Where's the diode in the fan control circuit? p1520 code 98 legacy

[gl2tosl2]

hi,

 

In short:

Where are the diodes in the fan control circuit? I can't find them.

 

In long:

This is all based on poking around with my multimeter and the wiring diagram in the haynes manual.

 

I have a 1998 legacy outback. The computer is giving a p1520 trouble code. According to this page, there's a problem with the fan control circuit. The passenger side fan (I think it's the sub fan) is not coming on.

 

If I remove the relay (inside the car), jumping the terminals turns the fan on. When the main fan is off, the control terminals at the relay both show 12 volts. When the main fan is on, the terminals show 12 and 3 volts. This leads me to believe that there is a problem with the diode in the circuit.

 

Any ideas?

 

thanks for any help.

[89Ru]

In addition to the series diode, there is also an evaporation thermoswitch in series with the blower motor relay. Since the blower motor isn't siezed, the diode is probably not toast from overcurrent, I'd check the thermoswitch first.

[gl2tosl2]

In addition to the series diode, there is also an evaporation thermoswitch in series with the blower motor relay. Since the blower motor isn't siezed, the diode is probably not toast from overcurrent, I'd check the thermoswitch first.

I just tried unplugging the switch, and no change in behavior.

[89Ru]

I just tried unplugging the switch, and no change in behavior.

 

I'm assuming you shorted across the thermoswitch connector terminals.

 

Anyway, I don't think the problem lies in either the diode or the thermoswitch since the fan is running.

 

First let me see if I understand your original post.

 

If I remove the relay (inside the car), jumping the terminals turns the fan on.

I am assuming you are jumpering the relay load terminals to turn the fan on.

 

When the main fan is off, the control terminals at the relay both show 12 volts. When the main fan is on, the terminals show 12 and 3 volts.

Is the relay in or out of the circuit here? I'm assuming in and you are measuring a 9 volt drop across the relay control side.

 

This leads me to believe that there is a problem with the diode in the circuit.

Which diode? The load side series diode or perhaps a diode inside a solid state relay, control side?

 

Check the relay's control side, if it is a coil then there should be low resistance across the coil. If it is solid state (optical control, which I doubt) then there is a diode inside the relay (perhaps this is what you are referring to in the orig. post) and your multimeter can check it in diode mode...I would expect 2 volts or so for an led-type diode.

[gl2tosl2]

I'm assuming you shorted across the thermoswitch connector terminals.

No, I just unplugged it. I didn't re-install the glove box yet, so it's still easy to get to. So I need to short two of the three connections? Which ones?

 

Anyway, I don't think the problem lies in either the diode or the thermoswitch since the fan is running.

 

First let me see if I understand your original post.

 

I am assuming you are jumpering the relay load terminals to turn the fan on.

Yes. I removed the relay and jumped the load terminals. The fan came on.

 

Is the relay in or out of the circuit here? I'm assuming in and you are measuring a 9 volt drop across the relay control side.

I took the relay out and measured the voltage on the control terminals. One showed 12 and the other showed 3. This is when the other engine fan was running.

 

Which diode? The load side series diode or perhaps a diode inside a solid state relay, control side?

I was refering to the control side. The wiring diagram I have (from the haynes manual) says that both fans should be controlled by one wire going into the computer. The diagram gives each their own diode feeding into that wire.

 

Check the relay's control side, if it is a coil then there should be low resistance across the coil. If it is solid state (optical control, which I doubt) then there is a diode inside the relay (perhaps this is what you are referring to in the orig. post) and your multimeter can check it in diode mode...I would expect 2 volts or so for an led-type diode.

It's a coil type. The control terminals have some resitance, I don't remember exactly what and can double check later, but I think it was between 100 and 200 ohms. I tested the relay by connecting the control terminals to +12 and ground and I hear/feel it click. The load terminals give no continuity when the control is open, and continuity when the control is closed.

Edited November 10, 2009 by gl2tosl2

[89Ru]

ok, sounds like the relay is good. Ignore what I posted about the load side diode (doesn't exist, the diode is control side only), sorry was in a hurry this morning. Thermoswitch is control side also.

 

The control contact that is 3V on the relay connector should go to ground through the fan speed switch in all positions except 'off'. Try connecting the relay back up and measure that same point again. You could also try shorting it to ground, the relay should close and the fan should come on. If this works, then there is a high resistance path back to ground in that return circuit, which like I said involves the fan speed switch.

 

If the a/c is working then ignore what follows. If not, then here's a bit of troubleshooting. On the thermoswitch connector there are three contacts. One is always 12V. With the thermoswitch disconnected, push the blue a/c button and you'll find another goes to 12V. Ground the remaining contact and the a/c should turn on. If this works, the diode may be bad as you have already suspected, or since the thermoswitch and fan speed control share the same return path, I would start there especially if both the a/c and the blower motor are not working. If only the blower motor isn't working, then its probably not the diode since its not part of that return path.

 

Hope this makes some sense.

[gl2tosl2]

The control contact that is 3V on the relay connector should go to ground through the fan speed switch in all positions except 'off'. Try connecting the relay back up and measure that same point again. You could also try shorting it to ground, the relay should close and the fan should come on. If this works, then there is a high resistance path back to ground in that return circuit, which like I said involves the fan speed switch.

