Fan Test

Jun 13, 2016 9 Replies

Hi,


The following Fan was used in several of our prototype systems



formatting link


The thing is that one of the FAN in one of our systems broke down due to unknown reasons ( mechanical or electrical ) . The FAN ran 5000 cycles.



I send the FAN to the manufacturer and the manufacturer said that one of the IC on the FAN was burned out. But they did not know , how and why?



So, now I need to test the FANS so they can fail. My question is what is the best way to do testing.



I am planning to run several FANS ( quantity : 60) at the same times with +24 V DC and measure current. How many of these FANS should I test?



I will be using temperature chambers and mechanical stress fixtures. Should I also measure airflow if yes, then how?



jess


scritto:

ld I also measure airflow if yes, then how?

IMHO air flow is not necessary, but measure current and speed. If current/a nd speed change from nominal (higher current and/or lower speed) then somet hing in the motor is going to break very soon.

Bye Jack

Hi,

How can I measure the speed of the Fan?

jess

a scritto:

ould I also measure airflow if yes, then how?

/and speed change from nominal (higher current and/or lower speed) then som ething in the motor is going to break very soon.

use photocell to count the blades passing.

Here's a MIL standard for electric motors:

formatting link

Look at method 403: Life expectancy. It's for AC motors, but should give an idea of what and how to monitor. Send me one, I'll do life test on it:

formatting link

You plan on testing 60 fans to destruction?

That sounds like a pretty ambitious, possibly costly, undertaking. There's got to be a better way...

To test RPM a photo tachometer will do the job - most are hand-held and hobby shops carry them to measure prop speed on model planes. But if you want to monitor speed on 60 fans there may be some way to measure current and count pulses. Small DC fans, energize the internal coils in sequence - that draws power in pulses from the power supply and those pulses can be measured and counted.

Many computer fans output a pulse just for that purpose or accept a signal to adjust the speed.

But again there's bound to be a better approach. The manufacturer should be the expert on testing their fans why not contact them for their test procedures?

Some years ago I built a power supply to control a fan based on temperature. I wasn't too concerned with absolute speed just so long as the temperature didn't go too high (and I had progressive alarms for that, one beep every 30 minutes if the ambient was 80F, 2 beeps @

85, etc. or continuous alarm to indicate a too high temp or bad sensor)

o unknown reasons ( mechanical or electrical ) . The FAN ran 5000 cycles.

f the IC on the FAN was burned out. But they did not know , how and why?

s the best way to do testing.

th +24 V DC and measure current. How many of these FANS should I test?

ould I also measure airflow if yes, then how?

these kind of fans may have a feedback signal that give the speed (PWM, squ are wave with speed proportional to frequency, ...). If it has it, then it' s easy ;)

Bye Jack

if there's not a speed wire, shine a light through the blades, measure the blink rate.

\_(?)_

But what are you going to do, run them till they die, how long will that take? Months, years? That why they test at accelerated levels, to simulate a few lifetimes in a few weeks. You need to run at elevated ambient temp, at least. High humidity and heavy clogged air flow.

The data sheet says that's what the yellow wire is for.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required