| 
  • If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • Buried in cloud files? We can help with Spring cleaning!

    Whether you use Dropbox, Drive, G-Suite, OneDrive, Gmail, Slack, Notion, or all of the above, Dokkio will organize your files for you. Try Dokkio (from the makers of PBworks) for free today.

  • Dokkio (from the makers of PBworks) was #2 on Product Hunt! Check out what people are saying by clicking here.

View
 

Fresh air for the little toy

Page history last edited by PBworks 15 years, 8 months ago

Fresh air for the little toy

 

USE AT YOUR OWN RISK

* __The modifications are not adequately verified and tested, so if you try to modify your own equipment, it's on your own risk.__

* __If you have any doubts, do not perform the modification.__

* __If you want to complain because any failure of your experiment don't contact me. I am just telling my little story here. You are aware.__

 

The unit is a "refurbished" one, so no warranty: no problem to modify it! So let's play on it.

 

Everybody knows the heat problem of the WGT634U. The hole unit get quite warm when operate, indeed it is necessary to keep the router vertical to favorite the natural cooling. The processor and other component get pretty hot at touch. While I was building my serial console I decide to solve the heat problem as well. Here my 2/cents solution:

 

My WGT634U
The box open

 

Definitely nicer than this one, serious cooling problem there!

 

First I did some research on internet on the chip used to regulate the voltage. Two AIC1577 are used to convert the 12V to suitable voltage level. Now some assumptions: one is probably for processor and logic circuit (I believe run at 3.3V like the serial consol). The other is for the USB port that requires 5V. The datasheet claim "Operating Input Voltage from 4.5V to 24V" so the voltage input can theoretically be as high as 24V (though, I suggest nobody to try to run the router at so high voltage) and as low as 4.5V (maybe some volt more).

 

I got a nice tiny fan from an almost dead laptop. The sacrificial unit was an old Pentium75. The fan and some connector have been useful for this project.

 

The fan is DC 5V 0.3W. Now the problem is where to find the 5V and an easy point for soldering. It is difficult to solder a new connector to a surface soldered board. One solution is to use the power from the USB port, but this requires soldering two wires on the bottom of the board and taking them to the top... long path. I than look at the AIC1577 datasheet, the Fig.17 show a typical circuit for "Large Power 5V/5A Step-Down Circuit". This seems the circuit used on the board. Output pin is on the inductor. See the red square on the picture. Easy to solder there! (I suggest to use flexible wires.) This is the positive terminal. Negative terminal can be taken everywhere on the board. Near the 12V connector there is the predisposition for a zener. The soldering point close to the network ports is the negative.

 

Here I solder the positive terminal
Inside the box

 

Now it's time to prepare the cover for holding the fan. I decide to place the fan directly over the processor with fresh air blowing from outside on the CPU. I also add a switch to make possible to turn the fan off. It can be a bit noisy if you like to sleep with your WGT on the bedside table. Moreover, adding a diode (0.7V drop) and two resistors (4.7 and 8.2 ohm did the trick for me) I have a 2 speed fan. The full speed fan is a bit noisy, but you can hardly hear the minimum speed (just above the stall speed). Can be a good idea to use a variable resistor to accurately tune the speed of the fan. Or even more advanced built a temperature control.

 

The fan and the switch from the top...
...and from the bottom

Two word on the serial consol. I use the MAX3233E from Maxim (sample). You do not need any external capacitor, just one optional to filter the voltage (that I used). They came in a nice plastic box that can be used to hold the circuit. The wires can pass through one of the ventilation holes between the cover and the transparent plastic. In this way the serial console can be connected without open the box, Pretty solution. And it works!

 

Under the cover
Serial console cable through the ventilation hole

The processor is the main source of heat. With the fan on (minimum speed) it stays cold all the time. The other components benefit as well from the fresh air. The whole unit is now up, running and fresh even in the warmest summer days.

 

The result

 

- Nico


A Critic:

There is probably forum place to say this, but I can't find it, so...

 

I would think it perferable to use a 12 volt fan, instead of a 5 volt fan. This must be connected in a different place. One could take 12 volts right of the adpater jack (the soldering will be a lot easier :-).

 

A 5 volt fan will cause the power regulator circuit to work harder, generate more heat, and decrease power available for USB - which is already a problem with the stock unit. Most 12 volt fans can tolerate running off unregulated power, and won't have any of the previous problems + I THINK the AC adapter is over-rated and has the extra power to drive a fan without affecting anything else.

 

Other approaches

Another WGT owner reports postive results from changing from the supplied 12 volt ac adapter to a 7.5 volt adapter. From the manufacturer spec sheets the power circuit will operate considerably more efficently and generate less heat. (If Netgear used the manuacturers example ciruit the minimum input voltage would be 6 volts.)

 

Do We Know What We Think We Know?

We think we know the CPU overheats. We base this on the fact the CPU gets very hot, Netgear refurbished the units with heat sinks on the CPU, and if we add cooling then unit works better.

 

However my unit came with a heatsink on the CPU and still overheats. Did Netgear get it wrong a 2nd time. Is the thermal cutout coming from somewhere else - like the power circuit? One reason to supect this is: presumably the units passed testing during development. I conclude either it was never testind inside the case or something changed after development. The case is Netgear standard, so I'm betting it was tested in a case. That leaves changing something after delevopment: software or the AC adapter.

 

Just wondering.

Comments (0)

You don't have permission to comment on this page.