a604 Limp, Rebuilt, Sensors replaced, solenoid replaced still Limps!

A bad TPS is easy to see with an analog ohmmeter, just connected the
ohmmeter to the TPS, move it slowly through the range of travel. The
ohmmeter needle should track the movement exactly, if the needle jumps
around particularly if it jumps when going back and forth over the same
spot, the TPS is bad.

What retailer did the sensors come from? Unfortunately these days you
can't really blindly trust the brands, since many brand name companies
manufacturer different levels of quality depending on the retailer that
orders
them.

Ted

 

Most independents just buy "remanufactured" TCM's which are, as you
probably can guess, nothing more than existing TCM's that have been
cleaned up a bit and reflashed.

They are not. You need scan tools that are prohibitively expensive (even
used)
for the DIYer.

But in any case, his TCM is probably flashed already and if not, there's
plenty
of flashed ones in the wrecking yards. What's the part number on the side
of it?

Ted

 

The sensors came from Advance auto parts. I hadnt thought to try an
analog meter on the TPS, the Haynes manual I have shows no info on how
to test it.

 

Oh trust me with it being an intermittient problem I wont celebrate
for a bit yet, I am just glad that its behaving.

The diode I'm looking at is part of a GM harness repair kit and is at
http://www.ecklers.com/product.asp?pf_id=44831&dept_id=1135

Other than the speed sensors are there any other sensors I need to be
concerned with?

I wish i still had my oscilliscope to watch for the spike but it went
bye bye about 6 years ago and I just cant justify buying another.

 

I had virtually the same experience with a late 90's Intrepid. Don't
know what transmission model it was. But I could have your post for you.

Car was under warranty. It spent 6 weeks on and off in the Chrysler
dealership - it's fixed; it's broken; it's fixed; it's broken.... I even
drove around for one weekend with the Chrysler "Co-Pilot" sitting on the
passenger seat. Chrysler replaced all kinds of parts.

In the end, the cause was high resistance on one pin in a multipin
connector. They replaced that portion of the wiring Harness (connector
and all) and that was the end of the problem.

Phil

 

Hi Ted...

I'd respectfully suggest that you re-consider your suggestion to
replace the existing zener with "any old diode"

A zener <> any old diode.

Take care.

Ken

 

Good job of figuring this out. Really tricky. Good discussion. I would
like to ramble a little since we are now discussing analog and a touch
of digital engineering in the automotive newsgroup.

"Two (equivalent) Zeners in series and in reverse order, in the same
package, constitute a transient absorber (or Transorb, a registered
trademark). They are named for Dr. Clarence Melvin Zener of Southern
Illinois University, inventor of the device." = Wikipedia.

I vaguely remember this when I was building stuff and worried about
spikes and surges. I gather the question here is allowing the signal
through but not the spike or surge. How big is that signal? +5 volts? A
Light Emitting Diode or LED would light up but that's a bit fancy.

About the oscilloscopes, an analog might be fine here. I have seen even
100/200 MHz old Tektronix 'scopes going for very cheap and they will
get the spike, yes? Now a good digital scope might what, record over a
period so one could go eat supper and then come back. I'm wondering,
maybe a dirt cheap signal probe might get it? Guess it's back to a
good, fast digital 'scope because you might want to record and inspect
the bugger at leisure.

I guess if this signal or surge is that bad, then a small isolation
transformer inserted here? That would be overkill but they really do
the job. Nothing gets by a good isolation transformer, well, maybe a
big lightning strike. Can you put one on a signal line though?
Breadboard this to see if it works?

Where's a good analog design engineer when you need one for a car? Most
engineers are digital so look for the mature guys for the analog which
is far more complicated than digital on a byte for spike basis

 

Yes, I know that. However there is really no need for a zener in this
circuit. Most likely some engineer specced something like a 20 or
so volt zener with the idea that it would be connected anode to positive
and cathod to negative, and when the EMF collapsed in the clutch
coil the zener would clip the hundred volt or so pulse. However, a
regular diode connected anode to negative, cathod to positive, does
the same job since when the EMF collapses it will flow backwards
and in that case the diode looks like a direct short. Diodes are used
in electronic circuits in parallel with relay coils all the time for exactly
the same reason.

A more specific explanation is here:

http://experts.about.com/q/Electrical-Engineering-1356/Protection-Diode.htm

I guess the Chrysler engineer that specced it was the same guy that wrote
the question in the the electrical expert. ;-)


Ted

 

Ahh my electronics coursework was over 15 years ago so I dont problaim
to be not rusty

Basically the tip I got mentioned a plymouth breze that had the same
issue. The dealer mechanic in question didnt post to the group so I am
assuming he had his reasons. Basically he stated something to the
effect that the collapse of the A/C compressor circuit could produce a
spike as high as 600 volts. That in the case he saw it was literally
spiking all the way back to the solenoids and activating them all.

I'm going to err on the side of caution and get the zener since thats
what was originally in the cicuit apparently. It all plays out very
sensible to me that the zener could have been going bad through all of
my limp incidents. Each time the car was hot, if the zener was going
bad and opening while hot all it would have needed was one time for
the A/C to cycle off and poof.

