Author Topic: Passive airflow ?  (Read 2022 times)

0 Members and 1 Guest are viewing this topic.

Offline Jan AudioTopic starter

  • Frequent Contributor
  • **
  • Posts: 820
  • Country: nl
Passive airflow ?
« on: December 25, 2019, 12:10:18 pm »
Hi, i am almost going to print my electronic synthesizer casing.
Does it help to add a few holes or a roster to keep the electro cool ?

If i search google for without fan, i only get stuff with fans.
So the air input has to be low, the output high placed, and air input close to the hottest parts.
Does it help without fan ?

Hoe do you keep your electronics cool in a passive way ?
 

Online iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Passive airflow ?
« Reply #1 on: December 25, 2019, 12:19:43 pm »
What is the power dissipation of your electronics?
Readers discretion is advised..
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 9333
  • Country: fi
Re: Passive airflow ?
« Reply #2 on: December 25, 2019, 12:22:43 pm »
Note that the passive convection cooling requires a large temperature differential to work. This means, you need to design the electronics to run (including proper lifetime analysis) at higher temperatures than with forced (fan) cooling.

In practice, it's hard to design the heating components to run much below 100 degC with passive cooling, while with fan cooling, it's fairly easy to design them to run at, say, 50 degC (assuming you specify maximum ambient temperature at around 30 degC). This is, of course, assuming that significant cooling efforts are needed at all.

The step #1 is to calculate the actual heating and approximate the cooling through the surface area of your case to see if you need the holes or not. If it runs cool anyway, why bother?

Often testing and measuring is much easier than doing even a crude calculation.

For example, if your components happen to run at, say, 60 degC with closed case, there is no point in adding holes because it runs cool enough anyway, and with such low temperature differential, air wouldn't circulate through the added holes anyway, so they would only bring dust in. But if you measure your components running at 120 degC, nearly destruction, adding a few tactical holes could easily bring it down.

Run the device long enough so that all air inside the case has time to heat up. This may mean hours, or over 10 hours in a large device.
« Last Edit: December 25, 2019, 12:24:59 pm by Siwastaja »
 

Offline jhpadjustable

  • Frequent Contributor
  • **
  • Posts: 295
  • Country: us
  • Salt 'n' pepper beard
Re: Passive airflow ?
« Reply #3 on: December 26, 2019, 08:20:31 am »
Aren't you concerned about tuning drift due to temperature coefficients?
"There are more things in heaven and earth, Arduino, than are dreamt of in your philosophy."
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 9333
  • Country: fi
Re: Passive airflow ?
« Reply #4 on: December 26, 2019, 08:47:18 am »
Aren't you concerned about tuning drift due to temperature coefficients?

Modern design practice in most electronic circuitry is all about designing stable circuits. Temperature regulation to avoid drifts will likely be more expensive than just designing it to be more stable to begin with.

Of course, in extreme cases, temperature coeffs can't be brought down enough, and more expensive temperature control is needed; typical examples include oven-controlled crystal oscillators or voltage references, but these apply in high-accuracy (often <10ppm) expensive lab instruments.

For an audio synth, frequency stability of a few hundred ppm is definitely good enough even for demanding users, and any crystal oscillator is capable of doing it over the full temperature range.

The question for the OP still stands, what are the losses? I would further say that a synth shouldn't produce very much heat, so if heating is a concern, I'd strongly suggest improving the design to reduce power draw, instead of increasing cooling.
 

Offline Jan AudioTopic starter

  • Frequent Contributor
  • **
  • Posts: 820
  • Country: nl
Re: Passive airflow ?
« Reply #5 on: December 26, 2019, 01:57:43 pm »
What is the power dissipation of your electronics?

 Total of 100mA.

Aren't you concerned about tuning drift due to temperature coefficients?

Yes, using voltage controlled oscillators and filter.

The question for the OP still stands, what are the losses? I would further say that a synth shouldn't produce very much heat, so if heating is a concern, I'd strongly suggest improving the design to reduce power draw, instead of increasing cooling.

I am printing a enclosure, and dont want the heat to build up inside.
Now i am trying to have some airflow and save some PLA filament.

