Please forgive me for not being able to speak English!

Introduction page
Tested ambient temperature -40°C .. +50°C

About 60°C+ and room temperature only 20°C
Please answer
谢谢！

Tested ambient temperature -40°C .. +50°C

This means unit will work OK in room temperature +50C

About 60°C+ and room temperature only 20°C

Room temperature is 20C, and it works fine. It will work fine with higher temperatures too. You should not pay attention to the CPU temperature in this situation. CPU temperature is allowed to be high. Unit will still work ok.

However, the temperature is much higher than the previous product! Some worry it is easy to be bad!
I am now down-converting and use temperatures above 50°C

The surface temperature is 50°C+ and the core temperature is 60°C+. I am very worried that it is easy to be bad!

40nm! Small size and no cooling holes，Design flaws?

My cAP ac is also hot for me. I feels high temperature when I touch the back of the case. Now I can't check it's temperature because cli command /system health print do nothing. I will check it when I'll be at home.

It doesn't matter how hot it feels. The hardware can handle it and will work fine.

Hmm.. Not sure is this a (known?) bug or something else, but with 6.41.3 on hAP ac^2 it seems that the system/health is not printing anything:
[x@y] > system health print

[x@y] >

Also Health-page in Winbox shows only buttons, not any actual data. Not so big deal for me, but just mentioned.

Hmm.. Not sure is this a (known?) bug or something else, but with 6.41.3 on hAP ac^2 it seems that the system/health is not printing anything:
[x@y] > system health print

[x@y] >

Also Health-page in Winbox shows only buttons, not any actual data. Not so big deal for me, but just mentioned.

hAP ac^2 does not have a built-in temperature sensor

Dear mikro, you got an excellent 750gr3 - a classy body, a radiator on cpu, microSD ... What prevented the body for hAP AC ^ 2 more and standard white? Why this wretched black body?Was it really so hard to do? Is it really difficult to do more with a radiator and a larger body? Maybe you should outwit your designers and marketers?

p.s. want me to make for you a free case for hAP AC ^ 2 in SolidWorks?

Guys...

Man said no need to worry as the hardware will handle it.I am sure they have tested this for a while before putting it in the market. If You are really annoyed about this then try to add some small radiator. However, the device comes with warranty so I wouldn't worry about it at all...

Best Regards.

Do you believe they tested the temperature at least a bit more carefully than the wifi?

I also worry about the heating which means inefficient power consumption. Take the power consumption, deduct the radios and compare to rb750gr3...

Looked at the disassemble picture, found that the heat sink is mounted on the back and is not installed on the chip! Is this design useful?

image
https://wikidevi.com/wiki/File:MikroTik_hAP_ac2a.jpg
https://wikidevi.com/wiki/File:MikroTik_hAP_ac2.jpg
Is this design a major cause of poor cooling?

Good hardware but poor overall package (software and hardware).
Now the logical question - was Mikrotik aware?
Are there AC2 with correct placement of the heatsink?
Will Mikrotik partners be replacing units with the mentioned "design" or the users will be forced to fix the problem by themselves.

I measured the temperature with a pyrometer without a box - 52-55 CPU, the radiator - 50-53 degrees and this without special load (I replaced thermal paste with MX-2) !!! The temperature in the room is 24-25 degrees, what will there be in the closed box? 60-65 + without load?

p.s. I will repeat - what prevented to make a normal white classic case for this device? This is essentially a future hit!

hi Keyko
You are the surface temperature, I believe that the core temperature is 70°C+
I want to know if this is a strategy for the retirement of the mikrotik plan?

Is the performance too strong? Do you need to use temperature to exercise plan retirement?
Hope MT can improve heat dissipation！

Economy "2 cents" on the body and cooling. It was possible to make a PCB more and quietly fit everything on the upper side, and below make an aluminum diffuser as in the RB951 + normal size radiator on the top side. Let it be more expensive for 10-15 $, but the user is calm and you + in karma.

image
https://wikidevi.com/wiki/File:MikroTik_hAP_ac2a.jpg
https://wikidevi.com/wiki/File:MikroTik_hAP_ac2.jpg
Is this design a major cause of poor cooling?

