Ask the Car People.SM


A Mechanic's Life - Tales From Under the Hood

2927 messages,  Last post on May 23, 2013 at 6:16 PM

You are in the Maintenance & Repair Forum.

What is this discussion about? Auto Repair, Hatchback, Truck, Sedan

     This topic is primarily for professional mechanics, current or retired, or ardent amateurs who would like to share the suprises, victories, tricks and challenges of working on the modern automobile. All Forums members are invited, of course, to ask technicians about their work, or comment on your own experiences dealing with mechanics.
 
If you have a maintenance or repair question about your vehicle, please use search to find one of our Maintenance and Repair discussions, or ask a question in Edmunds Answers.

Messages Page 66 of 294

#643 of 2927 Re: Chevrolet Cobalt, large evaporative system leak [steve_] by thecardoc3

Dec 10, 2012 (10:56 am)

Replying to: steve_ (Dec 10, 2012 10:02 am)
Then I'd go buy an OEM gas cap. If that doesn't work, I'd take it back to the dealer and get my money back (less a 20% restocking fee perhaps)
 
So then if they try to sell it to the next person and they realize that it's already been enjoyed once by someone else and they get upset because they have been sold a "used part" instead of a "new one". In that customers eye's someone is being a theif and ripping them off.
 
It's pretty simple from the point of view from the other side of the counter
 
 Yea, you might just as well keep the new cap and out your old one back in the box to return.... That's how it would look to another customer on your side of the counter. That's how it would look to you if you opened the box and found the cap you just bought had already been tried.
 
In real numbers, barely 1% of evap failures are caused by faulty gas caps, claiming anything else is intentionally misleading to the consumer. Now why would you want to do that?
 
 Loose caps are caught by the systems because today they run a specific series of tests that are triggered by an increase in the level in the fuel tank.
 
You have said numerous times that you want the cars to be able to 'self diagnose" more completely and to a certain extent that has happened. They have managed to make it pretty cut and dry how diagnostics should be performed and just how obvious blind guesses by untrained individuals are nothing more than that.
 
Feel free to admitt that you have no clue about how to really approach this anytime. The sad part is, what I use is a very simple, repeatable routine that has a technician arrive at the correct answer THE FIRST TIME, and EVERY TIME, no matter where the failure is. Guess you should be able to see why it was somewhat refreshing to have had a customer that was glad we were there to help him.

#644 of 2927 Re: Chevrolet Cobalt, large evaporative system leak [thecardoc3] by xwesx

Dec 10, 2012 (1:36 pm)

Replying to: thecardoc3 (Dec 10, 2012 4:16 am)
(Any ideas why I mentioned that right here?)
 
Make assumptions that something isn't a problem because it was already tested/replaced, and you're going to make the whole process that much more difficult for yourself.
 
What is the next step and why?
 
Assuming an understanding of the system (which you noted as step 1), I'd say that the next step is to replicate conditions that should cause that DTC to set, in order to verify that the OBD system is working correctly.
  
How would you proceed with testing this system to repair the problem?
 
Then, I'd test the parts of the system that fed into it to determine which part(s) was(were) out of spec.
 
Sorry, but since I haven't a clue how that system is looped, I can't get any more specific than that.
 
---
 
I'm curious about two things (in addition to your general solution methodology & the actual problem in this case): 1. How long did it take to diagnose, and then fix, the issue, and 2. Appx. much did the customer end up paying?
 
I'm glad to hear that the customer was appreciative in this case. He tried the other method, failed, and knew it was time for something better!

#645 of 2927 Re: Chevrolet Cobalt, large evaporative system leak [thecardoc3] by steve_ HOST

Dec 10, 2012 (2:16 pm)

Replying to: thecardoc3 (Dec 10, 2012 10:56 am)
I take "used" stuff back all the time and store repackage and resell it all the time. Go in to the back of most any store and you'll see a shrinkwrap machine they expressly use for this purpose. Bit of a red herring issue; if a parts store doesn't want parts returned, they can post a sign and not accept returns, and check boxes when they do take returns and make sure the part isn't scratched and dinged.
 
In real numbers, barely 1% of evap failures are caused by faulty gas caps, claiming anything else is intentionally misleading to the consumer. Now why would you want to do that?
 
In the real world many CEL lights come on because of loose gas caps. A $20 exercise at AutoZone or NAPA is cheaper than going to the shop.
 
If a customer called you and said their CEL was on but they couldn't stop by your shop for 3 days, would you tell them to check their gas cap and perhaps ask them when someone last filled up the car?
 
(As an aside, did you see where the most recent Ford recall for potential engine fires is going to be addressed by a software fix?).

