Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Cutting corners on automotive engines 9

Status
Not open for further replies.

enginesrus

Mechanical
Aug 30, 2003
1,003
0
0
US
Like the crankshaft snout that has no keyway cut into it, and requires a nice tight bolt to push on the harmonic balancer to keep the timing gear from rotating and throwing the valve actuating system out of time.
And best of all the bolt has a right hand thread (clock wise to tighten) and that end of the crankshaft turns the same direction, so you have not only dynamic movement from engine and firing pulses, but also any belt driven accessories like alternator, power steering pump and what ever else is belt driven tending to loosen that bolt. I personally don't think you would ever find such idiocy on an aircraft.
 
Replies continue below

Recommended for you

As a sparky, I wonder how the lamp can back-feed when the light switch is off. I am not saying that it is not possible, but in the days before computers, it would not be possible.
I don't remember the lamp being lit, 24/7

--------------------
Ohm's law
Not just a good idea;
It's the LAW!
 
I have my doubts that trying to backfeed the license plate lamp will work on a modern vehicle in which all the bulbs including that one are switched electronically through a lighting module or body-control module. It might work on an old skool vehicle with a hard-wired switch that you've left on.

The electronics designed to feed power to that bulb, probably don't conduct electricity backwards no matter what you try.
The signal that commands the bulb (or any other switched device including any relay-switched device) to turn on, is probably dead along with the battery, because it needs power from that battery to do that.
The thing that you need to back-feed power to (the power door lock circuit) is extremely unlikely to be on the same circuit as the license plate bulb. In fact, given the need for the power door lock circuit to function regardless of whether the lights are switched on or off, it's impossible for it to be on the same switched circuit.
 
hpaircraft said:
Getting back to the OP, an example of a keyless crankshaft disaster.


Pro tip: Click Settings and set playback speed to 1.5

That is a stupid design, but not necessarily the lack of a keyway makes it stupid ... the lack of a taper-lock, or some other similar keyless locking device that can function independently of whether the front pulley bolt is properly torqued, makes it stupid.

No mechanic expects the cam timing to jump simply by removing the center bolt for the front pulley!

VW diesel engine cam sprockets are on a taper, no keyway. They don't move if you remove the center bolt - they need a knock from the backside to free the taper. I wouldn't dare start the engine without that center bolt torqued, but the engine doesn't jump time by removing it.
 
waross said:
As a sparky, I wonder how the lamp can back-feed when the light switch is off. I am not saying that it is not possible, but in the days before computers, it would not be possible.
I don't remember the lamp being lit, 24/7

I’d assume that that light is operated by a relay that gets a signal from a body control module somewhere that says “turn me on.” And when you back feed into the bulb, your putting power into the “always hot” side of the system that the relays switch on and off. That gets power back to the brain bits as well.

Dunno. Seems like a hell of a thing to make up a YouTube video about if it doesn’t work. I’ve read the same thing elsewhere as well.

Here’s another video describing the same procedure:

And a Reddit thread where it worked:
 
On the other hand, we industrial sparkies never switch a neutral. We never even consider switching a neutral.
As a result, we assume that automotive switches are on the "Hot" side. That may not be the case,
I had an old Subaru that ran a common hot wire or fused hot wires to a lot of places an switched the grounded return.
It takes an extra wire, but that may be needed in any event for burn-out detection of lamps.
It's worth a try.
Just now the temperature is minus 17 C with a wind chill of minus 29 C.
Add to that, the weather channel has an advisory:
"EXTREME COLD TO COME:"
I may not be out in that field for some time to come.
Thinking about it some more:
If the switching is done by the BCM, then the BCM may monitor its own outputs.
If an output doesn't present 12 Volts when in the off state that would indicate a probable failed lamp or blown fuse.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!
 
BrianPeterson said:
The electronics designed to feed power to that bulb, probably don't conduct electricity backwards no matter what you try.
Spartan5 said:
I’d assume that that light is operated by a relay that gets a signal from a body control module somewhere that says “turn me on.”
I'd say these days it's likely to be a FET rather than a relay... and the body diode on a FET will allow for reverse current, so that may still allow the "apply power to light socket" trick to work.

