AFM Adjustment

Carbs & F.I.

Moderators: Sluggo, Amskeptic

Post Reply
Blue Moose
I'm New!
Status: Offline

Re: AFM Adjustment

Post by Blue Moose » Thu Jun 14, 2018 6:58 am

Thanks! That makes sense to me, and it's good to know that the lope is normal. Thanks to Colin's posts here and elsewhere about proper adjustment, I think I can get it done. I'm aiming for 13.5-6 at 65mph "cruise" on a flat road and 12.6-7 at wide-open-throttle accelerating (slowly) at that speed. At lighter throttle and idle I'm aiming for leaner (14+), but mostly looking for driveability.

I don't have a CHT gauge, but it sounds like it would be a good indication to have during this kind of tuning. I don't want my CHTs to get too high during test runs if I'm accidentally on the lean side. From what I'm reading here, a constant CHT of 400 is about as hot as anyone is comfortable with on VW heads. From my experience with aircraft, I can tell you that CHTs change almost instantly (within 3 or 4 seconds) in response to mixture changes. In fact, pilots use the CHT as a direct reading when adjusting the mixture.

Are there any noticeable drawbacks to having a ring gauge under the #3 spark plug? (It seems like it would actually lower compression slightly, but may also affect fuel burn if the plug is a little further out of the cylinder). It looks like the preferred gauge is the Dakota Digital, but that's going to run about $150. Has anyone tried one of the less expensive Swift or Falcon gauges from Aircraft Spruce (about $45)?

DugB
I'm New!
Status: Offline

Re: AFM Adjustment

Post by DugB » Thu Jun 14, 2018 9:53 am

I have the DD CHT (as well as their Fuel Pressure Gauge, on my '75 FI Campmobile) and it's good, but IMO the ring senders can only take so many removals/reinstallations of the #3 plug before they're kind of torqued. I already replaced my sender once, but it's an easy job limited to just 2 connections in the engine bay.

BTW, if you're looking for a cheap AFR meter to try, I have a complete PLX Devices Gen. 1 AFR setup I'd let go for low-$$. I recently switched to the LC-2 to verify the PLX numbers and they turned out to be right...so the swap was not needed. It was working well when I removed it from the van. PM me if you're interested.

- Doug :-)

User avatar
SlowLane
IAC Addict!
Location: Livermore, CA
Status: Offline

Re: AFM Adjustment

Post by SlowLane » Thu Jun 14, 2018 10:33 pm

Blue Moose wrote:
Thu Jun 14, 2018 6:58 am
Are there any noticeable drawbacks to having a ring gauge under the #3 spark plug? (It seems like it would actually lower compression slightly, but may also affect fuel burn if the plug is a little further out of the cylinder). It looks like the preferred gauge is the Dakota Digital, but that's going to run about $150. Has anyone tried one of the less expensive Swift or Falcon gauges from Aircraft Spruce (about $45)?
I have a 4-channel MGL Avionics TC-2 gauge that I use to monitor all four cylinders. At a price of $280, it's less that twice the cost of a DD for four times the functionality.

Alternatively, one of these multi-purpose gauges from Auber Instruments would do the same thing as the dedicated DD gauge at half the price, and could be re-purposed for monitoring something else if you decide that you're tired of watching CHT.

I've never liked the ring-terminal-under-the-spark-plug approach, because it just seems like a hack to replace an engineered gasket with a bent-out-of-shape electrical terminal. Instead, I bedded some thermocouple sensors into one of the head's crannies using a high-temperature aluminum epoxy compound. It's held up for 12 years now, and I don't have to fuss with it each time I replace the spark plugs.
'81 Canadian Westfalia (2.0L, manual), now Californiated

"They say a little knowledge is a dangerous thing, but it is not one half so bad as a lot of ignorance."
- Terry Pratchett

DugB
I'm New!
Status: Offline

Re: AFM Adjustment

Post by DugB » Fri Jun 15, 2018 5:37 am

I love the idea of embedding a sensor wire rather than using the ring terminal. The latter makes plug checks/changed such a pain. Do you feel like the temperature is accurately conveyed through the epoxy? Are you using Lab Metal or something like that?

Thanks!

