Can you hear the relay click when the ignition switch is turned on? If so, this sounding like broken wire or a grounding issue to me more than anything else. Or a DOA pump.
Can you hear the relay click when the ignition switch is turned on? If so, this sounding like broken wire or a grounding issue to me more than anything else. Or a DOA pump.
In reply to eastsideTim :
Interestingly, on this car the fuel pump doesn't do a priming pulse - it only runs the pump when the engine is rotating. But connecting the battery directly to the 87 pin on the relay socket - which should then go through the verified-good fuse and then on to the pump - does not engage the pump at all. So I'm leaning towards a bad ground or a chaffed wire somewhere. Before I go tearing things apart looking for that, though, I want to triple check just to make sure I didn't probe the wrong ports or something like that. It's a pretty simple system overall, so it shouldn't be too hard to narrow down exactly what's wrong.
In reply to wae :
Is the relay wired "backwards" so 30 is the output?
A lot of people did that, so you can power the pump through the NC contacts. GM even left a little red wire dangling underhood so you could do that.
In reply to Pete. (l33t FS) :
According to the wiring diagram, no. Pin 30 is attached to battery positive and pin 4 on the ECU. But who knows what they've actually done.
As I look at this, though, I'm starting to wonder if I'm looking in the wrong place... There is a box with fuses and relays in it on the right side of the car. One is labeled "ABS" and the other is labeled "Pump". I wonder if that isn't the fuel pump and there's another pump in the car somewhere. Like the hydraulic pump for the top. Maybe I should trust the wiring diagram a little more.
wae said:In reply to Pete. (l33t FS) :
According to the wiring diagram, no. Pin 30 is attached to battery positive and pin 4 on the ECU. But who knows what they've actually done.
As I look at this, though, I'm starting to wonder if I'm looking in the wrong place... There is a box with fuses and relays in it on the right side of the car. One is labeled "ABS" and the other is labeled "Pump". I wonder if that isn't the fuel pump and there's another pump in the car somewhere. Like the hydraulic pump for the top. Maybe I should trust the wiring diagram a little more.
I hope it is something that simple.
eastsideTim said:wae said:In reply to Pete. (l33t FS) :
According to the wiring diagram, no. Pin 30 is attached to battery positive and pin 4 on the ECU. But who knows what they've actually done.
As I look at this, though, I'm starting to wonder if I'm looking in the wrong place... There is a box with fuses and relays in it on the right side of the car. One is labeled "ABS" and the other is labeled "Pump". I wonder if that isn't the fuel pump and there's another pump in the car somewhere. Like the hydraulic pump for the top. Maybe I should trust the wiring diagram a little more.
I hope it is something that simple.
Just like working IT helpdesk back in the 90's... The problem is located between the keyboard and the back of the chair.....
In reply to wae :
That's not the relay for the fuel pump. That's ABS stuff.
The fuel pump and system relays are in the car, RH side footwell, behind the carpet.
My guess is, you don't have a hall sensor signal, which the ECU needs to see before it'll pull down the pump relay.
If you turn the key to the Run position, you should see the tach needle bounce just a little bit, once. If you don't see that, you most likely have a sensor issue. If the car is a Turbo, the hall sensor is in the distributor. If NA, it's got a crank sensor (at the pulley).
procainestart said:In reply to wae :
That's not the relay for the fuel pump. That's ABS stuff.
The fuel pump and system relays are in the car, RH side footwell, behind the carpet.
My guess is, you don't have a hall sensor signal, which the ECU needs to see before it'll pull down the pump relay.
If you turn the key to the Run position, you should see the tach needle bounce just a little bit, once. If you don't see that, you most likely have a sensor issue. If the car is a Turbo, the hall sensor is in the distributor. If NA, it's got a crank sensor (at the pulley).
^^ What he said about the relay location ^^
The hall sensor worked last summer when the car was last (very briefly) run.
Don't rely on that ignition switch to do anything meaningful. How are you getting power to various things? I apologize if I missed that earlier in the thread. When I was messing with it, I had unhooked the leads from the electrical portion of the ignition switch under the center console, connected the ones together that were necessary, and was cranking it from the diagnostic port under the hood.
