# 100LL and O2 sensors; your thoughts

Discussion in 'General Auto Conversion Discussion' started by pfarber, Oct 13, 2019.

### Help Support HomeBuiltAirplanes Forum by donating:

1. Oct 13, 2019

### pfarber

#### Well-Known Member

Joined:
Feb 21, 2019
Messages:
287
15
Location:
Pennsylvania
I just want to add to this as I've been digging for flight worthy 02 sensors for an auto conversion.

The O2 sensors get coated with lead, that's what 'kills them'. They cannot detect the O2 level in the exhaust as the lead literally covers the sensing probe. Its not a like a chemical reaction is 'eating away' the metal, its still there, buried under a coat of lead. The same thing that lead does do spark plugs. It doesn't eat the plug or corrode it, it just fouls it.

From Shells web site:

"The Tetra Ethyl Lead used for octane boost in the fuel naturally degrades to form Lead Oxide when it is burned. In reality it is this oxide which gives the octane boost. The problem is that Lead Oxide is a solid up to about 900 deg C which is well within the wall temperatures inside a piston engine.

In order to prevent these deposits from forming, a Lead scavenging compound is added to Avgas 100LL - this compound is Ethylene Dibromide. This scavenger is designed to react with the Lead oxide to form Lead Bromide which is more volatile - becoming a gas at around 200 - 250 oC. This is a low enough temperature to ensure that the Lead is removed from the engine as a gas end it subsequently goes back to the solid phase as the exhaust gas cools in the atmosphere."

From wiki:

"Leaded gasoline contaminates the oxygen sensors and catalytic converters. Most oxygen sensors are rated for some service life in the presence of leaded gasoline, but sensor life will be shortened to as little as 15,000 miles (24,000 km), depending on the lead concentration. Lead-damaged sensors typically have their tips discolored light rusty."

But 100LL has 8 times the amount of lead that car fuel does. So even if you get the bare minimum of 15,000 miles (at 60mph that's 250 hours, 1/8 of that is 31.25 hours). So you could most likely find an O2 sensor to last a 'normal' flying season, to be changed at the annual.

So it would appear that if you put the O2 sensor in a location that keeps it hot, and the additives in 100LL keep the lead gaseous. Now could you move the pre-cat O2 sensor close enough to the exhaust valve where the TEL is still gaseous? And will it survive the higher temps? Most O2 temps need to be at 300C+ to work, well above the 200-250C gaseous Lead Bromide temp.

Would a 20 or 50 hour remove and clean be enough? Or even just replace the sensor every 25 hours (I dare say most non-school planes don't fly that much). O2 sensors are $30-ish each.$60 per oil change is not a huge amount of money.

Ninja edit - NO, I DON'T WANT TO HEAR ABOUT RACE ENGINES. They are highly modified to burn much more gas to create much more power. And who knows what 'tricks' they incorporate to run the 8-10 laps on a short track.

2. Oct 13, 2019

### Hephaestus

#### Well-Known Member

Joined:
Jun 25, 2014
Messages:
1,175
255
Location:
YMM
Pick the right o2 sensor

Seriously some of the non heated automotive ones are cheaper than certified sparkplugs.

The one I just did in my Toyota lists at 18$on us Amazon... Then replace as a maintenance item like plugs or oil changes... proppastie likes this. 3. Oct 14, 2019 ### TiPi ### TiPi #### Well-Known Member Joined: Aug 25, 2019 Messages: 142 Likes Received: 138 Location: Mackay (AUS) W What is the purpose of the O2 sensor? If it is a critical part of the EFI, I wouldn't use any leaded fuel as it will also cause other issues in engines that are not designed for AVgas (sticking valves, burnt valves etc). If it is as a mixture indicator (AFR instrument with wide-band O2 sensor, just mount it away from the main exhaust gas stream to reduce the exposure to the lead. In a closed-loop EFI system, the ECM needs the O2 feedback pretty much immediately. As an AFR feedback for the mixture control, a time delay is of no consequence. O2 probes can't be cleaned from lead contamination. 4. Oct 14, 2019 ### Marc W ### Marc W #### Well-Known Member Joined: Apr 1, 2017 Messages: 296 Likes Received: 199 Location: Colorado NGK/NTK makes a sensor that is more tolerant of lead. Look at Ballenger for info on it. It is higher priced but may be worth it. There was also a Kitplanes article recently about using the Ballenger AFR monitor in an aircraft. 5. Oct 14, 2019 ### pfarber ### pfarber #### Well-Known Member Joined: Feb 21, 2019 Messages: 287 Likes Received: 15 Location: Pennsylvania If the lead bromide is above 200-300C then an O2 sensor close enough to the exhaust port where the gases are hot should not collect significant amounts of lead as its still a gas travelling at high speed. Most pre-cat O2s are still fairly far removed from the exhaust header usually on the downpipe. I'm sure some will deposit but not like if the sensor was in a 'cold' region where the lead would solidify and stick like tar to anything in its path. If anything a significant reduction should be realized. Running an EFI system open loop is madness. I think the fuel savings in a year would more than offset replacing O2 sensors at 50 hours. At 10gpm a 10% fuel savings (which closed loop would obtain easily) is 1gph. At$6/gal for 50 hours that $300 in fuel vs$60 for O2 sensors at an oil change. I'll take $240 (most likely more) in savings any day. 6. Oct 14, 2019 ### AdrianS ### AdrianS #### Well-Known Member Joined: Jul 5, 2014 Messages: 374 Likes Received: 226 Location: Australia It would be interesting to know if lead affect narrow band sensors more/less/the same as widebands. Many older automotive ECUs just have a narrow-band sensor, and run open-loop at full throttle. At least mine does. Narrow-band sensors are cheap enough to be a service item. 7. Oct 14, 2019 ### pfarber ### pfarber #### Well-Known Member Joined: Feb 21, 2019 Messages: 287 Likes Received: 15 Location: Pennsylvania O2 sensors for the LV3 are$30 from rock auto. I am not sure if the ECU can be set up to run off two or if it will need to see the post cat O2 (or even just put in a 'simulator').

