AFR reading on a 7Hp gen set

testing some tinkerings on my play arounds…

What effect does the water injection has on the AFR and how come the AFR changes when i throttle up or down…

This time no load on the generator, but next test will try to go max load possible


Koen; Can you talk to us next time to explain what is going on? I have never used an AFR meter so don’t know what it is saying. A higher number is richer or leaner, etc. TomC

1 Like

Koen, I could not see the afr so I don’t know what it said. However I did notice your nozzel is horizontal in this gasifier. You used to put them in vertically. Do you find one way advantages over the other? I have done both ways also but they were different sizes on different size engine’s and no data recorded . They both work well but could not say one is better from my work. What do you think?


Thanks for putting this up Koen , I am amazed at how much more engine speed you get from your water spray , I get a very small speed increase but I get no extra power from my generator when under full load at all .
It will be interesting to watch your next video with the engine under a load and then full load , will you have a inverter generator handy to do the test on ? be great to compare it to mine then .
Cheers for now

1 Like

I’m with you Tom. That is why I got an old fashioned analog meter instead of digital numbers meter.
8 is lean and 18 is rich.


Nope you have it backards. :slight_smile: the lower the number the less air is in the mixture this is reading the Oxygen in the exhaust mixture. The lower the number the less ppm of O2 it is reading and then the higher the more ppm it is reading (higher O2 in the mixture) In genera the Stoichiometric for gasoline is around 14.7 ppm. I have found for wood gas it is closer to around 15.5. This maybe different for charcoal systems.


I think it is the other way. 8 parts oxygen to one part gasoline is rich and 18 parts air to one part gasoline is lean. While we are on this, do I understand correctly that the AFR is actually a wide band O2 sensor and that 14.7 to 1 is stoichiometric for gasoline so the readout of 14:1 when running wood gas is actually around 1.5:1 air to wood gas?

OK, my meter doesn’t have numbers like the one I posted. it just says LEAN when the needle is all the way to the left and RICH when it is all the way to the right and I just A-S-S-U-M-E-D this was the same.


Hi Tom,

The Gauge set used is with a readout for gasoline.
The numbers indicate the amount (mass) of air against the amount of gasoline (mass)
Where as 14,4 to 1 stands for Lambda 1 = complete combustion with no excess air.

the lower number on the readout means: richer
The higher number means: leaner

Notice in the above clip, i point to the gauge, that if spray water in the nozzle, the AFR readout goes towards more rich.
Hence the water turns into energy :grin:

Also, on the bezel, you’l see a led moving around, for those that love the analog readout , it runs from green left to red richt over orange in the middle.

Next time i will try to install it with Lambda readout where 1 stands for complete combustion, more then 1 stands for lean and less then one stands for rich.

Testing to find at what level i can get the optimum power output at what settings.

PS: My work is not intended to tell people how to do things right. Its only a way to share some knowledge/experience that others might implement in their builds for their own benefits.


Hi Jim,

Yes this set is horizontal, for education reasons.
At small systems, for showing, its difficult to tell the difference since those shows are only short time lasting
I did find that, if using vertical nozzles, they can be smaller and the gas gets a little more performant, but again, it depends on a little more then that.

The main observance overall for my gasifiers is: getting rid of obstructions/restrictions and making the gasifier breathing easy, at minimum vacuum, works the best.

Also, keeping the charcoal at best quality and uniform sizing.


Hi Bruce,

Yes you’r correct. the 14,4 to 14,7 stands for Lambda 1 = stoichiometric mixture


Hi Koen , what type of meter are you using in the video narrow or wide band ?
I have a cheap narrow band one on order but have been told they are not as good as the wideband , but I am hoping this should be a good way of giving me an idea what is going on when my generator is being asked for a higher load , if I can see how the mixture is being affected .

1 Like

Hi Dave,

I am using the wideband sensor. The narrow band might work to, but i am unsure… maybe a reason to put it to the test… having those lying around here…

I have used the narrow band on my Chevy. No good.
Woodgas tends to work best at a slightly lean mix. Thats becouse there is litle difference in energy output between stoich and a bit lean, but a a leaner setting the cyl filling ratio is higher. So, for this reason, the narrow band sensor was out of range at best power seting.


Thanks Guys I may have to keep an eye out for a cheap wideband, but all the ones I have seen so far seem to be in the hundreds of $$ , if you hear of a online bargain shop selling them let me know .

Cheers for now


Everything KVL says here is correct.
To help understand “rich” means all of the oxegen is used up and on a gasoline HC fuel you need to cut back on the fuel in the mix. “Lean” would means free oxygen is still available un-used up in the exhaust stream. And probably on a gasoline HC fuel you could add more gasoline to the mix.
This can get tricky even on a refined gasoline fueled system. A cylinder misfire will pump through unburned (uncombined) gasoline HC’s and atmospheric oxegen as still free.
O2 sensor reading, you’d think Lean. No. Why vehicle OBDII mandated individual cylinder misfire detection as mandatory.

Yes woodgas mixes seem to give better power/more complete combustion as Lean measured apparent.
tree-farmer Steve Unruh


I run with a narrow band and like you have said runs best just off the meter while just cruising if I hit the throttle real quick my meter jumps up a little rich that’s what I look for.

If your just using it as something to reference to manually adjust Id just stick to a cheap narrow band. The only reason I use the wide band sensors is because a narrow band does not have enough voltage range to work with a logic controller. It can be done but the range is very small compared to a wide band where you have a full 0-5 volt range. I do not believe there is any benefit going to a wide band if just for reference use.


Thanks for your input Matt, I will just play with this narrow band for now , its only wanting to see what happens under different loadings that got me wanting to play and see .

1 Like

Copy paste. I liked to drive on 0.1volt.