So I guess I will go with the zener I mentioned earlier in the thread
and for giggles I am replacing the AC relay in case this relay has its
own zener (a lot of older bosch relays did I know). For less than $10
its cheap insurance.

The car today did just fine again, not proof positive but it is
encouraging especailly considering it was hotter today.

 

That's a good discussion but let me ask something here. I think I
understand some of your comments [it's been a while since I used zeners
or regular diodes] in regards to the OP's problem involved. The OP
fellow is looking at $2000 and possibly unable to get to work. Now a
zener diode is what? It's been a while since I bought them, but from
Digi-Key, I could buy 10 or 20 for a few dollars? Diodes are probably
cheaper and simpler to understand. But when looking at thousands of
dollars and the specter of not getting to work, why mess with a part
that can be purchased often for loose change, like a quarter or now 50
cents, even at Radio Shlack? Nothing in my car costs that little except
maybe a cap to the tire's air valve! I know, it's the challenge
involved. Is there any condition under which a regular diode would be
better than the zener diode? In this case, the zener is reversed to the
regular diode if I followed your cathode/anode comment but the
schematic or original usage calls for a zener, does it not?

 

No. You're half right. Whether a zener or non-zener is used, it would
be placed anode to negative, cathode to positive (keep in mind that the
terminology for the zener terminals is for the forward - non-zener mode
- conduction - same as a regular diode). Either diode would clamp
negative spikes generated by the coil turning off - either the regular
(non-zener) or the zener would be with anode to negative and cathode to
positive - when the positive wire tries to go negative, the diode goes
into foward conduction mode and clamps at just below the ground wire
(around -1 volt depending on the current it is absorbing).

With the zener reversed like you said, the positive wire voltage would
be clamped to the forward conducting voltage (around +1 volt) when it
was attempted to turn the coil on thus blowing the fuse for the driver
circuit and/or damaging the driver.

In reality, the Zener would never go into zener (reverse conduction)
mode in this application. It does have the secondary benefit of
clamping the positive wire to whatever the zener voltage rating is if a
*positive* spike were to come in from elsewhere (but it won't come from
this coil). But that's not why it was designed in for this application.

The technical rationale for using a zener for a back-emf snubbing
application is that they are, by design, optimized for *short*
*duration* *high* *current* situations in a small package (internally
mechanically and thermally designed for high *local* heating and
electrical stresses), whereas a regular diode is rated for current under
steady state conditions (high watts absorbed by thermal mass), and so if
it were to have the same (short-duration) current capability as the
zener's (short-duration) rating, it's package would be much larger, and
more expensive. IOW - to get the same high current rating in a regular
diode as what the zener can handle on a short-term basis, the regular
diode would be larger (and for the high volume manufacturer)
significantly more expensive.

If you get a 600 volt 1 amp regular diode, it should work, but a 20 volt
..5 or 1 watt zener would be preferred (and smaller).

Bill Putney
(To reply by e-mail, replace the last letter of the alphabet in my
address with the letter 'x')

 

On Mon, 22 May 2006 10:45:46 -0400, Richard Ahlquist

Well #@$#$#@!$#@!!@#$

I left my house to come to work this morning and made it about 1 mile
and it went into limp. It went into limp while i was coming to a stop
at a stop sign. I stopped, shut the car off and restarted. It never
went back into limp and I drove it all the way to wrok mostly on back
roads.

I tried the key trick for retrieving error codes (on-off-on ....etc)
and got nothing at all.

So as it sits its more intermittient than it was before but its still
limping.

The only thing totally different today than any other time was it was
cool, in the 60's and there was heavy dew this morning.

Is there any way to 100% find out what caused it to go into limp? Last
time the computer didnt show any codes.

If it were youre car what would you start with replacing at this
point? The TCM? Wiring? The A/C is still disabled so i guess that
wasnt the whole problem.

 

The "key trick" only allows you to read codes from the ENGINE
computer. The transmission computer is completely different.

I would not start replacing ANYTHING. What you need to do is find a
competent shop that actually knows what it is doing. Looks like this
is a wiring harness problem. But the problem needs to be properly
diagnosed.

Find a good shop, and just drop the car off to them. Tell them you do
not expect to get soaked for the work, but by the same token they can
have the car as long as they need to find the problem (within reason).
Give them a month if yo have to. They can use your car as back-fill
for times when things are slow.

On the plus side, if the problem is becoming more frequent, then it
should by the same token be easier to find.

If all else fails, then find a Five Star Dealer and drop it off to
them. A friend of mine had a weird problem with his Firefly stalling
on the highway at 80kmh! Just driving along fine, and sudden
"flame-out". Three shops did not do the trick! So I told him to take
it to GM. The catch??? If you schedule your repair far enough in
advance, then tell the dealer that you need a loaner car! My friend
needed the car to get to work. So called GM, and sure enough, he had
to wait a month, but they gave hime a loaner car. GM was able to
reproduce the problem - but it took them almost 4 weeks to fix! The
final computer print-out was over 5 pages long of the work performed.
They were so embarased that they only charged him for about 1/4 of the
work done - about $800. That $800 bill was much less than it would
have cost him to rent the loaner car for a month. He got the car back,
and it was indeed fixed - not to mention largely rebuild! lol.

