Autoseatbelts Fixed!

[96svxer]

I know there's alot of people out there that have the common, "gotta open the door before you turn the car off, or the seatbelts wont retract problem." And while searching have found very little advise on how to fix it. Well fixed mine after 2 months of tinkering, and the solution was.................drum role please.............a loose connection!

There's a 4 wire conenction directly below the seatbelt motor. Even though there's a "lock tab" to securely hold the connection together, there's a lot of play in it. The instant I pushed the two conenctions together, the seatbelts fired. Doesn't seem like they have been messed with, just looks like the connections were a poor engineering build. Zip Tied the connections together and the seatbelts have been working perfectly for a week now. NOW GO FIX YOUR SEATBELTS

[ensteele]

I will go try this right now. If it fixes it, I will bow to the east for this great suggestion. That is probably the only thing that I have not tried. It worked like yours, but now it will not go back. It will only go forward. I will let you know how it turns out. Thanks
:)

[96svxer]

Yee-ha.

[Green1995SVX]

I'll be interested to hear if this fixes it for earl!!

Mike

[svx1999]

Hey,
Now my Auto Belts did this a time or 3.. Kinda funny trying to get out when not expecting it. :) then all would retunr to normal.
it's comes and goes...

I think I've now just gotten so use to opening before I turn the ignition I dunno if it still has an issue or not.
BUT thank you very much for this suggestion.

Now Back in the old days of Yahoo there was mention of something like electrical contact grease called Cromlin. it's not really a grease but more of a eletrical stabilant.
VW has such a product for Audi and VW's. It's not cheap and no one knows what the name is..
I've heard too many swear by it and it's gota make ya wonder why Audi has something VERY similiar to it..
Protects contacts and enhances connection so they claimed.

I've searched and searched and never found it..
but I know it exist...

SVX1999

[Earthworm]

Hmm, I wonder if that stuff would help us with dim headlights?

[curlykicker]

Hey 96svxer, could you include a pic of the "4 wire conenction directly below the seatbelt motor"? I'm really anxious to fix mine. I've been riding around for the last month without a belt across my chest! :-P

The motor is not in the door right? It's located under the panel, below the rear window correct? SO, if that's right, I have to take out my rear seat - top and bottom?

[96svxer]

I dont have a pic but you dont need one, really easy to find. I just removed my bottom seat, left the top on, You have to recline it to get at one screw though I believe. Yeah the seatbelt motor is located right below the rear window. You'll see the seatbelt motor at the bottom near the door, cant miss it, if you do slap yourself. And there is a large four pin connection, the only one, feeding the motor.

Just an update though. My wife borrowed the car and magically the seatbelts seem to be an the frits again. It worked perfect for me for a few weeks, now they are working most of the time, wich is still better than none-of the time before. So Not sure if this is a permanent fix or not? Possibly another conenction loose? I will be looking at it again this weekend. It's easy so it's worth a try I guess.

[alia176]

Awright, I removed the rear seat and checked out the connector. I even tied a wire tie around it for more "snug" factor. Unfortunately this didn't cure my seat belt issue. Furthermore, I ran a self test with the diagnostic plug with pin #8 and that produced zilch. There isn't any fault codes stored in the memory for the belts.

Back to the drawing board for me! :confused:

Ali
'92 SVX

[96svxer]

Sorry to hear that. Yeah my seabelts worked great for about two weeks, and they are on the fritz again. Who knows. So I completly tore apart me door and got out my old multi-meter and starting measuring voltages and resistance (cant let those 4 years of college go to waste). Again, I wasn't getting any seat belt error codes either. I drew out a complete pin out of all pins/connection/voltages/resistances. Basically what was happening, long story short is my seatblet motor was'nt getting any voltage, when I applied 12 volts directly to the motor is fired and ran fine. SO that leaves up with either a bad limit switch (even though I didn't have a code) or a bad computer. I measure the resistance over the rear limit switch and it had some very funny readings, so I'm guessing the switch was going. Keeping in mind that my seatbelts would fire sometime, and not others. I'm ordering a new one from Kervin subaru, I'll let you know if it cures the problem. If you want I could e-mail all the measurments I took. Maybe it will help you out

[alia176]

Are there any relays that can be experiencing bad contacts in this circuit? After so many cycles, I wonder if one of the relays simply need replacement. I'll need to refer to the electronics manual to see what's up. Do let me know of your progress.

Thanks.

Ali

[96svxer]

That was my first thought, somebody told me of two relays located below the rear window that control the seatbelts, I found none. Unless these relays are located in a remote location there are none, but why would you need a relay on 12VDC? You probably dont. Limit switches dont require any current. And the motor is driven off it's own wire bundle, not sure if it requires a relay, doubt it though.

[alia176]

Here're some of my symptoms:

-the system does know that the door is open. It definitely knows when the door is closed
-the belt moves out of the way sometimes one sec to 30 sec pause after the door is open.
-coincedentally, if you manually release the belt, then it moves out of the way.