i don't think that moving the heatsink to the other side would make things different.
the CPU would be cooled more efficiently, i suppose.
but the heat is heat, it is radiated either by the cpu or by the heatsink. w/o improved ventilation the case will heat up, regardless where you place the heatsink.

on the other hand, you can cool stuff through the PCB as well. this way Mikrotik could use a flat heatsing, while on the other side they would have needed to follow the various component heights.

by the placement of heat transfer blobs i figure the heat generated by all the radios is also transferred to the heatsink.

also good to know, the heatsink has a mass, so if it gets heated up to a certain temperature, it will radiate it until the temperature difference between the environment and the heatsink disappears. there is a "buffering" effect in play. all we get from the heatsink is to collect the heat from multiple points and radiate it out more evenly on a greater surface.

also good to know, the heatsink has a mass, so if it gets heated up to a certain temperature, it will radiate it until the temperature difference between the environment and the heatsink disappears. there is a "buffering" effect in play. all we get from the heatsink is to collect the heat from multiple points and radiate it out more evenly on a greater surface.

The whole problem is that there are no vent holes in the black casing and there is no convection. What prevented a more spacious case? Yes, even the same standard white casing from hAP AC would be better. So the question is - why in the cAP AC the radiator is 2x more than the hAP AC? Hardware they have the same, but the cooling - different.

image
https://wikidevi.com/wiki/File:MikroTik_hAP_ac2a.jpg
https://wikidevi.com/wiki/File:MikroTik_hAP_ac2.jpg
Is this design a major cause of poor cooling?

The back heatsink is the official recommendation, directly from the maker.

Maybe i can start the theory of poor wireless performance possibly due to high temps.

On paper hap ac2 and cap ac are such nice products.... but the reality of things is.....

and with Mikrotik being silent and all.... isnt helpful.

Mikrotik is not silent. They admitted that they worked on the wifi improvements already. You should check the forum more carefully before you say that.

I don't understand this discussion. Why are you measuring the temperature? I already answered that this is OK and it will work fine even it hotter conditions.

Well. If not more, just because the heat means uselessly consumed electricity...

50-60°C for a possible home-router on a wooden desk or so? Realy? I guess i wont try that stuff. Im new on Mikrotik....sorry, but i have to think about that.

My hAP ac2 is in a room at 27C ambient, with very little airflow. I don't use it for wireless, but it is being used for routing with an involved QoS setup, and The Dude running. I'm having no problems with it.

The ARM CPUs are much more advanced than MIPS. You have to expect they will consume more electricity, and dissipate more heat. But the package is rated for high temperature, so until you have actual problems with it that you can prove are temperature-related, don't be concerned.

Sent from my Pixel XL using Tapatalk

Hi, I have no issue with mikrotik's statement that device is designed for the temperature, however, I would like to ask: How did you measure the cpu temperature?
When i do: I get empty result (both winbox or console while running ROS 6.41.3 ).

I have 4 devices, all acting same so I do not believe it is faulty device. Could you please tell me how did you find out the CPU temperature in hAP ac2 ?

Hi, I have no issue with mikrotik's statement that device is designed for the temperature, however, I would like to ask: How did you measure the cpu temperature?
When i do:

Code: Select all

/system health

I get empty result (both winbox or console while running ROS 6.41.3 ).

I have 4 devices, all acting same so I do not believe it is faulty device. Could you please tell me how did you find out the CPU temperature in hAP ac2 ?

Infrared temperature gun+taobao.com or amazon.com!

The whole problem is that there are no vent holes in the black casing and there is no convection.

The black case has vent holes. With this design, the router should be mounted vertically, with cables from bottom and ventilation holes from top, so that convection is better.

Ok, I have seen a lot of posts... Great, everyone is worried about high temperature.. I will test this device in 35+ degree ambient temperature. If it burns out, let it be...

Best Regards.

I tested hAP ac2 few hours, more than 51C with full load don't see (23C in room).
P.S. temperature measurment with infrared termometr

I have accidentally (while installing dust cover for USB port) noticed that my hAP ac^2 feels less warm (around 35 C) than around 2 weeks ago (when it felt more like 45 C).
Has MikroTik implemented some thermal improvements (e.g. shutting down/throttling unused CPU cores) for hAP ac^2 in recent RouterOS versions?
P.S. I have ordered several dust covers from eBay and this one fits best and looks decent.