#646 of 2927 It isn't wrong to do the job right! by thecardoc3

Dec 11, 2012 (2:43 am)

Replying to: steve_ (Dec 10, 2012 2:16 pm)
I take "used" stuff back all the time and store repackage and resell it all the time
 
Many returned items never make it back to a stores shelves. Too much of a risk of dissapointing the next customer. These items need to be sent back to the manufacturer who can inspect them, and then they must use another routine to re-release them or else just scrap them.
 
In the real world many CEL lights come on because of loose gas caps. A $20 exercise at AutoZone or NAPA is cheaper than going to the shop.
 
Not anymore they don't. My 2002 Ford Explorer has a check fuel cap light that is seperate from the check engine light. When you refuel that car (and many others just like it) the computer runs the large leak test and compares that result to stored test results from previous tests. The system then knows for certain if the newest result is different from previous test results and in the case of my Explorer would turn on the loose cap light and not store a pending code for the evap system. Now the next time the test runs and it detects a leak, then the system stores the pending code. Then only after the third time the test would find a leak it sets the gross leak code and now turns the light on.
 
Cars not equippped with the loose cap light set a specific code for a large leak, seperate from the SAE code and again it's because of weighted testing results that are compared to previous tests.
 
If a customer called you and said their CEL was on but they couldn't stop by your shop for 3 days, would you tell them to check their gas cap and perhaps ask them when someone last filled up the car?
 
No. The questions that I am going to ask are:
Is the car running normally? Are you having any trouble with it running rough, stalling, difficult to start or is or has the check engine light been flashing?
 
The moment someone touches the cap once the light has come on, the only thing that they really accomplish is make it impossible to prove if the cap's being loose caused the light to be on. To a technician it ends up being an intermittent failure condition and you can only assume the cap must have been loose, you don't actually know for certain if it was or was not.
 
By touching the cap, and now we get a "no trouble found" it's quite common for the light to come back on within the next few weeks and now it's "a second trip back to the shop" because they didn't fix it the first time.
 
BTW, with some cars that have ORVR (on-board refueling vapor recovery) systems you can leave the cap completely off and it doesn't care because the filler neck is sealed off down below closer to the tank. Cars that test themselves with pressure cannot test the filler neck with an ORVR system and it makes for a nice little puzzle for the technicians when I do as a hands on portion of an evap class that I teach. We make the monitor run with the gas cap off of the car and it passes. This does not work with cars that use vacuum to test themselves, they will catch the cap loose/missing.
 
As an aside, did you see where the most recent Ford recall for potential engine fires is going to be addressed by a software fix?.
 
No, haven't paid any attention to that at all. First we don't do recalls as an independent. The reflashes that we do but are totally software driven and pretty much a hands off operations outside of telling the computer yes, do the reflash and then exercising the ignition switch as the system commands.

#647 of 2927 Re: Chevrolet Cobalt, large evaporative system leak [xwesx] by thecardoc3

Dec 11, 2012 (3:23 am)

Replying to: xwesx (Dec 10, 2012 1:36 pm)
Assuming an understanding of the system (which you noted as step 1), I'd say that the next step is to replicate conditions that should cause that DTC to set, in order to verify that the OBD system is working correctly.
 
In a round about way that is correct. The code sets because the computer seals the system by closing the cannister vent valve and then opening the purge valve in order to pull the system into a vacuum. (about 8"-12" in water) Then the purge valve is then turned off and the computer watches the vacuum bleed up rate. By knowing the fuel level, and therefore how much space is air/vapor in the tank it can calculate if the bleed up rate is excessive (leak detected) or not (system sealed).
 
So the first step is to command the vent valve closed with the scan tools bi-directional 'controls. The vent valve will make an audible "click" when it opens and closes.
The Cobalts cannister vent valve could not be heard to be closing/opening.
That means the system cannot test itself. Two questions need to be thought of and then proven/disproven at this time.
If the problem is electrical, why isn't there a code for the vent valve's circuit?
The customer (allegedly) already replaced the vent valve so does that mean it's a mechanical issue with the "new" valve?
 
With these two questions in mind it's time to do pinpoint testing on the vent valve and it's circuit.
 
The purge valve is tested in a similar fashion, first command it to operate with the engine off and simply listen for it. If you cannot hear it, then specific testing for it, and/or it's circuit is required. If you can hear it then you need to start the engine and see if you can control the vacuum to the system. The Cobalt's purge valve passed these checks.
 
1. How long did it take to diagnose, and then fix, the issue, and 2. Appx. much did the customer end up paying?
 
The time to locate the exact failure was about half an hour, the time to actually repair the wiring harness damage that I found was about an hour, including doing the post repair testing that was required to make sure that the car would then pass it's monitor the next time that it ran. Our diagnostic/electrical labor rate is $115/hr. People often gasp at that but in reality a shop hanging brakes or doing suspension work or exhaust is realizing more profit per hour than we are with that diagnostic rate and they don't have to spend a dime to do that kind of work they way we do with scan tools, software and schools.
 
