56 captured repliesDiscussion pagination complete · Source accounts, not a confirmed diagnosis
THIS DISCUSSION / SOURCE-DERIVED OVERVIEW
Discussion summary
YOUR RESEARCHSAI Repair DIY
Leading reported evidence
Check with durametric tonight and sai readiness is a pass
Performed · outcome unconfirmed
No successful repair reported for this finding.
0
Reported fixes
0
Unsuccessful attempts
1
Discussions
Read evidence and context
Explicit resolution reports before identified causes, performed actions and suggestions. Within each level, matching component labels precede direct query mentions and inherited thread context; successful versus unsuccessful discussion counts, owner confirmations and independent discussion support determine order. This is not a probability of success.
Discussions that performed work without a confirmed outcome
1
0 original-poster resolution reports. Counts can overlap; they are not unique people or a success rate.
0 discussions have direct query context; 0 report a fix in that context. Other evidence may inherit the discussion’s subject.
Performed · outcome unconfirmed
Checked with Durametric tonight and SAI readiness is a pass.
Leading evidence: 'Check with durametric tonight and sai readiness is a pass' (no explicit attributable resolution; this is a reported action or suggestion). Across 1 matching discussions, the most frequently mentioned possible cause is 'fairly significant vacuum leak' (1 discussions; 0 report identifying it and 1 suggest it). The most frequently discussed action is 'Check /' (1 discussions). For that action, 0 discussions report a resolution and 0 report an unsuccessful attempt. No action has an explicit, attributable resolution report under the current rules. These are forum reports and suggestions, not a confirmed diagnosis.
Source reports, not a diagnosis or a success probability. Counts refer to discussions and can overlap.
Explicit resolution reports before identified causes, performed actions and suggestions. Within each level, matching component labels precede direct query mentions and inherited thread context; successful versus unsuccessful discussion counts, owner confirmations and independent discussion support determine order. This is not a probability of success.
Discussions that performed work without a confirmed outcome
1
0 original-poster resolution reports. Counts can overlap; they are not unique people or a success rate.
0 discussions have direct query context; 0 report a fix in that context. Other evidence may inherit the discussion’s subject.
Performed · outcome unconfirmed · Original poster
I then replaced the seal and verified it was seated.
Explicit resolution reports before identified causes, performed actions and suggestions. Within each level, matching component labels precede direct query mentions and inherited thread context; successful versus unsuccessful discussion counts, owner confirmations and independent discussion support determine order. This is not a probability of success.
Discussions that performed work without a confirmed outcome
1
0 original-poster resolution reports. Counts can overlap; they are not unique people or a success rate.
0 discussions have direct query context; 0 report a fix in that context. Other evidence may inherit the discussion’s subject.
Performed · outcome unconfirmed
3) Tested changeover valve via Durametric activation.
Explicit resolution reports before identified causes, performed actions and suggestions. Within each level, matching component labels precede direct query mentions and inherited thread context; successful versus unsuccessful discussion counts, owner confirmations and independent discussion support determine order. This is not a probability of success.
Discussions that performed work without a confirmed outcome
1
0 original-poster resolution reports. Counts can overlap; they are not unique people or a success rate.
0 discussions have direct query context; 0 report a fix in that context. Other evidence may inherit the discussion’s subject.
Performed · outcome unconfirmed
1) Tested SAI pump in car via Durametric activation and on bench via 12V supply.
Explicit resolution reports before identified causes, performed actions and suggestions. Within each level, matching component labels precede direct query mentions and inherited thread context; successful versus unsuccessful discussion counts, owner confirmations and independent discussion support determine order. This is not a probability of success.
Discussions that performed work without a confirmed outcome
1
0 original-poster resolution reports. Counts can overlap; they are not unique people or a success rate.
0 discussions have direct query context; 0 report a fix in that context. Other evidence may inherit the discussion’s subject.
Performed · outcome unconfirmed
I used multiple applications of WD40 to clean out the bearings then used 3 in 1 oil to lube as I did not have grease and it was after 10 pm when I started the project so no stores were open...
0 original-poster resolution reports. Counts can overlap; they are not unique people or a success rate.
