely not.

Here are the product lines I've found that are exact replacements (according to my searching on digikey and cross-referencing datasheets):

Linear Technology LTC3406 (I used this one on a TL-WR703N and I know it works.  Make sure to order the adjustable version.  I ordered LTC3406ABES5#TRMPBF )
ams AMS1234AD
STMicroelectronics ST1S12GR


This part is small, and if you don't have some experience with surface mount soldering, you will almost certainly break the unit.  You might also lose your replacement part somewhere, or fry it again in the future, so it is advisable to order more than one, just in case.

Hi everyone.Recently i've managed to burn my wr703n also. smile
Trying to power it through the hub with some generic AC-DC Power Adapter that outputed >6V instead of 5V (tested later) the LED blinked and the router didn't boot anymore.
Being a noob in soldering I've replaced the L14 and U5 in not the most delicate way..: http://tinyurl.com/mawbmor
Anyway I just would like to make it boot again.
So.. 1) was it the wright choice to get MB102 Breadboard Power Supply Module 3.3V/5V as an external 3.3V power supply
(to use with 6AA battery pack=9v) to solder it somewhere instead of internal converter?
2) How can I know with the multitester wheather the chips are not fried? Thanks in advance guys, really !

i buying 30 pcs AME5258 from a china seller,when get it,i will test it on wr703

The discussion might have continued from here.