The best face mask for motorcycle riding is the one still on when you arrive. An N95-type mask leaked 9% against 35% for a basic surgical mask (Steinle et al., 2018), but runs hottest. Lower it for over 29% of a ride and a surgical mask kept on matches it (our arithmetic on Steinle et al., 2018).
Part of our Masks vs filters guide.
At a red light in September haze, look at the riders around you. Masks hang hooked under chins, and the helmets are still on.
That chin is the whole problem. The mask that filters best traps the most heat against your lips, so in tropical air it comes down first. We have covered why even a good mask leaks on a rider before. This post compares masks on the test nobody publishes: heat, and the minutes it costs you.
Key takeaways
- Worn without fit training, an N95-type mask let 9% of fine dust inside, a PM2.5-rated surgical mask 22%, and a standard surgical mask and an Indonesian flat-fold mask 35% each (Steinle et al., 2018).
- After one hour, lip-area skin read 36.9°C under an N95 against 35.9°C under a surgical mask, in a 22 to 24°C room (Scarano et al., 2020).
- In the same study, N95 wearers touched and moved their mask about twice as often as surgical mask wearers (Scarano et al., 2020).
- On those leakage figures, an N95 worn for less than about 71% of a ride protects no better than a flat-fold mask kept on for all of it (our arithmetic on Steinle et al., 2018).
The research behind this
Peer-reviewed measurements this post is built on. Each card says what was measured, what was found, and what it cannot tell you.
How much four common masks leak on real faces
- What they measured
- Ten volunteers wore four disposable mask types, including a basic flat-fold mask bought in Indonesia, with no fit training. PM2.5 was measured inside and outside the mask during two 10-minute clean-up tasks.
- What they found
- Total inward leakage averaged 9% for the N95-type mask, 22% for the PM2.5 surgical mask and 35% for both the standard surgical and flat-fold masks.
- What it can't tell you
- A volcanic ash clean-up simulation with 10 people. Nobody was riding and nobody wore a helmet.
Lip temperature under an N95 and a surgical mask
- What they measured
- Twenty people wore a surgical mask for one hour, then an N95 for one hour the next day. Thermal cameras read the skin around the mouth in a 22 to 24°C room.
- What they found
- Lip-area skin read 36.9°C after the N95 against 35.9°C after the surgical mask. N95 wearers rated heat, humidity and discomfort higher and touched their mask more.
- What it can't tell you
- Seated, indoors and air-conditioned. Tropical riding air is hotter, so these are floor values.
Where N95 heat builds up
- What they measured
- Twenty healthy people exercised at a low to moderate rate for one and two hours in four N95 models, with core and skin temperature tracked wirelessly.
- What they found
- Core temperature barely moved, but skin under the mask warmed significantly. The authors link the hot feeling to that skin and to warm inhaled air.
- What it can't tell you
- A temperate lab with no helmet on top.
N95 against surgical in humid air
- What they measured
- Ten people did intermittent treadmill exercise in a climate chamber at 25°C and 70% humidity, wearing N95 and surgical masks.
- What they found
- Air and skin inside the N95 were hotter and wetter, and heart rate was higher, than with surgical masks.
- What it can't tell you
- Ten people in a chamber, walking rather than riding, with no helmet.
What cloth masks stop from diesel exhaust
- What they measured
- Three cloth masks and one surgical mask common in developing countries, tested in the lab against lab particles and diluted diesel exhaust.
- What they found
- Cloth masks filtered 15 to 57% of diesel particles, and the authors call them only marginally useful below 2.5 µm.
- What it can't tell you
- Fabric only. Air leaking around the edges on a real face is not included, so worn protection is lower.
What the studies actually measured
Start with leakage, because a mask's fabric rating is not what reaches your lungs. Steinle's team put four disposable masks on 10 volunteers with no fit training. They measured fine dust inside and outside each mask during two 10-minute clean-up tasks (Steinle et al., 2018).