The signal to tell the motor to run, I assume are: door trigger is open or the dome lights are on. Dome lights always come on as they should so I don't think it is telling the belt system to turn on. So there must be a sepate signal coming from the door trigger, which I assume is negative switching.

I mentioned relays because the motor may draw an amp or two, specially as the track wears due to age and not properly lubricated. Typically, relays are used to switch current but i don't know what's happening in this case.

The search continues....!

Ali

[96svxer]

I have the exact same problems as you. Yeah I agree with you on the relay part for the seatbelt motor, Usually though relays either work or they dont. Not much ground inbetween. The motor works on a reverse voltage switch. 12 volts goes forward, -12 goes backwards. Motor work perfectly when an external 12 volt system I wired up was applied. So it's just not getting 12 volts. Which I'm hoping isn't a fried computer somewhere. Or Cross your fingers, A floating ground. Just havent had a chance to trace the motors wires back to under the dash to see whats controlling it. Hoping my new limit switch cures the problem.

[alia176]

Just talking out loud here....

The purpose of the limit switches - stop the motor once they're tripped, otherwise there will be a chance of overtravel. In our cases, the motor does know when to stop, at least that's the case in my car. If the limit switch is bad, then the motor would stop halfway (thinking that it's reached the destination) or even try to force the seat belt mechanism past the stopping point. I don't think you're experiencing that, right?

Question:
Why do we have to open the door before shutting the IGN off in order to have the seat belt get out of the way?

The system works beautifully when the IGN is on but hap hazardly when the IGN is off. If you're looking at your circuit book, see where the power to the motor is originating during these two instances. I don't think it's anything big as the computer. If it was, then the entire seat belt system wouldn't work. My passenger side works fine as far as I know.

On a side note, relays do have issues with contacts burning/corroding due to high cycles. It's always a good idea to replace relays once they've reached a certain life cycle (years, heat/cooling cycles, humidity, etc).

On a totally unlreated topic, last night I found my spare tire floating in the trunk! Guess what I'm doing tonight!

Ali
:confused:

[96svxer]

There are three major wire bundles that control the seatbelts:

Connector 1 (Rear Limit switch bundle)
Yellow (Rear Limit switch 1 Signal)
Black (Rear Limit Switch 1's Ground)
Green (Rear Limit Switch 2 Signal (I belive it's a redunt switch for safety)
Black (Rear Limit Switch 2's Ground)


Connector 2 (Front Limit Switch Sensor)
Green (Forward Limit Switch Ground)
Black (Forward Limit Switch Signal)


Connector 3 (Seat Belt Motor wiring)
Black (Motor Forward)
Red (Motor Reverse)

--------------------------------------------------------------------------------
Basically with the key in the ignition both the yellow and green wire (the rear locked limit switches are reading 10V), and when this happens the seatbelt works, 12V to seatbelt motor)

With the key removed from the ignition the yellow and green wire read (.26V), and the seatbelt doesn't work, no 12V to seatbelt motor)

MY THEORY
So it seems like when the key is removed from the ignition, power is killed to the rear locked limit switch. With no power the seatbelt computer isn't recieving a signal from the limit switches saying the seatbelt it reatracted (locked) thus it doesn't fire the seatbelt motor to let you out. Seems logical.

I just wish somebody had an electrical schematic, it would make this so much easier. I'm going to rip apart my passanger side today to see what it does.

[alia176]

You know, I'll take a look at my ETM (elec troubleshooting manual) tonight and apply your theory. If I had a scanner, I could scan the pages!

Which direction (fwd or rev) does teh motor have to go to let you out of the car? I assume +12V?.

Ali

[96svxer]

All DC Motors to reverse direction you just apply a negative voltage through them. SO basically the wires going to the motor will be this:

When seatbelt is moving towards windshield
black 12V
red 0

When seatbelt is moving towards seat
black 0
red 12V

One goes makes the motor go forward the other makes it go backwards.

[curlykicker]

Originally posted by 96svxer
There are three major wire bundles that control the seatbelts:

Connector 1 (Rear Limit switch bundle)
Yellow (Rear Limit switch 1 Signal)
Black (Rear Limit Switch 1's Ground)
Green (Rear Limit Switch 2 Signal (I belive it's a redunt switch for safety)
Black (Rear Limit Switch 2's Ground)


Connector 2 (Front Limit Switch Sensor)
Green (Forward Limit Switch Ground)
Black (Forward Limit Switch Signal)


Connector 3 (Seat Belt Motor wiring)
Black (Motor Forward)
Red (Motor Reverse)


96svxer, where's Connector 2 located?

[96svxer]

Connector two is located with all the others below the rear window. Might have written the colors down wrong? It's the larger guage (smaller wire) two wire connector. it runs along the top of the window to the limit switch mounted near the windshield.

Well it seems my theory is wrong. My passanger side seatbelt works fine, and the rear limit switch reads 0V with the ignition off. Actually the rear limit switch has nothing to do with the seatbelt operation after the igniton is turned off. You can actually unplug the harness to it, and the seatbelt still works fine. So?????

