CO2 and sleeping while driving

[Ivan Seeking]

After almost falling asleep while driving the other day, I got to thinking about CO2. I started to wonder if due to normal respiration of the driver and or occupants, CO2 levels in a car can rise to levels that can inhibit responses or even lead to unintended sleep, and make driving unsafe. I was struck by how quickly I regained composure as soon as I opened the window for fresh air. Maybe cars are too air tight for our own good?

 

[SteamKing]

Spending time in any sealed enclosure without adequate ventilation is not recommended. Fortunately, you learned (or re-learned) this lesson without injury to yourself or others. Cars are built with multiple means to provide ventilation to the cabin. Please use at least one at all times, even if it's just leaving a window open a crack.

 

[Doug Huffman]

My VW owner's manual is clear that the HVAC vent should be open but for exceptional circumstances. For the usual position of the function control, recirculation is defeated.

 

[russ_watters]

Was the fresh air cold?

In winter, it would be very difficult to have a CO2 problem before a humidity/condensation problem. Most cars have an indicator that tells you whether the outside air damper for the HVAC is open or closed. In summer it should almost always be closed and in the winter it should almost always be open.

 

[Greg Bernhardt]

Maybe cars are too air tight for our own good?

Do you turn off the air coming from the outside?

 

[NascentOxygen]

Maybe cars are too air tight for our own good?

Maybe also too warm or too comfy to maintain a high state of alertness.

 

[mfb]

~1kg of CO₂ per day and person, or half a cubic meter. Let's assume the car has 3m³ of volume inside, then you get ~0.7% of CO₂ per hour (by volume, a bit more by mass). Driving for more than two hours without any ventilation is not a good idea, and with multiple passengers you'll certainly need some sort of air exchange.

I was struck by how quickly I regained composure as soon as I opened the window for fresh air.

I would expect temperature had a larger effect there.

 

[.Scott]

After almost falling asleep while driving the other day, I got to thinking about CO2. I started to wonder if due to normal respiration of the driver and or occupants, CO2 levels in a car can rise to levels that can inhibit responses or even lead to unintended sleep, and make driving unsafe. I was struck by how quickly I regained composure as soon as I opened the window for fresh air. Maybe cars are too air tight for our own good?

A slow build up of excessive CO2 will result in hypercapnia. The normal reflex to hypercapnia is to awaken so that you can reposition yourself during sleep.

Unless you are missing this reflex (a potentially fatal condition), what you are describing is not consistent with a hazardous buildup of CO2.

If there was a buildup of CO2, that may have been what awakened you.

 

[OmCheeto]

According to our government, a difference of only 1.4% in the level of oxygen means the difference between normal, and minimum required, to have no effect.

* 20.9 percent: Percentage of oxygen found in normal air. No effect.
* 19.5 percent: Minimum permissible oxygen level. No effect. ​

From my calculations, if you were in a 2015 Subaru Outback, with a very voluminous 100 ft³ of air, it would only take about an hour and 20 minutes before there were noticeable effects.

Of course, a smaller vehicle, and extra occupants, would reduce the time.

ps. It looks like mfb and I have come up with very similar numbers, so I'll not recheck my maths. (available upon request)
pps. Please use the homework template in the future.

 

[Ivan Seeking]

Normally I have the exchange open but noticed later that it was closed. It wasn't very cold out that night but it was raining. And it is possible that the windows not receiving forced warm air were fogged, but I don't remember. If they were I wouldn't have necessarily noticed.

I did find this

Results: Relative to 600 ppm, at 1,000 ppm CO2, moderate and statistically significant decrements occurred in six of nine scales of decision-making performance. At 2,500 ppm, large and statistically significant reductions occurred in seven scales of decision-making performance (raw score ratios, 0.06–0.56), but performance on the focused activity scale increased.

...100,000 ppm causes visual disturbances and tremors and has been associated with loss of consciousness; and 250,000 ppm CO2 (a 25% concentration) can cause death...

http://ehp.niehs.nih.gov/1104789/

I've never heard of it being potentially dangerous to drive without an air exchange. If true it would be news to a lot of people I'm sure. It could be argued to be a dangerous design flaw.

 

[russ_watters]

~1kg of CO₂ per day and person, or half a cubic meter. Let's assume the car has 3m³ of volume inside, then you get ~0.7% of CO₂ per hour (by volume, a bit more by mass). Driving for more than two hours without any ventilation is not a good idea, and with multiple passengers you'll certainly need some sort of air exchange.

This also assumes the car is completely air tight, of course, and it isn't anywhere close.

