Carb Ice Discussion

Huckster79

Pattern Altitude
Joined
Nov 22, 2018
Messages
2,322
Display Name

Display name:
Huckster79
Theres a good thread in Lessons Learned about a fellow POA Pilot that survived a brutal forced landing likely due to carb ice. I had a run in with it in primary training that turned out fine, but lets face it its a risk that is always there, for some of us more than others depending on what we fly. My Cessna 140 with a C-85 is a known "Ice Maker", so I'm very conscious of it, but I think there is more I and we could learn on the topic. What is some wisdom and practices you know of that could prevent what our brother Doug F went through?

My POH calls for carb heat from start up to take off. I've read several decades worth of NTSB reports on Cessna 140s and notices multiple engine failures after take off with no known explanation, no fuel starvation, nothing found mechanically wrong with engine. I'm wondering if these could be carb ice problems, that maybe the pilot didn't follow the original POH and run heat before take off. Even testing it in run up for just a second or two could make matters worse if you had a bit of ice, hit it for a second see the drop in rpm and right back off, driving ice even deeper into carb?

I will shut carb heat down in summer for taxi for filtration, but I do run it all through the run up and have it as my last item on checklist before take off. I also run it periodically through cruise for a minute or two here and there. I never turn it on for less than a good 30 seconds to a minute or so, as I believe a quick burst could be worse than anything. After reading Doug's story I am going to come up with a system so I hit it on a regular basis throughout flight.

I've also been told (By my flight examiner) that at a certain really low temp one shouldn't use it at all as it could put the temp into an area of temps ice is more likely. I've never heard a soul say that besides him. Is there anything to that?

What practices do you follow to prevent carb ice? Things to be mindful of about carb ice? Do you fly when temp and dew point close? When they are close do you do anything different?

Lets share some collective wisdom on this, it could prevent another story like Doug's possibly with a worse outcome.
 
The J3s I fly are also "Ice makers". Within the "ice range" of temperatures, if there is a high humidity condition, you have to fly with carb heat on, or you will make ice.
This has a secondary deleterious effect - it burns fuel faster. With only 11 gals usable, you need to spend a lot of time watching the fuel wire and playing with carb heat.
 
The J3s I fly are also "Ice makers". Within the "ice range" of temperatures, if there is a high humidity condition, you have to fly with carb heat on, or you will make ice.
This has a secondary deleterious effect - it burns fuel faster. With only 11 gals usable, you need to spend a lot of time watching the fuel wire and playing with carb heat.

My POH (really on the 140 an owners manual not a fully detailed POH) speaks of adjusting carb heat at cruise through trial and error of finding the sweet spot to keep ice away... So at cruise you play with it to find the sweet spot? Or full on and off periodically? Do you have any insight on how you find that sweet spot?

Couldn't you lean it out while running with carb heat on? Being hot thin air needs less fuel for the proper mixture ratio? Could that assist in the heavy fuel consumption? Just a thought...
 
Last edited:
Lycomings tend to be more resistant to carb ice than Continentals due to the location of the intake plenum within the oil sump.
 
How about adding a Carb Temp gage? I really like having one, it takes much of the uncertainty away.

Adding carb heat richens the mixture, because warm air is less dense than cold air. Play with the heat, lean the engine again.
 
How about adding a Carb Temp gage? I really like having one, it takes much of the uncertainty away.

I was just looking at a carb ice detector in AC Spruce, supposedly detects frost on a probe not just temp, as there could be plenty of time below freezing that no ice is building... Just curious if anyone uses one? I like the concept, especially flying an icemaker, I may put this on my upgrades list after harnesses...

I'm definetly going to seriously look at one or the other...
 
I was just looking at a carb ice detector in AC Spruce, supposedly detects frost on a probe not just temp, as there could be plenty of time below freezing that no ice is building... Just curious if anyone uses one? I like the concept, especially flying an icemaker, I may put this on my upgrades list after harnesses...

I'm definetly going to seriously look at one or the other...

