lane_a

Ok, after trying to clean SAI ports from exhaust manifold side and pouring Techron and carburetor cleaner from the top, and after replacing infamous check valve, I was still getting MIL lamp on fairly regular basis. The MIL was tripping at an interval from every couple of days to a maximum of about a week in-between. For the record, I’ve been using about a quart of oil every 1000 miles, or maybe a little less than 1000 miles.

I’m convinced that the SAI ports are plugged, but since I really wasn’t ready to do top end rebuild I took the electronic circuit drawing off the forum posting a couple of months ago and built and installed the circuit. So far, it’s been about a month, no MIL lamps.

So this is the good news, the circuit is doing a great job of keeping the MIL lamp dim. The bad news is that I haven’t been able to get a successful drive cycle to reset the OBD II monitor icons. Somebody posted a BMW drive cycle procedure on the forum a while back and I’ve tried to follow this procedure, but so far no luck. I’ve tried disconnecting the battery for four days thinking this might help clear the computer’s memory, etc., but this didn’t help.

Also, just in case this may help, the OBD II code reader shows the “non ready” monitors flashing. The flashing monitors are the Evaporative System Monitor, Secondary Air System Monitor, Oxygen Sensor Monitor, Oxygen Sensor Heater Monitor and Exhaust Gas Recirculation Monitor. Several other monitors are showing ok (not flashing).

Does anyone have any ideas about this? Since the electronic circuit is doing its job in not allowing the MIL lamp to appear, it seems unlikely to me that this circuitry would prevent the OBD II monitors from resetting. Nevertheless, I can’t get emission inspection until these monitors reset. Any help would be greatly appreciated. Thanks in advance.

g_murray

Hi,
I think you are being the guinea-pig here, for the rest of us, and I commend you for that! We are all going to be where you are, currently, some day.

Setting those infamous 'readiness-monitors' can be an absolute bear. Others have had 'success' (in a relatively short period of time) getting theirs set via the myriad of drive-cycles but it seems, like all things, YMMV. I could not, for the life of me, get mine to set -- and I tried with all my sanity and might -- to 'stick' to the drive cycles. I ended up having all 8 monitors 'set' within 700 miles of 'normal driving'.

Kudos on getting the electronic circuitry to work (which one did you use -- coz I posted TWO diagrams, if memory serves me). I do NOT think the 'non-setting' of the RM's is due to the OBD 'noticing' that you are faking out the signals from the O2 sensors. Naaah, that's giving OBD way too much credit.
YOUR job - -now that you have half your battle won (i.e. no CEL) is to perform the drive cycles so as to get the RM's set. Then you are good as gold!

Suggest u do a search on the web and on these forums for 'drive cycles' -- and git ur *** in gear ...metaphorically speaking!

G.

mborkow

do you have something like CA's "referee statio?" if TX does then you might try taking your car there.

i could be completely wrong here but, if the circuit prevents a failed signal from being recorded or transmitted might that same circuit not also block a passed/ready signal from being reported?. if the default for these readiness tests is "non ready" and a ready signal was being prevented by the circuit from being transmitted then the monitor might never enter the ready state.

Robin 993DX

FWIW, don't know how accurate this information is. Last time when I disconnected my battery and was trying to get the rediness code set for a SMOG check. I had a help from a dealer tech using the Hammer to reset some of the stuff that's resetable with the Hammer and then do a few simple drive cycles. He did mention to me that without reseting some of the items with the hammer it is almost impossible to duplicate all the drive cycles to get he rediness to reset.

epj993

The drive cycle works - if you can find the road to do it and you follow it exactly. OBD readiness codes will also reset when a series of cycles occur. Each monitor is different, but typically it takes 20 to 40 start-stop cycles which include full warm up and cool down. Thus, a quick trip to the local store may have you start twice, once to go there and once to return. But, if a full warm up & cool down does not occur then it may not count as a "cycle".

When I did my SAI flush, I could not find the road space to complete the drive cycle test (too much traffic around here). However, the monitors did reset after several months and many full warm up drives.

If you buy a decent scanner, there's usually a good description of how each monitor works, fault levels before triggering the CEL and number of cycles to reset. Ideally, we'd all have a Bosch Hammer but they're tough to find and expensive.

ecobb993

That's interesting that the BMW cycle (http://srlx.com/p-car/obdii_readiness.pdf ) doesn't work for some. This short routine has worked for me without fail.

I wonder if it has anything to do with transmission type (mine's a tip), year blt (mine's a '96), mileage, gas used, etc.,etc?

Or maybe this routine only handles enough of the ready codes to pass here in Georgia, but not elsewhere.

The CEL/SAI mystery continues.

lane_a

Thanks for the replies guys.

Note to g_murray. I didn't realize I was the guinea pig on this, but I really think it's working very well. Many, many thanks for the circuit drawing. It may in fact be the quickest and cheapest solution to, in my opinion, this great shortcoming of our 993s. And to answer your question, I used the second listed circuit, as it seemed a little simpler than the first. It was the one with only one switching transistor(PN 2N2222A). I bought everyting off internet for less than $20. I should've taken pictures, but didn't take the time to do so. Actually, I thought I'd probably botch the first attempt and end up building a second one and correct mistakes on the first. But, first one seemed to work out fine so far. One concern I've had though is whether the circuit will be a problem with Texas summer heat since I've mounted the circuit box in the engine compartment where it was most convenient to tie-into both the air pump relay wiring and the oxygen sensors. Time will tell. As far as resetting the drive cycle, maybe I just need to be patient.

