Micro Fan Guide: How to Size and Mount Small DC Fans

Size comparison of micro fans from 20mm to 92mm side by side

A micro fan is a small DC axial cooling fan, sized from 20mm to 92mm square, built for embedded electronics and compact equipment where a standard-frame fan will not fit. This guide covers that 20-92mm range, using our own DC brushless fan lineup, and it is our own working definition, not an industry standard.

What Is a Micro Fan? (And Where the Size Line Actually Falls)

We use “micro fan” to mean Proventofan’s smallest DC brushless models: a 20mm square frame up to 92mm, across twenty-seven distinct frame and depth combinations in that range (our full DC brushless line goes further, up to 200mm, but that is outside what this guide covers). SUNON, one of the manufacturers active in this space, markets its own micro fan line down to 8mm, as thin as 3mm. We don’t carry anything that small. If your project genuinely needs an 8mm to 19mm fan, this guide will not solve that, and we would rather tell you now than have you find out after a quote request.

Where Micro Fans Actually Show Up

So where do these show up? Three types of equipment are a natural fit at this size.

Mini PCs and compute sticks are the most literal fit. The whole selling point of a compute-stick-class device is a footprint smaller than a paperback, which leaves almost no room for a standard 60mm or 80mm case fan. A 20mm to 40mm fan tucked behind the exhaust vent, running off the same 5V or 12V rail as the board itself, is often the only option that physically clears the enclosure.

Network and storage appliances are the second group. Small switches, NAS boxes, and edge gateways run continuously in a closet or a rack shelf with no active airflow around them. These units generate less heat than a full server, but they also have less surface area to shed it passively, so a 40mm to 60mm fan earns its keep even at modest load.

Industrial control and instrumentation modules round out the list. Compact controllers, signal processing units, and sensor housings sit inside a larger cabinet but still concentrate heat in a small volume. Here the fan usually has to survive a wider temperature swing than a desktop application would, part of why bearing choice, covered later, matters more than it looks.

If your own project looks like one of the three above, the fastest way to confirm you actually need a micro fan rather than a standard-frame one is to measure the real clearance inside your housing, not its nameplate dimensions. Do this for the first time and you will likely be surprised how little room is left once the PCB, connectors, and cable routing are accounted for.

Finding What Actually Needs Cooling

Before you can size a fan, you need to know what is actually generating the heat and where the fan physically has to sit. Neither question is as obvious as it sounds once the design gets this compact.

The heat source is rarely the whole board. It is usually one of three things: a power conversion module (buck or boost converters run hot under sustained load), a main processor or SoC, or, in battery-powered gear, the battery management circuit during charging. Point the fan at whichever one is the bottleneck. Cooling the wrong corner of the board with an undersized fan wastes airflow that a repositioned fan would have used properly.

Mounting position is the other half of the decision, and it sets a hard ceiling on both footprint and depth before you even look at a spec sheet. Where the fan physically attaches and which way it blows are two separate questions. PCB-mounted fans are fixed to the board itself, usually on standoffs, but the fan can still face air toward a nearby component, away from it toward a vent, or along the board depending on orientation; you still need clearance on both the intake and exhaust sides, not just where the mounting holes sit. Enclosure-cutout mounting, where the fan sits in the housing wall itself, gives more flexibility on depth but demands a clean opening and a gasket or shroud, or air simply recirculates around the fan instead of through the enclosure.

One category worth flagging and setting aside is micro blowers, which pull air in through the top and push it out a side vent instead of straight through. They solve a real problem in thin, ducted designs. But Proventofan’s product line at this size is axial fans only, so if your design specifically needs blower-style side discharge, that falls outside what we can quote here.

Diagram comparing PCB-mounted micro fan versus enclosure cutout mounting
Micro fans mount either directly on the PCB or through an enclosure cutout, each with different clearance requirements.

Settle this before looking at a size at all. Is the fan going on the board, or in the wall? PCB mount caps you at whatever the board-to-lid clearance is, often under 10mm. Enclosure mount opens up more depth options but adds mechanical complexity that a lot of first-time designs underestimate.

Sizing Guide: Matching Micro Fan Dimensions to Your Design

Once you know where the fan sits and what it is cooling, sizing it is a matter of matching a real constraint, net clearance, against what the product line offers, not picking the smallest number on a spec sheet.

Key Takeaway

First narrow the list to frames that physically fit, then compare installed airflow, noise, and power draw among those that do. When size itself is the priority, pick the smallest of the frames that still met your requirements in that comparison, not one that didn’t.