The Detector just tells you that ice is present. The Carb Temp Gage tells you when your carb throat temp is in the ice-making zone. Wouldn't it be nice to know before the ice forms?
 
Same old answer: it depends on temperature and dewpoint. Even a Continental won't make ice if the air is dry enough. Pay attention to the carb ice charts. I have plenty of time behind small Continentals, and there are times to be very wary and times when it's not such a big deal, but you need to KNOW the factors that cause it and watch for them.

In very cold temps the moisture is all frozen. Typically lower than -20°C. If there are ice crystals in the air your engine will suck them in and cause no problem. With carb heat you melt them and the water travels farther up the induction tubing where it can re-freeze. The small Continental's induction tubes are near the cooling blast exiting under the cylinders and in cold weather they don't get much warmth from the cylinders. There are winter kits that include insulation sleeves for the tubes and a jacket for the oil tank.

AOPA had an article some years ago that said that several accidents right after takeoff were blamed on carb ice that formed during taxi, even in Lycomings. The engines couldn't generate full power on takeoff. It can take a measly 100 RPM loss to make for very marginal performance.

Google Carburetor Ice. There are plenty of good articles to study. A couple:
https://www.skybrary.aero/index.php/Piston_Engine_Induction_Icing

https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=23&ved=2ahUKEwj385TQ_szfAhUWo4MKHRzlB90QFjAWegQIDRAC&url=https://www.aopa.org/-/media/files/aopa/home/pilot-resources/asi/safety-briefs/sb09.pdf?la=en&hash=526443ED8798604D33A89490F116E9EE2C281F6C&usg=AOvVaw3WAaimtFBVOpa5YvWT6Jlo

These are basic articles. There are more complex treatments of the subject in textbooks. And there is plenty of misinformation on the 'net about carb ice, from people who themselves don't understand it.
 
Last edited:
Lycomings tend to be more resistant to carb ice than Continentals due to the location of the intake plenum within the oil sump.
ya but.....my Six ( Lycoming O-540...carb'd) did ice up a few times. My C-150 (TCM 0-200) would ice up on a moment's notice. So, if there is a carb "I" use carb heat routinely....especially when its warm and humid...like on a summer day. Winter weather usually drys out the air.
 
The Detector just tells you that ice is present. The Carb Temp Gage tells you when your carb throat temp is in the ice-making zone. Wouldn't it be nice to know before the ice forms?
The problem is that the carb temp gauge will be in the hazard zone most of the time. It can't tell you what the dewpoint is and it can't tell you if the risk is high or low.

Carburetors are very old technology. Back in the old days when people HAD to understand machines to operate them, they knew what to do. Now we live in a world of automated stuff including motor vehicles, and we've lost that knack. Pilots flying carbureted engines need to study the phenomenon and understand it. Gauges can help, but if it isn't being monitored you'll still get stung, and if it indicates high ice risk all the time and nothing happens, you'll start ignoring it.

The METARs always list temp and dewpoint. With those you can refer to the carb ice chart and see what the risk is for that day, and operate accordingly.
 
ya but.....my Six ( Lycoming O-540...carb'd) did ice up a few times. My C-150 (TCM 0-200) would ice up on a moment's notice. So, if there is a carb "I" use carb heat routinely....especially when its warm and humid...like on a summer day. Winter weather usually drys out the air.
Lycomings will build ice as fast as a small Continental on the first start of the day if the temp and dewpoint are close enough. The oil is cold for some time and contributes nothing to ice prevention until it warms up. You could take off with ice already in the carb. One thing to watch for: will it idle with the throttle all the way back? If it tries to die, you have ice already.
 
The problem is that the carb temp gauge will be in the hazard zone most of the time. It can't tell you what the dewpoint is and it can't tell you if the risk is high or low.

Carburetors are very old technology. Back in the old days when people HAD to understand machines to operate them, they knew what to do. Now we live in a world of automated stuff including motor vehicles, and we've lost that knack. Pilots flying carbureted engines need to study the phenomenon and understand it. Gauges can help, but if it isn't being monitored you'll still get stung, and if it indicates high ice risk all the time and nothing happens, you'll start ignoring it.