I don't think I understand the path here. Or my wiring diagram is wrong. I'll have to try shorting the control side of that relay to ground (while the relay's plugged in) tomorrow and see what happens.

 

According to my wiring diagram, the sub fan (not turning on) and main fan (turning on) relays have the same path to ground, except each has their own diode leading into the path. So wouldn't a problem further down the path affect both of them?

 

I'll try this stuff tomorrow.

[gl2tosl2]

If the a/c is working then ignore what follows. If not, then here's a bit of troubleshooting. On the thermoswitch connector there are three contacts. One is always 12V. With the thermoswitch disconnected, push the blue a/c button and you'll find another goes to 12V. Ground the remaining contact and the a/c should turn on. If this works, the diode may be bad as you have already suspected, or since the thermoswitch and fan speed control share the same return path, I would start there especially if both the a/c and the blower motor are not working. If only the blower motor isn't working, then its probably not the diode since its not part of that return path.

 

Hope this makes some sense.

 

I drove this morning and double checked, the a/c (not so pleasant when it's 40F out) and the blower motor seem to be working fine. The CEL came on in the same drive (I'd previously cleared the code). I checked the code when I got to school, and it was the same one, p1520. I won't be able to try shorting the relay return path (with the relay in) to ground until I get home.

[89Ru]

Sorry for the confusion, I mixed up the blower motor (cabin) and cooling fans (engine).

 

The cooling fans may be two speed...my FSM wiring diagrams show multiple relays for the fans (two main fan relays, two sub fan relays) but all are earlier model years. The fan motors are three terminal, my guess is that if one terminal is at 12V then its low speed, if both terminals are 12V then its high speed (the third terminal being fixed at ground)

 

Sorry I can't answer the location of the diodes. Earlier model year FSM's say the connector is orange and draw the harness in the engine compartment.

 

P1520 Radiator Fan Relay 1 Circuit High Input

 

My guess is this code refers to sub fan relay 1

My diagram shows the control for sub fan 1 being ignition +12V on one side, and the diode assembly on the other, where the ECM can pull the diode assembly common pin (2) to ground and close the contacts on sub fan relay 1. Until that pesky diode assembly is found the next best location to probe is the ECM. Do you have that diagram?

 

My FSM does not have diagnostics for this code unfortunately.

[gl2tosl2]

Sorry I can't answer the location of the diodes. Earlier model year FSM's say the connector is orange and draw the harness in the engine compartment.

Which model years?

 

P1520 Radiator Fan Relay 1 Circuit High Input

 

My guess is this code refers to sub fan relay 1

Yes. This fan relay is inside the cabin. The other fan relays are in the engine compartment.

 

My diagram shows the control for sub fan 1 being ignition +12V on one side, and the diode assembly on the other, where the ECM can pull the diode assembly common pin (2) to ground and close the contacts on sub fan relay 1. Until that pesky diode assembly is found the next best location to probe is the ECM. Do you have that diagram?

 

My FSM does not have diagnostics for this code unfortunately.

That's also what my diagram shows. I don't have a diagram that gives me pinouts at the ECM. I could probably figure out which is the right wire with some patience, but I'd be nervous about probing around the ECM like that.

[89Ru]

'95 - '97 show the diodes near the bulkhead connectors, driver's side. I took a quick look and didn't find them. I'll try to post a picture of the harness if I can figure out how.

[gl2tosl2]

'95 - '97 show the diodes near the bulkhead connectors, driver's side. I took a quick look and didn't find them. I'll try to post a picture of the harness if I can figure out how.

That would be awesome. My wiring diagram covers "95 and later models" which I think includes up to 99. So it'd be great to know where they are on your car.

 

I'm afraid I don't know what near the bulkhead connectors means. Is that near wear the wires come into the engine compartment? Or near where the wires go to the under hood fuse panel? Or does it mean at the front where the wires go to the fans?

[89Ru]

The bulkhead is the panel that separates the cabin from the engine compartment. Not having gone that far deep in the wiring, I'm not sure where to begin looking. I would not guess that the '98 is very different from the years prior. If you don't mind I try to post the diagram late tonight. Strange that the '97 wiring diagram doesn't show the diodes but they appear on the harness drawing...

[89Ru]

Here's a diagram of the '97 outback engine harness, hope it helps some.

 

http://i444.photobucket.com/albums/qq166/89ru/97_OBW_front_wiring_harness_dwg.jpg

 

http://i444.photobucket.com/albums/qq166/89ru/97_OBW_front_wiring_harness.jpg

 

[gl2tosl2]

Here's a diagram of the '97 outback engine harness, hope it helps some.

 

http://i444.photobucket.com/albums/qq166/89ru/97_OBW_front_wiring_harness_dwg.jpg

 

http://i444.photobucket.com/albums/qq166/89ru/97_OBW_front_wiring_harness.jpg

 

 

That's amazing. Thanks for posting this. I won't get a chance to dig around for it until this weekend. I'll let you know what I find.

