Anatomy of a Bosch MAF (long but lots of pictures)
04-02-2003, 12:57 PM
#1
Junior Member
Thread Starter
Join Date: Apr 2001
Posts: 1,503
Likes: 0
Received 0 Likes
on
0 Posts
Anatomy of a Bosch MAF (long but lots of pictures)
Finally got around to dissecting my bad Bosch MAF last night in an attempt to figure out why these things are going bad. Snapped some pics along the way and did find some interesting things out.
As a little history, the MAF was deemed bad due to VAG readings of a little more than 200 g/s. Low for a chipped or stock car. I tried cleaning but no change and decided to get the new MAF. VAG showed marked improvement and even measuring a few months later showed similar if not slightly better results. Haven't logged in a month or so, but last time was getting 250+ g/s typically.
<img src="http://pictureposter.audiworld.com/18887/current_maf.jpg">
As most may know, the sensor portion of the MAF is held in place in the housing by two security torx screws and has a tight fitting gasket. The sensor has no serviceable features, so you need to destroy it to take it apart - something I am very good at.
The best place to start is the little side panel that is held in place by some brown glue that is more rubbery than solid, so a dremel really doesn't help much - it just gets clogged up. I ended up using a x-acto knife and digging it out.
<img src="http://pictureposter.audiworld.com/18887/dscf0461.jpg">
With enough of the brown stuff out, you can pry the cover off and viola - the guts of the MAF.
<img src="http://pictureposter.audiworld.com/18887/dscf0469.jpg">
Here's where it gets a little interesting. The circuit card is covered in a semi-clear "goo" that is rather soft and almost seems like it was a resin system that didn't get enough catalyst (sorry for the terms - I am a composite engineer). I am thinking the "goo" is heat resistant and/or low heat conductivity. The right side is towards the actual thin film element and is attached to the circuit board with tinny little arcs of wires - very fragile looking. I'm sure the "goo" probably helps reduce shock and vibration damage to these.
The next part is a little painful since you have to pry the circuit board out to go any further (i.e. point of no return). The board is bonded onto a pretty thick piece of stainless steel sheet metal - which you can actual see through the back side of the sensor casing at one spot (at first I thought this was the thin film element). Once out of the sensor casing, the real thin film element is clearly visible.
<img src="http://pictureposter.audiworld.com/18887/dscf0472.jpg">
The film element was covered with very tiny spots and looked "dirty" - my take on this is that its contamination. None of my pics show enough detail or are too out of focus to post, just take my word for it. Using a q-tip and alcohol and some fairly aggressive rubbing, the element became cleaner - at least seemed cleaner. Without the rubbing, the contamination didn't budge - this is probably what made the "dunk and swish" method ineffective.
Since the whole thing was pretty messed up, I went a step further and tried to pry out the element - its very brittle and the edges break or chip easily.
Since the thing is destroyed I can not go back and re-test the MAF now that its cleaned - so I am not sure the aggressive clean worked - not that you could do it and replace it anyway. I tried to remove just the end housing around the element, but the halves are glued in place with clear RTV and you have to force it off. I was hoping this section might be removable to give access to the element for the more aggressive cleaning method. You might be able to clean with the unit together, but it would be a tight squeeze to get at the element inside the casing and you couldn't tell if you were cleaning well anyway because you can't see the surface only the side.
As a matter of curiosity, once the section is off, you get an idea how the airflow is guide past the element and out of the sensor casing.
<img src="http://pictureposter.audiworld.com/18887/dscf0475.jpg">
Not sure where to go from here.
My thoughts: the MAF thin film element is heated during operation, so contamination may be burned or melted on. I also suspect water droplets might be getting on it, evaporate, and leave deposits - similar to hard water spots on your paint, which is what the contamination kind of reminded me of. Whether or not that is the whole story is unclear.
The Hitachi MAFS uses the same thin film element and should be susceptible to the same issues, but they are obviously not. Is there something to the Hitachi MAF that alleviates contamination issues (see 6 below)?
I can't imagine the issue is with the circuitry - I saw nothing to make me believe there was an issue there - plus I read that MAF's degrade at different rates. I would think a circuitry issue would be more black and white (good or bad) instead gray like you might have with various levels of contamination.
Does anyone have a Hitachi MAF they would care to donate? - you won't get it back. I would like to see the differences.
Hope this helps in some way - it may just be for curiosity sake right now. I will leave with a little of the research I found in the past few days:
1. Many other makes and models have issues with the thin film MAF's (then why the heck do we have to have `em??? - seems like a conspiracy so Bosch can keep making $200 per MAF). The wire MAF's seem to be more robust because they have a burn off cycle at shut down or start up that burns off contamination.
2. Subaru's in particular have issues and have changed between wire and film MAF's across model years. Subaru owner's have claimed MAF failures are caused by contamination (oil filters, etc.), vibration (I guess the MAF is mounted poorly) and is often sited as the number one cause, and blow off valves shooting air back up the intake tract.