The METARs always list temp and dewpoint. With those you can refer to the carb ice chart and see what the risk is for that day, and operate accordingly.

My Carb Temp gage is almost always above the orange "danger zone" stripe. When it's in it, I pay more attention and run enough heat to get above it in IMC. But then, I have a Lycoming nit a Continental ice box.
 
The J3s I fly are also "Ice makers". Within the "ice range" of temperatures, if there is a high humidity condition, you have to fly with carb heat on, or you will make ice.
This has a secondary deleterious effect - it burns fuel faster. With only 11 gals usable, you need to spend a lot of time watching the fuel wire and playing with carb heat.

Does the J3 not have a mixture control? I can’t remember.

In most airplanes, you can actually reduce your fuel burn using carb heat as long as you lean after applying. It’s actually an old timers way (I think Lindbergh discovered it) for running radials lean of peak.
 
Interestingly enough, I’ve only ever experienced carb/induction ice in radial airplanes.
 
My POH (really on the 140 an owners manual not a fully detailed POH) speaks of adjusting carb heat at cruise through trial and error of finding the sweet spot to keep ice away... So at cruise you play with it to find the sweet spot? Or full on and off periodically? Do you have any insight on how you find that sweet spot?

Couldn't you lean it out while running with carb heat on? Being hot thin air needs less fuel for the proper mixture ratio? Could that assist in the heavy fuel consumption? Just a thought...

There is no mixture control in a J3. The plane is almost prehistoric.
Some days I play, some days it's on or off.
The Cub is pretty good (usually) about letting you know the carb is icing up.
That being said, I did get surprised last April when the engine sputtered and quit on downwind.
 
2 points;
1 the most dangerous part of carb ice is the little tin mufflers cool very quickly. If you don't get carb heat going quickly, there won't be any heat to remove the ice.

2 the smaller TCM engines based upon the C-series design and the 0-470 and up, are ice makers due to the remote mounted carbs, the 0-300 is not as prone to carb ice because it mounts the carb on the oil sump.
but still, IMHO a carb temp gauge should be mandatory.
 
There is no mixture control in a J3. The plane is almost prehistoric.
Some days I play, some days it's on or off.
The Cub is pretty good (usually) about letting you know the carb is icing up.
That being said, I did get surprised last April when the engine sputtered and quit on downwind.
Yeah, I thought that might be the case. I’ve never flown an actual J3, but did fly a Champ that had no mixture control. That would be a challenge with carb heat on frequently
 
Can someone post a link to an easy to print table for carb icing temp and dew point values? I looked at the two articles already mentioned but did not see any charts there. I would like to print one out and throw it in my flight bag to have on hand.
 
So we can get wind and temperatures aloft data, but there are no dew point values for temps aloft (or maybe I am just missing them). If the chart indicates icing potential using airport surface temp and dew point info, does that translate into conditions during cruise at 5000 feet? Does the relative humidity stay relatively constant with altitude?

I am guessing that all this info is on the Skew T, but I am not quite there yet. Still trying to pull the trigger for Weatherspork, but can’t quite do it.
 
2 points;
1 the most dangerous part of carb ice is the little tin mufflers cool very quickly. If you don't get carb heat going quickly, there won't be any heat to remove the ice..

That has always been on my mind, On most checklists I see apply carb heat as not the first item for after engine failure, in my mind it should be first due to that, and the fact it is a split second operation, trimming to glide may take more than 1/4 second as an example, and while trimming to best glide your mufflers are going to be cooling off the whole time.

In addition to what I mentioned in first post I try, and am now amending my pre-descent checklist to apply carb heat before starting my decent to the pattern before cutting any throttle... I would rather get a blast of more hot air from cruise than luke warm are after a nice easy long decent. I would assume has cooled them a bit too even in the "green" arc on decent as some of your rpm is coming from some windmilling action is it not? As in the engine isn't working as hard to maintain a given rmp in decent as it would be even in cruise...

