Testing 94 VAF meter - both off and on the car (1 Viewer)

[petegangies]

Hi all, I've got 4x VAF meters here, and I'm trying to test each of them to see if they are any good (all 2nd hand).

The problem is, I've found there are inconsistencies in the FSM with regards to which pins are what on the connector.

The first test I did, I believe tests the resistence and the IAT probe is functioning correctly. I'm doing this OFF THE VEHICLE. I understand that I still have to do some on the vehicle tests while manouvering the flap (measuring plate?). Are there any other tests I can do with this off the vehicle? The car will not start with this particular VAF fitted.

How can I definitevely test whether a VAF is good or bad?

My readings:
VS-E2 = 302.6KΩ
THA-E2 = 1.888KΩ at 27.3°C (and 1.3-1.7KΩ after hitting it with the hair dryer for a few seconds from about 10cm away, and temperature climbs slowly when sitting an icepack on the VAF). So can I safely assume the IAT is working correctly?
VC-E2 the readings are not super stable. During one test I got 285.7 to 286.2 and the second test I got 288.6 to 288.9.

Here are all the different references from the 1994 FSM pertaining to the VAF tests. See how the reference to pins changes between sections?

Why on some of the diagrams do the pin IDs change?

Are these the correct pin IDs?

 

[petegangies]

VS to E2


THA to E2

 

[petegangies]

VC to E2

 

[petegangies]

Current best price to have VAF "reconditioned" is $850 AUD. Though, neither of the 2 places will say exactly how they refurbish them. I'm curious as to whether the circuit board is somehow replaced, because I don't see the issue being fixed if a new resistor strip is not installed. The 2 main working parts are the intake temp sensor (IAT) and the resisitor strip that the potentiometer with contact runs along. That's where the majority of issues come from. Over time the contact wears down the carbon strip, creating grooves and wearing of the material, then you get drops in resistance in certain spots and there are no reliable signals back to the ECM at those points. The ECM will then default to a richer safer map for those points and what you end up with is poor performance, roughness, and bad economy. $850 is a lot of money if they are simply cleaning these inside and out and hoping they work better. I don't understand the secrecy in telling you what you're paying for, unless of course they aren't really refurbishing it to "like new" as claimed.

closeup photo of internals for those wondering..

 

[Malleus]

Hi all, I've got 4x VAF meters here, and I'm trying to test each of them to see if they are any good (all 2nd hand).

The problem is, I've found there are inconsistencies in the FSM with regards to which pins are what on the connector.

The first test I did, I believe tests the resistence and the IAT probe is functioning correctly. I'm doing this OFF THE VEHICLE. I understand that I still have to do some on the vehicle tests while manouvering the flap (measuring plate?). Are there any other tests I can do with this off the vehicle? The car will not start with this particular VAF fitted.

How can I definitevely test whether a VAF is good or bad?

My readings:
VS-E2 = 302.6KΩ
THA-E2 = 1.888KΩ at 27.3°C (and 1.3-1.7KΩ after hitting it with the hair dryer for a few seconds from about 10cm away, and temperature climbs slowly when sitting an icepack on the VAF). So can I safely assume the IAT is working correctly?
VC-E2 the readings are not super stable. During one test I got 285.7 to 286.2 and the second test I got 288.6 to 288.9.

View attachment 3846868

Here are all the different references from the 1994 FSM pertaining to the VAF tests. See how the reference to pins changes between sections?

View attachment 3846870

View attachment 3846871

View attachment 3846872

Why on some of the diagrams do the pin IDs change?

Are these the correct pin IDs?
View attachment 3846873

Why on some of the diagrams do the pin IDs change?

Because electrical stuff sucks. No other reason.

 

[Malleus]

Hi all, I've got 4x VAF meters here, and I'm trying to test each of them to see if they are any good (all 2nd hand).

The problem is, I've found there are inconsistencies in the FSM with regards to which pins are what on the connector.

The first test I did, I believe tests the resistence and the IAT probe is functioning correctly. I'm doing this OFF THE VEHICLE. I understand that I still have to do some on the vehicle tests while manouvering the flap (measuring plate?). Are there any other tests I can do with this off the vehicle? The car will not start with this particular VAF fitted.

How can I definitevely test whether a VAF is good or bad?

