Fan-cooled small box questionable design ?.

A commercial mixer amp in a tiny box. By gut-feeling I don't like the relative absence of air cooling and would like some justification to increase the , at least output vent size - there is space for much larger holes, but not necessarily aesthetic.

2 switch mode power supplies plus 2 100W rms D class amps and an 8 channel mixer in a box only 10x6x2.5 inches. The 2 SMPS are "cooled " by a 12V 40mm fan blowing the internal air, not positioned near any inlets, over the heatsinks. Inlet vents , if all 1/4 inch sockets should have tight fitting plugs in all of them, are 30 x 2mm diameter holes. The other fan is 12V, 50mm fan that pulls air over the 2 amps and exhausts directly to the outside but via 8 small slots .44 inch long and .08 inch wide that seriously restricts the flow out. The owner reports the exhaust as being hot rather than warm and i've taken some measurements with thermometer etc with relatively low continuous sine to external dummy load. I intend giving the owner a thermometer to take some readings in normal use over an hour of guitar performance. I fixed, with sealing tape, a large diameter tube around the exhaust area to direct into a very thin plastic rubbish sack, with no perforations, to time and measure the air flow until about 2/3 full and holding the floppy bag up as well so I'm assuming relatively little back pressure or whatever the term is. I did not think, while inside , to note any amp rating as well as 12V, nor make and model, but generally 50 mm fans would be about 8 to 18 cfm cubic feet per minute. But from my calculation over 3 minutes then the flow rate was only 0.78 cfm. This amp is designed so it can be transported and used , recessed into one of the speakers, so not much heat lost through the walls of the box I would have thought, I'm advising, interim, for the owner to use the amp removed from the speaker. Any comments on the adequacy of venting?
Reply to
N Cook
Loading thread data ...

Name names !

There's plenty of bad design around especially if it's from Asia. Proper thermal management is a discipline that's rarely taught to electronics engineers. I had to learn it by researching for myself.

Graham

Reply to
Eeyore

40mm

all

as

thermometer

to

time

up

term

cfm.

would

an afterthought. How much added strain is there on an axial fan by stopping down the free flow rate by more than 90 percent , leading to premature failure even without being in a hot airstream. I'm thinking of kitchen extractor fans which are relatively small motors for their power , relying on the passing airstream of human scale temperature range passing over them.

Reply to
N Cook

Justify to who?

Reply to
Meat Plow

Using vol of air per min = watts / (air density*specific heat*temp diff*60) in metric terms

60 seconds in minute , taking air density = 1.205, Cp= 1.05 and conversion factor of 35.3 for cu m /min to cf/m and measured outflow of 0.78 cf/m , however I plug these measured results in,scaling for higher demand, does not look good for temp diff at higher output and fan duress apart from the electronics which could be in circulation dead spots.

400Hz source sine fed into one amp only, delivering 12.5 watts continuously into an external dummy load. After 40 minutes the exhaust temperature was still increasing but very slowly, so cancelled then. So heat from 2 switch mode power supplies and one idle amp with no load and the other one active plus contributions from the voltage regulators and mixer circuitry etc. Temperature from bulb of thermometer laying directly over the exhaust vent rose to 42 degrees C from a room temp of 24 degrees C.

Reply to
N Cook

I avoid the use of fans, they draw a lot of crud into the device, particularly problematic if the mixer is used in a venue where smoking is allowed. If the thing is not failing in short order, the heat is probably not an issue.

Reply to
James Sweet

You get what ?

You won't get anything like a fan's rated cfm figure when the airflow's restricted btw.

So, is it overheating ? Put a thermocouple on the heatsink and see what temp it is.

Graham

Reply to
Eeyore

larger

channel

exhausts

exhaust

an

area

nor

cubic

one

removed

The other figure is 90W consumption printed on the box at the mains inlet, presumably 90W plus 2x100W of audio, ie 90 Watts having to be vented from the box at full audio output. Plugging 90W in the formula and with 0.78 cf/m then temp diff would be 54 above ambient or 78 degree C with presumably temperature hot spots above that inside.

Reply to
N Cook

No, the figure on the 'plate' is meant to be the actual power drawn.

Have you been able to measure 2 x 100W continuously ? It may have circuitry to stop you doing that, I've seen that trick before.

And what is this unit ?

Graham

Reply to
Eeyore

Do you know or recommend any app notes for a small class D amp, say

50W, running off a non automotive 12V lead acid battery?

martin

Reply to
martin griffith

it

12V won't do 50W with direct coupling.

I'd be inclined to use a SMPS to provide suitable +- supplies to do the job then use any of the many claas D designs thar are kicking around.

Graham

Reply to
Eeyore

it

then use

Thats what I thought, I'm using the Phirrips/NXPee BTL stuff at the mo, with the big Boost caps and they can get a bit warm, and haven't played with D amps. Any recommendation on TI devices etc. or what to avoid? As they all claim to be near perfect!

martin

Reply to
martin griffith

then use

What are these big boost caps ?

recommendation on TI

Honestly I haven't any experience of them.

But of course !

Graham

Reply to
Eeyore

I think National Semiconductor has some, take a look on their site.

Reply to
James Sweet

Not on this particular design.

But I conclusively proved that the vent holes on the LinkSys cable modem are WAAAAY too small. Creating all sorts of long term reliability problems.

And truly bizarre symptoms.

And easily compounded if you stack a router on top of it. Cisco denies the problem.

The only way I've found these to remain reliable is to lean them at a 45 degree angle with free flow above and below.

--
Many thanks,

Don Lancaster                          voice phone: (928)428-4073
 Click to see the full signature
Reply to
Don Lancaster

then use

Class H where they charge a couple of4700u caps to Vcc, and *add* them to the supply lines, to handle peaks, like TDA1560

recommendation on TI

martin

Reply to
martin griffith

inlet,

from

54

above

Its a Yamaha Stagepas 300 , made 2005 someone's pic, no scaling on any pics I found, but only 10x6x2.5 inch box

formatting link

Mosfets knocked out maybe by recovered punch-through in softened, suspiciously small size 470uF 85 degree C output filter cap. Interestingly the pic i've referred to above would suggest an enlarged grill area in the bottom left corner. The bit that looks like a bar code is probably and extended grill that is not on the one I am dealing with, the grill is only from the chrome handle to the extreme left corner. I will probably bend up the corner of the cover where the outlet vent is, propped up on a 1/4 inch stand-off, and add a bit of grill. So increase the airflow by double, decrease the temperature by half and reduce the strain on the main fan. If it was my amp I would probably cut a hole in the case where the internal fan is to ameliorate matters as well.

Reply to
N Cook

Do they indeed ? I wondered. It must have a slightly adverse effect on the audio output I'd have thought.

Graham

Reply to
Eeyore

audio output I'd

How do you ADD the charge from a capacitor to the supply, surely it`s just added reservoir capacity all the time? or have I missed something? Is there any disadvantage to having more capacitance, can you have too much smoothing?

Ron(UK)

Reply to
Ron(UK)

audio output I'd

It's to boost the supply voltage using a charge pump sort of idea.

Graham

Reply to
Eeyore

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.