0 discussions have direct query context; 0 report a fix in that context. Other evidence may inherit the discussion’s subject.
Suggestion
This case would seem to be an argument against the electronic hack to disable the SAI fault checking; if it's indeed caused by a vacuum leak, that's not completely contained to the SAI system and could have a negative effect on other systems like the intake resonance flap, etc.
0 original-poster resolution reports. Counts can overlap; they are not unique people or a success rate.
0 discussions have direct query context; 0 report a fix in that context. Other evidence may inherit the discussion’s subject.
Suggestion
While you're working on all this I recommend wiring the oil filler tube up to the engine cover so it's out of the way and you aren't constantly bumping it.
0 original-poster resolution reports. Counts can overlap; they are not unique people or a success rate.
0 discussions have direct query context; 0 report a fix in that context. Other evidence may inherit the discussion’s subject.
Suggestion
The total carbon footprint of replacing &/or repairing the system plus the aggravation of failing Smog Tests and performing vague Drive Cycles makes the entire repair of dubious merit.
0 original-poster resolution reports. Counts can overlap; they are not unique people or a success rate.
0 discussions have direct query context; 0 report a fix in that context. Other evidence may inherit the discussion’s subject.
Suggestion
The total carbon footprint of replacing &/or repairing the system plus the aggravation of failing Smog Tests and performing vague Drive Cycles makes the entire repair of dubious merit.
These are separate views of the same evidence. Frequency alone does not establish a fix, and these counts must not be added together.
fairly significant vacuum leak
No successful repair reported for this finding.
0
Identified cause
1
Suspected cause
0
Reported fixes
Read evidence and context
Discussions that suspect a cause
1
0 original-poster resolution reports. Counts can overlap; they are not unique people or a success rate.
0 discussions have direct query context; 0 report a fix in that context. Other evidence may inherit the discussion’s subject.
Suspected cause
After cleaning it it's quite clear that there's a big crack, so probably a fairly significant vacuum leak, which may have increased/decreased depending on temperature as the tank expands and contracts slightly.
Counts describe captured discussions, not unique people or a diagnosis. Author identities are not reliably stored. Exact duplicate evidence is counted once across matching discussions; one discussion counts at most once per action/state. Multiple repairs reported together remain a combination. Suggestions and failed attempts do not establish a fix. Matches cover whole discussions and can include related faults, model variants and incomplete threads.
56 eligible posts · 0 unverified posts · 1 discussions match the subject in their title or opening post.
Rennlist: 1 discussions
0 evidence records excluded for query context · 1 excluded by outcome qualification · 0 adjusted by outcome qualification.
1 acknowledgements have no verified action link; they do not establish a repair result.
Outcome without a verified action link
After installing the switchover valve, I activated it via Durametric to ensure it worked.
This is a relatively easy DIY that may save a failing secondary air injection motor, commonly referred to as the SAI. Removal and installation is easy and covered many times elsewhere, so I wont repeat it. I would only do this repair if I heard the tell-tale screech/squeal on SAI start-up. The noise is indicating the SAI has a problem anyway, whether its the bearings or fan discs, the unit may need service soon. The noise could be from the sealed bearings or fan discs.
After SAI removal, separate the mounting bracket from the SAI assembly.
Remove retaining clips from assembly case.
You can now see the motor shaft and fan discs retaining nut. Don�t remove the snap ring. Just remove the nut.
Remove all fan discs and spacers, carefully noting the order and position of each piece. I usually stack or lay the pieces in order so I don�t forget. You may want to photograph or make notes.
You should now see only the motor left in the motor case. Remove the two retaining screws.
Replace the retaining nut on the motor shaft and carefully secure in a vise. Gently pull the motor case from the motor. You will see that it is just a basic rotor/stator electric motor.
You can now see the sealed bearings on each end of the electric motor. The shaft end and support end.
Carefully remove the seal from the SHAFT END sealed bearing. I used a jewelers flat blade screw driver to remove the seal. Just be gentle with it and it wont break.
You can now see the exposed caged ball bearings. Mine were dry. I carefully packed the bearings with a lite quality grease while rotating the shaft. I then replaced the seal and verified it was seated. The seal cant come out later because a spacer rides against it.