I tried tracing the seatbelt motor wires back to see whats controlling them (they are a smaller guage wire so I thought they would be easy to trace.) I looked near the fuse box in the drivers side kick panel for about 1/2 an hour with no luck. Then decided to cut the tape around the loom to see which direction the wires are going...they are actually headed towards the back of the car???? so I rip out the trunk paneling, the wire loom heads from the bottom of the trunk, towards the top, and it looks like it runs along the bottom of the rear windshield. I assumed they were heading for the passanger kick panel, along with the passanger side motor wires. But could not find either in the kick panel. Withought gutting my cars interior in 15 degree weather, I'm not going to have any luck withought a wiring schematic.

[OrthopodSvx]

Dont Give up. I have been reading this thread with anticipation, hoping to cure my own lazy driver seatbelt. I feel that it is at a climactic point in the story and am eagerly awaiting the next installment.

[alia176]

A'right dude,

I got the schematic. Finally had time to work on the SVX tonight. I'll scan it for you, 96svxer, and send you or anyone else a copy. This system is very complex at a quick glance. There's the ecu for Automatic shoulder belt system and everything else is being from it. This ECU is located under the rear wiper motor (rear deck) area.

Once we both the schematic, we can talk about it more intelligently. Send me an email so that I can return it with a schematic or two.

Ali

[96svxer]

I guess thats explans why I lost trace of the wire loom under the rear window. But i swear the passanger side setbelt motor wires head towards the kick panel? Maybe they do a loop around?

Orthopodsvx, dude your post sounds dirty.......but funny.

All right but if I get frost bite...

thanks those schematics are going to help alot. Yeah I was surprised how complex this system actually turned out to be.

[alia176]

Dude,

Your profile doesn't have an email address and the link you have didn't take me anywhere. Send me an email and I'll send you the schematics. This goes for everyone who might want a copy.

Ali
alia176atyahoodotcom

[alia176]

Ok dudes,

I finally pulled my head out of my nether region and decided use my locker! Duh! Anyway, I posted the pdf schematic and a few jpgs of the automatic seat belt system. Hopefully this will help out a few people, including myself!

Party on,

Ali

[96svxer]

Looking over the schematic, looks basically what I thought it would. The passanger side wiring loom doese run forward, then runs across the floor into the trunk, not the kick panel. Those crazy subaru engineers. Other then a bad " automatic Shoulder belt control unit" everything in my car I have checked and seems like it's ok. Except there is a seatbelt switch I did not know existed, it's actually located inside the seatbelt buckle near the emergency brake. Kinda wierd that there's only a driver side one. I'm guessing there might be small internal relays inside the control unit that have gone bad explaning why only one side has gone bad? My money's on a bad control unit. Anybody have a working one I could "Barrow"?

PS: alia176 that link wasn't a website, that was my e-mail address. just paste into your email.

[alia176]

I'll take a closer look at this thing tonight at home. Too many damn meetings at work today!

You can easily rule out the ECU by switching the wires from the driver side to the passenger side. If the same problem occurs on the opposite side, then you can concentrate your troubleshooting further on the ECU. However, I suspect it's a relay or some other high wear item that's gone capoot! The driver side is a highly used side and bunch of other SVX are experiencing the same issue. Furthermore, if your ecu is bad then you would have recieved an indication of such from the seat belt warning icon in the instrument panel.

To be continued....

Ali

[alia176]

Just in case you're wondering where the power is coming from in the schematic:
FB-8 (#22) - always hot
Fb-28, 30, 29, MB-2 - ignition

Here's what I'm thinking for troubleshooting: you'll need a second person to open and close the door while you're at the truck with a DVM at the ECU. I'm going on the assumption that the driver and pass sides are symmetrical and identical in operation. Voltage and continuity should be identical from both sides.

step 1:
Check voltage at Pin #1 and #2 while your helper opens and closes the door with ign on. Basically you're trying to establish a baseline. Make a note of the voltage as the motor is operating. do the same thing for the passenger side, pin #6, #7. They should be identical. Now close the door, turn IGN off and do the same thing for both sides. See if there's any diff between the driver and pass side. If the pass side is working and driver side ain't we can proceed to step 2. If the voltage is ok but the motor is not turning on the driver side, then let's move down to the connector at the motor.

step 2:
when the seat belt is out of the way, check for continuity between ground and #17 for driver and ground and pin #13 for pass. Sometimes when I wiggle the shoulder belt, it moves. Could be a coincidence but notice that the shoulder belt sw and rear limit sw are in the same connector. Also try to stop the motor from moving toward the rear of the car by creating a jumper between gnd and pin #17. If this works, then we know that the rear limit sw is doing it's thing.

step 3:
the front limit sw is a normally closed (NC) switch. Check for cont between pin #1 and gnd for driver, #14 and gnd for passenger. Perhaps this is being flaky and telling the motor not to start moving once the door is open. Try to mimmick this by opening up the connection between pin #2 and gnd before the belt reaches it's destination (moving toward the front). At least you can prove to yourself that the belt forward movment logic is correct.