My readings:
VS-E2 = 302.6KΩ
THA-E2 = 1.888KΩ at 27.3°C (and 1.3-1.7KΩ after hitting it with the hair dryer for a few seconds from about 10cm away, and temperature climbs slowly when sitting an icepack on the VAF). So can I safely assume the IAT is working correctly?
VC-E2 the readings are not super stable. During one test I got 285.7 to 286.2 and the second test I got 288.6 to 288.9.

View attachment 3846868

Here are all the different references from the 1994 FSM pertaining to the VAF tests. See how the reference to pins changes between sections?

View attachment 3846870

View attachment 3846871

View attachment 3846872

Why on some of the diagrams do the pin IDs change?

Are these the correct pin IDs?
View attachment 3846873

Pete,

You can check individual components on the board, but there's nothing else that I know of that would be considered a "service test", and these are vendor level component checks, anyway.

Short of rigging up some homemade VAF/science project, and using that to test the operating conditions, I think you've done all there is to do.

 

[Malleus]

Keep at it though, you're on your way to becoming the world wide expert! Success just takes perseverance...and lots of luck.

 

[petegangies]

Pete,

You can check individual components on the board, but there's nothing else that I know of that would be considered a "service test", and these are vendor level component checks, anyway.

Short of rigging up some homemade VAF/science project, and using that to test the operating conditions, I think you've done all there is to do.

I agree.. my only hesitation is that I have 3 units that will run the vehicle with no CEL, yet all 3 are exhibiting the same issue with resistance drops.

Before I go forking out big dollars for a refurb, there's one more test I was suggested to try based on this response I received from an electronics guy on youtube:

"Hard to say for sure... keep in mind that there are two different styles of AFM used by Toyota. Download this PDF and check out page 20 to start with - https://www.autonerdz.com/yabbfiles/Attachments/TCCS__TOYOTA_COMPUTER_CONTROLLED_SYSTEM_.pdf - the VS and E2 pins do go to the potentiometer and thus you'd expect to to see clean numbers on your digital multimeter. However, with that said, keep in mind that the AFM circuit in you are dealing with is designed to function with a voltage input present (either from the ECU or directly from the car battery itself - in the case of the later, the ECU makes compensation to take into account voltage level fluctuations in the battery, whereas when voltage is supplied via the ECU, it does not). As such, the potentiometer is designed to work in concert with voltage, not strictly through resistance only. My advice would be to power the circuit up (as it would be in the vehicle during operation) and compare your voltage level output seen on VS (during the sweep) to the chart on page 20 - ideally with a digital Oscilloscope if you have one available to you. A voltage meter (even an analog one) would also be fine as well. I think so long as you do not see voltage spikes up or down, then I wouldn't worry too much about the specific resistance readings directly from the Potentiometer itself. The ECU is reading voltage, not resistance to calculate air volume going into the engine, so I would test voltage levels, not resistance levels (at least to start). It's possible that when voltage is running through the circuit, then you could see that output on the VS pin is fine. It's even possible that there is a capacitor or something on the ECU side which is acting as some type of filter, although I don't know for sure. So, I would set out to test voltage, not resistance on those AFM's and see what you can find going with that angle. The other option would be to test a known working AFM and see how the Potentiometer responds when isolated. All 3 being bad seems a tiny bit suspicious to me, but then again we are dealing with very old parts too."

So the next step is to test while powered on the vehicle.


E2 - Ground (Earth)
VC - 5V Power from ECU
VS - Airflow Signal Output
THA - Intake Air Temp Signal
E1 - Sensor Ground

Basic Voltage Check (Using a Multimeter) Set your multimeter to DC voltage mode (10V or 20V range).
Connect your ground probe to E2 (AFM Ground).
Connect your positive probe to VS (Signal Output).
Move the AFM flapper door slowly while watching the voltage reading.

Flapper Position	Expected VS Voltage
Fully Closed (Idle)	~0.2V - 0.5V
Partially Open (Light Throttle)	~1.5V - 2.5V
Fully Open (Full Throttle)	~4.5V - 5.0V

If VS voltage drops to 0V suddenly at certain points, the resistor strip may be damaged.
If voltage smoothly increases as the flap opens, the AFM might be OK.

Others have said, basically if you get the drops on basic resistance test, that's evidence enough, as they have tested other units that have no drops. Also hoping my multimeter is functions as it should (can't see why it wouldn't be).

 

[Malleus]

I wasn't thinking about on vehicle tests...all good advice about further testing. But I'm an ME, not an EE, so take that for what it's worth

I would seriously beg or borrow a DSO...it'll be a big help.