Here is my thing up to so far in openSCAD :
* synth.txt (7 kB - downloaded 61 times.)
* frontpanel.txt (7.03 kB - downloaded 62 times.)

I already have some airflow at the backside.
 

Online iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Passive airflow ?
« Reply #6 on: December 26, 2019, 04:24:03 pm »
What is the power dissipation of your electronics?
Total of 100mA.
The power dissipated by your electronics would be something like the current multiplied by a voltage.
For example your total current is 100mA, and the power voltage is 12V, your P = 0.1A * 12V = 1.2W.
Readers discretion is advised..
 

Offline Jan AudioTopic starter

  • Frequent Contributor
  • **
  • Posts: 820
  • Country: nl
Re: Passive airflow ?
« Reply #7 on: December 26, 2019, 04:44:52 pm »
It runs on 15v going to a 12v regulator.
 

Online iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Passive airflow ?
« Reply #8 on: December 26, 2019, 04:56:16 pm »
Thus your total power dissipation will be 15V * 100mA = 1.5W (aprox).
From that 0.3W comes from your 12V voltage regulator.
Readers discretion is advised..
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 9333
  • Country: fi
Re: Passive airflow ?
« Reply #9 on: December 27, 2019, 12:41:24 pm »
1.5W is nothing to worry about, I'm assuming it's not a miniatyrized device. If you have some hotspotting component, spread the heat using a small internal heatsink. The plastic case likely has enough surface area to dissipate 1.5W just fine.

As a comparison, think about a typical 100W laptop power supply dissipating at least 10W in a small plastic case. They are well designed, though, and use internal heat spreaders to use all the plastic surface as efficiencty as possible.

Adding excess holes only brings dust and, in worst case, metallic objects inside.
« Last Edit: December 27, 2019, 12:45:01 pm by Siwastaja »
 

Offline Jan AudioTopic starter

  • Frequent Contributor
  • **
  • Posts: 820
  • Country: nl
Re: Passive airflow ?
« Reply #10 on: December 27, 2019, 02:42:25 pm »
Ok i am leaving it like this and printing.
Just have to wait two days now.
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: Passive airflow ?
« Reply #11 on: December 27, 2019, 02:56:47 pm »
Here is my thing up to so far in openSCAD :
(Attachment Link)
(Attachment Link)

This is a fun "tower" shape. I don't know how crowded it will be inside, but looks like there will be plenty of room so heat should not be a concern. Thing to do would be to analyze which parts of your electronics will dissipate the most power and focus on those. I'd bet it's the voltage regulator of your PS if this is a linear one. Heatsink it properly. Should be enough.
 

Offline Jan AudioTopic starter

  • Frequent Contributor
  • **
  • Posts: 820
  • Country: nl
Re: Passive airflow ?
« Reply #12 on: December 27, 2019, 03:54:43 pm »
I have no room for heatsink, maybe i can attatch it upside-down.
Its only 3 volt, you really think it will be hot ?
 

Online iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Passive airflow ?
« Reply #13 on: December 27, 2019, 04:30:07 pm »
The voltage regulator will dissipate P=0.3W=(15V-12V)*0.1A

The temperature of its package is approximately

Tpack = Tamb + Rth*P, where the thermal resistance Rth depends on the voltage regulator's package type and the size/type of the heatsink.

For example with Rth = 20C/W (no idea what is yours)

Tpack = 23C + 20C/W*0.3W = 29C
« Last Edit: December 27, 2019, 04:33:56 pm by imo »
Readers discretion is advised..
 

Offline Jan AudioTopic starter

  • Frequent Contributor
  • **
  • Posts: 820
  • Country: nl
Re: Passive airflow ?
« Reply #14 on: December 27, 2019, 04:48:52 pm »
29 degrees = No heatsink needed, betters.
I had a 12v to 5v @ 100mAh it gets so hot you cant touch it to long pressed, that is another case.
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: Passive airflow ?
« Reply #15 on: December 27, 2019, 04:51:39 pm »
Not sure you got it. Obviously with no heatsink, it's going to get much warmer. Consider the thermal resistance of the regulator package alone in this case. At 0.3W it's probably going to be in the 40ish °C or something like that. imo gave an example, but I doubt the 20°C/W are for the package alone with no heatsink. Still not too bad, but something to consider.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf