Ignore Check Engine light at your peril!!

When people had reported the Check Engine light (MIL) on with an EVAP code
(P045x; P145x), I had been advising people to avoid the hassle and expense
of getting it fixed, if they didn't have a smog check to pass.

This is a potentially /expensive/ mistake.

My attention has been called to TSB A03-001, which covers just about all
models from '98 and up. It seems that corrosion in the EVAP system can
result in an electrical short that can damage the ECM.

I have just witnessed my first instance of exactly this having occurred. A
lady with her '99 Accord is now facing a repair bill of $1,560, part of
which is replacement of the ECM. She ignored the MIL and kept driving,
luckily not having done this because of anything I told her, but just
because she didn't feel like getting it looked at.

Upshot: If the MIL comes on, and the codes have anything to do with EVAP
(anything like P045x or P145x), DON'T IGNORE IT! GET IT FIXED, or have the
EVAP electrically unplugged from its power source!

 

i don't buy this. the ecm is protected against over-voltage and dead
shorts on all inputs and outputs. a fried solenoid is either doing open
circuit or dead short - neither are going to harm the ecm, it will
simply throw a code.

i think any shop saying the ecm needs to be replaced is taking her for a
very expensive ride.

and this code is usually fixed by simply replacing the gas cap.

 

i don't buy this. the ecm is protected against over-voltage and dead
shorts on all inputs and outputs. a fried solenoid is either doing open
circuit or dead short - neither are going to harm the ecm, it will
simply throw a code.

i think any shop saying the ecm needs to be replaced is taking her for a
very expensive ride.

and this code is usually fixed by simply replacing the gas cap.

 

i don't buy this. the ecm is protected against over-voltage and dead
shorts on all inputs and outputs. a fried solenoid is either doing open
circuit or dead short - neither are going to harm the ecm, it will
simply throw a code.[/QUOTE]

Well, now, you're assuming Honda didn't screw something up in the design
or manufacture of the ECMs from '98 on up.

Right now, that's a HUGE assumption. Let's see, it was the '98 model V6
four speed transmissions that started the whole "Honda can't build a
transmission to save its life, makes Chrysler look like geniuses" thing
that went on for 7 years and across two different models of transmission.

I'm with you that a properly designed and built ECM is protected, but a
shitty piece that saved somebody a half a penny per unit? That would be
Honda, from '98 on up.

 

i don't buy this. the ecm is protected against over-voltage and dead
shorts on all inputs and outputs. a fried solenoid is either doing open
circuit or dead short - neither are going to harm the ecm, it will
simply throw a code.[/QUOTE]

Well, now, you're assuming Honda didn't screw something up in the design
or manufacture of the ECMs from '98 on up.

Right now, that's a HUGE assumption. Let's see, it was the '98 model V6
four speed transmissions that started the whole "Honda can't build a
transmission to save its life, makes Chrysler look like geniuses" thing
that went on for 7 years and across two different models of transmission.

I'm with you that a properly designed and built ECM is protected, but a
shitty piece that saved somebody a half a penny per unit? That would be
Honda, from '98 on up.

 

Not in this case! American Honda very specifically says so in TSB A03-011.

Quote:
"The EVAP bypass solenoid valve can fail due to
corrosion. The solenoid valve may get water inside. If
the water contains road salt, the solenoid windings
could corrode, causing the valve to fail. In a few rare
instances, the corrosion could be severe enough to
cause an internal short in the solenoid valve, which
could damage the ECM/PCM. If this happens, both the
bypass solenoid valve and the ECM/PCM would need
to be replaced.
Vehicles driven in the Northeastern part of the U.S. are
more likely to have this problem because of the salting
of roads during the winter months. Vehicles driven
where salt is not used on the roads are much less likely
to have this problem."


They say "a few rare cases". Well this was one of them.

In this case, it can indeed kill the ECM.

 

Not in this case! American Honda very specifically says so in TSB A03-011.

Quote:
"The EVAP bypass solenoid valve can fail due to
corrosion. The solenoid valve may get water inside. If
the water contains road salt, the solenoid windings
could corrode, causing the valve to fail. In a few rare
instances, the corrosion could be severe enough to
cause an internal short in the solenoid valve, which
could damage the ECM/PCM. If this happens, both the
bypass solenoid valve and the ECM/PCM would need
to be replaced.
Vehicles driven in the Northeastern part of the U.S. are
more likely to have this problem because of the salting
of roads during the winter months. Vehicles driven
where salt is not used on the roads are much less likely
to have this problem."


