$9 Router Cooling (RT-AC68U Example)

[martinr]

A mechanical engine and a solid state processor are not comparable. As RMerlin stated.

A mechanical engine wears down even at idle speeds, depending on how long it is run and how many on/off (cold/warm/hot) cycles it endures before it's next 'tuneup'.

A solid state processor has a safe limit (125C in this case) within which it can run continuously at.

I agree to keep it as cool as possible; but if the specific chip you have runs stable without any issues at half the rated tdp, keeping it even cooler is for your benefit, not the router's.

Indeed. Themal and mechanical cycling are the real killers. My router happily sat at 80C for 18 months, and no doubt would have carried on indefinitely at that temperature. And if there were no temperature data we'd all be none the wiser (and wouldn't be unnecessarily tinkering, adding fans, drilling holes, disturbing heat sinks). I'm just glad there's no data for the cpu temperature in my TV or car ECU and especially my iPhone!

 

[ColinTaylor]

And for those of us in the UK over a certain age who remember the saga of the fan-less Amstrad PC-1512:

Alan Sugar said

I’m a realistic person and we are a marketing organisation, so if it’s the difference between people buying the machine or not, I’ll stick a bloody fan in it. And if they want bright pink spots on it I’ll do that too. What is the use of me banging my head against a brick wall and saying ‘You don’t need the damn fan, sunshine’?

http://www.i-programmer.info/history/machines/1364-alan-sugar-and-the-fall-of-amstrad.html?start=2

 

[martinr]

And for those of us in the UK over a certain age who remember the saga of the fan-less Amstrad PC-1512:

Alan Sugar said
http://www.i-programmer.info/history/machines/1364-alan-sugar-and-the-fall-of-amstrad.html?start=2

Fascinating, Colin. My brother, being an electronics journalist, will also enjoy reading that article. It just goes to show all the factors involved in a successful chip-based product, many outside the control of the manufacturer. Guess what I'm going to stick on the back of my router next to the fans? A couple of pink spots.

 

[martinr]

... . I'm just glad there's no data for the cpu temperature in my TV or car ECU and especially my iPhone!

I knew I shouldn't have joked about a fan for my iPhone:

From today's posting by Brian Krebs.

"Harrigan and Kelley said this apparently creates havoc with most of the applications built into the iPad and iPhone, and that the ensuing bedlam as applications on the device compete for resources quickly overwhelms the iPad’s computer processing power. So much so that within minutes, they found their test iPad had reached 130 degrees Fahrenheit (54 Celcius), as the date and clock settings on the affected devices inexplicably and eerily began counting backwards."

http://krebsonsecurity.com/2016/04/new-threat-can-auto-brick-apple-devices/

 

[red_pope]

I still use a small floor fan set at low speed for the summer. Since, I did relocated the TPLINK Adsl router, switches and the ASUS WiFi, to a small storage room, the fan is only use to boost the air circulation around the small room, there is no windows in that room. The Asus and the ADSL TPLINK does gives a small amount of heat, that is why it has huge heat sinks inside. I keep them dust clean, I did removed the carpet and use laminated wood floors in all the house, easy to maintain and clean.

 

[martinr]

I still use a small floor fan set at low speed for the summer. Since, I did relocated the TPLINK Adsl router, switches and the ASUS WiFi, to a small storage room, the fan is only use to boost the air circulation around the small room, there is no windows in that room. The Asus and the ADSL TPLINK does gives a small amount of heat, that is why it has huge heat sinks inside. I keep them dust clean, I did removed the carpet and use laminated wood floors in all the house, easy to maintain and clean.

Might be worth monitoring the room temperature with and without the fan running: if the air in the small room is not being refreshed because the room is well sealed then the heat generated by the fan's motor (and the fan itself) will only serve to raise the ambient temperature. You could be better off without the fan. A couple of vents in the door, top and bottom, might be more effective than the fan. Worth checking with a thermometer, anyway.

 

[TonyH]

Maybe getting another fan and mount the fan so that it will blow hot air out?

Blow in or Suck out, whatever. As long as air is moving in/out. When I encounter laptop, desk top or router running
hot, not to my liking. I have a habit of taking off the heat sink and take a look to make sure there is enough of thermal paste or thermal pad with good contact. Some times I apply high quality paste after removing original one. Some routers
seem to run hotter than others. So far I never experienced heat trouble with any router I used.

 

[martinr]

..... Some routers
seem to run hotter than others. So far I never experienced heat trouble with any router I used.

Indeed. And as Merlin said, who are we to decide what's too hot? We are only going off our subjective assessment: 60C is as hot as our hands can stand to stay in contact with a body; 100C is boiling water we use to make tea. All very nice, but not very objective. However, for those of us who admit to being tinkerers, it's plausible that a router run continuously at 60C might last longer than one run at 80C. But given we would have replaced it decades earlier for other reasons, it's all hypothetical supposition.