Then look at the support end sealed bearing. You will not be able to remove the seal but you can carefully separate it enough to �inject� grease into the bearing. You can use a needle-tip adapter on your grease gun or a large gauge syringe (18g or 20g) with lite grease.
I even put a little extra grease on the support end of the shaft because it is isolated in the closed housing end.
Verify the motor turns smoothly, reassemble, and install. You can test it prior to installation with a 12V DC source if you want to.
Some miscellaneous notes:
This will probably not help a seized sealed bearing(s).
This is NOT a critical complicated engine part and is not hard to DIY if you take your time and carefully note the dis-assembly.
I will run mine for awhile then take it apart again and see if my �extra� grease attracted more dust or dirt.
If my bearings would have been ruined, I still would not have trashed the unit. I would have rebuilt the electric motor itself, replacing the brushes and bearings which is much harder than this DIY.
Excellent job Krazy - detail in your instructions is really good - a true "how to" for each step - for example rather than just saying "remove the motor from the case" - you actually tell us how.
Thank you - (now don't get all mad at me if by chance mine requires SAI work in the future and I don't do it by myself - what I will be armed with though is the knowledge that this part is not something that only a NASA engineer can work on)
KrazyK,
Thanks so much. I had the loud whine noise followed by a wobbly wah wah wah when the SAI shut off. I followed your instructions to the letter and the noises are completely gone. I did take the liberty of spraying out every individual part with compressed air including the filter on the end which I also disassembled. There was a lot of dust in that small part! I used multiple applications of WD40 to clean out the bearings then used 3 in 1 oil to lube as I did not have grease and it was after 10 pm when I started the project so no stores were open... I will monitor it and if the noise returns I will disassemble again and use grease. Thanks again.
By SAI was screaming whenever I started up the car cold. I cleaned it out this winter thanks to these instructions. It was pretty straightforward.
I used an air-compressor to blow out (a lot) of dust from the motor, and cleaned and inserted teflon grease into the bearings.
Cleaned and reinstalled. The startup noise is gone.
A few minor additions to K's helpful Post:
The two machine screws that retain the motor are T-20.
If you want to clean the air filter there are 3 push clips to remove. I used a hook tool to gently lift the ears over the plastic pin head .Harbor Freight Pittsburgh - item#66836 pick hook set. To re-install these clips, I used an 8mm nut driver to push them in to position.
There are two foam filters inside .They were very dusty and the foam is quite brittle so be careful when knocking out the dust.
I used silicone grease on the bearings .The bearings are both marked "Made in England NMB 608S " You can find them by googling '608-2RS'. The bearings measure 8x22x7mm. A pair should be less than $20.I found some for $1 each on Flea Bay! The problem in replacing them would be the crimp on the motor case.
If when you try to reassemble the plastic case, the 5 spring steel clips may not fit .I just removed a non-essential 2mm circular lip at the center of the case.
Unaware of this thread, I just did this last weekend. Here are some pictures that might help others.
Keep the stack up of parts correct for reassembly after
Shaft end bearing
Support end
Any tips for getting removing the motor from the plastic housing? I didn't want to yank too hard given that it's 15 yo plastic.
For the time being, seems like greasing the shaft end took care of the problem.
I couldn't remove mine either. Just too tight to risk breaking something. I did lube the shaft end and the high pitched whine went away. Thanks to the OP and contributors for this nice DIY write-up.
I'm getting SAI fault codes on my car every couple weeks. The SAI pump sounds kind of "sick" sometimes when it runs, and it does not seem to run on every cold start. It actually sounds like maybe the fan discs are contacting the housing, because it makes a metallic ringing sound, like discs are occasionally rubbing the housing and then ringing down. I'm thinking maybe the bearings are so shot that it's able to orbit a little and touch the housing, or perhaps the discs are loose and shifting relative to one another? Or perhaps the motor itself is spinning but the fan discs are loose on the shaft and not spinning much?