In reply to procainestart :
I am getting a little bounce out of the tach signal, so I'm hoping that means that the ECU is seeing the engine turn. I just need to maybe go probe the correct relay this time and it should all become much less mysterious.
In my defense, every other car I've done fuel pump on, the relay was under the hood. So when I opened the hood and saw "Pump Relay", I pretty much just went for it. And in the picture in the FSM, the fuel system arrow for the relay points to a very similar place on the car that the ABS diagram points to for the ABS relays.
In reply to Powar :
I've got a strong clamp on all the "drive" position wires and then I'm clipping it to the battery voltage wire with a larger gauge jumper wire. To start, I've got another jumper wire that goes from the battery voltage wire to the starter relay wire from the starter as well. There's no significant voltage drop, so that should be good. When the pump didn't fire under cranking, I assumed either a bad ground or a bad relay and I think I'll find one or both of those at fault once I start checking the correct circuit.
It is a little odd, though, that the fuel sender isn't working either. But for all I know that's been broken for 30 years, so that may or may not be a symptom. I'll know more when I can take the kid out there again to chase down the correct relay.
In reply to wae :
Bounce is good -- you've got spark.
Ha! No defense needed: it's an old Saab, before they stopped being weird -- how can you be expected to know exactly how weird? ;-)
This works backwards, by the way: most of my wrenching experience is on 99s/900s. So when I work on normal cars, I'm often bewildered. It's pretty humbling, and sometimes frustrating, when most of what you know comes from pre-OBD cars with the engine in backwards, longitudinal FWD, chain-driven transmission, double-wishbone suspension, ignition on the floor...
wae said:In reply to Powar :
It is a little odd, though, that the fuel sender isn't working either. But for all I know that's been broken for 30 years, so that may or may not be a symptom. I'll know more when I can take the kid out there again to chase down the correct relay.
These break. You can sometimes epoxy them back together.
It may also be at the dash. The electricity is passed via a flexible plastic sheet with copper bonded to it. There are two press-fit plastic connectors on either side -- sometimes, they come loose. Could be some other failure or loose connection in there, too. Honestly, I would give the top of the dash a good Fonzerelli smack to see if it wakes up. This is how I avoid actually fixing it in one of my cars. But, if the sensor itself makes a decent rattling sound, then it's come apart inside and it's probably done.
In reply to procainestart :
That description sounds a lot like VWs from the era.
I have seen on three occasions where a terminal in the comb connector cracked and was not making good contact.
Bill, still got that bottle of Stabilant? 
We had a few minutes after school today to go out and check things out. We learned a very important thing in our 90ish minutes at the shop, my daughter and I. Apparently, the ability to use a multimeter is not one of those things that you're born with.
The first troubleshooting step that we took was that I had her put the multimeter on the fuel pump connector to see if there was voltage there when I cranked the engine. She reported no voltage. So that's where this whole troubleshooting adventure began. This afternoon, instead of digging for the fuel pump relay, we just went to the #30 fuse - which gets its input voltage from the relay - and checked for power there when cranking. Check.
Next, we connected a +12V to the "output" side of the fuse and listened for the pump. Nothing.
So we went to the pump and checked the ground for continuity from the earthing point to the fuel pump connecter. Check.
Next we jumpered the fuse again as before and checked from the fuel pump connector to the earth point. +12 volts.
So now we rather confusingly proved out that the wiring was fine. I used aligator clips on the connector to connect the multimeter and had her watch it again. Now we have a solid connection and we were getting good voltage when cranking. The problem was that she wasn't getting the probes up against the metal of the terminal. Lesson learned.
Now we were left with the suspicion that the fuel pump was Never Ever Worked. We applied direct batter voltage to the connector on the canister and got nothing. So out came the canister again.
With the canister out, we checked continuity between the plug on the outside top and the actual fuel pump itself, starting with the ground.
Nothing.
It turns out that the top of the canister is a male connector on the outside that goes to the car harness and another male connector on the inside that goes down to the fuel pump. Pulling that internal connector off revealed a ton of corrosion on the pins. After hitting it with some sandpaper and dabbing a little Stabilant on it, we now have continuity from the top of the canister all the way to the pump. After re-installing the pump we gave it another attempt.
