Symptoms of fuel pump O-ring problem?

[Oo-v-oO]

Our '02 H-6 Outback suddenly started running poorly tonight - it runs fine at idle and up to 1/8 to 1/4 throttle, but as soon as you ask for more power it just falls on its face. If you can get it to rev up to around 5K, it perks up some but still isn't running like it should. I at first suspected a plugged cat, but after driving it some I don't think this is it, as the break point between running OK and not is very distinct.

I haven't had a chance to connect a fuel pressure gauge to the fuel rail yet but I suspect that this may be a case of low fuel pressure, possibly caused by the O-ring popping out of the cap beside the fuel pump as I have seen in a couple of threads here. Most of the descriptions seem to indicate that the car just quit running, though, not running but at reduced power.

2001 subaru just shut down

Has anybody seen the infamous O-ring cause symptoms such as the ones I'm seeing, or if not any ideas where to look next?
Also, what pressure should I be seeing on the fuel rail at idle? The manual I have isn't very complete as far as diagnosis but it does call out a pressure range from the pump of 54-98 psi.

Thanks!

[monkei]

Sounds about right. Yours probably hasn't completely expelled the o-ring yet. Mine ran similar to that on the way back home from the gas station and wouldn't start after I parked it a few days back. Just fixed mine today by replacing my cracked cap (the thing that retains the o-ring) with one from a used 4-cyl pump. The good news is the fuel pump is pretty easy to access and remove to repair/replace.

[Oo-v-oO]

Well, the car wouldn't start at all for my wife this morning so she used my daughter's car to go to work. I hit seven stores on the way home - auto parts, (Napa included) hardware stores, and Lowe's - and nobody had a -138 O-ring in stock. When I got home I ordered a pack of ten Viton O-rings off of Amazon and they will be here tomorrow. Shipping was one and a half times the cost of the O-rings, but what are you going to do? I was lucky to get them overnighted, I only had 20 minutes to spare before it would have been too late to get them by tomorrow.

Looks like I get to dig into it tomorrow after work. Hopefully that's all it is.

[Oo-v-oO]

Got it apart - one of the tabs on the metal cap is cracked partway where they folded it vertical. Manufacturing defect! The O-ring did not appear to be swollen but I will replace it with a new one I got that is made of Viton so that shouldn't be a problem in the future.

Taking a break for supper now, will post some pictures I took of it later.

I am going to try to wrap a hose clamp around the tabs once I reassemble the cap, hopefully that will help it stay together.

[monkei]

Originally Posted by Oo-v-oO

I am going to try to wrap a hose clamp around the tabs once I reassemble the cap, hopefully that will help it stay together.

That worked for me for a few months, but overtime the crack grew and the o-ring pushed out again despite the hoseclamp. I would definitely plan on replacing the cap sometime soon before it leaves ya' stranded.

[Oo-v-oO]

Pretty poor pictures for the most part, but it was dark and cold. Click the picture for a large version.

Here you can see the O-ring popping out:



A couple of pictures of the bend and crack in the metal tab:









I ended up putting a hose clamp around the three tabs once it was installed on the pump assembly body. I tightened it up just enough to prevent it from slipping off. Hopefully the crack doesn't grow.

Car runs great now! This really wasn't that bad of a job. Beats crawling under the car and dropping the tank to get to the pump, like I had to do with the Bonneville my wife used to have. That was the pits.

[Oo-v-oO]

So I take it the only way to get a replacement cap is to cannibalize another pump?

If so, is the cap on a 4-cylinder the same as the cap on the pump on an H6?

[monkei]

Yep, gotta take one off of a pump assembly, unless you can find a source selling replacement caps (Lemme know if you do!). Assuming the used pump I bought on ebay was correctly described as a 2.5L pump, then yeah, the caps are the same.

[Oo-v-oO]

Just as a followup...

Car still runs great and has all the power it should, but does not start immediately if it has been sitting off for a while. I had my wife try turning the key on and off a couple of times before starting it after it had sat and it did fire right up then, so it appears that the fuel pressure is bleeding off over time.

My suspicion is that the replacement Viton #138 O-ring I put in is just a whisker smaller in cross-section than the (probably metric) original and it is allowing gas to weep past when the pump is off. The Viton is a slightly higher durometer than Buna, at 75 vs 70 and that may not be helping matters, especially with the cold temps we have been having.

I'm inclined to leave it alone for the time being and try to source a replacement cap. I kept the original O-ring and will probably put it back if I can find a cap - I really don't want to mess with the cracked one I have now if I can help it.

I searched car-part.com for a fuel pump and they didn't have any listings whatsoever.

Found a couple of pumps listed on eBay for $90 in a junkyard in VT, about a 2 hour drive from me. They did not have actual pictures of the pumps, and the listing was somewhat unclear so I sent them an E-mail asking if it was just the fuel pump itself or the entire assembly via their listed contact address the day before yesterday but have not heard any response back from them as of yet.

Next best choice I have found is from Advance Auto Parts, at $338.09. They currently have a 15% off promotion going on so that would bring it down to about $288. Free shipping, too.

Not sure what I'm going to do yet - waiting to see if I hear back from the junkyard. My wife relies on the car for work so uptime is very important.

[1 Lucky Texan]

OK, I was gonna try to draw somthing in Paint, but I'm gonna try to just describe an 'approach' to a fix.

There are 3 molded tabs on the chamber right? A screw type 'hose' clamp, positioned above them, should be secure. Now, suppose 2 more screw clamps were prepared in such a way that they were were cut apart and secured behind/under the clamp on the chamber, at 90degree intervals, but extended under the cap and crossed in an X. Then they could be tightened to pull UP on the cap. Certainly some extremely secure ways to attach everything (maybe pop rivets?), positioned so the screw mechanisms clear each other, etc. is critical. And i can only suppose the screw clamps would stay secure. Unless there are some type that can be safety wired or otherwise 'locked', there's probably still some risk. Might be at least as secure as the original design!

kinda like garter belts hanging down, then passed under the cap in an X, tightened, to pull up on the cap.

does this make sense?