No, this particular board won't get very hot. But it would be a feature if it reported proper air flow in the crate that it controls.
We'd report local board temp of course, and the crate power supply voltages. That's easy.
Haven't looked at the schematic or app note, but piezo's work as both receivers and transmitters, even if they produce two types.
You can't say much about them if you don't.
Don't be a dill. Ultrasonic air-speed sensors are as old as the hills.
Read the schematic and explain it to us.
Simple: It's amplifying and detecting low-frequency noise from fan blades and air turbulence (a result of airflow) using PVDS piezo film sensors. As in, it's a fancy kind of low frequency microphone amplifier with a rumble detector.
They use PVDS in guitar pickups, it's very effective and easy to use. You can even buy it by the A4 sheet and cut it with scissors.
Happy?
No. Spinning fan blades make noise even if air flow is blocked.
More, usually.
It would be possible, maybe even practical, to analize the voltage and PWM into a fan and the tach output, and compute pressure drop or flow.
That wouldn't help my case, where I don't own the fans, but it might be useful elsewhere.
A small fan used in tach-only mode, PWM=0, would be a cheap flow or DP sensor.
The fan blade noise is probably much higher frequency than the rumble of vortex shedding airflow around items in the case. These sensors look to be optimised for this very low frequency rumble. A full schematic would give you more idea what frequency range they're looking at, but I suspect it's in the 2-10Hz range, not the 100's Hz that you'd get from fan blades.
A cheap electret microphone, located close to fan blades, makes a gigantic signal.
My, you really are as stupid and have as low an IQ as several group members are already saying about you.
The "start at high speed" mode has *noting* to do with removing clogging debris. That is nothing more than a false illusion you made up in your head because you are not intelligent enough to reason out the actual purpose for the high-speed start.
The high-speed start is because whatever model fan Dell sourced to build into your machine needs some seconds of 'full power' applied at startup in order to get spinning from zero RPM.
Further, high speed air-flow from the fans at 100%, but in the usual air-flow direction, simply will not dislodge clogging debris. The faster airflow, in the same direction as always, will only further embed any 'clogging debris' into whatever they are stuck upon.
The fans would have to start in reverse airflow direction for a short time to have any hope of "dislodging clogging debris".
Gosh, you're nasty. And probably wrong; nasty and wrong correlate.
I'll take that as "no."
This question asks about air flow temperature sensors but does not identify the maximum tolerance nor whether the majority of the air flow is used for cooling which requires it to be turbulent at the PCB surface with a plenum overhead.
So the heat flow design should be defined before a solution may be defined. MOST PC towers have poor heat flow over the PCB since the air height is much greater than the maximum component height. Thus the wind blows over the "treetops" rather than under. The cooling rate difference is significantly better with a plenum tightly over the board to increase surface air velocity which is far more important than volume air flow.
Given the lack of details, I can recall two solutions.
1) "Fan Fail" detect on a Shindengen triple Power Supply 1985 which used a "Fan Fail" logic signal derived by a heater resistor on the fan exhaust to a thermistor switch-controlled relay.2) Lambda AC-DC PSU active cooling on the transformer open frame which I modified with a Mylar plenum air flow just above the hot parts with a thermistor epoxied to the transformer to control fan speed from 0 to
100%. when the transformer case temp is between 40 to 50'C.Both fans would have to fail and PSU loaded at 100% @ 40'C ambient before the thermistor registers 55'C and fan fail.
Conclusion:
- self-heating triggers the alarm.
- Any thermal sensor works.
- but turbulent air flow air speed is better for collecting heat ( >=1 m/s) and laminar flow is more efficient for air flow with exhaust then cu.m. or CFM is irrelevant.
TS
If you want the rumble but not the fan blades, you want good LF response. How does the electret go at 5Hz?
I did it in a CAMAC crate. When the fan blade sweeps close to the mic, it makes a huge sharp spike.
But tachs are better.
I'm not sure if you're being deliberately dense here, because you're clearly smart enough to understand, I must assume you're just unwilling.
*air flow* instead of fan blades, you need to chuck away anything above 20Hz and look for the real LF stuff that comes from vortex shedding.That will be what the TI thing does, based on their choice of large PVDS pads, instead of electrets. This much has been obvious to me from the beginning, but apparently not to you.
Never miss a chance to be obnoxious. That's what most people are here for.
My CAMAC crate used microphones to sense fan failure, not flow. They were AC shaded-pole fans without tachs. Fans are more reliable now. The sleeve bearings on the AC fans tended to gunk up.
The TI note didn't say which sensors were used. Do you know?
It also didn't say how the flow sensing actually worked.
Why two sensors? You somehow understand the things that I don't.
Clifford Heath knows more than John Larkin. This is not a statement that John Larkin could process, so it isn't addressed to him.
We do try to educate John Larkin from time to time, but he is complacently ignorant, and writes it off as people trying to insult him.
I'll look these up.
One thing that strikes me is that some Minco units I looked at had thin silver wire leads, whereas the TO-92 leads will be kovar, and maybe the PTFD RTDs as well.
Yes. Probably best to feed the RTD using constant-current sources.
I would not depend on any single data point, as this data is likely to be noisy and impulse-ridden.
Some kind of moving-window median filter is needed at the least, to prevent random false positives and false negatives.
A minute. That ought to suffice.
What if the airflow is blocked by a cleverly-placed random piece of trash, preventing the fans from doing anything useful?
Probably need to also sense the base temperature of the SMD RTD regardless of air flow, to trigger shutdown in time.
Joe Gwinn
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.