In any event, those of you that have done this, do you think mine sounds serviceable, and/or that such behavior could cause poor enough airflow that it would trip the fault code? Guess there's no way to know without taking it apart. It seems there's not much harm in trying though, before I start ordering new parts. Any major caveats to be aware of other than keeping the parts in order?
What year is your car? You should be able to tell once you put it on the bench and test it with some 12v power (jump starter). If you have Durametric, it can trigger the pump.
A new pump is like $1k+ but there are cheaper alternatives from Bosch.
KrazyK and others have submitted DIY instructions with photos of the SAI pump overhaul. There are some problems if you go deeper than KrazyK did.The spinning plate can scrape and be noisy. Even if you fix the pump perfectly, there are many other problematic parts to the system. Most of those parts are very time consuming and tedious to reach. The diagnostics ar equally fraught. I have done all of this several times because I was curious but it was a ridiculous amount of time. The SAI system only operates for around 90 seconds at cold start an rarely during normal operation and it's practicality is debated. The total carbon footprint of replacing &/or repairing the system plus the aggravation of failing Smog Tests and performing vague Drive Cycles makes the entire repair of dubious merit. That is why some people just do the ROW flash or use the test device offered by Particlewave. Neither are ideal solutions but they are practical expedients until you have time to fix/upgrade the entire SAI system.The green-police wince at this but they have yet to offer a practical alternative that has earned the acclaim of other enthusiasts.
2001, I can trigger it with Durametric and it runs, just sounds nasty. Not sure if that alone is likely to be causing my codes or if there are likely to be other issues with switchover valves, etc.
My guess is you have both P0491 and P0492? There's a recent thread here for more info
https://rennlist.com/forums/996-foru...0492-help.html
Assuming you have codes for both banks, it's likely you have a problem with the vacuum, electrical changeover valve, or the two mechanical valves. You can just disconnect the SAI pump hose from the pump, activate the pump and feel if you have air, a lot of air.
Re the pump, best is to work on it on the bench. If it's just the bearing, adding some grease like in this DIy thread will help for sure. If you want to replace the pump, you have a few (practical) options:
1) Get a low mileage one from qualityporscheparts on eBay (search for 99660510400). This one is $150
https://www.ebay.com/itm/01-Boxster-...wAAOSweW5VSrsl
2) Look up "0580000023" and "0580000022" on ebay and get a used or new one
https://www.ebay.com/itm/0580000023-...IAAOSwbWZaYbSb
The cheaper new ones are probably from China but that doesn't mean they are not good (no experience here though)
Some more data:
1) Tested SAI pump in car via Durametric activation and on bench via 12V supply. Runs reliably every time, but does make an occasional pinging/scraping noise like the pump vanes are scraping the housing every now and then, causing them to ring like a bell. Blows air hard.
2) Disassembled SAI pump to inspect internals. Everything looks flawlessly clean and feels tight. Actually looks like it is brand new even though it has a 2000 production date. I don't see any evidence of any bearing looseness or scrape marks, so I can't explain the intermittent noise I'm hearing from it. I feel like it's functioning fine.
3) Tested changeover valve via Durametric activation. Can hear it clicking about once per second while activated, like it is opening and closing. Not sure if this is the standard "activation" protocol. I get identical behavior from the intake resonance flap changeover valve if I activate it via Durametric, so I kind of think this is the standard "activation" protocol, that it just cycles it open/close once per second or so. This was done with the engine not running so I can hear it click.
4) Pulled off a vacuum hose from resonance flap changeover valve just because it's easily accessible. Feels like good solid vacuum with engine running (will suck onto my finger and not let go). Very non-scientific.
So to me the remaining possibilities are:
1) The combination valve behind the alternator (haven't gotten to this to test/inspect it since I'd have to remove alternator)
2) A vacuum leak in the SAI system somewhere (seems unlikely, all my vacuum hoses are surprisingly nice-looking and supple, no signs of drying out and cracking)
3) Mechanical failure of the changeover valve (just because Durametric makes it click doesn't necessarily mean it's opening and closing properly)
4) Clogging of a vacuum line
5) Clogging of the SAI pipes or ports on the engine.
Other thoughts?
Agree with your assessment so far. Do the codes point to both banks? You can run more test based on
this
thread.