3. I found an A4 owner who was able to "short circuit" his MAF and have it run at what the computer thinks is maximum air flow all the time.
4. This is from a Bosch guide book, "The hot-film air-mass meter combines all measuring elements and the control electronics on a single substrate. In current versions, the heating resistor is located on the back of the base wafer, with the corresponding temperature sensor on the front. This results in somewhat greater response lag than that associated with the hot-wire meter".
5. The newer hot film is slower to react than older hot wire (why the heck are we using films - back to the wire!!!)
6. The Bosch web site has a diagram that shows a deflector screen to deal with particles and contamination - where is ours???
<img src="http://pictureposter.audiworld.com/18887/hotfilm_image3.jpg">
7. One emissions article recommended on wire types using a toothbrush and throttle body cleaner. Then blow off with compressed air and follow with brake cleaner. I thought these things were delicate!!! (see page 7 on http://www.gatewaycleanair.com/mechanic/air4_4/air4_4.pdf)
8. A Subaru article mentioned, "It is not possible to clean these sensors" (so which one is it???)
9. Another good article http://www.wellsmfgcorp.com/counterpoints/Counterpoint3_2.pdf
All for now. Enjoy.
As a little history, the MAF was deemed bad due to VAG readings of a little more than 200 g/s. Low for a chipped or stock car. I tried cleaning but no change and decided to get the new MAF. VAG showed marked improvement and even measuring a few months later showed similar if not slightly better results. Haven't logged in a month or so, but last time was getting 250+ g/s typically.
<img src="http://pictureposter.audiworld.com/18887/current_maf.jpg">
As most may know, the sensor portion of the MAF is held in place in the housing by two security torx screws and has a tight fitting gasket. The sensor has no serviceable features, so you need to destroy it to take it apart - something I am very good at.
The best place to start is the little side panel that is held in place by some brown glue that is more rubbery than solid, so a dremel really doesn't help much - it just gets clogged up. I ended up using a x-acto knife and digging it out.
<img src="http://pictureposter.audiworld.com/18887/dscf0461.jpg">
With enough of the brown stuff out, you can pry the cover off and viola - the guts of the MAF.
<img src="http://pictureposter.audiworld.com/18887/dscf0469.jpg">
Here's where it gets a little interesting. The circuit card is covered in a semi-clear "goo" that is rather soft and almost seems like it was a resin system that didn't get enough catalyst (sorry for the terms - I am a composite engineer). I am thinking the "goo" is heat resistant and/or low heat conductivity. The right side is towards the actual thin film element and is attached to the circuit board with tinny little arcs of wires - very fragile looking. I'm sure the "goo" probably helps reduce shock and vibration damage to these.
The next part is a little painful since you have to pry the circuit board out to go any further (i.e. point of no return). The board is bonded onto a pretty thick piece of stainless steel sheet metal - which you can actual see through the back side of the sensor casing at one spot (at first I thought this was the thin film element). Once out of the sensor casing, the real thin film element is clearly visible.
<img src="http://pictureposter.audiworld.com/18887/dscf0472.jpg">
The film element was covered with very tiny spots and looked "dirty" - my take on this is that its contamination. None of my pics show enough detail or are too out of focus to post, just take my word for it. Using a q-tip and alcohol and some fairly aggressive rubbing, the element became cleaner - at least seemed cleaner. Without the rubbing, the contamination didn't budge - this is probably what made the "dunk and swish" method ineffective.
Since the whole thing was pretty messed up, I went a step further and tried to pry out the element - its very brittle and the edges break or chip easily.
Since the thing is destroyed I can not go back and re-test the MAF now that its cleaned - so I am not sure the aggressive clean worked - not that you could do it and replace it anyway. I tried to remove just the end housing around the element, but the halves are glued in place with clear RTV and you have to force it off. I was hoping this section might be removable to give access to the element for the more aggressive cleaning method. You might be able to clean with the unit together, but it would be a tight squeeze to get at the element inside the casing and you couldn't tell if you were cleaning well anyway because you can't see the surface only the side.
As a matter of curiosity, once the section is off, you get an idea how the airflow is guide past the element and out of the sensor casing.
<img src="http://pictureposter.audiworld.com/18887/dscf0475.jpg">
Not sure where to go from here.
My thoughts: the MAF thin film element is heated during operation, so contamination may be burned or melted on. I also suspect water droplets might be getting on it, evaporate, and leave deposits - similar to hard water spots on your paint, which is what the contamination kind of reminded me of. Whether or not that is the whole story is unclear.
The Hitachi MAFS uses the same thin film element and should be susceptible to the same issues, but they are obviously not. Is there something to the Hitachi MAF that alleviates contamination issues (see 6 below)?
I can't imagine the issue is with the circuitry - I saw nothing to make me believe there was an issue there - plus I read that MAF's degrade at different rates. I would think a circuitry issue would be more black and white (good or bad) instead gray like you might have with various levels of contamination.