Anyway, the fact we all run custom firmware is indication that we are, at heart, tinkerers!

 

[TonyH]

Indeed. And as Merlin said, who are we to decide what's too hot? We are only going off our subjective assessment: 60C is as hot as our hands can stand to stay in contact with a body; 100C is boiling water we use to make tea. All very nice, but not very objective. However, for those of us who admit to being tinkerers, it's plausible that a router run continuously at 60C might last longer than one run at 80C. But given we would have replaced it decades earlier for other reasons, it's all hypothetical supposition.

Anyway, the fact we all run custom firmware is indication that we are, at heart, tinkerers!

From my working day experiences, super computers, water cooled mainframes, 90C is shut down temperature. If things running too hot, noise(shot noise) level increases, components become sluggish. Hooking up a 'scope to a circuit will show that. In electronics anything heat is not a friend. Simply compare the specs. temp. range of any component for military version and consumer version. In the '70s I made an electronic CD ignition replacing conventional coil and point. After couple failures, I had to replace all the parts with mil-spec. ones which solved the problem. Good thing I installed a switch so I could run on OEM ignition or the other. Did not get stranded on the road side.

 

[mgeoffriau]

Just did a fan mod on my ASUS RT-N66U last night. I was tiring of the 2.4Ghz radio dropping connections constantly. My reported temps weren't outrageous at all (in fact, I'd say at 57° C and 55° C, they were pretty low compared to others), but I've already eliminated as many other factors as I could. So, if it is a hardware issue, I'd rather try a $10 fan first before investing in a new router.

I purchased this USB 5V fan: http://www.amazon.com/dp/B00G05A2MU/?tag=snbforums-20

I marked out a pattern on the top cover of the ASUS case using a fan template, then drilled a bunch of ventilation holes. I mounted the fan on top of the case, blowing out. Got everything mounted back up on my network board, and powered it all back on.

The results:

Radio    Before Fan    After Fan (High)
2.4Ghz   57° C         40° C
5 Ghz    55° C         39° C

A few additional thoughts about this project:

1. I know there's plenty of debate about the need for more cooling, and that several people with much more insight than myself have weighed in on the opposite side. Like I said above, I was running out of options for addressing the 2.4Ghz radio problem, and this is a low-investment stab at a solution. If it doesn't work, no big loss.

2. I don't claim to be smarter than an ASUS engineer...but for a passively cooled router, the physical layout is pretty disappointing. The router is designed to lay flat or sit upright, but all the vents are on the back and sides. Additionally, the mainboard almost seals off the heatsink (which is mounted on top) from those vents. And, of course, heat rises.

3. Just a few changes would (probably? maybe?) make a difference -- putting some vents on the top so that the heat radiated by the heatsink can escape directly rather than being absorbed into the plastic case. Or making the case slightly larger so that there is better airflow from the front of the case to the back of the case. Maybe ASUS tested this stuff and had a reason for doing it the way they did. But it's still surprising to me. I understand the reasons to do passive cooling instead of active cooling, but not the reasons to do passive cooling in the manner they did, unless it was really just an accountant's decision that it was good enough and they weren't going to retool the case after dropping the fan from the first version.

4. 120mm barely fits on the case, as you see. I would probably do this again with either a single 92mm fan or dual 80mm fans. My purpose for using a 120mm fan was to move as much air as possible as quietly as possible, and I did manage to make it fit.

5. The fan switch has 2 intermediate speed settings, so I plan on testing those lower speeds to see how it affects the temp/stability of the router. The router is in our work/project room, so absolute silence isn't necessary, but less noise and power usage is always good.

6. I tested the fan's power draw using a little USB power meter and found that it draws less power than the external USB drive I previously had plugged in to the router, so I'm not too concerned about the load on the router's power supply. That said, I did order one of those ASUS EEE-PC power supplies just in case, since I have seen several posts about the original power supply crapping out.

 

[martinr]

....... In the '70s I made an electronic CD ignition replacing conventional coil and point. After couple failures, I had to replace all the parts with mil-spec. ones which solved the problem. Good thing I installed a switch so I could run on OEM ignition or the other. Did not get stranded on the road side.

So did I! Can't remember much about it except it had a silicon-controlled rectifier, and I did have problems with it, and eventually I gave up on it and installed a Sparkrite. So it's only taken me 45 years to discover what the problem was. Thanks, Tony.

 

[mgeoffriau]

5. The fan switch has 2 intermediate speed settings, so I plan on testing those lower speeds to see how it affects the temp/stability of the router. The router is in our work/project room, so absolute silence isn't necessary, but less noise and power usage is always good.

Bumped the fan speed down to the medium setting yesterday evening and have been monitoring temps. Seems like it's staying pretty steady at 41° C (2.4 Ghz) and 40° C (5.0 Ghz). I'll take an extra degree for the reduction in noise and power draw. Will try the low speed setting later and see what that does to the temps.