Blue Moose
I'm New!
Status: Offline

Re: AFM Adjustment

Post by Blue Moose » Fri Jun 15, 2018 5:09 pm

I appreciate both of those recommendations! Do you recall what thermocouples you used?

User avatar
SlowLane
IAC Addict!
Location: Livermore, CA
Status: Offline

Re: AFM Adjustment

Post by SlowLane » Fri Jun 15, 2018 6:56 pm

DugB wrote:
Fri Jun 15, 2018 5:37 am
I love the idea of embedding a sensor wire rather than using the ring terminal. The latter makes plug checks/changed such a pain. Do you feel like the temperature is accurately conveyed through the epoxy? Are you using Lab Metal or something like that?
I used a product called Pyro Putty 1000 manufactured by Aremco Products
. I bought it from McMaster-Carr or Grainger or some outfit like that, but I see it can now be ordered from Amazon.

While some may claim that the ring terminal is the only place to get accurate measurements, I was able to diagnose a pre-ignition problem I was having right after rebuilding the engine with this setup. I would see temperature spikes on my gauge (it shows a bar-graph) that lasted for less than a second. That's a good enough result for me. I've entertained the notion of installing a ring terminal on #3 cylinder to compare the reading there with my chosen location, but just haven't been motivated to go through the hassle,
'81 Canadian Westfalia (2.0L, manual), now Californiated

"They say a little knowledge is a dangerous thing, but it is not one half so bad as a lot of ignorance."
- Terry Pratchett

User avatar
SlowLane
IAC Addict!
Location: Livermore, CA
Status: Offline

Re: AFM Adjustment

Post by SlowLane » Sat Jun 16, 2018 12:11 am

Blue Moose wrote:
Fri Jun 15, 2018 5:09 pm
I appreciate both of those recommendations! Do you recall what thermocouples you used?
I used Omega's WT series K-type thermocouples bonded to the heads, connected with K-type miniature connectors to 4-pair multi-stranded extension wire running to the gauge.

Oh, and if you do decide to roll your own CHT gauge setup, be aware that in thermocouple world, the red wire is the negative wire. Quite important to remember when wiring up the connector assemblies
'81 Canadian Westfalia (2.0L, manual), now Californiated

"They say a little knowledge is a dangerous thing, but it is not one half so bad as a lot of ignorance."
- Terry Pratchett

72Hardtop
Old School!
Location: Seattle, WA./HB. Ca./Shizuoka, Japan
Status: Offline

Re: AFM Adjustment

Post by 72Hardtop » Tue Oct 23, 2018 6:42 pm

DugB wrote:
Tue Jun 12, 2018 11:17 am
From Turbo by Garret:
“So what is meant by a rich or lean AFR? A lower AFR number contains less air than the 14.7:1 stoichiometric AFR, therefore it is a richer mixture. Conversely, a higher AFR number contains more air and therefore it is a leaner mixture. Leaner AFR results in higher temperatures as the mixture is combusted.”
Only to a certain point...The hottest AFR area is in the range of 14:1 - 15:1 (WOT/under load).

Everything lean of 15:1 is cooler running (part throttle condition). The goal is to NOT be under load when lean cruise tuning. Your foot dictates load.

16-17:1 during cruise (part throttle cond) is cooler than 12.5 -13.5 at cruise (part throttle).

In a bus keep the mechanical advance at 28 degrees full (hose off) ~3200+rpm's
1972 Westy tintop
2056cc T-4 - 7.8:1 CR
Weber 40mm Duals - 47.5idles, 125mains, F11 tubes, 190 Air corr., 28mm Vents
96mm AA Biral P/C's w/Hastings rings
42x36mm Heads (AMC- Headflow Masters) w/Porsche swivel adjusters
71mm Stroke
Web Cam 73 w/matched Web lifters
S&S 4-1 exhaust w/Walker 17862 quiet-pack
Pertronix SVDA w/Pertronix module & Flamethrower 40K coil (7* initial 28* total @3200+)
NGK BP6ET plugs
002 3 rib trans
Hankook 185R14's

User avatar
MountainPrana
Getting Hooked!
Location: National Elk Refuge, Jackson WY
Status: Offline

Re: AFM Adjustment

Post by MountainPrana » Fri May 17, 2019 4:49 pm

Amskeptic wrote:
Wed Apr 13, 2016 9:33 am
Factory economy was 16.7 mpg mixed driving.