They say "a few rare cases". Well this was one of them.

In this case, it can indeed kill the ECM.

 

Sorry, typo. The correct TSB number is A03-001.

 

Sorry, typo. The correct TSB number is A03-001.

 

i don't disbelieve the existence of the tsb, but several of my friends
are embedded systems engineers. the probability of failure for an
engine module like this, being as survival of dead shorts and open
circuits - the two outcomes of solenoid failure - is built in from day
one, is next to zero. far smaller than the likelihood of misdiagnosis
and resorting to "we can't figure out why the code keeps setting so it's
got to be the computer".

just like the diagnosis in tsb 97-025, they blame the thermostat because
whoever wrote it didn't bother to do two fundamental things:

1. understand the computer logic that goes into energizing the lockup
solenoid - several conditions need to be met - one of them being that
the gear selector switch is making contact.

2. deal with the logic of this being an issue only on one of their
automatics, not all the vehicles with this exact same thermostat. the
transmission selector switch is the only differentiator between the
conventional auto, the cvt auto, and the stick. if it really was the
thermostat, /all/ vehicles would be affected. fix the switch and the
problem disappears immediately and permanently, even with what was
previously a "defective" thermostat.


getting back to this case, i have one of these accords. i have
experience with this exact issue, and i'm telling you for fact - the
code sets each time the gas cap loosens. why it loosens, i don't know,
but it does. fix the cap, and your codes disappear - no broken or
leaking solenoid, and no new ecm.

 

i don't disbelieve the existence of the tsb, but several of my friends
are embedded systems engineers. the probability of failure for an
engine module like this, being as survival of dead shorts and open
circuits - the two outcomes of solenoid failure - is built in from day
one, is next to zero. far smaller than the likelihood of misdiagnosis
and resorting to "we can't figure out why the code keeps setting so it's
got to be the computer".

just like the diagnosis in tsb 97-025, they blame the thermostat because
whoever wrote it didn't bother to do two fundamental things:

1. understand the computer logic that goes into energizing the lockup
solenoid - several conditions need to be met - one of them being that
the gear selector switch is making contact.

2. deal with the logic of this being an issue only on one of their
automatics, not all the vehicles with this exact same thermostat. the
transmission selector switch is the only differentiator between the
conventional auto, the cvt auto, and the stick. if it really was the
thermostat, /all/ vehicles would be affected. fix the switch and the
problem disappears immediately and permanently, even with what was
previously a "defective" thermostat.


getting back to this case, i have one of these accords. i have
experience with this exact issue, and i'm telling you for fact - the
code sets each time the gas cap loosens. why it loosens, i don't know,
but it does. fix the cap, and your codes disappear - no broken or
leaking solenoid, and no new ecm.

 

if there is any issue, and i assign a very low probability to that, it's
hardware non-conformance, not design.


regarding the transmissions - i don't think there's much wrong with the
mechanical design per se, but i think the bean counters royally
misunderestimated the effects of their manufacturing execution.

i believe the problem is that they switched from carburized to flame
hardened gears - the latter being much cheaper to make. but these
cheaper gears are also incapable of making the same hardness on the
running surfaces, thus they spall, resultant swarf clogs the cooler
channels, and then the hydraulics fail. spalling of these gears is a
known issue, so i don't believe this was an engineering oversight,
purely a financial decision. and one i suspect that ties in to another
classic bean counter hot button - that of cars "lasting too long" -
because their customer service on the problem has been so bad and ties
in with a fundamental shift in honda attitude.

back in the day, honda's management understood that customers were loyal
because they were happy with the fact that their old honda had never let
them down. these days, business management schools don't teach about
the value of brand loyalty, just about how to calculate increased profit
if turnover can be increased by reducing vehicle lifespan. thus the
literalistic bean counter has a double incentive to mandate a known
defective transmission - cheaper to build, doesn't last, make the
vehicle uneconomic to repair by shafting the customer on the price of
the new transmission and by keeping spare parts off the market, so they
get to sell another vehicle. they think.

gross miscalculation. just like when they stiffed the previously
die-hard "enthusiast" market with the macpherson civics. even if they
fix their mistakes today, it'll take a decade, if ever, before they get
brand loyalty back. and if kia/hyundai ever release a hatch with
wishbones and engine options, they never will.