 

[petegangies]

I've re-tested the resistance after a friend of mine suggested the OL/drops could possibly be coming from the meter if it changes ranges. So was told to set it to one fixed range, either 2k or 20k and see if they still happen. So that's what I did (I think), and this is the result:



Seems now figures are stable and no drops to zero. One final voltage test with the unit hooked up to the vehicle and if that passes, then we're all good with this particular air flow meter, and I can cross that off as being the issue.

 

[jpoole]

Nice work digging into your AFM, I'm pretty sure you'll be able to figure out how to reliably test and service these with a little more digging.

Since I'm not sure how common the following ideas are I'll share a few adjustments that I've had success with on similar, though earlier, AFMs in the past.

One change that I've made on multiple AFMs (early 80s Bosch / VW) is to adjust the contact that's on the end of the arm so that it runs in a new track/position on the board. That along with a little cleaning is often as good as replacing the board w a new one. I can't tell what your options are for making that adjustment on the VAF above, as it's slightly different from the ones I've adjusted, but you can likely figure it out or share more photos for ideas. You want to tread lightly because you don't want to change the pressure of the contact on the board or the way that the contact rides on the board, only where it rides on the board.

The section of the track that I can see looks to be in good condition but it's possible that there is relevant wear hidden by the arm. You can move the arm to fully open by hand and take another photo of the track if you opinions on it's condition based on visuals.

The following changes likely aren't relevant in your case but I share them in case they end up being useful as you try to understand the AFM and learn about how it was likely setup when new to get it into spec before installation.

You can change the spring tension (big coil spring under the arm) by moving the black wheel a few index points one way or another. This adjustment changes the fuel curve and can allow you to increase or decrease fuel at lower rpms or upper rpms independently. This can be useful to add or remove fuel at any point in the acceleration curve to dial in air fuel ratios at all points of the rpm range. I recommend doing this only if you have a wide-band air fuel ratio to see exactly what your changes are doing and if you understand how this is going to work with the computer / fuel injection system on the vehicle in question, etc. Some systems allow you to run based solely on AFM input, without O2 sensor feedback compensating for adjustments, which can allow you to dial adjustments like this in under load (driving), etc. which, along with dialing in timing can be very productive on older / early fuel injected engines with more primitive ignition and control systems.

Another change that you can make is moving the position of the head in relation to the spring by loosening the allen bolt and then moving the head independent of the spring tensioned base. This richens or leans out all points on the curve equally vs the independent curve change you can achieve with the spring tension adjustment. If you mess with spring tension you often then need to also move the whole curve up or down by adjusting arm position. Again, don't mess with this unless you have the right tools and know what you are trying to accomplish. If you move this head you also typicall have to change the arm that triggers the park switch so that it turns on/off at the right time. The park switch in many systems will trigger the fuel pump and can be a useful part of simulating operating conditions with an FI system so that you can test fuel rail pressure, etc. though I don't know if that works with the odb1 1fzfe systems.

Good luck

 

[petegangies]

Nice work digging into your AFM, I'm pretty sure you'll be able to figure out how to reliably test and service these with a little more digging.

Since I'm not sure how common the following ideas are I'll share a few adjustments that I've had success with on similar, though earlier, AFMs in the past.

One change that I've made on multiple AFMs (early 80s Bosch / VW) is to adjust the contact that's on the end of the arm so that it runs in a new track/position on the board. That along with a little cleaning is often as good as replacing the board w a new one. I can't tell what your options are for making that adjustment on the VAF above, as it's slightly different from the ones I've adjusted, but you can likely figure it out or share more photos for ideas. You want to tread lightly because you don't want to change the pressure of the contact on the board or the way that the contact rides on the board, only where it rides on the board.

The section of the track that I can see looks to be in good condition but it's possible that there is relevant wear hidden by the arm. You can move the arm to fully open by hand and take another photo of the track if you opinions on it's condition based on visuals.

The following changes likely aren't relevant in your case but I share them in case they end up being useful as you try to understand the AFM and learn about how it was likely setup when new to get it into spec before installation.

You can change the spring tension (big coil spring under the arm) by moving the black wheel a few index points one way or another. This adjustment changes the fuel curve and can allow you to increase or decrease fuel at lower rpms or upper rpms independently. This can be useful to add or remove fuel at any point in the acceleration curve to dial in air fuel ratios at all points of the rpm range. I recommend doing this only if you have a wide-band air fuel ratio to see exactly what your changes are doing and if you understand how this is going to work with the computer / fuel injection system on the vehicle in question, etc. Some systems allow you to run based solely on AFM input, without O2 sensor feedback compensating for adjustments, which can allow you to dial adjustments like this in under load (driving), etc. which, along with dialing in timing can be very productive on older / early fuel injected engines with more primitive ignition and control systems.