I've never heard of it being potentially dangerous to drive without an air exchange. If true it would be news to a lot of people I'm sure. It could be argued to be a dangerous design flaw.

It isn't dangerous. Cars are nowhere close to tight enough for CO2 buildup to be a problem.

You breathe out a lot more water than CO2 -- much faster than you'd notice high CO2 content, it would practically be raining in the car.

 

[Doug Huffman]

As I recall, and it has been many years now since I was last at sea as responsible crew or irresponsible civilian, the minimum [O₂] concentration on submarines then was 17.5%!

When we were at quiet-ship for long durations, [O₂] was minimized to keep everyone in their bunks, groggy and miserable. Watch standing was a challenge. The roving watches would fill a drum liner from the O2 discharge and huff from it, but only for the psycho-ilogical boost.

 

[.Scott]

As I recall, and it has been many years now since I was last at sea as responsible crew or irresponsible civilian, the minimum [O₂] concentration on submarines then was 17.5%!

rgl.faa.gov/Regulatory_and_Guidance_Library/rgFAR.nsf/0/

Sec. 91.211

Supplemental oxygen.

(a) General. No person may operate a civil aircraft of U.S. registry--
(1) At cabin pressure altitudes above 12,500 feet (MSL) up to and including 14,000 feet (MSL) unless the required minimum flight crew is provided with and uses supplemental oxygen for that part of the flight at those altitudes that is of more than 30 minutes duration;
(2) At cabin pressure altitudes above 14,000 feet (MSL) unless the required minimum flight crew is provided with and uses supplemental oxygen during the entire flight time at those altitudes; and
(3) At cabin pressure altitudes above 15,000 feet (MSL) unless each occupant of the aircraft is provided with supplemental oxygen.

At sea level, the O2 partial pressure is 0.2095 atmospheres.
At 12,500 feet MSL, the partial pressure is 0.5950 x 0.2095 = 0.1247 atmospheres (equivalent to 12.47% O2 at sea level).
At 15,000 feet MSL, the partial pressure is 0.5363 x 0.2095 = 0.1124 atmospheres (equivalent to 11.24% O2 at sea level).

http://en.wikipedia.org/wiki/Barometric_formula

 

[Ivan Seeking]

This also assumes the car is completely air tight, of course, and it isn't anywhere close.

It isn't dangerous. Cars are nowhere close to tight enough for CO2 buildup to be a problem..

With inhibited function at possibly as low as 0.1% CO2, I don't see how you can support that statement.

 

[russ_watters]

With inhibited function at possibly as low as 0.1% CO2, I don't see how you can support that statement.

Cars are not very air-tight...but perhaps more to the point, do you really think that if this were a real issue it wouldn't already be well known? Ivan, you're blind-guessing here. You have an idea in your head and you are chasing it for no reason. Don't let your imagination get the better of you here.

It was warm and humid in your car, which does cause people to feel drowsy. Then you opened the window and the cold air hitting your face woke you up. That's it. This isn't rocket science.

 

[russ_watters]

Here's an interesting article that says the old VW Bug was made too air-tight (on purpose?), which made the doors hard to close and made hurt your ears:
http://auto.howstuffworks.com/under-the-hood/diagnosing-car-problems/body/car-air-tight.htm

And video:

 

[Ivan Seeking]

Cars are not very air-tight...but perhaps more to the point, do you really think that if this were a real issue it wouldn't already be well known? Ivan, you're blind-guessing here. You have an idea in your head and you are chasing it for no reason. Don't let your imagination get the better of you here.

It was warm and humid in your car, which does cause people to feel drowsy. Then you opened the window and the cold air hitting your face woke you up. That's it. This isn't rocket science.

I just don't take arm waiving arguments as convincing; especially given the calculations already done and the paper cited. Frankly, I expect more from PF. And I did cite impaired function and not just a loss of consciousness as a point of interest. I'm not arguing that I fell asleep due to CO2. It may have been a blood sugar issue. But as soon as I did the math in my head I couldn't help but wonder.

Also, anyone who has ridden in a closed car with a smoker may question just how much air exchange takes place during normal driving.

 

[Doug Huffman]

@.Scott, yeah, I know I have climbed Mount Whitney's 14,500 feet twice, once the easy way and once the hard way. Anyone seriously mountaineering in the Sierra Nevada knows to acclimate for a week before pushing.

 

[RonL]

A car in motion can have some serious air flow conditions, that produce a vacuum on parts of the body. Depending on seals and insulation, sucking air out of the cabin is likely if vents are closed or windows not opened a little.

 

[mfb]

This also assumes the car is completely air tight, of course, and it isn't anywhere close.