I guess I need to decide now a temp gauge or ice detector to put on the to do list...
 
Lycomings tend to be more resistant to carb ice than Continentals due to the location of the intake plenum within the oil sump.
That is generally true. Moreover, some installations of a given Lycoming engine are more ice-resistant than others. I've never experienced carb ice with an O-360-A4M in a PA-28, yet the same engine in my 172N will make some ice if conditions are right.
 
That has always been on my mind, On most checklists I see apply carb heat as not the first item for after engine failure, in my mind it should be first due to that, and the fact it is a split second operation, trimming to glide may take more than 1/4 second as an example, and while trimming to best glide your mufflers are going to be cooling off the whole time.
It shouldn't require more than a couple seconds to do both.
 
It shouldn't require more than a couple seconds to do both.

Oh I agree neither are lengthy activities, but my thought is those are precious seconds of losing heat... I still can pull a knob quicker than adjusting a trim wheel...
 
To me it's really easy, if the engine starts running crappy and you have carb heat, pull it, pull it hard and all the way. If the engine runs worse after you pull it then you probably have just fixed the problem, give it time, don't push it back in.
 
To me it's really easy, if the engine starts running crappy and you have carb heat, pull it, pull it hard and all the way. If the engine runs worse after you pull it then you probably have just fixed the problem, give it time, don't push it back in.

I don't disagree on the basics of what you say, but after reading Pauls story in lessons learned and reading NTSB reports on take offs with engine failures for no apparent cause, I think its something we could discuss on best practices more than just the basic: if it runs rough pull it. Paul said he had little to no warning... These engine failures at take off with no apparent causes are those could be possible carb ice... A discussion on what temp is too low to use it, etc...
 
On my Fly Baby with a C85 Continental, I've had only one definite case of carburetor icing in ~20 years. Most of which is probably because I learned on an O-200 Cessna 150, and the instructor made sure I was applying heat aggressively. On for taxi (except with visible blowing dust), on after the runup, on for the initial throttle-up for takeoff, on before any power reduction.

The one time I got it was flying in the foothills. It was completely unexpected. I turned on the heat and turned towards lower elevations. Both seemed to do the trick.

Ron Wanttaja
 
I don't disagree on the basics of what you say, but after reading Pauls story in lessons learned and reading NTSB reports on take offs with engine failures for no apparent cause, I think its something we could discuss on best practices more than just the basic: if it runs rough pull it. Paul said he had little to no warning... These engine failures at take off with no apparent causes are those could be possible carb ice... A discussion on what temp is too low to use it, etc...

I flew skippers many years ago in New England and it seemed many days were ripe for carb ice. I remember a few times seeing the rpm drop (fixed prop) then pulling the carb heat for a few minutes and it would get back to normal. Part of the run up for those airplanes was to pull the carb heat during the runup, after the mag check I believe, looking for an rpm drop, then pulling back to idle, pushing the carb heat off and looking for the subsequent rpm increase. I don't fly an airplane with carb heat now, but one of the first things I was taught to do with a balky engine is immediately pull the carb heat if it has it. As Tom said, the heat dissipates quickly if the engine isn't running correctly, so you don't have long to fix it.
 
So we can get wind and temperatures aloft data, but there are no dew point values for temps aloft (or maybe I am just missing them). If the chart indicates icing potential using airport surface temp and dew point info, does that translate into conditions during cruise at 5000 feet? Does the relative humidity stay relatively constant with altitude?
A given air mass has fairly consistent moisture levels. Passing through a front can change that, but otherwise you can assume the same dewpoint. You can also forecast cloud bases from it. The temperature drop is about 2 degrees per thousand feet, so if the dewpoint is 6 degrees below the ground temp the cloud bases will be at 3000' agl.
 
