AFM Adjustment FAQ

[SGKent]

[BusjeCA]

OK, my first post. I'm still subject to smog here in CA, and have a (nearly complete except EGR) 79 FED type 2. As per Ratwell's advise, I have a LM-2 permamently hooked up. I have found a LOT of misinformation how to adjust the AFM. I changed EVERYTHING and measured all speeds, below is my procedure.
The basic info is correct, there are three adjustments you can do:

1/ wiper (CW is leaner)
2/ spring (black cog, CW is leaner)
3/ Air bypass (screw on the side, CW is richer)

Some things to consider At high RPM, (lots of air going through) the flap is near or at its maximum deflection (REGARDLESS of the cog setting, wiper setting and bypass setting. The, the largest effect at high RPM on air/fuel mixture is therefor the WIPER. The deflection angle is not a linear function with AIR/RPM but in fact a curve steep at low RPM and shallow at high rpm. Thus at high RPM use the wiper. From measuring highway driving I know that the wiper is the most important thing to change if your van is running too rich or too lean.

So to start, set the air bypass in the middle and use the cog to set the O2 at say 14.5 stationary, 900 RPM. Go drive 60 and measure the O2. Highway driving should be around 13.5 or a little richer if you love your exhaust valves, a little leaner if you hate to buy fuel. Accelerating it'll go down to the 12s, which is fine. If you can, test if there is any stumble uphill, and if if so make it richer with the wiper (SMALL adjustments). See the figure below how the wiper changes the RPM/fuel.

If you are happy with the highway air fuel mixture, weld the wiper in place (i.e. no more touching).

Two adjustments left 1/ bypass and 2/ cog. Most of this can be done in the driveway. Consider that the cog spring is long, and a small adjustments does not really change the force/angle, but it does change the force with which the flap is held back. The flap is very responsive at low RPMs (the flap is near perpendicular to the airflow and middle RMPs (flap still partially in the airflow). Since the linear response of the spring is nearly unchanged, the cog adjustment shift the fuel air curve left and right at low and medium RPM.
So keep the bypass where it is and set 2000 rpm at 14.5 (or so). Back to 900 RMP. If the O2 fine (you are done).
If too lean, use the bypass to make it a little too rich (14.0). At 2000 rpm it will run too rich so set it where you want it (say 14.5) with the cog, usually one to three teeth CW. The effects of these setting is changing the angle of the curve between 900 and 2000 rpm. The bypass is more effective with the flap largely closed, less effective at 2000 rpm. The cog affect both about equally (see fig.). Keep going through this little circle until all is within specs.
Take it out for a drive and test the O2. You'll find highway driving unaffected (IF YOU KEPT YOUR FINGERS OFF THE WIPER). It might run a little rich at 2000 rpm (15 in second, 25 in third) (the engine is working a little harder), so back to the driveway, and turn the cog a little leaner, and re-set the stationary back to where you want it with the bypass.

I do this once every two years, and the van passes smog without CO or NOX problems.

[airkooledchris]

Thank you for the additional details BusjeCA

Do you track your head temperatures as well? If so, what have you noticed in regards to AF mapping and your head temps?

[AZ Landshaper]

Innovate NEW 3844 MTX-L Wideband O2 AFR UEGO gauge kit w/ Sensor Included ( as seen on TV).

Is something like this sufficient to determine my A/F ratio in my effort to tune my AFM?

I'm gonna start w the 60 grams setting and then use the 3844.


Responses appreciated.
_________________
Support Small Business.

-------------------------------------------------------
85 Weekender w/ EJ22
Previously
64, 71, 72, 73, 76, 81, 84, 85 & 87 Campmobiles and Westfalias
and a 67 bug.

[SGKent]

that will work but have someone else call out the numbers while you are driving on the tuning sessions.
_________________
“Most people don’t know what they’re doing, and a lot of them are really good at it.” - George Carlin

[AZ Landshaper]

Roger that Merlin.

Thanks
_________________
Support Small Business.

-------------------------------------------------------
85 Weekender w/ EJ22
Previously
64, 71, 72, 73, 76, 81, 84, 85 & 87 Campmobiles and Westfalias
and a 67 bug.

[Busstom]

Wow! Slightly intimidating device

When disassembling my stuff, the lid just fell off my AFM. Sealant was a clear sign somebody had been in there. I haven't touched anything in it. Still in teardown mode, but already found issues (such as no small inner seal on the #1 injector!) that were causing my crazy timing/burned head problem. I plan to detail that stuff in my "Bizarro" thread, since that's what led to my current (and sad) state.

But something else I saw in here, injector colors. I believe mine are white, and I have a hydraulic lifter engine. Displacement still unconfirmed, I'm chipping away at it night by night. I don't recall seeing the definitive answer on which color injectors are for which displacements...it was probably up in there somewhere. Bentley manual on the way, via Amazon.

[AZ Landshaper]

AFR kit has arrived. Can't wait to get it installed.

_________________
Support Small Business.

-------------------------------------------------------
85 Weekender w/ EJ22
Previously
64, 71, 72, 73, 76, 81, 84, 85 & 87 Campmobiles and Westfalias
and a 67 bug.

[airkooledchris]

What would you go about adjusting if your low RPM speeds were rich and your high RPM speeds were lean?

Ideally, id like the opposite.