Another change that you can make is moving the position of the head in relation to the spring by loosening the allen bolt and then moving the head independent of the spring tensioned base. This richens or leans out all points on the curve equally vs the independent curve change you can achieve with the spring tension adjustment. If you mess with spring tension you often then need to also move the whole curve up or down by adjusting arm position. Again, don't mess with this unless you have the right tools and know what you are trying to accomplish. If you move this head you also typicall have to change the arm that triggers the park switch so that it turns on/off at the right time. The park switch in many systems will trigger the fuel pump and can be a useful part of simulating operating conditions with an FI system so that you can test fuel rail pressure, etc. though I don't know if that works with the odb1 1fzfe systems.

Good luck

Unfortunately I've already undone the allen key bolt and swapped over the pentionometers between 2 of my units. One of the units was broken from dad trying to undo the 2 screws on the plug and tugging.. and the other unit was apparently a "refurbished" one from 2016.

I quickly discovered that the contact now does not sit in the original position (I dont even know what the exact original position is), but have tried to fiddle around with holding the pentionometer where I think the arm should sit and then tightening at the same time to stop it moving. I've also noticed the little contact (which is a bunch of tiny metal bristles like a broom), are also separating in certain positions as the contact sweeps along the resistor strip.. so either the strip has some worn spots on it, or I've accidentally bent the contact arm down. I tried playing with that gently too. Since I have no idea whether I have done this correctly or stuffed up the original settings (likely), I will opt not to use that specific air flow meter on the vehicle. The one I'm currently testing at the moment has never been opened and is a second hand Denso unit.

The close up image of the unit above is not actually mine, it was one I found online. These are photos of one of mine open:

 

[jpoole]

There's probably a spec for how far the "park indicating" switch should open when the afm is at rest but I'd guess you would be good achieving a minimal open gap when parked so that as soon as the vane opens at all that switch closes and the fuel pump can engage, etc. You could likely check the angle that the unopened unit closes that switch at and then match it on the unit that has been opened. I've seen afms so far out of adjustment that the park switch would never open, etc. and they still worked but obviously the tune was off or the ecu was having to compensate a lot.

I'd assume that any of your afms can be fixed and put back into service. You are in a nice situation of having likely good / unmodified example(s) to use for comparison if you choose to setup the other units.

 

[skw6r]

Hi all, I've got 4x VAF meters here, and I'm trying to test each of them to see if they are any good (all 2nd hand).

The problem is, I've found there are inconsistencies in the FSM with regards to which pins are what on the connector.

The first test I did, I believe tests the resistence and the IAT probe is functioning correctly. I'm doing this OFF THE VEHICLE. I understand that I still have to do some on the vehicle tests while manouvering the flap (measuring plate?). Are there any other tests I can do with this off the vehicle? The car will not start with this particular VAF fitted.

How can I definitevely test whether a VAF is good or bad?

My readings:
VS-E2 = 302.6KΩ
THA-E2 = 1.888KΩ at 27.3°C (and 1.3-1.7KΩ after hitting it with the hair dryer for a few seconds from about 10cm away, and temperature climbs slowly when sitting an icepack on the VAF). So can I safely assume the IAT is working correctly?
VC-E2 the readings are not super stable. During one test I got 285.7 to 286.2 and the second test I got 288.6 to 288.9.

View attachment 3846868

Here are all the different references from the 1994 FSM pertaining to the VAF tests. See how the reference to pins changes between sections?

View attachment 3846870

View attachment 3846871

View attachment 3846872

Why on some of the diagrams do the pin IDs change?

Are these the correct pin IDs?
View attachment 3846873

Were you able to confirm through your testing that these are the correct pins?

 

[petegangies]

Were you able to confirm through your testing that these are the correct pins?

yes, pins are correct. My meters are passing the off-vehicle tests. Have not tested on the vehicle though.

 

[skw6r]

My meter is passing off vehicle test except I have open loops and resistance drops when sweeping. I'll need to get a multimeter that supports manual range selection to see if it's the multimeter giving false readings. If the meter checks out, I'll proceed to on vehicle testing. Definitely curious to see your results. Dealing with CEL 24, 25, 26 and running very rich. Local yota shop believes it lies with IAT reading somewhere between VAF, harness, and ECM.