The N95-type mask let in 9% on average. A surgical mask designed for PM2.5 let in 22%. A standard surgical mask and a basic flat-fold mask bought in Indonesia each let in 35%. Volunteers also said the N95 felt uncomfortable and harder to breathe through.
A second team tested retail masks bought in Beijing on volunteers in a diesel exhaust chamber. Inward leakage ranged from 3 to 68% while sitting, and only one mask averaged under 10% (Cherrie et al., 2018).
Cloth sits lower still. In the lab, cloth masks stopped 15 to 57% of diesel particles before any edge leakage was counted (Shakya et al., 2017).
Then heat. Scarano's team filmed 20 people's faces with a thermal camera after an hour in each mask. Lip-area skin read 36.9°C after the N95 and 35.9°C after the surgical mask, a statistically significant gap (Scarano et al., 2020). They note that skin above 34.5°C can start to feel uncomfortable (Scarano et al., 2020).
Roberge's team found the same pattern during exercise. Core temperature barely changed, but covered facial skin warmed significantly (Roberge et al., 2012). At 25°C and 70% humidity, Li's team measured hotter, wetter air inside the N95 and a higher heart rate than with surgical masks (Li et al., 2005).
None of these rooms was as hot as a Jakarta afternoon, and none put a helmet over the mask.
Ride with filtered air
Filters haze PM2.5 on every ride
Easi Breezi mounts on your scooter or motorbike and feeds H11 HEPA-filtered air into your helmet. Bike-powered, no batteries, no charging.
Why haze season sends every rider back to the same mask advice
The advice this month is the advice of every haze season. TIME reports health agencies recommending an N95 mask outside, with more than 12 million people affected and 600,000 hectares burned in August alone (TIME, 2026).
Indonesia's Health Ministry counted 174,694 acute respiratory infection cases from August to 21 September, more than 40,000 of them in toddlers. Its advice was to use masks (Jakarta Post, 2026). Central Kalimantan's index hit 1,919 on 2 September (TIME, 2026).
The stakes are not new. A model of the 2015 haze estimated 100,300 excess deaths across Indonesia, Malaysia and Singapore (Koplitz et al., 2016).
What the advice skips is the rider. It assumes the mask stays on. This year's smoke arrives in an El Niño dry season, the hottest riding weather of the year, and heat is what takes masks off faces.
The best face mask for motorcycle riding, compared on heat
Here is every common option scored on the studies above. Cells we have no measurement for say so. The last column is our engineering judgement, not a study result.
| Option | Leak or filter figure | Heat under it | Still on after 30 min at 33°C? (our judgement) |
|---|---|---|---|
| Cloth mask or neck gaiter | Fabric stops 15 to 57% of diesel particles, before edge leaks | Not measured in these studies | Likely, but it filters little |
| Standard surgical or flat-fold | 35% inward leakage | Lip skin 35.9°C after 1 h, cooler than N95 | Likely |
| PM2.5-rated surgical | 22% inward leakage | Not measured in these studies | Likely |
| N95-type respirator | 9% inward leakage | Lip skin 36.9°C after 1 h; hotter, wetter air inside | Least likely of the four |
| Powered filter in the helmet | No on-rider leakage figure published yet | Nothing sealed against the skin | Nothing to pull down |
Table figures: leakage from Steinle et al. (2018), cloth from Shakya et al. (2017), heat from Scarano et al. (2020) and Li et al. (2005).
Read down the heat column and the ranking flips. The best filter on paper is the one your hands will fight with most. For a wider look at gear beyond masks, see our anti-pollution gear comparison.
A mask pulled down for a third of the ride loses to a cheaper one kept on
Filtration is not protection. What reaches you depends on two things: how much the mask leaks while worn, and how much of the ride it is worn at all.
Put simply, protection equals (1 minus leakage) times the share of the ride the mask stays up. This is our arithmetic on published leakage figures, not a measured result.
Pull an N95-type mask down for longer than this and a flat-fold mask kept on the whole way protects you as well.