Off to a mtg...to be cont...

Ali

[96svxer]

I think all that you said was good, I've already done it. And after doing it I have realized that alot of that isn't nessasary (only found out after I did it).

The rear locked limit switch has nothing to do with the seatbelt retracting after ignition is turned off. (try it, disconnect the passanger side rear limit switch, I guarantee the seatbelt will still retract....So we can 99.99% rule that switch out.

I think we are looking for the same things, just in a different way.

I too thought I'd just switch the sensors from the passanger side to the drivers side, unfortunantly withought cutting and splicing your harness this is impossible. And just moving the sensors is impossible because they are mirror images of each other. Plus I measured all the sensors and both sides matched. Basically the problem is the ecu doesn't spit out 12V to the seatbelt motor to make it spin when it should. Why nobody is sure yet.

Also you think it's a relay thats gone bad, I'm kinda leaning towards the ecu.Well took the ecu apart. (yeah that was not fun) The instant I took it apart I noticed 4 small square devices, that reminded me of relays I used in college. Ran the part number through google. (NEC MR331-N3) Well NEC Makes relays. I'm guessing there are two per motor (One to switch voltage forward, the other to switch the voltage negative). No way to tell if they are good, they are in a printed circuit board. Only way to tell is to swap a known working ECU into place. And I dont have one!

My only question with it being a relay is, why does the seatbelt work when the ignition is on? You would think if it's a relay it wouldn't work at all?

[fasteract]

I've been following this discussion with interest since I have had the "syndrome" of "driver's belt will retract if key "ON", door open" for a while. One thing I will add is that I have removed the door panel twice and gotten the mechanism to function correctly by adjusting, pulling the wire bundle near the forward upper door area as it dives into its sheet metal burrow. However, as soon as I reattach the door panel, it reverts back to fail mode. Is there a connector in this area ? Has any had this similar experience or wants to try it and comment ?

Cheers,

Craig

[96svxer]

I'm not sure what you are refering to as forward upper door area (Maybe above the dashboard?) but the locations of connectors are:

Three connectors for the seatbelt switches located at the bottom of the rear window near the floor. See Alia176 locker for pics. These connectors then run directly to the seatbelt control unit directly below the rear windshield. Thats it.

Yeah I too had luck if I was messing around with the switches/bundles I could get it to work for a few days...actually thats how this thread got started but nothing permanent.:confused: SO thats why I ohmed ever single wire from the connector back to the module..looking for a weird reading...no luck all wires aboubt .5 ohms which seems normal...so not convinced I used my continuity tester to double check...all the wires were continous, no breaks. I'm just confused :confused: :confused: :confused:

[alia176]

Glad to hear that you found the relays. I'd love it if you could snap a pic or two of the ECU and I can stick in my locker too for reference.

Do the relays have see through housing? If so, can you compare the operations of the driver and pass relays to see what's happening? It'd help to have the schematic for the relays too I imagine.

With regards to your question as to why things don't work with IGN off, I can only assume that constant +12vdc isn't going to the appropriate places.

From the schematic, there are two sources of CONST 12vdc: FB-8 and MB-2. The FB-8 is going to: key warning switch, security elec module, time control unit, and front/rear door latch sw. MB-2 is going to only Pin 3 on the ECU. Curious, do you see +12vdc with IGN off on any of the contacts (NC or NO) of the relays? How about the coil?

Are you still driving this car daily? If so, my hat's off to ya for putting up with all of these inconveniences!

Ali

[fasteract]

Sorry...meant the wiring bundle located on the driver's door, panal removed, near the front, upper portion, ...interesting that next message mentions the door latch sensing switch as being on of the controls with 12v with IGN. ON. Subaru mechanic here said my door latch switch was probably the problem...is it located inboard from the latch mechanism as I think was mentioned in a prior post ?

[96svxer]

Glad to hear that you found the relays. I'd love it if you could snap a pic or two of the ECU and I can stick in my locker too for reference.

I uploaded them into my locker, feel free to steal them. The relays are the 4 black square things.

Do the relays have see through housing? If so, can you compare the operations of the driver and pass relays to see what's happening? It'd help to have the schematic for the relays too I imagine.

Nope they are opaque. We couldn't be that lucky. I'm sure there's no schemtic for the relays. It's a printed circuit board, nothing user fixable on it. Just throw away. No terminals to measure from either. Only way to tell is to swap in a known working ecu :mad:

With regards to your question as to why things don't work with IGN off, I can only assume that constant +12vdc isn't going to the appropriate places. From the schematic, there are two sources of CONST 12vdc: FB-8 and MB-2. The FB-8 is going to: key warning switch, security elec module, time control unit, and front/rear door latch sw. MB-2 is going to only Pin 3 on the ECU. Curious, do you see +12vdc with IGN off on any of the contacts (NC or NO) of the relays? How about the coil?