You can see if air is actually pumped into the exhaust streams based on the O2 sensor readings during cold start.
Yes, codes point to both banks, P0410 and P1411. Secondary Air Injection system signal implausible for each bank. I clear the code and it comes back a couple weeks later. Has happened 3 times now. I can try logging at startup to see what the O2 sensor readings look like.
Post #6 here explains how the O2 readings should look like
https://rennlist.com/forums/996-foru...l#post11746570
Btw, you said your vacuum hoses are supple. Are those stock? Stock ones are hard plastic....just want to see if it has been mosified.
You guys are getting into way more than I did. Mine sounded like a power saw when ever it ran. Greased the bearings and its run fine for a few months now. For the cost a tube of grease and 30 minutes of your time, I think it's worth doing to see if it fixes the problem. Sounds like you already have it out. Sorry if missed anything.
Yeah, my lines are hard plastic, I was referring more to the rubber elbows that join the plastic bits together. I'd expect those to all be kind of dry and flaky/fragile by the age of 17, but they all look and feel like nice gooey new rubber, so I'm really kind of doubting there's a leak somewhere. Perhaps the reservoir though, or some other plastic bit.
What are the conditions under which the SAI pump should run? Mine doesn't fire up on EVERY cold start, I'd say just "most". Like this morning it did not, but the ambient temperature was pretty warm...does it need to be below a certain temperature outside to be a "cold" start?
For now my inclination is to just drive it and see if the issue continues to come back, or until I have another reason to start taking things apart. Whatever extra bit of pollution may be happening in the first 90 seconds after a start is paltry compared to what I'm doing to the environment by tracking the car
Update: I took this apart last night, targeting the mechanical combination valve, the electronic switchover valve, and the vacuum reservoir, since all of those things reside behind the alternator and it makes sense to replace them all while it's apart. Combo valve and electronic valve seem okay. I found a crack in the vacuum reservoir, so I'm hoping that proves to be the issue. It was very difficult to see at first, but I noticed a berm of dirt in one area of the bottom of the reservoir, caused by vacuum suction pulling dust/dirt up against the tank in that one area where there was a crack. After cleaning it it's quite clear that there's a big crack, so probably a fairly significant vacuum leak, which may have increased/decreased depending on temperature as the tank expands and contracts slightly.
The bracket that holds the electronic valve was also cracked, allowing that valve to float around and just be supported by vacuum hoses and wiring. I don't think this has an effect, but I'll secure the new one in place better so it can't rattle around.
I'll upload some photos of it all tonight, and report back on how it runs after reassembly.
Great update. I suggest that you double check/test the valves because if you have to go back in ......
Another suggestion is to use brass fittings to connect the vacuum tubes and use new ,high quality USA-mad silicone tubing. I wrote this up some time ago but IIRC you can find the tube and fittings on EBay or Amazon.
The old plastic T's & EPDM tube will be brittle . .You may need to improve the connection where the vacuum tube enters the intake pipe.
If all else fails there is an electronic hack to eliminate the CEL while you continue diagnosing
Ok here is the "after":
The red arrow shows the broken bracket where the switchover valve is supposed to attach. I just attached it to the pipe shown with wire ties to keep it in place, rather than floating on the vacuum lines and wire harness like it has been for who knows how long, while still functioning. I don't see any harm in that. I considered using a hose clamp but decided if that vibrated it could erode the pipe, whereas plastic wire ties won't.
To replace the combination valve you can get pretty much any 10mm wrench on the left screw (green arrow), but you'll need a very stubby one to get on the right one (orange arrow). Anything longer than about 4" and you won't be able to swing it, and you're going to want a 12-point box wrench so you can swing it 30 degrees at a time; there isn't room to move it the 60 degrees at a time required for a 6-point wrench, since you're constrained on the left by the valve itself, and on the right by the dipstick tube. You'll need some very deft fingers and/or some tricky magnetic tools to avoid dropping the screws, and to get them threaded to put everything back together. Don't forget to replace the metal gasket that goes between the combination valve the distribution piece it attaches to; the old one on mine stuck to the distribution piece but came off cleanly when picked with a razor blade.