Dan - Owner
Footwell%20Animation%20Tiny.gif
 
TugboatEng said:
HPaircraft, there is only a problem because they didn't read the manual. I wouldn't consider that an engineering disaster.

Yeah, no. I write manuals for a living, and this one gets a pass from me.

I also do a lot of DIY car work, and on a recent job I had to remove and replace the timing cover on a Toyota 2AZ-FE motor. I also installed thread inserts for the head bolts and replaced the head gasket, water pump, and serpentine tensioner, but that's an AZ motor tale for a different thread.

In the automotive world, there is a realm of standard practice that mechanics can reasonably expect to encounter and deal with. Mechanics can reasonably expect anything that falls outside that realm to come with a warning label. That's just how shit gets done in the real world, and for the most part it works fine.

When I did that 2AZ-FE head gasket, I was able to accomplish it as expeditiously as FWD packaging allows because it was all within the realm of standard practice. I only had to look at the manual a couple times to get the general sequence, check for gotchas, and learn the important torque values. The two places that fell outside of the realm of standard practice had warning stickers stuck to the engine that warned not to touch this sensor, and to use only Iridium plugs. It even had painted links on the timing chain that, when aligned with the marked teeth on the cam and crank sprockets, resulted in correct cam timing. Oh, and the ignition trigger wheel had a big F on the front face to help you put that on right. The drive sprockets for the cam and oil pump chains, as well as the trigger wheel, were all indexed by a single keyway and key that kept everything in line.

And, yes, FoMoCo has probably covered its ass by documenting in the FSM for their motor that the cam drive is secured by the clamping force of the harmonic balancer bolt. Yay Henry. But by not having a simple warning sticker on the timing cover near the balancer, they have left a lot of mechanics and DIYers out in the cold.
 
I only had to look at the manual a couple times to get the general sequence, check for gotchas, and learn the important torque values.

Why'd you look for gotchas if what you were doing is standard practice and you not supposed to have to look?
 
No normal mechanic who expects to earn a living is going to studiously check every step in the service manual before doing it on every job no matter how small, and that's the level of diligence it would take to avoid disaster in that case. Perhaps a Ford specialist mechanic might have been to a training course on that engine and known never to remove that bolt without first removing the valve cover and locking everything in place ... but a normal general purpose mechanic or DIY faced with a bad balancer is going to knock that bolt out, swap the part, then check the book for the torque spec. Oops.

Manufacturers have set times for performing procedures. Good luck achieving those if every step has to be carefully looked up beforehand to avoid disaster.

Bad design. Taperlock would have avoided this. Assembly sequence designed in a way that encouraged the right way to do it is another way but it's hard to stop someone from removing a bolt. Warning label bare minimum but doesn't account for language barriers or incomprehensible symbols.
 
LOL, whatever. If you can' be bothered to read the service manual to ensure you do it right then that's on you and not the manufacturer, because they documented how to do the repair so you could do it correctly.

I highly doubt it would require very much effort looking at the manual before you knew there were special steps involved.
 
Decals are for manufacturing, safety, or certification, not a how-to for service and subject to change (read: often incorrect).

FWIW, if a tech is struggling to make book time on the majority of tasks then he's a lousy tech. The system is designed around people with median skills who are following a manual and encountering common issues, not a super-tech by any stretch of the imagination.
 
Hello,

What are the opinions on engines with timing belts and pistons that have no valve relief pockets? Timing belts have a finite life and if they snap, the result is bent valves and general destruction to the top end of the engine. Is the absence of pockets to save the machining costs?

Kyle
 
There is a halfway. Changing an oil filters generally don't require extensive study of the service manual. By the time you're breaking loose a connection on the crank then it's time to do a manual review for the gotchas. Maybe the manual was well laid out, maybe it wasn't.

Safety decals are inked in blood because 'RTFM' isn't sufficient. Event when labeling the exhaust pipe 'HOT'. I disagree that the procedure change would violate the decal. The pulley wouldn't decide to stay on the shaft 'cause it's Friday and a new procedure was released.

Tech failed, engineer could have out smarted him.
 
They definitely do not. Valve pockets just mean they needed more piston crown to hit the compression number than valve clearance would allow when valve timing is correct. They don't mean that if the valve timing is wrong, things won't clash.
 
Status
Not open for further replies.
Back
Top