...so dumb question.. but why not put the air filter further upstream and just run the engine with carb heat anytime condition X prevails? Sure it will get more rich, so just lean it out.. what am I missing?
 
I once had a BCS garden tractor that would ice up, had to get rid of it, it would run about 15 minutes and quit, the carb looked like a snow ball
 
...so dumb question.. but why not put the air filter further upstream and just run the engine with carb heat anytime condition X prevails? Sure it will get more rich, so just lean it out.. what am I missing?
Warm air is less dense, so less power, and more heat.
 
Carb ice is one of those things the more you talk about it the more you think your going to die every time you start the engine. In another thread I wrote about recent suspected icing I had and dealt with before it dealt with me.
@Sinistar had a good addition about locking the MP and making it part of the hard scan at level altitude, any drop indicates possible early icing. I like that and will try to integrate into my scan. I recently watched a yT vid of a 152 icing up on full power takeoff. Never made it off the ground luckily. He mentioned a long ground idle may have set him up for that.
 
...so dumb question.. but why not put the air filter further upstream and just run the engine with carb heat anytime condition X prevails? Sure it will get more rich, so just lean it out.. what am I missing?

The warmer air reduces the power the engine can make.


Sent from my iPhone using Tapatalk
 
Warm air is less dense, so less power, and more heat

The warmer air reduces the power the engine can make.
Right, but you lose what, 100 RPM? I'll take that over an engine failure.. and it's not like you would run it with it always on..

as noted above, once you recognize that you have carb ice you have a short amount of time to get it off
 
Right, but you lose what, 100 RPM? I'll take that over an engine failure.. and it's not like you would run it with it always on..

as noted above, once you recognize that you have carb ice you have a short amount of time to get it off
Depends on what you are doing/phase of flight.

Leaving it on in cruise might lose you a few knots. Basically like running LOP.

Full carb heat on takeoff can make the difference between climbing and descending into terrain.
 
The Detector just tells you that ice is present. The Carb Temp Gage tells you when your carb throat temp is in the ice-making zone. Wouldn't it be nice to know before the ice forms?
As another poster said the needle is in the warning area for probably 9 or 10 months of the year here in Minnesota. So that gauge provides information but it is so highly dependent on another factor - the visible moisture or the invisible high humidity.

So for visible moisture its 'sometimes' more obvious, but not always. Freezing rain - carb ice time...but why am I up? Snow - maybe? Clouds in the distance - I have no damn clue it is visible moisture? Haze - ??? Mist or Freezing Fog - yeah, don't be flying either. So that carb temp gauge is in the zone but doesn't mean anything until you couple it with the not so obvious other factor of the moisture in the air.

I would happily trade the Carb temp gauge for any indicator of ice starting to form. Heck, all it need to be is a flashing red light that with a self test for the bulb. Other than that, on every occasion of carb ice we have noticed it on the MP or the engine sounds off a bit. And the only verification has been to pull carb heat and watch RPM drop then raise back somewhat and then remove carb heat. We have never had a case where we had to leave it on continuously.

Another poster indicated if suspected to pull it on and pull it on hard. I have been taught to pull the heat on but not to jerk it but rather to make it take like 1 second that way a big chunk of ice doesn't dislodge and block it.

The MP makes for a fantastic indicator...but you have to fly level :)

I want a ice indicator, unless a ice indicator can not sense the very start of if forming then the MP is the next best indicator.

The one that worries me the most is carb ice during taxi and waiting to takeoff.
 
I've also been told (By my flight examiner) that at a certain really low temp one shouldn't use it at all as it could put the temp into an area of temps ice is more likely.
Here is the classic picture of temp / dewpoint and carb ice probability
nyc02fa025_1.jpg
:

The thing is, absent visible moisture, raising the temperature with carb heat moves you horizontally to the right - any heat moves you to a reduced risk. The only exception is if there is some form of water (rain, snow, ice, cloud) that you can evaporate to increase the absolute humidity, and even then, you have to raise the dew point faster than you raise the temperature in your induction system.
 
Back
Top