Well I think the door switch actually feeds the Time control unit. Not vice-versa. You only want the key in the igniton to chime when the doors open. Kinda have to read that section from right to left. They just use the door switch for two different things (the key chime, and the seatbelt) Do you know that FB-8 and MB-2 are constant 12V? Have you measured them for sure? I have not. I have no clue whats feeding those fuses...so I'm not sure if they are constant voltage, impossible to tell from the schematic. WHat the heck does MB-2 stand for? Do you have the schematic of that section of the electrical system...ie...what feeds those fuses? No way to measure any kind of voltage off these relays, no pins.

Are you still driving this car daily? If so, my hat's off to ya for putting up with all of these inconveniences!

Yes I am. I'm a gosh-darn saint for putting up with this car. I'm not going to lie, I am having nightmares in my sleep about this system. I hid my matches and gasoline can, just so I wouldn't get tempted to stage an unfortunante and tragic "accident"

Fasteract : Probably not your door switch. If your dome light comes on when you open the door, it probably means your door switch is working.

[Funpilot]

Guys, its like you were driving my car. Same problems, you'all are a lot more savy in these matters than I, please keep up your detective work, cant wait to see how this comes out!

[96svxer]

Well I havent gotten a darn thing done in the last few weeks. Kervin Subaru wont return any of my calls. I sent them one e-mail and two calls so far in two weeks. They say since Mike left they "dont know" how to look up SVX parts and they "will get back to me." Oh well they lost my business. Too bad Mike helped me out on a few problems. Will start working on it again once I recieve my parts from another company.

[alia176]

Same story here. I've been a little busy with the wife's Legacy wheel bearing project and travelling for work. We'll keep the thread alive until 96svxer's seatbelts are working again!!

Seeing as how most of us have the same problem, I'm incline agree on the relay being the culprit. I wonder how difficult would it be to desolder two relays and swap them around? Basically swap just the one relay from the pass side to the driver side. This way, now the pass side would be experiencing the same issue if this theory proved correct!

Just a thought. You could purchase an ECU from a junk yard but chances are it might be in the same condition as ours. I'll try to find a local SVX with whom I can swap in the ECU for a few minutes to see if the problem goes away. Assuming of course that he doesn't have the same problem!

Ali

[96svxer]

I think it would be impossible to swap relays I thought of that too. Printed circuit boards are impossible to fix. They aren't like a normal board where you can heat up the sotter and remove it.

Yeah found a used ECU...but then thought better of it. Why buy as used one that may or may not work?

Oh no keeping the tread alive till I fix my belts.....well if this trouble shooting thing doesn't work, I have another trick up my sleeve. Might just build my own seatbelt circuit. But we'll see.

[96svxer]

Well I havent gotten a darn thing done in the last few weeks. Kervin Subaru wont return any of my calls. I sent them one e-mail and two calls so far in two weeks. They say since Mike left they "dont know" how to look up SVX parts and they "will get back to me." Oh well they lost my business. Too bad Mike helped me out on a few problems. Will start working on it again once I recieve my parts from another company.

opps minutes after posting this got a reply from Kervin. Looks like I got the last rear-limit switch in the country.....and no more seatbelt ecu left in existance. :( May have to try to find a used one:(

[Uncamitzi]

that I've been following like an episode of "24" Earl said he was going out to check his...(which one??) and never returned...

Earl.... You OK????:confused:

[96svxer]

Well finally figured out what was wrong with the seatbelts. They have been working perfectly for a week now. Gota get back to work but will post more later.

[alia176]

I'm waiting with breathless anticipation !

[Aredubjay]

Originally posted by 96svxer
Well finally figured out what was wrong with the seatbelts. They have been working perfectly for a week now. Gota get back to work but will post more later.

Work? WORK!?!? Where are your priorities man?! ;) :D

[96svxer]

All right! Fine....It's the ECU.....Now let me get back to work! And thanks for ruining my surprise!


I wrote up a little Seatbelt Manual and I will post it up later. It's got a lot of goodies that I dont think anybody knows about.....

Edit: Opps I guess I shouldn't yell at the president...ahhhh...sorry sir!:)

[96svxer]

Well it seems I have deleted my write up, too bad it was pretty informative. I have uploaded some pics to my locker if you want to check them out. But I doubt I will have time to re-write the manual again. But The ECU was in deed bad. I swapped in the new one and everything worked, but to make sure I swapped the old one back in and everything stopped working. But the ECU was $242. Any question give me a shout. Hope this helps!

[nickhawaii]

Aloha gang,

I have been dealing with the same problems with my 92 LS L. Suddenly, the drivers side cut out all togther -- it stays in the open position -- I bet it is the ECU. I hope that I can get another one, FHI has to keep us in parts until 2007 -- after that we are on our own. Alia176 and 96svxer thanks for posting all the great tech info.

Nick

[96svxer]

Well subaru stopped making the autoseatbelts in 94 (i think) so parts are scarce. Actually I would assume your problem isn't the ECU. I would assume it's a bad rear locked limit switch in your case. I would manually crank the seatbelt back to it's rear locked limit switch and open the door, if it fires...I would start at the switch. I had the same problem. (basically just have had bad luck with the seatbelts all together, but everything has been working fine for a few months now.) In stead of buying a new limit switch assembly I just opened mine up replaced both limits switches (both were bad) and sottered them in. Has worked great and only cost me $5. If you need help shoot me off some questions. I know the seatbelt system by heart.