Also to mborkow. I've not heard of referee station here in Texas, but it's something that I need to check out. Thanks for the reminder about this. I'll certainly get serious about this if I can't get the drive cycle to reset.

planeguy67

Good on you, Lane, for successfully building the, errrr, SAI enhancement device and sticking it to "the man".

I just did the SAI flush and my ports were pretty clear, so I don't think I'll be needing the circuit for a while (knocked on so much wood my knuckles are sore). But I am still very curious how you hooked it up. If you have the time (I know we're ALL busy), I am looking for answers to these questions:

1. The circuit receives its power from the 12 volt signal energizing the air pump. Where is this wire and how did you tap it?

2. Did you just ground the circuit to the body frame or tap a ground wire?

3. If I remember correctly, the circuit has to bypass 4 O2 sensors. Were these wires easy to identify and tap into?

4. Did you make an effort to hide the device?

As an aside note, the circuit leaves out a fuse, but I would definitely add one just in case something in the device shorts out. This will help protect the car's electrical system. I'm calculating a 5 amp fuse should be plenty.

Thanks for the info.

lane_a

planeguy67

It's been a little over a month, so I'll try to come up with the right answers to your questions. Also, you must understand that I'm not an electrical engineer, I know just enough to follow the circuit diagram.
1. Yes, the circuit gets its 12 volts from tapping into the wire from the air pump relay leading to the pump. I found the correct colored wire per wiring diagram and found a point on the engine side of the large multi-pin connector that connects to the fuse/relay box in the engine compartment. The air pump relay is right there in this box also, but by staying on the engine side of the connector the circuit box can come out with the engine with no need to cut the wiring.

2. As per the circuit drawing, the circuit grounds through the o2 sensor ground.

3. Yes, there are 4 o2 sensors to connect the circuit to. The wiring from the o2 sensors penetrate the engine shroud at the rear of the engine compartment on each side. The wiring is fairly easy to identify, access and tap into. You can identify the o2 wiring by tracing from the o2 sensors from the underside of the engine shroud to a connector located on top of the shroud. There are four color coded wires from each sensor and your looking for the black wire from all four sensors and the gray from one. It's my understanding that the gray is the ground wire. To make the connection I didn't cut the o2 wiring, I just skinned back the insulation and soldered the tap wire. However, by doing it this way I just had to re-tape connection with electrical tape although use of heat shrink tubing would be much preferred. I was just reluctant to cut the wiring.

4. I mounted the circuit in a small black electronic "project" box from Radion Shack (no affiliation). The texture of this plastic box looks similar to the housing for the engine compartment relay/fuse box, so I just stuck it with velcro next to the fuse box and it's completely inconspicuous, except for the wiring leading from the o2 sensors on the right side. I still need to do something better with this wiring.

If this system continues to prove out, I'll post some pictures and more description on the forum at a later date. In the meantime, if anyone else has questions I'll be glad to try and answer. Oh yes, I'd better go home this evening and check that everything I've said above is correct. As I said, it's been over a month and the memory isn't what it use to be. Will advise any corrections tomorrow if needed.

planeguy67

Thanks, Lane. Great explanation... who needs an engineer???

Oh, duh... next time I'll look at the schematic before I ask a stupid question.

Thanks again.

enjoytheride

Where might one find the schematics for building said device?

SwayBar

First of all, great job building and implementing the circuit!

I think a sound experiment to perform next would be to remove the new circuitry to see whether or not you can get the OBD II to reset properly.

If removed, and performing the recommended drive-cycles successfully resets the codes, then *maybe* the new circuitry is causing a problem.

If you still have a problem resetting the codes without the new circuitry, then it would seem to be 'normal' behavior as some have posted. Most importantly though, it would remove suspicion that the new circuitry is introducing a new problem.

2ndof2

lane_a

You ever get the codes to reset?

993inNC

I did two things recently, and not sure of their relevance on what I am about to say but here it is. My codes would not reset when following the set of instructions given to me by the test computer without throwing a host of the same codes (SAI, random miss.....) So having read something here before about A/C and its effects on idle I did the following. A) left my A/C on so as soon as the car started the system was on and doing whatever it is to the idle control it does. B) it just so happened that I was taking the car out of town and did roughly 150 miles each way at relatively constant speed (80-85). There were only a couple of short intervals that the car stopped(still running),started off again (traffic light) and sat idle in between. When I got back from the trip I noticed that the CEL had not popped (usually happens within 60 miles after a reset). I went to the same test machine as part of the same trip and sure enough.....every one of the readiness codes had reset INSPECTION COMPLETE
Again, not sure how or if the ride (which was at speeds higher than the prescribed) or the A/C remaining on had any effect, but it worked!

g_murray

That's GREAT news.

I can only surmise that as a result of all your 'previous' drivings ...that the 'system' had accumulated "almost enough" "info" - so to speak - about the goings-in in the emissions system.
Say a 9 on a scale of 1 to 10.

THIS particular drive, literally, 'put it over the edge' (again so to speak) - in removing all doubt that all the systems were, indeed, OK. (Read: you got a 10)

Voila ...all readiness monitors set to 'OK'.

Personally, (and I hope I'm not offending anyone here), the obd-ii system - for older cars like ours -- seems to work with the same (lethargic) pace as some government agencies that I've had to deal with. They EVENTUALLY do get the job done -- but, boy, you'd swear sometimes that they're getting paid by the hour.

G.