Spec Sheet Red Flags: When a Fan Datasheet Doesn’t Add Up

A datasheet is two documents stapled together: an engineering record and a
sales document. When the two agree, you have a part you can specify. When they
disagree, the difference tends to surface later — in a thermal budget that has
to be re-opened, or in a field return that nobody planned for.

None of the ten red flags below needs a laboratory to catch. Each one is an
internal-consistency problem: the sheet contradicts itself, states a number
without the condition it depends on, or quotes a figure that cannot be true
alongside the others on the same page.

1. Airflow quoted with no system context

“Maximum airflow” is measured at free delivery — the fan moving air with
nothing in front of it. It is the right-hand end of the P-Q curve, where static
pressure is zero. No enclosure works that way. The moment there is a filter, a
heatsink, a grille or a bundle of cable in the path, the fan moves to a
different point on its own curve and delivers less than the headline figure.

A sheet that leads with a free-air number and never shows the curve is
telling you the easiest number it can measure, not the number your design
needs. See
how to
read fan curves and find your operating point.

2. No P-Q curve, or a curve that does not match the headline

A curve is the one part of a datasheet that cannot be assembled from
marketing copy. If it is missing, treat every airflow and pressure figure on the
page as unverified.

If it is present, read it against the table. The maximum airflow on the table
should sit at the pressure-zero intercept. The maximum static pressure should
sit at the flow-zero intercept. If the table’s headline numbers sit somewhere
inside the curve rather than at its endpoints, the sheet is mixing operating
points — and the figure you were about to design around may be one the fan only
reaches under conditions you will never see.

3. A life expectancy with no basis stated

“40,000 hours” is not a specification. It is a number that becomes a
specification only when four things are attached to it: the failure criterion,
the temperature (and the measurement point for that temperature), the
confidence level, and the statistical model.

L10 bearing life, MTTF and MTBF are three different quantities with three
different definitions, and they are not interchangeable. A sheet that quotes one
of them without naming which — or without a temperature — cannot be compared
against a competing part, because you do not know whether you are comparing
like with like.

4. A life figure that ignores the bearing type

Sleeve and ball bearings do not share a life model, and they behave very
differently at the two ends of the temperature range. A sheet that quotes a
single long life figure while leaving the bearing type unidentified — or while
listing “sleeve or ball” as though the two were equivalent — is omitting the one
variable that dominates the outcome.

5. Noise without a measurement condition

Sound pressure level is meaningless without three qualifiers: the distance,
the position relative to the fan (inlet side, outlet side, or free field), and
whether the reading was taken in an anechoic chamber or in a duct.

A dB-A figure with no distance is not a specification. Two suppliers can
quote the same number and be describing conditions 2× apart in distance alone,
which is a difference of roughly 6 dB — larger than the gap most projects are
arguing about.

6. Speed as a single number with no tolerance

Nominal speed tells you nothing about the spread. A rated figure should
arrive with its tolerance, and the tolerance should be tight enough to matter:
if the design depends on delivered flow, it depends on the slowest unit that
will ever ship, not on the nominal.

7. A “voltage” that is not an operating window

A nominal voltage is a label. What the design needs is the range over which
the fan will still start, still run, and still deliver — including the
worst-case start-up condition, which is harder than steady-state running because
the motor has to accelerate the impeller from rest.

A sheet that lists only the nominal, or that lists a range without stating
what is guaranteed at the extremes, leaves the most common field failure in
industrial DC fans — failure to start on a sagging supply — completely
unaddressed.

8. Certification marks with no standard and no file number

A mark on a page is a claim. What converts it into evidence is the standard
the part was tested against and the certificate or file number that can be
looked up independently.

This matters most where the mark carries a legal obligation in your market.
Ask for the document, not the logo.

9. Protection claims without an IEC 60529 level

“Waterproof”, “sealed”, “IP-protected” — none of these is a specification.
Ingress protection is defined by a level (the two digits of IEC 60529) and by
the test arrangement that level implies.

It is also worth asking which surfaces were tested. In practice, the failure
point in an IP-rated fan is rarely the housing; it is the cable entry. A report
that does not describe how the cable entry was sealed and tested is describing a
different part from the one you will install.

10. Numbers that do not survive basic physics

This is the check that needs no reference data at all, and it catches more
optimistic sheets than any other.

The air power a fan delivers at an operating point is the volumetric flow
multiplied by the pressure rise. Divide that by the electrical input power and
you have the overall efficiency. Efficiency is a ratio of output to input, so it
cannot exceed one: a fan cannot deliver more air power than it draws from the
supply.

Now take the sheet’s own three numbers — flow, pressure, input power — and
put them through that calculation. In a real axial fan the result sits well
below one, because a small direct-drive fan loses a substantial share of its
input to motor losses, bearing drag and aerodynamic inefficiency. If a datasheet
implies a figure at or above one, then the three numbers are not describing the
same operating point. Something on the page has been quoted from a different
condition than the others.

You do not need to know the true efficiency to run this test. You only need
to know it must be below one.

What a datasheet you can actually specify from contains

• An airflow/pressure curve drawn to scale, with the table’s endpoints on it
• Airflow and pressure stated at an operating point, or the curve supplied so you can find yours
• Speed with its tolerance
• Operating voltage range, including what is guaranteed at start-up and at the extremes
• Current and input power at rated conditions, and the condition identified
• Life figure with model, temperature, failure criterion and confidence named
• Bearing type, named specifically
• Noise figure with distance and position stated
• Certification standards and file numbers, not just marks
• Ingress protection level to IEC 60529, with the test arrangement described
• Connector, lead length, wire gauge and pinout, unambiguously defined
• Operating and storage temperature ranges, with humidity

A second sheet is not a comparison

Two datasheets that both look complete still may not be comparable. The
useful discipline is to restate both parts in the same terms before you decide —
same operating point, same failure criterion, same measurement conditions — and
then compare. If a figure cannot be restated in common terms, that is itself a
finding about the supplier’s documentation.

Our
RFQ guide sets out how to
ask for a quotation in a form that produces comparable answers rather than two
pages of numbers that cannot be placed side by side.