The portion of Proventofan’s DC brushless fan line this guide covers runs from a 20mm square frame up to 92mm, in depths from 7mm to 38mm depending on size, on 5V, 12V, or 24V depending on the frame.

Size comparison of micro fans from 20mm to 92mm side by side
Micro fan sizes range from 20mm to 92mm across Proventofan’s DC brushless fan lineup.
Frame Depth shown Airflow (H-speed, 12V, max free-air CFM) Max static pressure (mm-H&sub2;O) What it means for you
20×20mm 10mm 1.89 7.33 Fits almost anywhere, but airflow is limited. Worth it only when space is the binding constraint
25×25mm 10mm 2.90 6.93 The usual step-up from 20mm when you need meaningfully more airflow in about the same footprint
30×30mm (5-blade) 10mm 4.23 4.47 A noticeable airflow gain over 25mm at a comparable depth
40×40mm (7-blade) 10mm 6.40 3.31 A common ceiling for compute-stick and small-appliance designs
60×60mm 10mm 17.17 3.15 Airflow jumps sharply here. Check whether your enclosure can take it before settling for a 40mm
92×92mm (Standard) 25mm 60.25 5.94 The upper edge of what we treat as micro. Past this, standard DC axial fans usually make more sense

Table source: Proventofan product specifications. All figures in a given row are H-speed variants at 12V from the same fan, same variant, and same speed setting.

Both columns in each row come from the same fan, same variant, same voltage, and same speed setting, but not from the same moment on its performance curve. Max free-air CFM is measured with nothing blocking the fan; max static pressure is measured with the outlet fully blocked. A real installation lands somewhere between the two, at whatever point your system’s resistance curve crosses the fan’s own curve, so treat both numbers as the outer edges of what that fan can do, not a combination it delivers at once.

The 25mm figures reflect the higher-output of two 25×25×10mm variants we carry under that model code, the one running at 13,000 RPM rather than 10,000 RPM. Ask us which variant fits your project if the difference matters. Other depths exist at several of these frame sizes too. A 30×30×7mm model moves meaningfully less air than the 10mm version shown here, while a 40×40×28mm model moves meaningfully more, alongside 5V and 24V options and the L and M speed steps. Full variant tables live on the DC brushless fan product page. If you are hunting for something below 20mm, down toward the 10mm class some manufacturers list, that is outside our current lineup.

Reading that table, airflow does not scale in a straight line with frame size. Going from 40mm to 60mm nearly triples free-air CFM, while going from 20mm to 25mm adds only about 1 CFM. That’s not a quirk of this particular line. It is blade area and RPM headroom compounding, and it means the size jump that looks small on paper is not always small in the result.

Here is how that plays out in practice, in a representative design scenario, not an actual customer project. Say a design has 65mm by 65mm of usable board area and 12mm of depth behind the exhaust vent, and needs somewhere in the 15 to 18 CFM range to hold the board under its thermal budget at worst-case ambient; those numbers are the scenario’s assumptions, not catalog figures. A 92mm frame would clear the airflow target on paper, but its real 92×92mm footprint, straight from the sizing table above, will not fit inside a 65×65mm opening no matter how much depth is available.

A 60×60×10mm fan is a better starting point. Its real 60×60mm footprint and 10mm depth both clear the assumed space, and its cataloged 17.17 CFM free-air rating sits close to, though just under, the 18 CFM upper end of the assumed target. Two things still need checking before calling it done. First, body dimensions fitting is not the same as installation fitting; mounting screws, standoffs, and clearance for the intake and exhaust still need to be confirmed against the real enclosure. Second, free-air CFM is a ceiling, not a guarantee, and real installed airflow will land lower still once system resistance is factored in.

Important Note

If 18 CFM turns out to be a hard minimum rather than a target, 17.17 CFM does not clear it, full stop. A larger frame will not fix this in the assumed 65×65mm footprint either; 70mm and 92mm frames both exceed it. At that point the fix is not a bigger fan, it is more available space, a shorter or wider air path, or a lower thermal budget to hit, whichever the rest of the design can actually give up.

We are not the first people to run into the tradeoff between footprint and airflow at this scale. Back in 2003, an engineer posting under the handle steverde on the Eng-Tips forum had a specific 1U rack space, 1.75 inches, and found the 40mm fans on hand were still a touch too big for that particular chassis, not a claim that 40mm never fits a 1U box in general. Another member, marshell, pointed out that 25mm fans existed, but that free-air output dropped to somewhere around 1.9 to 2 CFM at that size, a figure that still sits inside the range our own 25×25×10mm models cover today, 1.2 to 2.9 CFM depending on variant. Two decades on, the physics has not changed. Shrinking the frame buys you clearance, and it costs you airflow, in that order.