[gl2tosl2]

sooooo, In looking into this I noticed that the radiator let out steam from the middle. I took it to a trusted repair shop and they pressure tested it, said nothing showed up. They drove it and said that it's got a bad head gasket. They put in some of that subaru additive stuff. That didn't help. The leak is worse and it appears to be spraying coolant out of the radiator cap.

 

We've only had the car since september. Recap, 98 outback wagon, with 170k. I've done major car repairs before, but I'm not comfortable working on this engine yet and I don't have time to do major car work before december 10th (I have a big due date then). I don't have a heated garage at my disposal, so it'll be too cold for me to work on the car until spring.

 

What I'm trying to figure out is, is it worth paying the shop to work on the car, or try to come up with something else?

[johnceggleston]

has it been over heating ?

are there bubbles in the overflow tank.?

do you have black gunk in your radiator overflow bottle?

this is the first time i have ever heard of a bad head gasket engine throwing a code.

 

if the fan is not working then i would think that could cause it to over heat. and any overheating could contribute to bad head gaskets even if they started out good.

[gl2tosl2]

has it been over heating ?

are there bubbles in the overflow tank.?

do you have black gunk in your radiator overflow bottle?

this is the first time i have ever heard of a bad head gasket engine throwing a code.

 

if the fan is not working then i would think that could cause it to over heat. and any overheating could contribute to bad head gaskets even if they started out good.

I don't think the head gaskets threw the code, the fan is definitely not working properly. I'm still not positive as to why.

 

It hasn't overheated that I've noticed. My wife drives it more than I do, but she would typically notice such a thing.

 

There was a small amount of black gunk before the additive, now there's lots. I don't know if the additive would affect this.

[89Ru]

Sounds like a few issues going on

 

Radiator leaking from the core...but passes pressure test at shop...with obvious coolant spraying from the cap...a different shop to resolve conflicting info?

 

Black gunk in the overflow is exhaust byproducts in the coolant, this is classic head gasket failure symptom. Changing radiators is easy even with frozen fingers. If you do it search around for 'burping' procedure. Get a new rad cap too.

 

How does the oil look? Coolant in the oil is bad for bearings.

[gl2tosl2]

Sounds like a few issues going on

 

Radiator leaking from the core...but passes pressure test at shop...with obvious coolant spraying from the cap...a different shop to resolve conflicting info?

 

Black gunk in the overflow is exhaust byproducts in the coolant, this is classic head gasket failure symptom. Changing radiators is easy even with frozen fingers. If you do it search around for 'burping' procedure. Get a new rad cap too.

 

How does the oil look? Coolant in the oil is bad for bearings.

I just double checked so let me revise. I don't see *any* black gunk in the coolant. Nor do I see an oily residue on top of the coolant. But there is some residue on the sides of the overflow reservoir.

 

I changed the oil a couple of weeks ago and it didn't show any signs of coolant. I just checked the dipstick and it was normal fresh oil color.

 

I originally took it into the shop thinking it needed a radiator and I could save my self a few hours/trips to the parts store hassles. I've swaped out radiators in other cars before, so I'm pretty sure I could handle this one.

[johnceggleston]

ignore the head gasket part for now, address the code and fan problem. maybe the radiator and cap problem. that may be your only problem.

 

rad caps are cheap.

[gl2tosl2]

ignore the head gasket part for now, address the code and fan problem. maybe the radiator and cap problem. that may be your only problem.

 

rad caps are cheap.

update. I spoke with the shop. They tested the coolant and it had exhaust gas in it.

[gl2tosl2]

So now I'm doubting my earlier diagnostic work, and I also doubt that there is a diode in this circuit. The original fan operation I was seeing was *lo* speed of the main fan. I don't think I've seen the ECM turn on high speed this whole time.

 

So the original problem is, CEL comes on with code p1520.

 

I got a set of the service manuals from the subaru download site. The wiring diagram in there doesn't show a diode in the circuit. I can't find one that is plugged into the circuit. I've done everything short of open up the covered part of the wiring harness, I *have* managed to look at every inch of the wiring harness between the plug going to the ECM (F-45) and both fuse blocks.

 

I pulled plugs and looked at a bunch of stuff today. Basically, if I get to the connector F45 and ground the control wire, *both* fans come on. If I ground the control wire going into the cabin fuse block, both fans come on. If I pull the relay for the main fan, and ground that control wire, the sub fan comes on. I *think* this means the wiring/circuits from F-45 out to the fans is good, and that this means I should be looking back towards the ECM. Does this seem right?

 

If I removed the ECM, would I be able to open it up and inspect it, I might be able to spot if there's a burned out connector or resistor or something like that that way. I have a little bit of experience with circuits and soldering, so I might be able to spot something.

[89Ru]

Maybe swap out an ECM from a u-pick salvage yard...

[gl2tosl2]

Maybe swap out an ECM from a u-pick salvage yard...

I'm going to try to pick one up this weekend. The only thing I can come up with is maybe the ECM has a bad ground, but that doesn't seem likely to me. It'd be interesting to hook a voltmeter up to the pin while driving, but I'm not sure I can pull it off. We'll see how much time I have this weekend.