VS - E2 2.63kΩ
VC - E2 278Ω
THA - E2 2.64kΩ

 

[OrangeCrusher]

I spent a lot of time trying to fix the VAF in my rig as well. I had two examples of the China knock off version, and never could get the engine run completely right with it across the whole rpm and load range. There was always something that was either too rich or too lean.

Eventually I decided - screw that old junk! I ordered a very common MAF for 2000s Toyotas (Corollas etc, was able to get an aftermarket MAF for $20), and mounted it in the VAF housing. I glued the flap into the fully open positon to not have it interfer.
I know - questionable mounting position for the MAF sensor given these need a undisturbed airflow to work well - but given how crappy the currently remaining VAF options are, I had little to lose.

I then hooked an Arduino microcontroller to it and started tuning the voltage conversion.

I got the initial values by hooking the MAF-fi-fied-VAF in series with the existing VAF, and measured a few points. After that I spent a few hours driving around and tuning the conversion formula and signal smoothing function. At idle, the MAF signal fluctuates quite a bit. I don't know if that is normal or caused by the suboptimal mounting position in the VAF housing. However I was able to get that under control with some signal post processing magic.

I have driven it about a 1000 miles now under different conditions - everything from city traffic, 80mph interstate, to Golden Spike in Moab. For the first time since I own this rig, the engine is actually running right under all conditions! Such a joy to drive it now, being able to rev all the way up to 5k and having smooth power without stumbles

Can share more details if people are interested. I plan to clean up the wiring in the next few weeks once I'm 100% confident it's all working correctly. I hope I can make it look basically stock - the microcontroller should fit into the upper part of the VAF housing, and the MAF sitting on the bottom of the VAF housing is not very visible.

 

[jpoole]

@OrangeCrusher I'm impressed and think that what you've come up is a valuable alternative to the VAF fix/replace cycle that folks get stuck in as early 80 owners. I'm sure that a separate thread with a more detailed write-up would be appreciated by the DIY crowd but it seems like you could also have a marketable product on your hands should you be up for creating a few more of these and offering them at a price that makes that work appealing. Nice work!

 

[Bigtoepfer]

When I first got my 80 my VAF was supposedly bad. Bought a $100 amazon option, and managed to get it home from the shop that had done my headgasket. Still had a lean CEL occasionally.

Took apart my factory VAF and cleaned it, put it back on and things were better. Still occasionally a CEL for lean code. Swapped O2 sensors and no more code.

I still have the China/Amazon VAF for emergencies, but the basic test off the vehicle reads pretty much the same as the factory one. So I'll just pray it keeps working for a long long time. I like this idea of making a MAF with an arduino though and I'd subscribe to that post for future options.

I did find most of my stumble was caused by a vacuum leak at the brake booster though. Eventually changed it and fixed the vacuum issue. The 1FZ doesn't run perfect like a brand new motor, but doesn't give me any issues either for something with 248k miles on it. And I was getting 14mpg with 35s at 70mph on the high way so I can't really complain too much.

 

[skw6r]

When I first got my 80 my VAF was supposedly bad. Bought a $100 amazon option, and managed to get it home from the shop that had done my headgasket. Still had a lean CEL occasionally.

Took apart my factory VAF and cleaned it, put it back on and things were better. Still occasionally a CEL for lean code. Swapped O2 sensors and no more code.

I still have the China/Amazon VAF for emergencies, but the basic test off the vehicle reads pretty much the same as the factory one. So I'll just pray it keeps working for a long long time. I like this idea of making a MAF with an arduino though and I'd subscribe to that post for future options.

I did find most of my stumble was caused by a vacuum leak at the brake booster though. Eventually changed it and fixed the vacuum issue. The 1FZ doesn't run perfect like a brand new motor, but doesn't give me any issues either for something with 248k miles on it. And I was getting 14mpg with 35s at 70mph on the high way so I can't really complain too much.

I grenaded my first VAF after removing those stupid screws. I got a junk yard one and my once occasional CELs are now consistent. Bench tested the VAF and everything is within spec according to the FSM. Next step will be testing the harness connectivity and then ECM readings. I’ve also got new vacuum and PCV lines on the way. Will replace those while I’m at it.

I’m getting 8 mpg on 35s at altitude.