 

if there is any issue, and i assign a very low probability to that, it's
hardware non-conformance, not design.


regarding the transmissions - i don't think there's much wrong with the
mechanical design per se, but i think the bean counters royally
misunderestimated the effects of their manufacturing execution.

i believe the problem is that they switched from carburized to flame
hardened gears - the latter being much cheaper to make. but these
cheaper gears are also incapable of making the same hardness on the
running surfaces, thus they spall, resultant swarf clogs the cooler
channels, and then the hydraulics fail. spalling of these gears is a
known issue, so i don't believe this was an engineering oversight,
purely a financial decision. and one i suspect that ties in to another
classic bean counter hot button - that of cars "lasting too long" -
because their customer service on the problem has been so bad and ties
in with a fundamental shift in honda attitude.

back in the day, honda's management understood that customers were loyal
because they were happy with the fact that their old honda had never let
them down. these days, business management schools don't teach about
the value of brand loyalty, just about how to calculate increased profit
if turnover can be increased by reducing vehicle lifespan. thus the
literalistic bean counter has a double incentive to mandate a known
defective transmission - cheaper to build, doesn't last, make the
vehicle uneconomic to repair by shafting the customer on the price of
the new transmission and by keeping spare parts off the market, so they
get to sell another vehicle. they think.

gross miscalculation. just like when they stiffed the previously
die-hard "enthusiast" market with the macpherson civics. even if they
fix their mistakes today, it'll take a decade, if ever, before they get
brand loyalty back. and if kia/hyundai ever release a hatch with
wishbones and engine options, they never will.

 

That's the polite way of saying that they ignored the engineering
recommendations when it came to the manufacturing part of the
process--and no doubt because it saved them a buck or two per unit.

Making a mistake is one thing; what counts is how you recover from that
mistake.

Honda didn't recover from that mistake. Instead, they stuck their heads
in the sand for years and tried to ignore it.

I'm sure the engineers specified a material with a certain hardness such
that it behaved a certain way over time, and the beancounters--having at
the time just recently been given free reign to "make us more
money"--felt comfortable ignoring those engineering specifications. I'm
sure the beancounters discovered the magic "will it last 3 years through
the warranty?" specification that saved them a buck or two per unit, and
went with that.

And in the end, their choice has cost them dearly.

Honda engineers are (or used to be, anyway) brilliant. Ignore them at
your own peril.

yep.

What's the Genesis coupe like?

 

That's the polite way of saying that they ignored the engineering
recommendations when it came to the manufacturing part of the
process--and no doubt because it saved them a buck or two per unit.

Making a mistake is one thing; what counts is how you recover from that
mistake.

Honda didn't recover from that mistake. Instead, they stuck their heads
in the sand for years and tried to ignore it.

I'm sure the engineers specified a material with a certain hardness such
that it behaved a certain way over time, and the beancounters--having at
the time just recently been given free reign to "make us more
money"--felt comfortable ignoring those engineering specifications. I'm
sure the beancounters discovered the magic "will it last 3 years through
the warranty?" specification that saved them a buck or two per unit, and
went with that.

And in the end, their choice has cost them dearly.

Honda engineers are (or used to be, anyway) brilliant. Ignore them at
your own peril.

yep.

What's the Genesis coupe like?

 

to be fair, engineers and bean counter work hand in hand - and always
have. from an engineer's perspective, there's not much technical
challenge in making something that works - the challenge is making it
CHEAP but still able to survive the design objectives. if that design
objective includes a "bathtub curve", i.e. life limitation, the tech
challenge becomes significant and you have to throw substantial r&d at it.

it's ironic that saving money costs more don't you think?

i don't think it was a mistake - i think it was a business decision.
mistakes get recalled and properly fixed. this has been an exercise in
"customer re-education", i.e. trying to get honda customers to align
their expectations with detroit customers. afterall, detroit customers
/expect/ their transmission to fail after a while - they've been
brainwashed into thinking it's an ok routine maintenance item, and they
open their wallets accordingly. if honda can realign their customer
expectations to match detroit, honda think they can tap into the
millions of dollars a year detroit makes selling "routine" transmission
replacement. and honda are trying to take it a step further by
emulating bmw, and keeping replacement parts off the market. you can't
buy a honda transmission now - you can only exchange it, which keeps
aftermarket rebuilders out of the game, stops rebuilders improving the
build and re-selling transmissions that last properly.

it's a little too large for my taste, and coming in at $22k base with
macpherson struts, it's not there for anyone wanting to have fun - it
smells of drivers in their 50's who can't afford a bmw.