Sure.

You breathe out a lot more water than CO2 -- much faster than you'd notice high CO2 content, it would practically be raining in the car.

That is a good comparison. ~30g of water per cubic meter at 100% relative humidity, so 100g H2O would make it rain. Give or take a factor of 2 depending on temperature and car size.

 

[256bits]

With inhibited function at possibly as low as 0.1% CO2, I don't see how you can support that statement.

Where did the inhibited function come from?
1000 ppm CO2 is generally unoticable.
Symtoms of drowsiness at that level are also generally unsupported, although some people do seem to indicate such a symptom.

OSHA allows a 5000 ppm maximum for an 8 hour workday, and that is even low for toxic effects to become apparant.
Which would of course include include increased resperation rate, since your breathing is affected by CO2 levels and not O2 levels in the blood.

ASHRAE standards set it to around 700ppm, as being indicitive of insufficient air exchange ( accumulation of other human volatiles would be more noticable than CO2 buildup )

I would suspect that, perhaps,
While driving, a somewhat sedentary person heat rate becomes lower, less frequent respiration, and a possible lower core temperature, similar to sitting in the living room and becoming drowsy. Physical inactivity in other words.

Although, I have become drowsy while driving, and it seems the best sleep in my life I will ever get is coming along. Stopping, getting fresh air, and running around the car works for a while, but 5 minutes into the drive, the drowsiness re-occurs. What is going on with that, I have no idea, since an elevated CO2 level should have been purged from the interior, if that is/was the culprit.

Perhaps the research is not complete, and there actually is a "drowsiness" effect during a range of elevated low levels of CO2 that people can be susceptable to, much like the stufiness of a meeting in a conference room.

 

[Doug Huffman]

Highway hypnosis.

I learned to drive in a 1960 Land Rover 88" 1200 cc MOWOG and never got sleepy in that noisy, cold, unstable, slow rattle trap, even on long Western states trips. Our plush-mobile, at that time, Volvo PV544 was smooth and quiet and sleep inducing.

 

[Ryan_m_b]

I've just spent some time trying to find any guidelines/regulations on ventilation in cars. I can't find any with a casual google search but I'd be shocked if they didn't exist, at the very least there's probably industry data somewhere on the natural airflow in cars. I agree with those saying CO2 build up is very unlikely, given the ubiquity of cars and long distance driving this would have been a famous issue discovered and solved long before now.

 

[russ_watters]

I just don't take arm waiving arguments as convincing; especially given the calculations already done and the paper cited.

The calculations look nice, but since they don't take into account the most important factor, they really aren't worth much*. And you didn't provide a relevant paper.

Look, if you really want to know for sure, CO2 monitors start at about $120 on Amazon. I'm always looking for an excuse to pick up a work relevant tool, so I may get one too.

*No offense guys: it looks like a nice lower bound calculation, but when it scares the OP into taking it too seriously, it can be counterproductive.

 

[russ_watters]

ASHRAE standards set it to around 700ppm, as being indicitive of insufficient air exchange ( accumulation of other human volatiles would be more noticable than CO2 buildup )

That's 700 ppm over ambient, which is about 500-700 itself.

Everything else is good. Yes, the 1000 ppm threshold was far too low to notice the kind of effect Ivan observed.

 

[russ_watters]

That is a good comparison. ~30g of water per cubic meter at 100% relative humidity, so 100g H2O would make it rain. Give or take a factor of 2 depending on temperature and car size.

People exhale about 2.5 grams/min, so even if the car started completely dry and the windows were at room temperature, it would be raining inside in about 40 minutes. More likely; half that (starting at room temp and 50% RH or with cool windows).

 

[256bits]

That's 700 ppm over ambient, which is about 500-700 itself.

Everything else is good. Yes, the 1000 ppm threshold was far too low to notice the kind of effect Ivan observed.

Thanks for the correction.

 

[dragoneyes001]

Here's an interesting article that says the old VW Bug was made too air-tight (on purpose?), which made the doors hard to close and made hurt your ears:
http://auto.howstuffworks.com/under-the-hood/diagnosing-car-problems/body/car-air-tight.htm

And video:

I was going to bring up the VW as one of the very very few cars that had been made too air tight. almost all other makes of cars have a near built in level of leakage including through the back seat into the trunk which allows the doors to close without arm breaking strength to do it and not popping the passengers eardrums.

 

[dragoneyes001]

Highway hypnosis.

I learned to drive in a 1960 Land Rover 88" 1200 cc MOWOG and never got sleepy in that noisy, cold, unstable, slow rattle trap, even on long Western states trips. Our plush-mobile, at that time, Volvo PV544 was smooth and quiet and sleep inducing.