[michael]

This thread has been a great read guys, nice work. I'm still wondering tho, why can't you change out the relays. If it was put together can't it be taken apart? What if you cut the housing off the relay and glued it back when you are done. It is the relays right? not the ECU itself.

[kuoh]

That's generally the case. The SVX is pretty old, so I doubt that they used or needed multi layered PCB, in which case, it's a simple matter to use a solder wick or sucker and remove the existing solder on the relay pins. Finding exact replacement relays might not be as easy, but one can probably rig a similar relay in place.

One thing I didn't see in the thread is whether they tested the relays. While it's not as easy as doing it out of circuit, it is still possible. Simply measure the trigger and output pins on the PCB and see if they change states as expected. If the trigger pin doesn't change state when it's supposed to, then it may be the relay driver components that need looked at.

KuoH

If it was put together can't it be taken apart?

[96svxer]

Wow this thread is a blast from the past. Alot of people seem to be reading it. It's helped out lots of people. Too bad it's not a sticky yet.
I work with relays everyday. Every relay has a switching rating to them. (how many times they can cycle withough burning up). No relay lives forever. It makes sense that the problem is the relay, and not something else, because the driver side door is opened far more times then the passanger door (thats why the drivers side seatbelt stops working, and not the passanger side.) But again I have no definant proof.
Although I have not actually tested the relays (I still have my old module if any body is interested in testing it), I feel very comfortable that it is indeed the relays.
Again I dont believe the relays can be removed. It is on a printed board, no real sotter to mention of, if I remember right. Also I did a web search on the relays model and could find nothing. It must have been a subaru part only. I looked to find relay drawings to show the internal switchings of the relay but couldn't find anything. I did find similar looking relays, made by the same manufacture, but in the long run is it really worth it? Even if there was some way to get the relays out, and find a replacement one that would fit, had the exact same pin spacing, pin layout, life expectancy, voltage, somehow managed to sotter it in, it somehow it didn't come loose during the cold winter months, and hot summer days. I just bought a new one, and couldn't be happier. It's been years and no problems.

[kuoh]

It depends on whether we're talking a difference of a few tens of dollars or several hundred dollars. If you could pick up a new module for under $50, then sure, it probably would be simpler to just replace it. But if it's going to cost $250 when it's only a $5 part, then I wouldn't be so willing to give up. It should simple enough to find a replacement relay that has a similar coil spec as the original, the pin spacing isn't crucial, as one could easily put the relay elsewhere and run wires, it just won't look as pretty. As for it coming loose in changing weather conditions, that shouldn't happen with proper solder connections which are not being strained on a regular basis.

KuoH

I did find similar looking relays, made by the same manufacture, but in the long run is it really worth it? Even if there was some way to get the relays out, and find a replacement one that would fit, had the exact same pin spacing, pin layout, life expectancy, voltage, somehow managed to sotter it in, it somehow it didn't come loose during the cold winter months, and hot summer days.

[michael]

Wow this thread is a blast from the past. Alot of people seem to be reading it. It's helped out lots of people. Too bad it's not a sticky yet.
I work with relays everyday. Every relay has a switching rating to them. (how many times they can cycle withough burning up). No relay lives forever. It makes sense that the problem is the relay, and not something else, because the driver side door is opened far more times then the passanger door (thats why the drivers side seatbelt stops working, and not the passanger side.) But again I have no definant proof.
Although I have not actually tested the relays (I still have my old module if any body is interested in testing it), I feel very comfortable that it is indeed the relays.
Again I dont believe the relays can be removed. It is on a printed board, no real sotter to mention of, if I remember right. Also I did a web search on the relays model and could find nothing. It must have been a subaru part only. I looked to find relay drawings to show the internal switchings of the relay but couldn't find anything. I did find similar looking relays, made by the same manufacture, but in the long run is it really worth it? Even if there was some way to get the relays out, and find a replacement one that would fit, had the exact same pin spacing, pin layout, life expectancy, voltage, somehow managed to sotter it in, it somehow it didn't come loose during the cold winter months, and hot summer days. I just bought a new one, and couldn't be happier. It's been years and no problems.


I'd be interested in your old one if you have no use for it. Mabey I could find a way to use the passenger relays of mine in the driver relay of yours. It is so unfair that subaru can charge us these prices for a seatbelt ECU if it were a GM it would be about $20. It's not a guided missle for god sake! And a new steering wheel, Don't get me started. :mad:

[dannmarr]

I read in some old threads that someone used the rear window switch to control the seatbelts, bypassing all the relays etc. Does anyone know how to do this or has done it?

[96svxer]

PM me your mailing address and when I get back from California I will mail it out to you. I'm not sure people understand how tight of a space there is to deal with. But I wish everybody luck in trying to repair it.