 

[Eric Lieb]

Those are really good temps that you can overclock it.
You want the CPU to be kept below 80C for longetivity reasons (and to reduce heat build up)
Wifi chips below 50C will be stable.

My 2.4 ghz is sitting at 57C and my 5ghz is at 65 C... I take it I should buy a cheap cooling pad and place the router ontop of it.

AC66R btw

http://www.amazon.com/dp/B008RN1YZM/?tag=snbforums-20

Probably going to buy one of those and see how it works.

 

[elorimer]

My 2.4 ghz is sitting at 57C and my 5ghz is at 65 C... I take it I should buy a cheap cooling pad and place the router ontop of it.

AC66R btw

http://www.amazon.com/Fosmon-Notebook-Cooling-Alienware-Playstation/dp/B008RN1YZM?ie=UTF8&dd=VqP-PNNs0sRe2ulz4vPQtQ,,&keywords=usb cooling pad&qid=1461165987&ref_=sr_1_12&refinements=p_96:10155283011&sr=8-12

Probably going to buy one of those and see how it works.

The description says "Pulls" hot air out of the laptop. So the airflow is down, not up, for that pad. I have an N66, not an AC66, but perhaps you have a line of openings on either side along the top edge, and openings in the bottom, and on the beveled portion of the bottom. There is a large heatsink across the top over the cpu, switch and radios, and a plate across the bottom that might act as a heatsink. I don't quite see how the convection pattern works, but this cooling pad would reverse it, and unless there is airflow across the top heatsink, how will the temperatures be affected?

I got a cheap downflow for my 87 that didn't do much; I replaced it with an upflow that does much better.

 

[Eric Lieb]

The description says "Pulls" hot air out of the laptop. So the airflow is down, not up, for that pad. I have an N66, not an AC66, but perhaps you have a line of openings on either side along the top edge, and openings in the bottom, and on the beveled portion of the bottom. There is a large heatsink across the top over the cpu, switch and radios, and a plate across the bottom that might act as a heatsink. I don't quite see how the convection pattern works, but this cooling pad would reverse it, and unless there is airflow across the top heatsink, how will the temperatures be affected?

I got a cheap downflow for my 87 that didn't do much; I replaced it with an upflow that does much better.

The solution is simple, turn it upside down. Based on the photos the pad looks like it can be placed either blowing or drawing. I will see when I get it.

I also find it a little difficult to believe it would "pull" air out of the laptop, almost every laptop I have taken apart using bottom draw and side discharge (thru heat pipes) to cool a laptop, pulling against the laptop fan would not be a great idea.

 

[elorimer]

The solution is simple, turn it upside down. Based on the photos the pad looks like it can be placed either blowing or drawing. I will see when I get it.

I also find it a little difficult to believe it would "pull" air out of the laptop, almost every laptop I have taken apart using bottom draw and side discharge (thru heat pipes) to cool a laptop, pulling against the laptop fan would not be a great idea.

This one, maybe. Many are solid on the bottom, and push the air out the sides so it doesn't cook your....

Anyway, the one I got was downdraft and out the sides. Couldn't tell from the description and threw it out. After extensive dremel work to see if it could be salvaged. Also, I was trying to salvage an NAS with inadequate convection cooling, so downdraft was a problem. A big footprint lets me put other things on it--as would the three fans. The x-wings don't cover much.

 

[sfx2000]

Always try to exhaust air, not push air in....

Where it comes from, where it goes - not material - some like top, some like bottom, but at the end of the day, move the air, and the best way is to pull the hot air out...

 

[sinshiva]

Always try to exhaust air, not push air in....

Where it comes from, where it goes - not material - some like top, some like bottom, but at the end of the day, move the air, and the best way is to pull the hot air out...

since they're adding a fan, why not go the extra mile and add a filter and overpressure the case

I personally have had more success dropping temps by pushing with these small cases, actually, but I agree that the accumulation of particulates will eventually negate the benefits of the fan.

 

[Eric Lieb]

since they're adding a fan, why not go the extra mile and add a filter and overpressure the case

I personally have had more success dropping temps by pushing with these small cases, actually, but I agree that the accumulation of particulates will eventually negate the benefits of the fan.

That is why I own a shopvac! I never recommend blowing air in (like Canned Air) unless you can open it up so that you are sure the dust escapes and doesn't just get forced deeper in.

 

[Eric Lieb]

I just got the fan cooler I ordered, they blow up not down (so it isn't drawing out it is blowing in).


Edit: It isn't loud but you can hear a buzz from the fans... I would say it is working tho based on the temp drops I am seeing. Currently at 54/61 which better than before.

I am currently at 48/56 and it still may be dropping. I would say this was a success.