I've driven cross country at least a dozen times average 3500 miles per trip did it in 52 hrs with 10 hrs sleep along the way. highway hypnosis is a very real problem especially at sunset and sunrise when the light is changing the white lines become as bad as a metronome ticking away.

the seating position in cars is another culprit since we inevitably set our seats as comfortably as possible because we expect to be driving for a long distance.

temperature adds to the drowsy effect because we get the car to emulate our favorite environment. drivers literally adjust everything to be as comfortable as possible there not a lot of guess work as to why we get fatigued while driving.

 

[russ_watters]

Thanks for the correction.

Er, I looked into it a little more and even that is out of date...and it turns out that the 1000 ppm was a previous guideline as well:

ASHRAE 62.1/1989:

“Human occupants produce carbon dioxide, water vapor, particulates, biological aerosols, and other contaminants. Carbon dioxide concentration has been widely used as an indicator of indoor air quality. Comfort (odor) criteria are likely to be satisfied if the ventilation rate is set so that 1000 ppm CO₂ is not exceeded. In the event CO₂ is controlled by any method other than dilution, the effects of possible elevation of other contaminants must be considered.”

...This level is not considered a health risk but is a surrogate for human comfort (odor).

ASHRAE 62.1/2001 - 6.1.3:

Comfort (odor) criteria with respect to human bioeffluents are likely to be satisfied if the ventilation results in indoor CO₂ concentrations less than 700 ppm above the outdoor air concentration.

http://www.trane.com/commercial/Uploads/PDF/520/ISS-APG001-EN.pdf
https://www.ashrae.org/File%20Library/docLib/Public/20100608_62_1_2007_g_r_t_final.pdf

This language has since been removed entirely, probably because it gives the false impression that CO₂ levels are acutally part of the standard, when in actuality they are just used as a proxy. Still, the appendix provides the dilution calc whereby the ventilation rate standard (15 cfm/person) is determined. And as such, in demand controlled ventilation, CO₂ measurement is still used to determine occupancy.

Not that this is a big issue, but it gets even worse due to the fact that different states/localities incorporate different versions of the standard and update them at different rates.

 

[russ_watters]

So, this question piqued my interest enough that I'm going to geek-out on it a bit. For those of you who don't know, I'm an HVAC/energy engineer and my specialty, if I can be said to have one, is lab ventilation. So I'm constantly thinking about issues like this one. Further, I'm always looking for an excuse to buy a new tax-deductable tool for my work, so I bought a carbon dioxide logger (also logs temp and humidity) to test the issue.

Background/Standards:
OSHA TWA-PEL: .5%/5,000 ppm (8-hour Time-Weighted Average Personnel Exposure Limit)
Noticeable breathing elevation/headaches: 2%
Normal Spacecraft: 0.5%
Apollo 13: "Approaching 2%"

5,000 ppm is the relevant limit, but I include the other information for reference.

Setup:
I will be driving my normal routes to and from work (as well as others), with the CO₂/Temp/Humidity logger in the passenger seat, at 15 second intervals. My car is a Kia Optima SX (interior volume: 3.33 cubic meters/118 cubic feet), which has automatic climate control. I keep it in semi-manual mode, with only the temperature controlled automatically. I'll do a humidity purge (open all the windows for a few seconds) at the start of each test. Air will be supplied to the front and floor vents: the climate control will not allow the windshield vents open without the outside air damper being open. I'll conduct full-length trials with the OA damper open and closed and note any potential upset conditions when they occur (such as getting stuck behind a school bus). Drive times vary from 30-60 minutes and I have two different routes, to work is all "city" driving, home is a mixture of open and congested highway. I'll be alone in the car for all trials.

Hypotheses:
1. My primary hypothesis is that at no time will a closed OA damper result in CO₂ levels exceeding the OSHA PEL of 5,000 ppm. My prediction is a steady state of 300-500 ppm above ambient (reached within the time limit of the test), but it may be higher in slow speed driving and/or lower in high speed driving.

2. My secondary hypothesis is that being behind a bus or truck with a low exhaust and the OA damper open will have a substantial impact on concentrations, approaching and perhaps even exceeding the OSHA PEL for short periods of time. Even now, when I have the OA damper open and get behind a bus, I close it. For the test, I'll resist that urge.