Good news! The fuel pump pumped!
Bad news! The fuel line shot out of the top of the canister and it sprayed gasoline everywhere.
Upon looking at some pictures, I think there's another bit of the canister missing maybe. I think there might be a piece that is supposed to screw on to a boss on the top and then hold the pump and return lines in place. It should be fairly easy to make something to replace whatever was there originally. I just need to look at these pictures a little more closely and I should be able to figure it out.
So good news, overall! I should be able to get something made tomorrow and we can try again to fire it up, this time without the shower!
That's the top of the fuel canister that needs to have some sort of bracket. After a little bit of paperboard aided engineering:
And with a trim screw from a Ford Focus:
Good fix on that pump and an even better job getting the kid out to run the multimeter. I've probed those same leads on my own SAABs more than once and know it's easy to get wrong.
After school we ran out to the shop. It took a few attempts but it started up! We even drove it around the parking lot a little bit so we could put it nose-in and get the lift under it.
It was bogging a little bit and it's got the check engine light on, but those are problems for later.
Tonight I'm going to get an order together for tune up parts, oil filter, belts, and hoses.
Mezzanine said:Good fix on that pump and an even better job getting the kid out to run the multimeter. I've probed those same leads on my own SAABs more than once and know it's easy to get wrong.
Other than making that bracket, I am making a point to not do any work on the car without the kid. Not that I expect that this is going to become a passion for her or anything, but I want her to at least have some basic understanding. If that becomes something that she enjoys as a hobby, great. But I just don't want her to be completely unable to do any sort of troubleshooting of anything.
Note to myself for later: The top didn't want to operate. It looks like the system won't work if the handbrake isn't on. The indicator in the gauge cluster wasn't illuminated even when the brake was on, so the first thing to look at is the switch for the handbrake under the console. That should be a two-fer.
The last weekend was the Senior Retreat, so my daughter wasn't available to come out and do any work on the Saab. The only free time I had on Monday was a little bit during the day so we ran over to eastside Tim's place to work on Operation Swappatire. The Saab needs some new 15" rubber. Tim's S10 is sitting on some 5 year old 15" Hankooks (I think?) with about 3 miles on them or something. He has these 16" rims that had their tires get a little crispy so he wanted to get new 16" rubber for the S10, mount them on the bigger rims, and put those on the truck. So his offer was that I could have the tires if I'd take care of getting the burned up tires off the rim and the new ones on. So we met up at the junkyard on Saturday to help cut a section out of an Integra and then we transferred the 15" and old 16". We went over to his place on Monday to pick up the new 16" tires and I ran them to the local tire store yesterday morning to have the new rubber mounted and balanced and some new valve stems installed. The two remaining parts of the operation are to get the new rubber over to Tim and to have the 15" tires unmounted from his wheels, mounted to the Saab wheels, and then his wheels returned to him.
I have also apparently finally solved the issue with FedEx not wanting to just leave my packages at the shop! On Monday and Tuesday, they dropped off some boxes that have new plugs, wires, rotor, cap, belts, wipers, and an oil filter or two. But the first thing I want to take care of is the ignition switch.
We got in there last night and hooked the switch back up just to see in what way is is failed. It seems like there are at least some connections because we were able to get it to power the car on and even turn the starter if we wiggled things just right. It doesn't snap back from the start position, though. So we got the trim all pulled so we can pull the thing out of there and take a look:
Saab uses these weird security nuts to hold the assembly down to the tunnel and I had expected to be able to use a hammer and a drift to make those turn. No such luck. So I started hacking up an old 11mm socket to make the tool that will grasp that nut and make it turn. It was getting late, though, and while the socket is *almost* there, it needs a little more clearancing so that's a job for tomorrow or Friday. I've also heard that I can take a grinder to those three rivets and pop the top off without unbolting it. And that's really tempting. But I'm going to try to make the special socket first. And when I put it back together, it's getting regular old nuts.