While you're working on all this I recommend wiring the oil filler tube up to the engine cover so it's out of the way and you aren't constantly bumping it. It's in the way of this job, and known to be fragile, so it's best to just get it completely out of the way so it avoids abuse. Also be careful with the plastic vacuum lines. They seem pretty robust but I'm sure it wouldn't be hard to crack one, and when you're moving them around it's easy to lose track of how they're connected, so take some notes or photos so you know how they all reconnect.
When removing the alternator, I was unable to get the famous bushing on the back to move by hammering on the screw as all the DIYs recommend. I beat the life out of it and it just wasn't moving. I came up with an alternative method; I put a 2-jaw puller in place to push on the screw while pulling on the front right alternator mounting tab. This allowed me to apply pressure in a controlled manner rather than just beating on it, and it opened up the alternator bracket enough to get it loose. All you're looking for is 1/2 a millimeter or so, and then you can rock the alternator up out of its mounting. Be extremely careful doing this; too much pressure and you could snap the mounting tab off the alternator. Once it was off the car I used a vise and a socket to pop the bushing back a mm or so to make it easier to reassemble. Based on how much pressure I had to apply to get it to move in the vise, I don't think any reasonable amount of impact was going to make it move. I don't like swinging hammers that hard around expensive parts anyway.
Before installing the combination valve and reservoir, I tested them via a homemade vacuum pump, AKA my lungs and a piece of spare vacuum hose. After installing the switchover valve, I activated it via Durametric to ensure it worked. Would hate to button all this back up and then find out one of the new parts was DOA!
Here's the offending piece, the vacuum reservoir:
You can see there's a crack along the perimeter of the "floor", almost 25% of the way around. For what it's worth this is on the area of the perimeter closest to the mounting flange. This was creating a pretty massive vacuum leak.
After replacement the car runs fine, no CEL so far. I believe I detect a smoother and slightly faster idle. Before my idle would be around 750 RPM and bounce up and down +/- 50-70 RPM. Now it seems to be more like ~790 RPM with only a +/- ~20 variation. Very little data so far though, I've only driven the car about 5 miles since the repair. I'll report back if the CEL comes back, or in a couple months if it seems to be gone for good.
Maybe it would be prudent to put a fillet of epoxy in that interior corner all the way around the bottom on a new part? But who knows, that could introduce more thermal expansion/contraction stress and make things worse.
This case would seem to be an argument against the electronic hack to disable the SAI fault checking; if it's indeed caused by a vacuum leak, that's not completely contained to the SAI system and could have a negative effect on other systems like the intake resonance flap, etc. If it was a fried SAI pump and one was looking at hundreds of dollars to replace that, then I could see the argument for patching it, but it still seems risky that it could mask other problems that actually affect performance or health of the engine.
Now if only we could find a place to mount it and reroute the vacuum hoses, would be the end of the plastic reservoir forever and its location way back in there. Tempting...
https://www.summitracing.com/parts/sum-g1461
As part of the SAI hack you plug the Vacuum hose where it enters the intake.The Summit link is excellent becuse it is the same price as the plastic oem style part.
If your SAI readiness is set (can be read by a generic OBDII scanner or Durametric), your SAI system has passed the DME tests so it should be fine. In fact, the DME monitors the SAI system at least twice during every cold start and car stopped when engine is running so if it has not flagged anything, your SAI system is healthy.
VERY IMPORTANT note for anyone who comes across this thread looking to rebuild their SAI. The nut holding the fan disc assembly is reverse threaded - at least it is on the �99 cars. I chewed up the screwdriver a lot on the end of the motor shaft before I figured this out.
OK. It turns out MY99 cars have a different SAI pump than later cars. The electric motor is different and I could not figure out how to access the bearings to clean and lubricate them. I will probably buy and complete used SAI pump to replace mine.
When bench testing the SAI air pump it is worth noting that a pump in good condition gives quite a 'kick' when it starts. The momentary in-rush may be 20A+. The air should blast out .There are some Youtube videos for a similar pump on Mercedes that demonstrate how forceful the air flow should be.