So $60 for new O2 sensors every year vs the extra fuel burn of open loop. I guess if you tune the ECU's map on a dyno for the limited cruise RPM you might get a little better GPH open vs closed loop, but I'm seeing number of 10-20% better closed vs open. How well will the ECU handle higher density altitude with an open loop map? Or the much colder temps at say 10,000 feet? Again, you fly 30 hours a year, the money is on your side to run 100LL and just make the O2s a consumable. But if you find the sweet spot where O2s don't lead up then you might double or triple that number. If you can run an O2 sensor for 3 years that darn near$1000 in your pocket.

8. Oct 14, 2019

### proppastie

#### Well-Known Member

Joined:
Feb 19, 2012
Messages:
3,789
1,096
Location:
NJ
being experimental can run un-leaded auto fuel....just use a little 100LL when on a cross-country.

gtae07 likes this.
9. Oct 15, 2019

### TFF

#### Well-Known Member

Joined:
Apr 28, 2010
Messages:
11,632
3,269
Location:
Memphis, TN
Why not have three or four maps set for certain density altitudes. Just switch as you climb. I would want a open mode switch anytime I felt it was acting up. It could also be takeoff or sea level map.
Technically if the computer knew the right answers with the O2 it will because of you mapping the details so it knows what page to read. It will take a bit to get right

I know I would not want any OBDII near an aircraft not set by a factory.

10. Oct 15, 2019

### pfarber

#### Well-Known Member

Joined:
Feb 21, 2019
Messages:
287
15
Location:
Pennsylvania
OBDCII is not the problem. OBDC is for nothing more than emissions problems that a consumer can use to fix their car without a trip to the stealership. And limp home mode is a GOOD THING.

In a non-ECU motor cams/cranks can break, gear bolts can shear, mag's can break, basically catastrophic failures with little to no notice. None of the critical engine stats are monitored. Sure there is a oil temp/press/ammeter etc in the cockpit, WWAAAYYYY off to the side, safely out of eyesight.

Now consider an ECU (and again OBDCII is not even in play at this point). The cank and cam positions are known, the oil temp/press/level are measured. The alternator voltage is measured, there are knock sensors to adjust timing and fuel is precisely metered and delivered in, or near the cylinder. If ANY of those sensors go out of spec you get a light (CEL).

Now you can argue that 'the crank position sensor can fail' and a $20 part stops my motor. Well, an$8,000 crank can shear off and stop your motor. I'll put my money on the $20 hall effect sensor every time. They just really don't break that much. If the motor starts to operate outside of norms, you get a warning (CEL). So what would you rather have? A 'something is broke so were gonna reduce everything to make sure you can get somewhere' or be blind and only know whats up AFTER a failure? Bottom line even a car motor that's never been touched can run 100,000 miles with little more than oil changes. A certified engine has so many checks (even an annual checkup) that they should never break, yet they do. 11. Oct 15, 2019 ### proppastie ### proppastie #### Well-Known Member Joined: Feb 19, 2012 Messages: 3,789 Likes Received: 1,096 Location: NJ often after when there is a problem......the more I work on them the more I believe "do not fix it if it's not broke" 12. Oct 17, 2019 ### Terrh ### Terrh #### Well-Known Member Joined: Nov 12, 2016 Messages: 73 Likes Received: 10 Location: Windsor, ON, Canada / Detroit, MI, USA. I know you said you don't want experience with race engines, but that's 95% of the experience I have, and most of those are "street" engines that just get raced and see leaded fuel as a consequence of that: with how cheap widebands are now you can just change them annually if they last that long. Replacement ones for my AEM gauge (I believe they are a bosch sensor) are sometimes as cheap as$15 on amazon "open box" which usually just means damaged packaging.