PS Pics of it in my locker, you can read the part number of the Relays (kinda) It will give you an idea of what you have to work with

[kuoh]

That's not bad at all, there's plenty of room for a mini relay. In case you didn't know, the relay soldering happens from the bottom side of the board. All you have to do is measure the relay's dimensions and the pin centers, and I'll bet you can easily find a replacement from Digikey or even Radio Shack if it's a popular size.

KuoH

I'm not sure people understand how tight of a space there is to deal with. But I wish everybody luck in trying to repair it.

[billb]

I have this annoying problem too. I've been watching with interest and I hope that you guys figure this thing out. I'd love to just replace a couple of relays on a PWB. I have access to solder suckers and skilled people. If someone can find the part, I'm sure I can replace it.

[dannmarr]

I got word from SVX JAZZ in how to fix the seatbelt problem.
This is the cheapest alternate fix, try it!
Basically all you are doing is disconnecting the rear window switches and using them to control the seatbelts.
Step one; Remove the back seats
Step two; remove the screw with the washer holding the side panel
Step three; pull back and slightly lift side panel (just enough to reach behind)
Step four; unplug connectors to the rear window(green) and the seatbelt motor(white)
Step five; switch the connections (make sure the feed end is connected to the motor)
Remember, the connectors are the same type for both the window and the seatbelt motor so you can’t make a mistake
try the switches before you close up everything.
You’re all done!
I've included photos with the directions.

[alia176]

Most excellent writeup. Thanks for taking the time to do this!

I am going in a different direction finally! Found a '94 SVX with manual belts so I'll be switching the automatic to manual in the next few weeks. I was never a big fan of auto seat belt and have been waiting for this opportunity. I still need to fix the door limit switch as per the instruction layed out in this thread :rolleyes:

I'll post a writeup with pics shortly in the future.

Regards,

Ali
'92

[michael]

I have this annoying problem too. I've been watching with interest and I hope that you guys figure this thing out. I'd love to just replace a couple of relays on a PWB. I have access to solder suckers and skilled people. If someone can find the part, I'm sure I can replace it.

The part no longer is produced, I contacted the mfgr. and was told they "went out with the do-do". However it is aparently one of 4 relays that go bad, if some one could find out which relay does what we could get a used one and replace the bad one with another.

[kuoh]

Cool, I always wanted automatic rear windows! :p

KuoH


Step five; switch the connections (make sure the feed end is connected to the motor)
Remember, the connectors are the same type for both the window and the seatbelt motor so you can’t make a mistake

[SVX JAZZ]

Kuoh! Hows it been going! My wife and I have a 14month old son now so I don't get to post as often. Yes, I never put down the rear windows and a cheap fast fix was to plug the switch in to the seat belt motor. I liked it and have kept it this way for the past year. Hope all is well.

P.S. The hard start problem went away after I replaced all my injectors. Couldn't remember if I mentioned this or not. Take care.

[kuoh]

Congrats on being a daddy, looks like you'll need to update your sig, unless you were talking about a 14 month old pet. ;)

Luckily, I don't have any serious problems with the seatbelts, other than the driver's lap belt not turning off the dash indicator when buckled. My 92 is still hard to start without using the gas pedal, but at least it cranks every time now with the new ignition switch. Maybe I'll think about the injectors after I get all the struts replaced.

KuoH

[bboileau]

First off I would like to thank everyone in this thread for all the useful information.

I found cheap way to fix the ECU without replacing anything.
Got to thinking about the relays and I personally have never seen one have and intermittant failure, they either work or they don't and mine was still working some of the time. However I have seen lots of cold solder joints. Ripped apart the ECU, resoldered all the relay and and connection pins and VOILA!!! Success...worked like a charm.

Hope this helps.
And thanx again.

[Earthworm]

Welcome to the site!

So you're saying the relays are ok and it's just a loose connection fixed by soldering?

[bboileau]

That's what I found anyway, there are only 4 relays in there....I'm assuming two for motor forward and two for motor reverse....wouldn't have anything to do with power source with the accessory on or off. So if the relays are working with accessory on that would mean they are okay.

Over use you can get a crack in a solder joint or a lifted pad due to thermal migration, resoldering the connections or looking for a lifted pad and jumpering it fixes the problem...well it did for me anyway.

[ensteele]

Welcome to the site. :) :)

[leiavoia]

Easy question: How do i access the ecu? How do i get the panel off without ruining anything? There appears to be 4 little "things" that hold it in, but they don't come out easy and i don't want to break anything.

Thanks for your help.

[96svxer]

Nobody has answered yet huh? Well I will tell you what I did......not sure it was the correct way.

I removed one "clip" using a prying method with both a flat head screwdriver, putty knife, and needle nose pliers. Once I got one out I just grabbed the entire panel with my hands and pulled down. It popped right out. It did after looking at the other side of the panel put small cracks in the panel. The carpet covers it up so you cant see it. It re-installed perfectly, snapped really tight, you'd never know.

I have since sold my SVX and dont visit here often but if nobody else is helping you...(I know the feeling) please PM me and I will try and help.

