Concerning CLR Discovery

[jay mitchell]

I thought about that plastic barrier reducing the air flow on the inside, but I figured most of the heat is going to escape through the metal plate

Since there are heat-generating components that are not mounted to the plate, there is a significant amount of heat that is not conducted to it. That heat is transmitted to the air inside the bucket.

and the additional air volume would eventually warm up to a very similar temp with or without it.

That's correct. "Eventually" could be quite some time, however, and there will be a small amount of air exchanged with the outside world due to the pumping effect I referred to above.

The best solution - one I called for as part of the "MKII" revisions - would be to ventilate the plate. That would have been feasible during manufacture; it would be a huge project for a DIY.

 

[Chocol8]

My assumption was that the heat from the air would mostly transfer through the plate slowly.

A few strategically placed holes and a small blower wouldn’t be too hard to retrofit. It would take very little air turnover to make a significant difference.

 

[jay mitchell]

My assumption was that the heat from the air would mostly transfer through the plate slowly.

Unfortunately, it doesn't work out like that. I had an experience that helps illustrate the situation. A CLR I had been sent to service would completely stop passing signal after it had been powered up for several hours and then wouldn't reboot. It would only wake up after it had cooled down for a similar amount of time. I removed the module, connected it to my load bank, and left it on for several days with a signal applied and a scope monitoring the output. It never shut down; the added air circulation with the module out of the bucket was enough to make the difference. It turned out to be a bad power amp module.

A few strategically placed holes and a small blower wouldn’t be too hard to retrofit.

By all means, have at it. If you do, please take photos and share your project here. There may be others who will do likewise.
Edit: The placement of ventilation openings should exploit natural convection and create airflow across all the electronics, whether the speaker is used as a wedge or (rotated 90 degrees) for pole-mount or backline use. For that reason, the openings should be diagonally opposite each other, e.g., one on lower left, the other on upper right. There are areas on the plate where holes can be drilled that are consistent with this requirement.

It would take very little air turnover to make a significant difference.

If you're willing to do the work, that would be the best solution by far.

 

[Chocol8]

I may take a crack at the ventilation project. I will pull it apart again tonight to take a look at potential drilling locations and places to tap into DC voltage, and what voltages are available. I have an extra DC step down module, so I can step down anything under I think 55v, otherwise, if I need to tap into AC, I will probably need to order something.

 

[jay mitchell]

I may take a crack at the ventilation project. I will pull it apart again tonight to take a look at potential drilling locations and places to tap into DC voltage, and what voltages are available. I have an extra DC step down module, so I can step down anything under I think 55v, otherwise, if I need to tap into AC, I will probably need to order something.

The power supply provides unregulated bipolar DC voltages., +/- 21V in the case of the SMPS400A400, +/- 23V for the SMPS400A180. I suggest you try passive ventilation first. That may be sufficient to keep things from getting too hot.

 

[Chocol8]

Edit: The placement of ventilation openings should exploit natural convection and create airflow across all the electronics, whether the speaker is used as a wedge or (rotated 90 degrees) for pole-mount or backline use. For that reason, the openings should be diagonally opposite each other, e.g., one on lower left, the other on upper right. There are areas on the plate where holes can be drilled that are consistent with this requirement.

I had a useless Zoom call, so I shutoff my camera and pulled the amp. The smallest 5v blowers I have will not fit on the upper part of the plate without drilling through the heatsink, or relocating the power switch to below the power socket. Even if I 3d print a duct, I don't think there is room over there to fit the blower in any orientation because there is not much space between the power supply and the bucket, especially with the metal plate on the bucket over the PS. I need about 6mm more space due to the plate being recessed.

I am thinking the forced air from the always on blower will overcome any minimal convection, so I am probably going to mount the blower on the middle left over by input 1, just below the heat sink. Air intake can be done with a few holes or slits to bring air across the parts I want, plus create enough turbulence to not have anywhere that will collect heat.

The PS documentation shows aux power of about 20v across pins J5.3 or J5.7 and J5.5 but those are in use by the input board. @jay mitchell do you see any issues with me tapping into that power?