You can carefully get a cruise ratio closer to high 12 low 13 area, but the new benchmark is to perform a full throttle pull in 4th gear from 55 mph to 70 mph.
Set wiper to 11.7 at 55 and see if it leans to 12.3 at 70.
Check to see if the fuel cuts off at total decel from 70 down, you should see a 22 or so that "switches" back down to 15-16 or less when rpms drop below threshold.

Sure wish my LM-1 had not been stolen out from under me . . .
Colin
Colin, Is the 11.7 set with the full throttle enrichment switch hooked up or disconnected? Thanks for this wonderful article and the updates that you have added along the way. Since I'm running that 9590 cam with 8.0:1 compression would these numbers still be applicable?

Hope all is well for you!

Tim

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Re: AFM Adjustment

Post by Amskeptic » Fri May 17, 2019 8:27 pm

MountainPrana wrote:
Fri May 17, 2019 4:49 pm
Amskeptic wrote:
Wed Apr 13, 2016 9:33 am
Factory economy was 16.7 mpg mixed driving.

You can carefully get a cruise ratio closer to high 12 low 13 area, but the new benchmark is to perform a full throttle pull in 4th gear from 55 mph to 70 mph.
Set wiper to 11.7 at 55 and see if it leans to 12.3 at 70.
Check to see if the fuel cuts off at total decel from 70 down, you should see a 22 or so that "switches" back down to 15-16 or less when rpms drop below threshold.

Sure wish my LM-1 had not been stolen out from under me . . .
Colin
Colin, Is the 11.7 set with the full throttle enrichment switch hooked up or disconnected? Thanks for this wonderful article and the updates that you have added along the way. Since I'm running that 9590 cam with 8.0:1 compression would these numbers still be applicable?

Hope all is well for you!

Tim

Hooked up!
Colin
BobD - 78 Bus . . . 112,660 miles
Chloe - 70 bus . . . 206,845 miles
Naranja - 77 Westy . . . 116,898 miles
Pluck - 1973 Squareback . . . . . . 55,570 miles
Alexus - 91 Lexus LS400 . . . 84,465 miles

User avatar
Johnooot
I'm New!
Location: Toronto, Canada
Status: Offline

Re: AFM Adjustment

Post by Johnooot » Mon Jul 29, 2019 7:28 am

Thanks for this adjustment procedure Colin.

Is the procedure valid for an '82 Vanagon with California emissions? The reason I ask is that with California emissions I wonder what role the O2 sensor will have when making the adjustments. Can the procedure be followed as written or will there be a longer lag time in each adjustment as the O2 sensor stabilizes the idle / readings? Anything else to consider while maing AFM adjustments?

While on the topic of a California emissions setup, does this setup typically runner higher CHT then the Federal setup?

John

User avatar
SlowLane
IAC Addict!
Location: Livermore, CA
Status: Offline

Re: AFM Adjustment

Post by SlowLane » Tue Jul 30, 2019 12:56 pm

So, a California-emission Vanagon in Toronto? Wow, lucky you...
Johnooot wrote:
Mon Jul 29, 2019 7:28 am
Is the procedure valid for an '82 Vanagon with California emissions? The reason I ask is that with California emissions I wonder what role the O2 sensor will have when making the adjustments. Can the procedure be followed as written or will there be a longer lag time in each adjustment as the O2 sensor stabilizes the idle / readings? Anything else to consider while maing AFM adjustments?
The O2-enabled lambda feedback loop will walk all over any attempt you make to fiddle the AFM on a California-emission engine. It is continually adjusting the mixture about once per second to keep it shuttling back and forth either side of the 14.7:1 stoichiometric ratio. Any adjustment you make to the AFM to get a richer or leaner mixture will be compensated out by the feedback loop, until your attempts at adjustment drive it so far out of whack that the system cannot compensate. At which point you are up the proverbial creek without a paddle, unless you diligently kept track of every adjustment made so you could put it back to square one.