Apparently, getting the shifter mechanism out of an auto is a bit easier than on the mtx. There's a little set screw that comes out of the control cable, you pop off a c-clip, and then remove a couple screws and the shifter comes loose from the control cable. All that after you've been able to make your special socket to remove the three stupid bolts that hold it down. Our efforts were rewarded though:
With the shifter out, we could pull the switch and open it up. Yuck.
I don't know if it was spilled coffee or water coming in through the holes in the top, but there was a lot of corrosion in the switch.
Once it was all apart, we went and got a new spring for $0.89 at the hardware store to replace the springs that were supposed to keep the contacts in... Uh... Contact. We cleaned it up and greased it a bit and got it back together. The switch works now! Well, mostly. It's still a little sticky on the return from the start position, but it's close enough for now. I think the main spring in there is just weak from age and corrosion, but it's not the type of spring that the local hardware store carries. I may try to find something from McMaster. But it's good enough for now and I didn't have to spend $140 on a used one.
We also were able to get the parking brake switch to work correctly and the top is down! So that system is also working.
Later tonight we may go back out and see about getting the interior put back together, get the wheels off, and start changing the oil and plugs and stuff.
Post a pic of the spring. I have a bunch of random and weird ones and I can see if I have a match
In reply to jfryjfry :
I appreciate that! It's a bad news good news situation though:. Bad news is that I've got the switch all back together and didn't take pictures or measurements of the spring. Good news is that I think that working the switch more had been getting the dielectric grease spread around and that spring seems to be doing okay!
After my daughter's shift last night we went back to the shop and put in a couple hours. Whomever designed the plastic heating duct that's responsible for getting warm air in the foot well there needs to be slapped with a dead fish. There are two little nubs that completely interfere with the shifter assembly and make it frustratingly difficult to extract and impossible to install.
On the auto, you're supposed to unbolt the assembly from the floor, rotate it, separate it from the selector rod and the bracket that forms the bottom of the shifter, and then remove it. Fine. With installation being reverse of removal, of course, that means you have to line up the bottom bracket and the shifter while it's on its side. But with that stupid heating duct in the way, there's no doing. The solution I found was to mate the two vertically and carefully thread the screws up from the bottom. It's finicky, but doable. Twice, in fact, since the first time I failed to engage the rod into the bushing on the bottom of the year selector so I had to take it out and do it again.
As a theft prevention method (I assume) the have the bolts for the cover over the switch come in from the bottom with some splined rivnuts at the top. They also use those funky triangle nuts to hold the assembly down. Of course the threads on the triangle nuts couldn't be M8x1.00 - I don't have anything that is that pitch, but I suspect it's 1.25 if that exists? - so I tossed the carriage bolts and went with some regular M8x1.0 nuts and bolts to hold the shifter down. I also drilled out the bracket and threaded it for the same size so I could bolt the switch cover on from the top. In the future, if the ignition switch needs to come out, all I have to do is pull the center console, take out the three bolts on the cover plate, pull the electrics, and unscrew the two screws on the switch itself.
I also gave the console plastics a quick wash. In the future, I will probably pull the shift selector off again and slip some LEDs under the indicator panel since the existing light is a bit anemic. But for now, the two major issues are resolved: fuel pump is replaced and ignition switch is working.
There is something so satisfying about taking a part that is considered not repairable and instead of spending $140 to fix it, cleaning it up, greasing it up, and dropping under a dollar on repair parts for it. Sure, I spent some time doing that and it wouldn't make commercial sense, but it was an educational experience and rather cathartic.
Also, it looks way better with the top down:
Tires are mounted and ready to be bolted on. I also had some time during a conference call today to fix the interior light switch. It wasn't making good connections so I figured I could clean it up.
It opens up by prying the rocker out with a flat blade:
It was pretty nasty
A little brushing and some sandpaper, though and it becomes shiny copper once more
And now it works perfectly.
I also unplugged the Saab Gaurd control unit since something is drawing down the battery and I always start with alarm systems.
Good call on the SaabGuard. I've had a _lot_ of these cars and never had one of those units that was fully functional.
You can do the same sort of servicing on the window switches. I had to do the ones in my sedan recently.
You'll need to log in to post.