Here is the judgment call that matters more than the table: when your net clearance falls between two standard sizes, do not automatically round up. Check whether a mid-depth variant of the smaller frame gets you close enough to target first, before you commit to the next size up, and the noise, power draw, and mounting changes that come with it.

Once you have a target frame size narrowed down, the DC Brushless Fan Selector tool runs the exact voltage, depth, and airflow options against your numbers without cross-referencing tables by hand.

Voltage, Airflow, and Noise: Reading the Datasheet the Right Way

A fan’s spec sheet gives you a starting point, not a guarantee. Three numbers get misread more often than any others at this scale: free-air CFM, rated voltage, and dB(A), and each needs a bit of context before you can use it.

Free-air CFM, the number in the table above, is measured with nothing blocking the fan on either side. Put that same fan behind a filter, a heat sink, or a tight enclosure opening, and real airflow drops to wherever the fan’s own performance curve crosses your system’s resistance curve, not the free-air number on the label. This matters more at micro scale than on a 120mm fan, because the inlet opening on a 20mm or 25mm frame is small enough that even modest obstruction eats a large share of the fan’s usable pressure.

Rated voltage is not the same as the voltage range the fan runs across in practice, and the spread is not identical model to model. A 12V-rated 20×20×10mm fan in our line operates across 10.8V to 13.2V. Step up to the 30×30×7mm frame, and the same 12V rating covers a wider 6.0V to 13.5V band. If your supply sags under load, or your design runs the fan at reduced voltage for noise control, check the actual operating range for your specific frame rather than assuming every 12V fan behaves the same way. One basic wiring precaution applies regardless of frame size: confirm polarity and connector orientation before powering up, since reverse voltage can damage a brushless DC fan’s driver circuit, and never connect or disconnect the fan while the board is live.

Noise is the trickiest of the three to compare across a spec sheet, because dB(A) values only mean something at the distance and speed they were measured at. We’re not going to pretend every datasheet in this space uses the same test distance. Before comparing a micro fan’s dB(A) rating against a component from a different supplier, check what distance both numbers were measured at. A rating that looks quieter on paper can simply have been measured further away.

None of this is a new problem. Back in 2002, an engineer named Chris George asked on the Control.com automation forum how to drive a small 5V, 12V, or 24V DC fan off a 4-20mA control loop, and the replies did not fully agree on the details. What that thread actually settles is simpler than a universal rule: matching a control signal to a fan means checking that specific fan’s drive method, its real operating voltage range, and what feedback it can give you, not assuming one approach works across every small DC fan.

Bearing Type and IP Rating: Matching Duty Cycle and Environment

Bearing type and IP rating rarely get asked about at the start of a project, and they are usually the reason a fan fails early when they get skipped.

Sleeve and ball bearings trade off in a fairly predictable way. Sleeve bearings cost less and run quietly at moderate speed, but they generally need the fan’s rotational axis mounted horizontally to avoid uneven wear, and they tend to wear faster in continuous, high-temperature duty. Ball bearings cost more and can run a bit louder, but they tolerate any mounting angle and hold up better under sustained load and vibration. The full lifetime numbers and test conditions behind that tradeoff are covered on our sleeve bearing versus ball bearing guide; we will not repeat the hour figures here, because they depend on load and temperature in ways that do not compress into one sentence.

Say your design’s installation orientation does not match a given model’s recommended orientation, or the equipment runs in a cabinet that vibrates continuously during operation. That’s exactly the case where bearing type stops being a minor spec and starts being the difference between a fan that lasts and one that doesn’t.

Cross-section diagram comparing sleeve bearing and ball bearing micro fan construction
Sleeve and ball bearing micro fans trade off differently on cost, noise, and duty cycle.

IP rating deserves one honest caveat: the fan’s own IP rating is not the same as your finished device’s IP rating. A fan rated for dust and splash resistance can still let moisture into your enclosure through the mounting cutout, the cable gland, or a housing seam that has nothing to do with the fan itself. Sealing the fan into a rated device is a system-level job, and the fan’s datasheet number is only one input into it. For the general selection process beyond this size range, our DC fan selection guide covers the fuller framework.

Mounting position, sizing, voltage range, and bearing type: settle those four in order, and a model number falls out of the decision rather than starting it.

Frequently Asked Questions

What is a micro fan?

A micro fan is a small DC axial cooling fan built for space-constrained electronics, generally in the 20mm to 92mm square range. There is no single industry standard defining the term, so different suppliers draw the line in different places. In this guide, “micro fan” refers to Proventofan’s DC brushless fan lineup at this size, which starts at a 20mm frame.