I have accidentally (while installing dust cover for USB port) noticed that my hAP ac^2 feels less warm (around 35 C) than around 2 weeks ago (when it felt more like 45 C).
Has MikroTik implemented some thermal improvements (e.g. shutting down/throttling unused CPU cores) for hAP ac^2 in recent RouterOS versions?
P.S. I have ordered several dust covers from eBay and this one fits best and looks decent.

I also feel this way. After the latest update, the temperature is much lower.

After the update, Hap AC2 temperature dropped by 10°C, even though the weather is hot today (at room temperature 30°C)

After the update, Hap AC2 temperature dropped by 10°C, even though the weather is hot today (at room temperature 30°C)

Did you update the firmware?

After the update, Hap AC2 temperature dropped by 10°C, even though the weather is hot today (at room temperature 30°C)

Did you update the firmware?

yes

通过我的 DUK-AL20 上的 Tapatalk发言

What I said before was wrong. The temperature is still very high, and it is the same as initially.

What I said before was wrong. The temperature is still very high, and it is the same as initially.

I happens immediately? Or after a couple of hours?
During a download/streaming session?

For anyone that is interested, hap-ac2 is a drop-in fit in the classic white cases.
I mounted mine in a RB750gl case.

Only thing missing is the mode button (next to the usb), which I don't use anyway.
Even the leds are in the right location.

Clearly Mikrotk had this as an option in mind, just as they did with hap-lite (tower vs. classic case)

I too bought this unit - my first mikrotik, but I do not think this temperatures are acceptable.

I have the ROS 6.42.6, I have downclocked the CPU to 488Mhz, and have increased the antenna gain to lower wireless power consumption. And yet the unit is probably close to 50C on the plastic to the touch(I don't have anything to measure it right now) in a room around 24-28C.
I even go as far as asking the seller if the warranty will void if I drill holes in the case Yeah well of course it will, but since I can't find one of the white cases I consider doing that, because this case can't dissipate 15w of energy, and if I want to run it "stock" without AC on, this to me is fire hazard. Going to even ignore how good for electronics is to run above 70C.

Just got it. Excellent device.
It is indeed very hot. I understand it works, but having the outside temp well over 50°C is not that fine. No electronics like that high temp. Having it in a standing position is much better (lower temps) but still hot. This black enclosure really needs some additional venting holes. I'm thinking about replacing the board into my old RB750G white box Or just drilling some extra holes on the black one.
I've already replaced the thermal paste with Arctic MX-4 (it lowered my 3770K temps with a passive Scythe Ninja by 5-8°C) so it probably helps here too. A bit.

@cszolee79 - How did you open the plastic case?

@cszolee79 - How did you open the plastic case?

Open the back with a screwdriver or whatever you prefer like so.
https://imgur.com/a/yDsg3dI

I upgraded to the hAP ac^2 for the faster CPU over the hAP ac, because my QoS rules were hitting the CPU limit.

So I'm using the CPU quite a lot, and have been running this for months in an ambient temperature of 29 C without direct airflow.

It feels hot, but it must have been designed for a high operating temperature, or it would not still be running.

It feels hot, but it must have been designed for a high operating temperature, or it would not still be running.

Heat does not kill electronics immediately. It happens over years. So it will work for, say, 2-3 years before it goes dodo.
My RB750G worked for ten years (then it cost €0.5 to replace the 3 capacitors and now it's back in business).
Overheating TP-Links died in 1-3 years.

Do you think it is worth buying this model? I would not want the router to constantly overheat

It does not "overheat," in that I have not noticed any performance problems when the router is "hot."

As others have said, the heat may contribute to premature failure of the device, but that is conjecture at this point.

Mine is serving a 150M/150M Internet connection with periods of high traffic utilization, a large list of egress QoS rules, and it's working perfectly for me. I bought this for the faster ARM CPU over the MIPS CPU in the RB962 (hAP ac) model that was routinely hitting 100% CPU utilization due to the QoS rules.

Suppose a hypothesis with a short life of the router due to overheating is false. then such question:
How much worse is the signal in real conditions compared to the 951 series router?

Sure, the Wi-Fi performance is worse, and by design. Apart from the initial issues that they have fixed, it specifically has lower EIRP and fewer chains.