The entire notion of fiddling the mixture to achieve some sort of rich or lean running condition is obviated by having a feedback loop. There is only one reason that I can think of for making any adjustments to an AFM on a lambda-equipped system, and that is to get it "in the ballpark" while running open loop (ie. O2 sensor disconnected), so that the feedback system doesn't have to work too hard to achieve its programmed rich-lean-rich-lean dance around 14.7:1.
While on the topic of a California emissions setup, does this setup typically runner higher CHT then the Federal setup?
Having run both setups on the same vehicle, I've not noticed any appreciable difference. Engine load (headwinds, hills, leadfoot) has far more impact on CHT.
'81 Canadian Westfalia (2.0L, manual), now Californiated

"They say a little knowledge is a dangerous thing, but it is not one half so bad as a lot of ignorance."
- Terry Pratchett

User avatar
Johnooot
I'm New!
Location: Toronto, Canada
Status: Offline

Re: AFM Adjustment

Post by Johnooot » Tue Jul 30, 2019 7:06 pm

Ahhh. So that might explain why I am seeing some strange things.

1) I've had an Innovate MTX-L AF gauge for about a year and always wondered why it's bouncing all over the place while cruising at highway speeds. (https://photos.app.goo.gl/N8otdfK9sCefLATT7 ignore the digits due to camera frame rate. Just see the segments dance all over the place). When the engine is stone cold the AF gauge is pretty much solid at 8 to 10 as I would expect. Once it warms up the AF gauge starts its bouncing dance! [The AF gauge sensor is located in the test port on the engine side of the CAT .. which is an empty CAT. I’ve had a smoke test done to look for vacuum leaks. Replaced the throttle body to get rid of the major one. No change. Only a “very small” vacuum leak where the oil-fill tube transitions from metal to plastic.

2) Recently added a Dakota Digital CHT gauge: sensor under spark plug #3. If I am under 55 MPH the temps stay under 400F. Any attempt to get to 55 or over and it quickly climbs to 410 and 420. Downhill or a good tailwind and it stays under 400 more or less.

3) I've read that with California-emissions, that if you disconnect the O2 sensor, the ECU will go into "limp-mode" and force the system into a rich condition. So on a whim I disconnected the O2 sensor to see what the CHT would register. Well, the AF gauge immediately pinned to 22.7 (which is the max) and pretty much stayed there regardless of speed. (with the odd quick fluctuation to 17ish). However the CHT now improved a fair bit. More or less I could cruise delicately at 60 - 65 and keep the temps in the 395 to 405 range - for sure an improvement. This does not make sense. The CHT should have risen a fair bit?

4) Gas mileage through a number of tankfulls has not changed if the O2 sensor is connected or not – either side of 20 MPG (US) over 2,800 mile trip I did this month. Starts fine. Idles well after start. Power is normal. Hydraulic valves checked this spring. Only other anomaly is that I have a fast idle when fully warmed up. However Idle is normal when warm if I disconnect the O2 sensor. Weird!

QUESTION#1: Assuming that mechanical and vacuum advance are OK (still needs to be checked) do these symptoms suggest that I need to adjust the (A) STATIC and maybe (B) DYNAMIC AFM adjustments to get it close in open-mode?

QUESTION#2: In a lambda system, would the modification of re-routing the TEMP2 sensor through a potentiometer have the desired effect of making the system rich? In other words, does TEMP2 override the O2 sensor in closed-loop mode?

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Re: AFM Adjustment

Post by Amskeptic » Wed Jul 31, 2019 3:49 pm

Johnooot wrote:
Tue Jul 30, 2019 7:06 pm
Ahhh. So that might explain why I am seeing some strange things.

1) I've had an Innovate MTX-L AF gauge for about a year and always wondered why it's bouncing all over the place while cruising at highway speeds. (https://photos.app.goo.gl/N8otdfK9sCefLATT7 ignore the digits due to camera frame rate. Just see the segments dance all over the place). When the engine is stone cold the AF gauge is pretty much solid at 8 to 10 as I would expect. Once it warms up the AF gauge starts its bouncing dance! [The AF gauge sensor is located in the test port on the engine side of the CAT .. which is an empty CAT. I’ve had a smoke test done to look for vacuum leaks. Replaced the throttle body to get rid of the major one. No change. Only a “very small” vacuum leak where the oil-fill tube transitions from metal to plastic.

