Test or Replace?

A recent Best of podcast Ray said (somewhat facetiously probably) there was too much testing going on in the car repair business, and in fact the best way to a solution was to replace rather than test.

What do you think? Too much testing, make a guess and get on w/it? Or not enough testing, b/c replacing on a guess becomes too expensive?

Me, I prefer the more testing method.

Agreed. More testing means less “parts tossing” to solve the problem.

How many people have posted here after their mechanic did $2000 worth of work and the problem is still there.

How many perfectly good starters have been replaced because of a bad battery, cable or ground?

2 Likes

If you are working on someone else’s car for money, you do very little testing and a lot of parts replacing because:

  1. If anything else fails within a month or so, you will likely be blamed, so replace everything you touch to minimize the chance of a comeback.
  2. No one wants to pay you for head-scratching time. If the customer pays $300 for you to go through the diagnostic trouble matrix and $50 for a part, they will not be happy. they would rather pay $200 for new parts and $150 for labor.
  3. Most shops are marking up parts 100% these days. Your service manager will like you better if you replace everything you touch.
  4. In my state, you have to give a price quote before you touch the car. The only way to do that without getting burned or called a thief is to quote replacement of every part related to the symptom - starter AND battery AND alternator AND ,

When I am working on my own car, I figure that if someone is going to be diagnosing the problem by replacing parts, it might as well be me. I pay less for the parts when I buy them myself. Also, replacing parts proactively (e.g. all the suspension bushings rather than just the one that feels loose) means that I am only in there tearing it apart once, and there are a whole bunch of parts that I don’t have to worry about again for a long time. Also, I don’t want to clean up an antifreeze or brake fluid mess just to do it again next year.

1 Like

Just a DIY guy, but I’d just say that it depends. A car at 100K miles or more throwing something like a P0171/74? Sure, you can jump through hoops and check wiring and O2 sensor signals and search for reasons for lean/rich. But at that point, and with those miles? Just throw in new sensor and then spend testing time after that nothing lost IMHO.

But just to go to the other end - a P0420? Absolutely check first. Cats aren’t all that cheap. Look at O2 sensors. Check for exhaust leaks between O21 and O22.

Misfire codes? IDK - how old are the plugs? Due? Just replace and see what happens. Then test. Trans codes? ALL ABOUT THE TESTING!

So I just say it’s situational.

2 Likes

If intermittent, testing will not always yield results so replace if you don’t know what else to do. Sometimes you win and sometimes you lose. Mechanics are human and get paid by the hour.

That’s not just cars. It’s pretty much everything. Started with computers decades ago where the cost to troubleshoot a computer board exceeded the cost of a new board.

1 Like

On radio show from 30 years ago? They may have been referring to spark plug wires and light bulbs.

Big change from 30 years ago would be OBDII, helping narrow down what needs to be tested or replaced.

Interesting. An ongoing thread here is about a poor accel problem on a widely sold 1990’s econobox, noticed immediately after replacing the front crankshaft seal. The dealership shop insists to start by doing a valve clearance job, for which they quoted the price. They may be thinking the problem is likely associated w/the valve timing, & they’ll likely discover the problem while doing the valve clearance job, and the fix will be easy b/c a lot of stuff will already disassembled, so a no-charge. The customer will be happy to have the problem corrected, and the shop will collect the valve clearance fee.

I think this is the one where the op has been repeatedly told to just have a shop replace the crank sensor but continued to take it to shop after shop for a diagnosis of an intermittent problem. About $150 later we would see if that was the problem or not. I rest my case. Change the dang sensor.

As a note I took mine to a shop due to a miss on hard acceleration and told them to check the crank sensor. They said all was fine until the next day the cracked sensor came apart on the way to work. $80 tow, $500 for balancer and sensor, and lost a vacation day. I expected better testing.

I used to listen to the boys shows but really it was just entertainment with some useful information between the jokes. I don’t read rays column but suspect after a while it is just trying to fill space. There is another guy on Saturday morning with much better diagnostic information but can only get him when I’m in South Dakota. Think he is out of Georgia.

The other side …

“I just spent 3 months to fix P0017 problem, now it fixed. Share my story not only to you but also anyone have same issue. As you spent, I changed solenoid, crankshaft sensor so also spent about $1500 which is not worthwhile.”