Limitations:
Humidity will be a significant limitation and this test would be better in the summer, when air conditioning removes it. I checked, and my car will not allow the air conditioning to run in the winter, even in manual mode. So I may have to abort some of the tests. Adding more people would add more CO₂, but in winter, having 4 people in a car will cause condensation in seconds, so it isn't possible. If the steady-state CO₂ level is too low with just me in the car or if I can't keep the OA damper closed long enough for SS to be approached, the test may have to be repeated in the summer.

 

[russ_watters]

Here's trial 1, with the OA damper open:

Weather was about 35F with a 25F dewpoint, which matches the humidity in the car pretty closely. For trial 1a, I suspect I breathed on the logger when turning it on -- I'll try to avoid that in the future. It's also possible that's from backing-through my exhaust plume on the way out of the driveway. Anyway, the concentration seemed to follow the traffic on my way to work. I didn't get behind a bus though, so I'll try that a few more times until I do. Again, the trip to work is in "city" traffic and the trip home is on a highway. Concentrations follow speed and traffic, as expected.

Tomorrow will be a bit different from the plan, as I am traveling a long distance to a client's location. I'll try to leave the damper closed as long as possible. It will be snowing though, so that will make the fogging on the windshield worse.

 

[zoobyshoe]

I was going to bring up the VW as one of the very very few cars that had been made too air tight. almost all other makes of cars have a near built in level of leakage including through the back seat into the trunk which allows the doors to close without arm breaking strength to do it and not popping the passengers eardrums.

Late in the game VW fitted the Beetle with vents behind the rear side windows that allowed air out. The point wasn't to avoid popping eardrums, interestingly, but to finally address complaints about the Beetle's heating system. The cabin was always so airtight it wouldn't allow more air, hot or cold, in.

11) The "crescents" located behind the rear quarter windiows are also part of the heater system. It is well known and documented that Beetles are sealed well and pretty air-tight. The crescents aid in cabin pressure relief. Without a means of relieving cabin pressure, the cooling fan no longer has the ability to move air once the cabin becomes pressurized. Air travels from areas of high pressure to areas of low pressure, so once cabin pressure equals the output of the fan, air flow ceases. Before the addition of the crescents, VW recommended cracking a window open to allow the heater to function more efficiently. The drawback to this was that the hot air would escape the cabin before having a chance to warm most of the interior. By positioning the crescents at the rear of the vehicle, warm air entering the cabin in a direction toward the front of the vehicle would circle the front of the vehicle, then head toward the rear due to the location of the pressure relief. This allowed for more consistent and thorough heating of the interior.

http://www.thesamba.com/vw/forum/viewtopic.php?t=434051

So, all cars must have a modicum of ventilation for the heating system to work, if they take outside air, heat it, and introduce it into the cabin. The alternative would be for air that is already in the cabin to be removed from the cabin, put through a heater, then returned to the cabin. The former would prevent undue CO2 buildup but the latter wouldn't.

 

[dragoneyes001]

the fun thing with the really old VW's was they could float for quite an extended period. this is still used in places to cross wide rivers where the bug is allowed to float out and work its way across by judiciously not flooding the engine (some off roaders use snorkels).

I'm really interested in the results from russ's tests. since the first run looks to be an average around 800ppm i'd estimate an all closed run shouldn't surpass 2kppm

 

[russ_watters]

I'm calling this trial 3 due to the different driving profile:

Outside air was 20F, Dewpoint 14F.
For this trial, I was driving a long distance to a client's site during the second worst of the three snowstorms this past four days (oddly, "The Storm of the Century" was third-worst, even though none was worth writing home about...). Sorry it isn't scaled the same as the previous, but the profile is enough different it would have been tough to scale similarly without missing some information.

As you can probably tell, I drove for a while with the OA damper closed, then opened it to purge when the windows fogged-up enough to bother me. I drove far enough that it did get a bit cooler as I drove, but much of the variation in concentrations (of both CO₂ and moisture) is definitely due to driving speeds: The first 20 minutes or so was very slow due to local snow, and then I got on a highway, which wasn't too bad (probably averaged 50 mph). As discussed previously, since the human body exhales much more water vapor than CO₂, the water vapor reached steady state quickly. The steady state is lower at higher speed due to the wind speed making the windows colder. As condensation increases, it also slows, so that explains why the steady state is not quite steady.

The CO₂, did not reach steady, state, as I feared. And the peaks don't look anywhere close to steady state, either. As it is, the high was above my prediction, though not yet anywhere close to the OSHA limit (which was also part of the prediction). One can speculate that in heavy/slow traffic on a calm day, it is conceivable that the OSHA limit could be reached.

It is going to be very dry tomorrow, so I may be able to go longer than the 15-20 minutes in the above test. But I think I'll probably need to repeat this in the spring.