2) Recently added a Dakota Digital CHT gauge: sensor under spark plug #3. If I am under 55 MPH the temps stay under 400F. Any attempt to get to 55 or over and it quickly climbs to 410 and 420. Downhill or a good tailwind and it stays under 400 more or less.

3) I've read that with California-emissions, that if you disconnect the O2 sensor, the ECU will go into "limp-mode" and force the system into a rich condition. So on a whim I disconnected the O2 sensor to see what the CHT would register. Well, the AF gauge immediately pinned to 22.7 (which is the max) and pretty much stayed there regardless of speed. (with the odd quick fluctuation to 17ish). However the CHT now improved a fair bit. More or less I could cruise delicately at 60 - 65 and keep the temps in the 395 to 405 range - for sure an improvement. This does not make sense. The CHT should have risen a fair bit?

4) Gas mileage through a number of tankfulls has not changed if the O2 sensor is connected or not – either side of 20 MPG (US) over 2,800 mile trip I did this month. Starts fine. Idles well after start. Power is normal. Hydraulic valves checked this spring. Only other anomaly is that I have a fast idle when fully warmed up. However Idle is normal when warm if I disconnect the O2 sensor. Weird!

QUESTION#1: Assuming that mechanical and vacuum advance are OK (still needs to be checked) do these symptoms suggest that I need to adjust the (A) STATIC and maybe (B) DYNAMIC AFM adjustments to get it close in open-mode?

QUESTION#2: In a lambda system, would the modification of re-routing the TEMP2 sensor through a potentiometer have the desired effect of making the system rich? In other words, does TEMP2 override the O2 sensor in closed-loop mode?

Any oscillations you experience are the ECU trying to chase the mixture around while the sensor comes in late and tries to compensate for the ECU's trying to compensate for the sensor's trying to compensate the ECU's trying to compensate for the sensor trying to compens ....

DISCONNECT THE SENSOR.

Adjust the AFM according to the write-up.

Use the fuel pressure regulator/decel valve hose nipple on the intake plenum to check your work. You want the slight "finger-off" vacuum leak to result in a modest drop in rpm.. Use the mixture screw to trim.

Adjust idle speed to 1,000 rpm.

Now plug the sensor back in so your engine can enjoy lean hell AND stable operation.
Colin
BobD - 78 Bus . . . 112,660 miles
Chloe - 70 bus . . . 206,845 miles
Naranja - 77 Westy . . . 116,898 miles
Pluck - 1973 Squareback . . . . . . 55,570 miles
Alexus - 91 Lexus LS400 . . . 84,465 miles

User avatar
SlowLane
IAC Addict!
Location: Livermore, CA
Status: Offline

Re: AFM Adjustment

Post by SlowLane » Thu Aug 01, 2019 9:02 am

WAIT! STOP! Don't do anything to your AFM until you've made sure that the rest of your engine's systems are functioning correctly. The AFM is tempting to fiddle with because it makes one feel masterful, but it really should be the last thing you consider touching. They are sealed from the factory for a reason.

By your own post you don't know if the distributor advance/retard are working within spec. Get a Mity-Vac with gauge and figure out if they are. Clean and lube your distributor's mechanical advance if it seems the least bit sticky.

Check the operation of your deceleration valve with the Mity-Vac. On a CA-emission vehicle it should open at 15-17" Hg.

Do you have an EGR valve? Is it operating correctly?

You've already indicated that your van runs very well, so why would you want to compromise that by messing about with the AFM, simply because the AFR gauge is doing weird shit? I stongly suspect that you have an exhaust leak somewhere that is diluting the exhaust with fresh air and causing your AFR gauge's sensor to read an overly lean condition. Do you have two O2 sensors, one for the gauge and the original one for the ECU?

Narrow-band oxygen sensors and three-way catalytic converters are complementary technologies: having one without the other doesn't really make much sense. Some recommended reading: https://www.arb.ca.gov/research/seminar ... mooney.pdf
'81 Canadian Westfalia (2.0L, manual), now Californiated

"They say a little knowledge is a dangerous thing, but it is not one half so bad as a lot of ignorance."
- Terry Pratchett

Post Reply