[airschooled]

[jerryherb]

Would all this EXACT (60g test, etc) calibration info apply to a 2.0 AFC vanagon? i noticed a PO was already inside my AFM per RTV around the cover. my van gets 12-16 mpg cty/hwy so thinking its probably running rich, but temps can go up to 400 when under load (65 slight incline even when cool out) evens out around 350-375 at idle. it also idles low when summer hits (90F+/700RPM or so) when hot leading to oil light flicker. where would i hook up the wideband sensor on this and how would the mixture need to be adjusted for a deleted EGR valve?

Another question: I also have an 85 1.9 WBX where AFM was "touched" to run stinkin rich. is the flapper weight needed to open the same (60g) and would one adjust it the same but with the narrowband O2 disconnected (with wideband in its place for testing)? I know the ECU on digijet runs around .56V (internal reference voltage) once O2 is disconnected, but would the ECU then enrich/lean the mixture once narrowband O2 is reconnected making the fine tuning the AFM pointless? If there is another place i can find the WBX calibration/spec procedure please point me to it.

i did read a few adjustment threads here in thesamba and Amskeptic's visually colorful procedure on Itinerant air-cooled. It appears the aircooled experts are a lot more knowledgeable about this AFM pandora's box. TIA

[MrPolak]

I don't know if this was discussed, but 60g is 60ml of water. If you have a small measuring tube you can make yourself a near-60g weight.

Also coins:

Cent: 2.500 g
Nickel: 5.000 g
Dime: 2.268 g
Quarter: 5.670 g
Half Dollar: 11.340 g

[cellobus1]

I'm reviving this thread bc I want to clean out the bottom of the AFM, which is accessed by removing the screws that are facing us in the above photo. I'd like to know if anyone has done this, and what kind of tool they used to get the screws out. They're jeweler's screwdriver-size screws, and they're REALly in there. I'd also like to know what kind of sealant would be best for reassembly. Bentley briefly touches on this at the end of 4.9 of the FI chapter, but is typically tight-lipped about details. Any advice about what parts may come flying out after the cover is off would be helpful also. (The symptom I'm trying to cure is that the vane is getting stuck open, especially on uphills at a stop sign).

Thanks!
Andy
_________________
1976 unrestored daily driver Standard bus, "Stella"
formerly, 1959 Standard bus
formerly, 1973 transporter which was 4 years old when it taught me to drive

[SGKent]

[Clatter]

[Ozzieman]

Has anyone measured a faultless AFM's resistance across the whole resistive track, or measured the voltage of pin 7 throughout the flaps range (when connected to ECU) and mapped it out?
I suspect it to be non linear right?

Reason I'm asking is I have two AFM's that are worn on the resistive tracks and spares are getting hard to find.
I was thinking of replacing the potentiometer with a absolute position sensor connetced to an arduino/ESP32 (or similar) and then a mosfet to regulate the outgoing voltage.
This would have the benefit of no wear and also the possibiltity of tweaking the curve electronicallly (while driving) depending on what the specific engine wants thru O2/lambda, or adjusting for altitude.

I've used this cheap absolute position sensor/encoder in other projects and it's very accurate and stable:
https://www.seeedstudio.com/Grove-12-bit-Magnetic-Rotary-Position-Sensor-AS5600-p-4192.html

Mosfet: https://www.seeedstudio.com/Grove-MOSFET.html?quer...r_products

To fit the parts one could 3d-print a taller lid in place of the original one.
_________________
Kleinbus -78, 2.0

[SGKent]

get a hold of Richard Atwell. He was working with some people who are leap years ahead on this issue. The problem is that the AFM has many pins out and each one provides voltage to the ECU simultaneously You have to track all those voltages at once, not just pin 7. The measurement that matters is voltage and not just resistance.

I tried replacing a worn out board with a new one but it did not work right. The process Bosch used was to put the AFM in some kind of a special test rig then laser etch each resistor on the board to trim the voltage to whatever their specs were. Those specs are not to be found anywhere. Many have looked and all have failed. The new boards out there have not been trimmed with a laser yet and that is why they don't work. Richard Atwell was working with some folks who had a higher tech solution.
_________________
“Most people don’t know what they’re doing, and a lot of them are really good at it.” - George Carlin

[Ozzieman]

Ok, I'll send him an email.

I was looking at this image and it looks like it's only pin 7 of the AFM that's sending a signal (1.75-7V) to the ECU (except for the temp 1 value at pin 27):

_________________
Kleinbus -78, 2.0

[busdaddy]

There was another similar idea here on TS somewhere in the last year or two, but I can't remember where I saw it now, maybe Vanagon?
_________________
Rust NEVER sleeps and stock never goes out of style.

Please don't PM technical questions, ask your problem in public so everyone can play along. If you think it's too stupid post it here
Stop dead photo links! Post your photos to The Samba Gallery!

Слава Україні!

[Ozzieman]

Cool, I did a quick search, but didn't find any thread. I'll give it som time later on.

Today I hooked the AFM that's not in the bus to 12V on pin 9 and 7.5V on pin 8 along with ground on pin 6. A multimeter was connected to pin 6 and 7. Then I set a camera over the AFM and the multimeter and took a photo of them both every couple of degrees (give or take).

By importing the images into Illustrator I could determine how much the angle was shifted between shots and thereby make a spreadsheet with all the values and then print it as a diagram/curve. It's a pretty even curve and I guess if I even out the irregularities I can create an application that outputs the corresponding voltages pretty easily in arduino.

I will do the same with my other AFM when I have time!

_________________
Kleinbus -78, 2.0