I actually don't use the Wi-Fi on my hAP ac2. I have it enabled for worst-case management connections only; but I feed multiple RB962 (hAP ac) into it to serve the premise Wi-Fi.

But the wireless performance is a completely separate topic than the heat.

It doesn't matter how hot it feels. The hardware can handle it and will work fine.

In my case it does matter.
My router is sitting by the window and would stop working when under sun light.
I drilled the holes to help cooling. Will see how it goes.

you have a black router under direct sunlight? well... no wonder it gets hot.

Instead of drilling holes, even simple piece of white paper would help more

Looks like the holes helped
I think paper would not help THAT much
Anyway, the point here is that the temperature on the contrary DOES matter and the hardware can not handle very hot temperatures as was claimed above.

Nobody ever claimed that temperature does not matter.
Temperature does matter and it is stated in the specs: "Tested ambient temperature -40°C to 50°C" That means, it is guaranteed to work, as long as the temperature around the router does not go over 50°C. By putting it on direct sunlight you violated this condition, because there was additional directly applied heat (sunlight hitting black surface).

Drilled holes increase airflow, therefore decrease the temperature - obviously that helps, but it is unnecessary as long as you keep the ambient temperature in specified range.

1. Normis from Microtik support in post #7 says "it does not matter"
2. It was not DIRECT sunlight. two glass window was in between
3. Direct sunlight does not increase AMBIENT temperature

1. Normis from Microtik support in post #7 says "it does not matter"
2. It was not DIRECT sunlight. two glass window was in between
3. Direct sunlight does not increase AMBIENT temperature

1) It doesn't. Well, up until 50 celsius.
2) This is the definition of "direct sunlight".
3) Of course it does. It heats the case, that in turn heats the air inside. Your room may not be at 50 degrees, but certainly the air inside the router was. "Ambient temperature", in the manual context, does not equals to "room temperature". It equals to "temperature of the air surrounding the electronics". Since You decided to cook it on direct sunlight, the temperature limits were exceeded.

I have one of these units. It survived the Rio de Janeiro summer (37 positive celsius on the shade). I just didn't put it to cook on the sun.

The specs says "Operation temperature -40..50"
My understanding it is not the temperature inside next to the chip, it's the air temperature outside the unit.
The room temperature was 25 when it stopped working. I doubt sunlight through a window made it over 50.
Maybe my router is just faulty, but it does not stand up to the claimed specs.
I see this as a design mistake, the case is just too small, they seem to fix that in hap ac3

What makes you think the 25C degree temperature was the cause? It could have been any number of things, a power issue or a manufacturing issue (warranty case).

The specs says "Operation temperature -40..50"
My understanding it is not the temperature inside next to the chip, it's the air temperature outside the unit.
The room temperature was 25 when it stopped working. I doubt sunlight through a window made it over 50.
Maybe my router is just faulty, but it does not stand up to the claimed specs.
I see this as a design mistake, the case is just too small, they seem to fix that in hap ac3

Well, Your understanding is wrong. In this context (for ANY product) what count is the temperature of the air surrounding the device. It really doesn't matter if the room is at 19C, if you just put the poor router inside the chimney, with the fire roaring.

I can't say if sunlight through the window made it above 50, and your unit may well be faulty. What I can say is that the sun it didn't help it.

Do the test:
Put it on the sun, and see if it locks up.
Stick a sheet of paper on the window, to shade the unit. See if it locks up.

After this simple tests we will be able to say it the sun is affecting it. But one thing I guarantee you: my unit worked flawless with ambient temperature above 34C. I didn't test the 50C mark, but I saw it working above 34C ambient. And I mean the room was above 34C. The unit was on the shade, though.

It started when I moved it by the window.
It was locking up when under sun light (coincidence?)
After 2nd lock up I drilled the holes and it never happened again.
These simple events make me think the overheating was the issue.

It started when I moved it by the window.
It was locking up when under sun light (coincidence?)
After 2nd lock up I drilled the holes and it never happened again.
These simple events make me think the overheating was the issue.

It probably was. I didn't care much for the color choice on this model: to me a white case would be better. But I still believe that the "50C ambient temperature" is correct. What I think happened is that the black case got warmed up by the sun, and the "ambient temperature" (the air inside the case) exceeded 50C. When You drilled the extra roles the air got better flow, and everything worked out.