the old civic/crx concept put honda well and truly on the map and worked
for drivers of all ages. cheap to get into, cheap to run, highly
reliable, and even though it wasn't particularly powerful in stock
config, fun to drive. but because the basic platform was good, and
because more powerful engine options were bolt-in's, the civic platform
dominated the enthusiast market for nearly two decades and thus
generated huge brand loyalty when their drivers graduated to
newer/bigger/more profitable models. when honda dumped wishbones, you
couldn't make the civic platform a decent handling car even if you
wanted to, power then became the differentiator - and subaru [and
mitsubishi] ate their lunch.

honda seem finally to have paid a little attention with the crz, but
it's pretty freakin' bland frankly. it's much too expensive, doesn't
handle, and shows they're still not getting what was a real simple
formula - buy a base civic for $13k. put $10k into
engine/suspension/brake improvements, and for $23k, you have a car that
handles, goes, and more importantly ensures you and your family buy
another of the same brand. [recognition of this is where the toyota
"scion" brand originated.] today, you spend $23k on a honda and you
have nothing and there's nothing you can do with it even if you wanted
to. who is going to be loyal to that?

honda's only relief is that hyundai/kia don't seem to get it either.
they day they do, it's game over. subaru, mitsubishi and toyota will
all eat it too.

 

to be fair, engineers and bean counter work hand in hand - and always
have. from an engineer's perspective, there's not much technical
challenge in making something that works - the challenge is making it
CHEAP but still able to survive the design objectives. if that design
objective includes a "bathtub curve", i.e. life limitation, the tech
challenge becomes significant and you have to throw substantial r&d at it.

it's ironic that saving money costs more don't you think?

i don't think it was a mistake - i think it was a business decision.
mistakes get recalled and properly fixed. this has been an exercise in
"customer re-education", i.e. trying to get honda customers to align
their expectations with detroit customers. afterall, detroit customers
/expect/ their transmission to fail after a while - they've been
brainwashed into thinking it's an ok routine maintenance item, and they
open their wallets accordingly. if honda can realign their customer
expectations to match detroit, honda think they can tap into the
millions of dollars a year detroit makes selling "routine" transmission
replacement. and honda are trying to take it a step further by
emulating bmw, and keeping replacement parts off the market. you can't
buy a honda transmission now - you can only exchange it, which keeps
aftermarket rebuilders out of the game, stops rebuilders improving the
build and re-selling transmissions that last properly.

it's a little too large for my taste, and coming in at $22k base with
macpherson struts, it's not there for anyone wanting to have fun - it
smells of drivers in their 50's who can't afford a bmw.

the old civic/crx concept put honda well and truly on the map and worked
for drivers of all ages. cheap to get into, cheap to run, highly
reliable, and even though it wasn't particularly powerful in stock
config, fun to drive. but because the basic platform was good, and
because more powerful engine options were bolt-in's, the civic platform
dominated the enthusiast market for nearly two decades and thus
generated huge brand loyalty when their drivers graduated to
newer/bigger/more profitable models. when honda dumped wishbones, you
couldn't make the civic platform a decent handling car even if you
wanted to, power then became the differentiator - and subaru [and
mitsubishi] ate their lunch.

honda seem finally to have paid a little attention with the crz, but
it's pretty freakin' bland frankly. it's much too expensive, doesn't
handle, and shows they're still not getting what was a real simple
formula - buy a base civic for $13k. put $10k into
engine/suspension/brake improvements, and for $23k, you have a car that
handles, goes, and more importantly ensures you and your family buy
another of the same brand. [recognition of this is where the toyota
"scion" brand originated.] today, you spend $23k on a honda and you
have nothing and there's nothing you can do with it even if you wanted
to. who is going to be loyal to that?

honda's only relief is that hyundai/kia don't seem to get it either.
they day they do, it's game over. subaru, mitsubishi and toyota will
all eat it too.

 

give me a Fit Si.

It's more fun to drive a slow car fast...

 

give me a Fit Si.

It's more fun to drive a slow car fast...

 

macpherson front and torsion beam rear won't /allow/ you to drive it
fast dude. and the si has the same output as the base, only it's
heavier with all the accessories...

i'm hanging on to my 89 civic hatches for my fun wheels for the time
being.

http://www.indyscca.org/SoloFiles/SoloResults/2011/FunEvent/2011_indy_fun_event_040311_raw.htm

couple in the top 10, with some serious dough placing well behind...