I’m sure a lot of you have seen prefab monitor cooling fans being sold by various vendors. I’ve seen them for quite a while and for a long time wondered how well they work.
If your monitor has been on for an hour or so, it’s very easy to find out how much heat the picture tube can generate. Just put your hand over the vents.
CRT monitors are cooled by convection. There are vents in the bottom and top and as the components inside get warm, the air inside heats up and rises out the top, drawing cold air in through the bottom vents.
Is this enough ventilation?
I wondered how hot the inside of my monitor got so I decided to measure it.
The thermometer I used was a $9 generic indoor/outdoor thermometer that I got from Radio Shack. It has an encapsulated thermistor on the end of a nine foot cord.
I drilled a small hole in the top of my monitor directly above the neck of the picture tube and placed the sensor just inside the case.
After an hour of operation the internal temperature of the monitor had reached 40C – ambient temperature was 23C.
My monitor is a 21″ Trinitron and the replacement cost would be sky-high. It’s not under warranty anymore, either. I decided that the best way to insure longevity would be to lower its operating temperature and hopefully extend component life. I thought some sort of cooling fan would work wonders with the temperatures, so I set out to find the best option.
The prefab coolers have several characteristics I see as flaws.
First, they simply clip into the vent slots atop the monitor’s shell. This cannot seal well against the case, which allows much air to be drawn into the fan from the outside air rather than from inside the monitor.
Second, it is very difficult for their small fans to pull much air through the cooling vents in the shell; the cooling vents are pretty narrow.
Finally, they run on 12 volts DC. The way they get this power is from a plate that fits into an open expansion slot in the case of the computer.
On the inside of the computer’s case, a power wire from the computer’s PSU is connected to the expansion slot plate. The monitor cooler has a six foot wire that runs all the way from the other side of the expansion slot cover to the monitor cooler. This runs the fan whenever the computer PSU is on.
This wire is ungainly and adds to the rat’s nest of cables that most of us have behind our computers.
I thought of a few improvements that I could make. I wanted to embed a fan inside my monitor. I wanted that fan to run from a power source inside the monitor, and I only wanted it to run when the monitor’s filament was on.
All this would get the fan and its wiring out of the way, work more efficiently, and only work when the monitor was creating a lot of heat.
My first task was to find a fan that would not cause electrical interference with the monitor. Some fans have motors with electromagnets strong enough to distort the picture of the screen if they’re too close, and I didn’t want that.
Since the monitor is pretty big and I didn’t want any more noise than necessary, I decided to use a 120x38mm fan.
I had three candidates on hand:
1) The NMB model is a Low Noise Boxer and is rated at 88CFM at 37dBa, 120VAC
2) The Comair-Rotron is a Whisper XL series and is rated 51CFM at 32dBa, 120VAC
3) As a last resort, I also decided to keep a Panaflo H1A handy in case the AC motors of the other two candidates generated EMI. The Panaflo is 103CFM at 45dBa, 12VDC.
I wired up the fans and tested them all for EMI by placing them atop the monitor while they were running. None of them made any distortion in the picture so I decided to use the Rotron fan as it was the quietest of them all and 50CFM would be more than enough to drop the temperature quite a bit.
Now that I found my fan, the real tricky part was next; the wiring.
I wanted the fan to only be on when the picture tube heater was on because the heater and its associated power supply are the components in the monitor that generate almost all of the heat.
The heaters of picture tubes are almost always run from a 6.3 volt power source, and it is almost always DC.
The best way to create an effective switch was to use a relay with a low voltage coil, one that would close at 6 volts. This way I could switch the 120 volt fan motor with the filament voltage.
I took the shell off of my monitor and was faced by the EMI/RFI shields inside it.
I removed these light-gauge aluminum shields and they revealed the small backplane circuit board that the CRT plugged into.
I took my multimeter and went to work trying to find the two pins on the board that the tube heater filament plugged into. I looked closely at the circuit board and found a common ground, then I found a pin blatantly marked “HEATER”. This was fortunate.
I set the multimeter to read low DC voltage and checked the voltage between the heater pin and the chassis ground. It was exactly 6.3 volts DC.
I soldered a wire to the heater pin and a wire to the chassis ground.
Now that I had the power to actuate the relay, I needed to find a 120 volt power source for the fan.
After some quick investigation I found the only 120 volt source inside the monitor: the input to the monitor’s low voltage power supply board.
The white and brown wires are the leads going from the IEC power in connector to the LVPS board.
Here is the LVPS board tilted out of the monitor.
I located the pins for the 120 volt connector on the back of the LVPS board and soldered a wire to each of them.
I ran the 120 volt power wires and the wires from the filament outside of the EMI/RFI shields and put the shields back on.
I used a relay with a 12 volt coil, but made sure that it would actuate at 6 volts. I mounted the relay on the outside of the shields with double-sided foam tape, then I connected the wiring to it:
Blue – 120v from LVPS
Green – 6.3vdc from filament pins
Red – Relay
Yellow – Switched 120vac out to fan
I used a pigtail to connect the relay’s switched power to the fan. The pigtail had bare wires on one end and a molded plastic connector for the .120″ blade terminals of the fan on the other end.
I used a 120mm hole saw to cut a hole in the middle of the top of the shell. I drilled holes to bolt the fan to the case with and then I bored a larger hole with my dremel just next to the fan.
This hole needed to be large enough to fit the molded connector of the power cord through, because it was already attached inside the monitor. I had to mount the fan itself externally because the EMI/RFI shields didn’t allow me enough room to mount it internally.
I fed the power cord through the shell, put the shell back on and viola, job done.
I turned the monitor on and the fan started up. When the monitor went into sleep mode the fan stopped. I could hear the relay click it on and off. Perfect!
I hooked everything back up and disconnected the fan for a bit. I allowed the monitor to warm up to its normal 40C operating temperature for an hour. Then I plugged the fan cord back in.
After five minutes of the fan running it was down to 26C, just 3C above ambient. Thanks to Delta Electronics and their whining fans, I can’t hear my monitor fan over my computer and it does nothing to affect the picture on the screen.
Elapsed time for project: 2 hours.