Source: our arithmetic on Steinle et al., IJHEH, 2018Worn properly, Steinle's N95-type mask stopped 91% and the flat-fold mask 65% (Steinle et al., 2018). Divide 65 by 91 and you get 0.71. So the N95 must stay up for at least 71% of the ride to win.
Against the PM2.5-rated surgical mask, which stopped 78%, the bar is higher. The N95 must stay up for 86% of the ride to come out ahead (our arithmetic on Steinle et al., 2018).
Is pulling a mask down that common? Scarano's N95 wearers touched and moved their mask about twice as often as surgical mask wearers, sitting still in a cool room (Scarano et al., 2020). The authors concluded it is better to wear a surgical mask correctly than an N95 that keeps coming off.
Be honest about your own habit. If your N95 comes down at every red light, a surgical mask you never touch may protect you more.
Work out what your mask delivers
Pick a mask and set how much of the ride it stays on your face. The tool uses only the four leakage figures from Steinle's study.
Protection delivered on your ride
Leakage from Steinle et al. (2018). Wear time is your estimate. Illustrative only.
Your mask
Your mask, as worn
Flat-fold mask kept on the whole ride (65%)
Your N95-type mask, worn for 60% of the ride, stops about 55% of the particles over the ride. That is less than a flat-fold mask kept on the whole way (65%).
| Mask | Mean inward leakage | Wear time needed to match an always-on flat-fold mask |
|---|---|---|
| N95-type | 9% | About 71% |
| PM2.5 surgical | 22% | About 83% |
| Surgical or flat-fold | 35% | 100% |
A sealed mask under a helmet traps its own heat
The better a mask seals, the less air moves across your lower face. Masks block the three ways skin sheds heat: moving air, sweat drying and radiation (Roberge et al., 2012). That is why covered skin warms while your core does not.
Every breath out adds more heat. Exhaled air is warm, wet and about 5% carbon dioxide (Birgersson et al., 2015), and it fills the space between mask and lips. In a computer model of an N95, that space averaged about 3% carbon dioxide over a breathing cycle (Birgersson et al., 2015). You breathe part of it straight back in.
A full-face helmet removes the last cooling path. The chin bar blocks the breeze that would otherwise pass over the mask's outer face. That point is our engineering inference, since no study we found measured a mask under a helmet.
Why the best-sealing mask comes off first
A passive mask has no way to flush that space. In the same model, a small fan and one-way valve cut the average carbon dioxide from about 3% to about 0.3% (Birgersson et al., 2015). That was a simulation, but the principle holds: moving air clears the breathing zone, and a mask has no fan.
Cloth and neck gaiters feel cooler, but that is because they seal and filter little. Stopping 15 to 57% of diesel particles in the lab is a weak start before edge leaks (Shakya et al., 2017).
More El Niño years mean more hot haze seasons in a mask
Climate models suggest extreme El Niño events could become twice as frequent under greenhouse warming (Cai et al., 2014). That is a trend across this century, not a forecast for any single year.
Our own reading, not a figure from the paper: a rider commuting today should plan on at least one more mask-advisory haze season before 2031. It will most likely come in El Niño heat, when masks come down most often.
So the mask question will come back. You can check today's smoke across the region on our live haze map for riders.
Frequently asked questions
What is the best face mask for motorcycle riding?
The one you can keep on for the whole ride. An N95-type mask leaks least when worn, 9% against 35% for a basic surgical mask (Steinle et al., 2018). If heat makes you lower it for more than about 29% of the ride, a surgical mask kept on does as well (our arithmetic on the same study).
Is a KF94 or KN95 better than a surgical mask on a motorbike?
These are N95-type designs, and the studies here tested N95-type masks, not every label. Expect similar filtering and similar heat. The same wear-time rule applies.
Does a face mask for a motorcycle helmet fit under a full face helmet?
Most flat and fold masks fit, but the chin bar can press them and break the seal. The chin bar also blocks the breeze over the mask, which is our engineering reading rather than a measured result.