1.5hrs at $115 with the sales tax and he was out the door with the car working correctly for about $185. Consider that he had already spent over $400 while failing to fix this over the last year.
 
Now one of the things we don't do is go on a treasure hunt on the cars. I don't go over the whole car trying to find a grocery list of services to sell. We go straight in at the problem that the car came in for, and straight back out. I wouldn't be able to keep a job at a chain store working like that, I wouldn't be selling enough to keep the management happy.

#648 of 2927 Re: It isn't wrong to do the job right! [thecardoc3] by steve_ HOST

Dec 11, 2012 (7:49 am)

Replying to: thecardoc3 (Dec 11, 2012 2:43 am)
The moment someone touches the cap once the light has come on, the only thing that they really accomplish is make it impossible to prove if the cap's being loose caused the light to be on. To a technician it ends up being an intermittent failure condition and you can only assume the cap must have been loose, you don't actually know for certain if it was or was not.
 
Heaven forbid if I need to get to your shop and I'm two hours away and my fuel light just came on too.
 
The Escape recall was interesting to me because the only other reflash fixes I remember had to do with driveability issues, like shift points. Being able to reduce the risk of an engine fire with software is a pretty good trick and illustrates how the new cars aren't your father's Oldsmobile.

#649 of 2927 Re: Chevrolet Cobalt, large evaporative system leak [thecardoc3] by qbrozen

Dec 11, 2012 (8:23 am)

Replying to: thecardoc3 (Dec 11, 2012 3:23 am)
Our diagnostic/electrical labor rate is $115/hr. People often gasp at that
 
Hell, that's cheap compared to a BMW dealer.
 
I didn't get to answer your question originally, but I did just go through this on my volvo. Its a '98 and I've already had extensive experience with another '98 some years ago. Code was the massive evap leak. First thing I did was go to the dealer and buy a gas cap because, in my experience with this car, it is what was stated above: the most common issue.
 
Cleared the code and drove for about 5 days before it came back. There are 2 more common things on this car. One being the purge valve and the other being clogged vent. Before testing the valve, I decided to get underneath and inspect the vent. No clog. But then I inspected the canister and after removing some brackets in the way, I found a split air line. Removed, trimmed, reinstalled, and cleared the code again. So far so good, but I'll want to see a couple of weeks without a light before I feel I've solved it.
 
Now, I know you aren't happy with my process. Should I have jacked up the car and gone through the whole troubleshooting first? Maybe some people should. But if $24 for the gas cap saved me the 30 mins I spent on fix #2, it would have been worth it to me. And, certainly, $24 total spend is a helluva lot less than a professional shop diagnostic fee. And, no, I'm not returning the cap because I know it to be such a common failure point that I might as well keep this one and know I'm good for about 4-5 years.

#650 of 2927 Re: Chevrolet Cobalt, large evaporative system leak [thecardoc3] by xwesx

Dec 11, 2012 (11:05 am)

Replying to: thecardoc3 (Dec 11, 2012 3:23 am)
he was out the door with the car working correctly for about $185.
 
That's a great deal for him. Plus, he's a happy customer that will be back (but only when another problem comes up that stumps him). If I could have found a local mechanic that could solve my '96 Outback's issues a decade ago, I would have stumbled over myself trying to get to that place! As it was, I spent around that same $400 for no resolution at all.
 
Also, thanks for sharing more information on the process - it's very interesting stuff and clearly illustrates the investment required (time, tools, and education/training).

#651 of 2927 Re: Chevrolet Cobalt, large evaporative system leak [qbrozen] by Mr_Shiftright HOST

Dec 11, 2012 (11:10 am)

Replying to: qbrozen (Dec 11, 2012 8:23 am)
Where I live common labor rates are at $140/hr now. I'm fine with that, if they are efficient AND correct.

#652 of 2927 Re: Chevrolet Cobalt, large evaporative system leak [thecardoc3] by srs_49

Dec 11, 2012 (11:50 am)

Replying to: thecardoc3 (Dec 11, 2012 3:23 am)
...to actually repair the wiring harness damage that I found ...
 
So, what caused the damage to the wiring harness?
 
And didn't you tell us about another problem (SUV liftgate?), that people were throwing BCMs at and that turned out to be a wiring harness issue?

Messages Page 66 of 294

To POST a message, please Sign In.

Advertisement

Browse by Category

Browse by Vehicle
   View All Vehicles

Browse by Board
Browse by Topic
View All Topics

Edmunds Community

Advertisement