What size micro fan do I need for my device?

Start with the net clearance inside your enclosure, not the equipment’s outer dimensions, since connectors, cabling, and the PCB itself all eat into available space. Match that measurement against the frame and depth options in the sizing table above, then check whether the airflow at that size meets your thermal target. If your clearance falls between two standard sizes, check for a mid-depth variant before rounding up to the next frame.

Why not just use a bigger fan and duct the airflow down?

You can, but a duct or reducer adds its own pressure loss, which pushes the fan’s real operating point down its own curve regardless of frame size. Whether that tradeoff is worth it depends on the actual available space, the target airflow, noise limits, and input power of each option compared side by side, not on frame size alone. Run the numbers for both before deciding.

What is the difference between sleeve and ball bearing micro fans?

Sleeve bearings cost less and run quietly at moderate speed but generally need the fan’s rotational axis mounted horizontally and wear faster under sustained heat. Ball bearings cost more, tolerate any mounting angle, and hold up better under vibration and continuous high-temperature duty. Full lifetime data and test conditions for both bearing types are covered on our sleeve bearing versus ball bearing guide.

Who typically specifies a micro fan in a design?

Micro fans at this size are almost always specified by an electronics or mechanical engineer working on embedded, portable, or compact equipment, not by a consumer buying a desk fan. The decision usually follows a thermal or EMC review once passive cooling is no longer enough. If you are designing a compute stick, a network appliance, or an industrial control module and hitting a thermal wall, this is the audience this guide is written for.

Do micro fans run on 5V or 12V?

Both, depending on the frame size and your equipment’s existing power rail. Proventofan’s smallest frames, 20mm through 30mm, are typically 5V or 12V only; 24V options start appearing around the 40mm frame. Match the fan to whatever rail your board or power supply already provides rather than adding a separate regulator just for the fan.

Does a micro fan need an IP rating?

Only if dust or moisture can actually reach it, which depends on where it sits in your design, not on the fan alone. A fan mounted well inside a protected enclosure, with no direct path to the outside environment, may not need its own IP rating; one mounted in an exposed cutout facing outdoor or wash-down conditions usually does. Either way, the fan’s IP rating and your finished device’s IP rating are two separate numbers; sealing the whole assembly is a system-level job.

How much airflow does a small DC fan like a 20mm or 25mm model move?

A 20×20×10mm fan in our lineup moves up to about 1.9 CFM free-air at its highest speed setting, and a 25×25×10mm model moves up to roughly 2.9 CFM depending on the variant. Those are free-air numbers, measured with nothing blocking the fan. Real airflow inside your enclosure will be lower once you account for filters, heat sinks, or a tight inlet opening.

Is there an official size definition for a micro fan?

No. There is no universal industry standard defining exactly where “micro fan” starts and stops. Different manufacturers use the term for different size ranges, and some overlap with “miniature” or “compact” fan categories. This guide defines it as Proventofan’s own DC brushless fan range, 20mm to 92mm square, and says so plainly rather than implying a standard that does not exist.

When is passive cooling a better choice than a micro fan?

Passive cooling, heat spreaders, graphite sheets, or simply more enclosure surface area, is usually the better choice when your heat load is modest, you need silent operation, or you cannot tolerate a moving part that will eventually wear out. Add a fan only once you have confirmed passive cooling cannot hold your components under their rated temperature at worst-case ambient. A lot of designs add a fan by default when the numbers never actually required one.

References

Disclaimer

The sizing, voltage, and bearing guidance in this article is a starting framework, not a substitute for reviewing your own thermal and mechanical requirements. Specific model numbers, CFM figures, and static pressure values reflect Proventofan’s current product data at time of publication and may change as the lineup is updated. Confirm current specifications before finalizing a design. For certification requirements specific to your industry or application, contact us directly to confirm whether a given model qualifies.

Ready to Size a Micro Fan for Your Design?

Send us the following, and we will tell you which frame sizes in our DC brushless fan line actually fit, and flag anything in your requirements that falls outside what we carry, before you spend time on a formal quote request:

  • Your enclosure’s net internal clearance (footprint and depth)
  • The voltage rail you are working with
  • Any noise limit, including the distance you measured or need it at, since dB(A) numbers only mean something alongside that
  • Environmental exposure your design needs to handle: temperature range, dust, or moisture

For a fast first pass on your own, the DC Brushless Fan Selector tool runs these same numbers against the full lineup.

Share this guide

Keep reading

Related Guides

View all guides →

Request a Quote

Tell us your fan requirements and we’ll get back to you with a suitable solution.