Why does my mask feel so hot when I ride?
Covered skin warms while your core temperature barely changes (Roberge et al., 2012). Every breath also refills the mask with warm, wet air.
Is a cloth mask or neck gaiter enough in haze?
No. Cloth masks stopped 15 to 57% of diesel particles in the lab, before edge leaks (Shakya et al., 2017).
What to do this week
- Count your pull-downs. On one ride, notice every time the mask comes off your nose. That number decides your protection more than the label does.
- Match the mask to the ride. A short, cool ride suits an N95. A long, hot one may suit a surgical mask you never touch.
- Check the air before you leave. Look up a live PM2.5 reading, since the level changes hour to hour.
- Take the filter off your face. The Easi Breezi unit runs off the bike and pushes air through an H11 HEPA filter that stops 95% of PM2.5. Nothing is sealed against your skin, so there is nothing to pull down at a red light.
A mask's rating is measured on a face that keeps it on. In haze-season heat, plan for the face you have.
Sources
- Steinle S, Sleeuwenhoek A, Mueller W, Horwell CJ, et al. (2018). The effectiveness of respiratory protection worn by communities to protect from volcanic ash inhalation. Part II: Total inward leakage tests. International Journal of Hygiene and Environmental Health 221. doi:10.1016/j.ijheh.2018.03.011
- Scarano A, Inchingolo F, Lorusso F (2020). Facial skin temperature and discomfort when wearing protective face masks: thermal infrared imaging evaluation and hands moving the mask. International Journal of Environmental Research and Public Health 17. doi:10.3390/ijerph17134624
- Cherrie JW, Apsley A, Cowie H, et al. (2018). Effectiveness of face masks used to protect Beijing residents against particulate air pollution. Occupational and Environmental Medicine 75. doi:10.1136/oemed-2017-104765
- Shakya KM, Noyes A, Kallin R, Peltier RE (2017). Evaluating the efficacy of cloth facemasks in reducing particulate matter exposure. Journal of Exposure Science and Environmental Epidemiology 27. doi:10.1038/jes.2016.42
- Roberge R, Benson S, Kim JH (2012). Thermal burden of N95 filtering facepiece respirators. Annals of Occupational Hygiene 56. doi:10.1093/annhyg/mes001
- Li Y, Tokura H, Guo YP, Wong ASW, et al. (2005). Effects of wearing N95 and surgical facemasks on heart rate, thermal stress and subjective sensations. International Archives of Occupational and Environmental Health 78. doi:10.1007/s00420-004-0584-4
- Jeyaretnam M (2026). Why Southeast Asia is facing a haze crisis. TIME, 17 September 2026. time.com
- Indonesian Ministry of Health, via The Jakarta Post (2026). Respiratory problems climb past 170,000 cases in Indonesia. 24 September 2026. thejakartapost.com
- Koplitz SN, Mickley LJ, Marlier ME, et al. (2016). Public health impacts of the severe haze in Equatorial Asia in September to October 2015. Environmental Research Letters 11. doi:10.1088/1748-9326/11/9/094023
- Roberge RJ, Kim JH, Coca A (2012). Protective facemask impact on human thermoregulation: an overview. Annals of Occupational Hygiene 56. doi:10.1093/annhyg/mer069
- Birgersson E, Tang EH, Lee WL, Sak KJ (2015). Reduction of carbon dioxide in filtering facepiece respirators with an active-venting system: a computational study. PLOS ONE 10. doi:10.1371/journal.pone.0130306
- Cai W, Borlace S, Lengaigne M, et al. (2014). Increasing frequency of extreme El Niño events due to greenhouse warming. Nature Climate Change 4. doi:10.1038/nclimate2100
Written by Ash, mechanical engineer and founder of Easi Breezi, building an active HEPA filtration system for motorcycle helmets (patent pending). Based between Hong Kong and Bali, riding daily in the traffic this blog writes about.