[kbarrett]

Another re-solder success.

Pulled down panel in top of the trunk, removed seatbelt relay box ( large box on left side ), disassembled it, and re-soldered relay pins in back, as well as the connector pins.

It looks like someone used as little solder as possible ... added just a little, and used enough heat to get a proper joint.

Seat-belt retractor motor works fine now, without having to bugger up the rear-window controllers.

[SVX JAZZ]

Another re-solder success.

Pulled down panel in top of the trunk, removed seatbelt relay box ( large box on left side ), disassembled it, and re-soldered relay pins in back, as well as the connector pins.

It looks like someone used as little solder as possible ... added just a little, and used enough heat to get a proper joint.

Seat-belt retractor motor works fine now, without having to bugger up the rear-window controllers.
Too much trouble - much easier to plug in the rear window switch. ;) Honestly, glad you got it to work. I may try that later.

[Minjin]

I opened up mine and the solder joints for the connectors and relays look fine. I'll reflow them just to be sure. But assuming that they aren't the problem, I looked over the board. The only thing that jumped out at me was maybe some bad caps. A few of them look like they might have leaked. Can anyone identify these markings?

http://i7.tinypic.com/6cihqg5.jpg

http://i11.tinypic.com/4knc2zt.jpg

http://i13.tinypic.com/63lgrwp.jpg

Sorry for the lousy pics.

edit: I know that they are aluminum electrolytic surface mounts. 16v. I'm just not sure of the capacitance. 10 uF? 16 uF?

[techy101]

YAY. I tried the re-soldering, and it seems to have worked like a charm. So +1 for this fix. IMHO it should be a sticky.

[CDG]

I think this should be a sticky somewhere, it took me most of the afternoon to find this particular part of the thread.
I plan to do the same with my bot shortly, and will try to post a pic of the relays, as the pic above is not the relays, but Caps.

[kwren]

That was my first thought, somebody told me of two relays located below the rear window that control the seatbelts, I found none. Unless these relays are located in a remote location there are none, but why would you need a relay on 12VDC? You probably dont. Limit switches dont require any current. And the motor is driven off it's own wire bundle, not sure if it requires a relay, doubt it though.

These cars are loaded with relays!

[pavanbabut]

These cars are loaded with relays!

Huh!... talk abt relays.... my Alfas are fully loaded with relays.... you touch something in it, it is connected to loads of relays before it actually starts working :D, svx is far better in this area (min use of relays comparitively).

-Pavan.

[flyby]

hello people,
im having a problem with my seat belt were it would get stuck going forward when i open the door....im able to tap it on the bottom and it retracts forward like normal but recently my seat belt has been spewing wire noodles all on my back driver side window leaving it all grease when i open the door...when i close the door the track wire goes back in.

Anyone having problem with the auto seat belt getting stuck when you open the door were you have to tap it in order for it to retract forward?

[CDG]

Yes, I too am suffering from a lagging seat belt motor(or whatever) on the drivers door.
Mine is intermitent, but will always move, after some banging aswell.
There is a fix, but haven't tried it yet.
There's a workaround, hooking it up to one of the rear window switches on the drivers door.

[ensteele]

Good to see you back Chris! :)

[CDG]

I've been so busy, but try to get here as often as I have time to peek around. :D

[kwren]

I've been so busy, but try to get here as often as I have time to peek around. :D

:)..............

[CDG]

Ya, and its time to Aircare the baby again.
Wish me luck.

[SuberNatural]

To add to this thread some...

Today I noticed that the response of the seat belt was slightly different in response to 2 conditions:

1) when I turn the car off
2)when I turned the car off (AND removed the key)

Upon taking note of this I decided to play with the key/ignition. If I removed the key before opening the door, 5 times out of 6 it would give the undesired "delay" before the seat belt retracted. So I of course tried the opposite and turned the car off, opening the door before removing the key. The "delay" before retraction was still evident, but it was a much shorter delay.

I also noticed that when I "touched" the key while it was still in the ignition (in the OFF position WITH the door open) I could affect the delay. Often the seat belt would retract as soon as I "touched" the key.

So I went ahead and tried another option. I turned the ignition OFF and removed the key before I opened the door. Then tapped and the ignition or jiggled it with my finger and that ALSO shortened/eliminated the "delay" before the seat belt retracted.

To prove to myself that I'm not crazy lol, I would like a few of you who are experiencing this problem to try this and report back!

Thanks,
Mike

[K_Dub]

If you've got 4 relays and a bit of spare time you could give this a shot. My roommate built mine a new brain with existing wires plus a 12v constant run from under the dash (ignition I think). That feeds to a open relay at rest, so you won't be draining the battery while you're parked. We also installed a fuse in line to prevent things getting too hot, but I've had this "mod" for a couple months without fail.

If anyone sees something scary in there, let me know cause I haven't got the mind for this stuff. That's what the roomie is for.;)

[walmsley]

That worked for me. How cool is that! I can make my seatbelt work by opening and closing my rear window. I'm not going to tell the wife how to work the belt. That should keep her out of the car.