Real world testing with a wideband in a 100LL application is easy. Just put a bung and standalone wideband gauge in and see when it starts to act up. A good standalone gauge is the AEM 30-4110. I've got a used one here I could sell you for $100 if you want. Probably should have a new sensor in it. My experience with leaded car fuel (mostly turbo blue 110 oct) is that it can foul wideband sensors as quickly as 20-30 gal of burned fuel. I have not found a way to make them work again after they foul, and I have tried many methods including glass bead and wire brushing. My sensor mounts pretty far down the exhaust, I have a twin turbo straight 6 and it's mounted after the two exhaust pipes merge, probably 4 feet after the exhaust port. I run the leaded 110 because it's SO much cheaper than unleaded 110 ($10/gal vs \$22/gal, Canadian) so it just makes more sense.

13. Oct 17, 2019

### Dan Thomas

#### Well-Known Member

Joined:
Sep 18, 2008
Messages:
4,855
2,070
You wouldn't get even 25 hours. Your car cruises at 25% power or a bit more. Your airplane cruises at 75% and climbs at 100%. Since fuel burn is directly related to power output, the vast amount of lead going into the exhaust would coat that sensor real quick.

14. Oct 17, 2019

### pfarber

#### Well-Known Member

Joined:
Feb 21, 2019
Messages:
287
15
Location:
Pennsylvania
Thank you for playing, but no.

Even race car engines (high RPM, massive flow, about 700-1000HP) last longer the 'even 25 hours'.

And you missed the first post. The lead is kept gaseous above 250C be a fuel additive. My though is to move the O2 sensor closer to the head to ensure that 1. the lead in the exhaust are still in gaseous form after combustion, 2. The exhaust flow is fast enough to keep gaseous lead from sticking, and 3. the O2 sensor is kept hot enough to melt any lead that did attach to the sensor.

So do you have ANYTHING that supports your position about anything you said?

15. Oct 18, 2019

### pictsidhe

#### Well-Known Member

Joined:
Jul 15, 2014
Messages:
6,849
1,843
Location:
North Carolina
Maybe someone should look at the temperature that O2 sensors run at. They don't even work until they are way over 250, that's why they have heaters... Lead does not smother the sensors, it poisons them.

16. Oct 18, 2019

### proppastie

#### Well-Known Member

Joined:
Feb 19, 2012
Messages:
3,789
1,096
Location:
NJ
Sounds like a plan.....report back on how it turns out.

17. Oct 18, 2019

### trimtab

#### Member

Joined:
Apr 30, 2014
Messages:
10
6
Location:
rocky mountains, rocky, usa
You can extend the life of an O2 sensor by increasing the cross count time. This isn't something most people are capable of or motivated to do.

Another approach is an optical system, but these are all roll-your-own unless you need to interface it to an ECM or FADEC. It isn't difficult, but again, it is beyond what most people are capable of or motivated to do. The optical system can be self cleaning, very fast, and very accurate.

18. Oct 18, 2019

### pictsidhe

#### Well-Known Member

Joined:
Jul 15, 2014
Messages:
6,849
1,843
Location:
North Carolina
Oh, I have better things to do. But I do know that they have be way over 250 to work as O2 sensors, I just forget the number.

19. Oct 18, 2019

### pfarber

#### Well-Known Member

Joined:
Feb 21, 2019
Messages:
287
15
Location:
Pennsylvania
No, there is no erosion or corrosion of the O2 sensor. From the first post (a wiki quote):

In order to prevent these deposits from forming, a Lead scavenging compound is added to Avgas 100LL - this compound is Ethylene Dibromide. This scavenger is designed to react with the Lead oxide to form Lead Bromide which is more volatile - becoming a gas at around 200 - 250 oC. This is a low enough temperature to ensure that the Lead is removed from the engine as a gas end it subsequently goes back to the solid phase as the exhaust gas cools in the atmosphere."

The lead in the hot exhaust cools and returns to a solid, and sticks to the O2 sensor, clogging the pores of the sensor material. The O2 sensor is fine, but just can't sample the exhaust flow.

20. Oct 18, 2019

Joined:
Apr 28, 2010
Messages:
11,632