Does anyone have a Hitachi MAF they would care to donate? - you won't get it back. I would like to see the differences.
Hope this helps in some way - it may just be for curiosity sake right now. I will leave with a little of the research I found in the past few days:
1. Many other makes and models have issues with the thin film MAF's (then why the heck do we have to have `em??? - seems like a conspiracy so Bosch can keep making $200 per MAF). The wire MAF's seem to be more robust because they have a burn off cycle at shut down or start up that burns off contamination.
2. Subaru's in particular have issues and have changed between wire and film MAF's across model years. Subaru owner's have claimed MAF failures are caused by contamination (oil filters, etc.), vibration (I guess the MAF is mounted poorly) and is often sited as the number one cause, and blow off valves shooting air back up the intake tract.
3. I found an A4 owner who was able to "short circuit" his MAF and have it run at what the computer thinks is maximum air flow all the time.
4. This is from a Bosch guide book, "The hot-film air-mass meter combines all measuring elements and the control electronics on a single substrate. In current versions, the heating resistor is located on the back of the base wafer, with the corresponding temperature sensor on the front. This results in somewhat greater response lag than that associated with the hot-wire meter".
5. The newer hot film is slower to react than older hot wire (why the heck are we using films - back to the wire!!!)
6. The Bosch web site has a diagram that shows a deflector screen to deal with particles and contamination - where is ours???
<img src="http://pictureposter.audiworld.com/18887/hotfilm_image3.jpg">
7. One emissions article recommended on wire types using a toothbrush and throttle body cleaner. Then blow off with compressed air and follow with brake cleaner. I thought these things were delicate!!! (see page 7 on http://www.gatewaycleanair.com/mechanic/air4_4/air4_4.pdf)
8. A Subaru article mentioned, "It is not possible to clean these sensors" (so which one is it???)
9. Another good article http://www.wellsmfgcorp.com/counterpoints/Counterpoint3_2.pdf
All for now. Enjoy.
Trending Topics
04-02-2003, 01:41 PM
#8
Do you have a picture of the backside of that film? or a better zoom?
Only the center part of it needs to be cleaned.
Here's how it works:
Assuming you have metal strips '|' for thin strip, '|||' for thick. '#' for edge strips
It basically looks like this:
# | ||| | #
The edge strips just keep the temperature at edges of the sensor in tune with the body. The center ||| is a heater element pad. The two | | thin strips around it are the temperature probes.
If the heater element is dirty (covered with oil film, dirt, residu,etc), it may not heat up the air properly and the two | | strips will not detect a temperature differential between each other. Likewise, if the two strips are dirty, they won't detect temperature correctly.
It works by having the center of the two | | strips hotter than the edge. When air moves, it cools one side down more than the other - as the other side will see air heated by the center strip. When there is no air moving, both strips will detect the same temperature - like radiant heat. The difference between the two | | is converted to a voltage and that is how the air mass is calculated.
Here's how it works:
Assuming you have metal strips '|' for thin strip, '|||' for thick. '#' for edge strips
It basically looks like this:
# | ||| | #
The edge strips just keep the temperature at edges of the sensor in tune with the body. The center ||| is a heater element pad. The two | | thin strips around it are the temperature probes.
If the heater element is dirty (covered with oil film, dirt, residu,etc), it may not heat up the air properly and the two | | strips will not detect a temperature differential between each other. Likewise, if the two strips are dirty, they won't detect temperature correctly.
It works by having the center of the two | | strips hotter than the edge. When air moves, it cools one side down more than the other - as the other side will see air heated by the center strip. When there is no air moving, both strips will detect the same temperature - like radiant heat. The difference between the two | | is converted to a voltage and that is how the air mass is calculated.
04-02-2003, 01:51 PM
#9
AudiWorld Member
Join Date: Feb 2003
Posts: 521
Likes: 0
Received 0 Likes
on
0 Posts
From what I remember the Bosch is 0-5v and the hitachi is 5-0v...
The lower voltage at higher airflow may keep the temp down and hence keep the sensor contaminantion down. Just a thought.
04-02-2003, 01:54 PM
#10
Junior Member
Thread Starter
Join Date: Apr 2001
Posts: 1,503
Likes: 0
Received 0 Likes
on
0 Posts
Kind of figured that...also, here's a few more fuzzy ones
The problem is that you can't get to any part of the element without disassembling the thing - which destroys it.
Too much flash pic:
<img src="http://pictureposter.audiworld.com/18887/dscf0471.jpg">
At an angle but out of focus pic:
<img src="http://pictureposter.audiworld.com/18887/dscf0473.jpg">
Too much flash pic:
<img src="http://pictureposter.audiworld.com/18887/dscf0471.jpg">
At an angle but out of focus pic:
<img src="http://pictureposter.audiworld.com/18887/dscf0473.jpg">