So find a DC 5 Star dealer, and talk to the service manager. See if
they can do the same kind of deal for you! Book in advance, get a
loaner, and let them clunk away. Just make sure that you set a limit
with the manager - say $800? and tell them that if the bill will be
more than the agreed amount, that they MUST get your permission to
proceed. A lot of service forms acutally have a section for this right
on them. If you fill it out, it covers your butt.

Good luck with it my man.

I had a 1988 Olds Cutlass that had in intermittent flame-out. Never
did get it fixed, just traded it in. Fortunately for me it did not
flame-out while they were test-driving it! )))))

 

And don't forget to put a resistor in series with it.. ... the current
spikes in any circuit of this type are enormous and will usually "smoke" a
diode or zener in very short order. Something that will clamp the surge at
the level the semiconductor can handle.. these spikes are VERY short and
the "ring" fades fast so I would suspect 1 watt resistor would be enough..

Been many years since I have been involved in this stuff so.. YMMV.

Ted

 

Where is this zener located in the A/C harness? I have a 94 Caravan
that has the 41TE in it (A604) that is acting kinda funny. It's not to
the extent of your problem, but for a few dollars (I probably have a
few zeners in my parts box), it's one step to get past.

 

On Fri, 26 May 2006 07:59:54 -0400, Richard Ahlquist

Ok called AutoZone, they said they would be happy to hook an ODB II up
and read the codes for me. I went down there and the tranny went into
limp 2 times on the way. When I got there he brought out their store
Acctron ODB II scanner hooked it up, turned on the key and got link
errors the 3 times he tried it. So he sent me off saying I need to
find an electrician. They also balked initally when i told them it was
a 95 saying it needed to be ODB II...

Drove it back and it went into limp. Neurtal, killed engine, restart
and it was fine the remainder of the way back. I am at a loss here.

Could it be something else electrical like battery/alternator? Should
I just replace the blasted TCM or is the engine computer also likely
having issues since both times my car has been scanned for codes in my
presense it had communication issues?

 

The one mechanic who has replied to me about this question indicates
its either on the A/C compressor or actually in the cluth coil. One
place I found this;

http://www.dormanproducts.com/cgi-b...=&start=&vfrom=&prev=&vnext=&category_list=:0

havent had a chance to look at my compressor yet to see if that would
work. Note that link is for a 95 stratus you will need to see if they
have one for your van.

 

Hi Richard...

Can you tolerate yet one more suggestion from an old retired
electrical guy? If so, I think I have what may be a good one,
but please don't try it unless and until you hear a blessing
from one of the mechanics here. I'd hate to contribute to your
problems...

If so, here goes

We know that limp mode forces the transmission to downshift to
second. Can we also safely assume that limp mode will force
an UPshift from first to second?

If that's a safe bet, then how 'bout parking your machine about
a half inch away from something immovable and un-damagable.
Only a half inch or so, so nobody can possibly get between
the car and the "thing" (bridge abutment?)

Take off your tie, no long sleeves, put the 4 year old safely
indoors. Chock the wheels firmly - even backwards, in case
the foolish car decides reverse gear is a good idea. Parking
brake on firmly.

Start the motor, put it in drive, and with a long stick or
handle end of a screwdriver poke, push, prod, and twist a
bit each of the cables, connectors, sensors, etc. Do a
little bit of percussion on each of the sensors. Do the
same with the tcm. (Remember that if you get near to the
knock sensors a reaction from the engine is expected and a
good thing)

If you don't get finished within 3 or 4 minutes, shut her
down, let it cool, and start again from where you left off.

Obviously what you're looking for is something that will
cause it to shift... I'm sure you'll hear it if and when it
does

If and when you find something reliably repeatable, you'll
be awful close to the solution methinks

Now hopefully one or more of the mechs will either bless
or condemn this idea.

Take care.

Ken

 

Ya know, I just woner....

Electricity plays such a big part in things these days...

My 94 Acclaim acted really weird once. I was scared to crap that the
thing was going to need a major repair by the behviour.

But when I took it to my local shop, they found an intermittent short
in one of the battery cells!

$100 and I was on my way, my problems never came back!

Considering they could have soaked me something terrible... they now
get almost all my business.

Have you taken a close look at the battery? Over time crap drops off
the plates - CONDUCTIVE CRAP. It builds up in the bottom of each cell.
If the height of the crap reaches the bottom of the plates, then you
get a short. If you are lucky, it just shorts and you can easily find
the problem. If you are not lucky, it is an intermittent short!

Such a short can wreak havoc with any and all electronics - including
your ECM and TCM!

Hey, keep it simple! This should be easy to check. But even if you
just wanted to swap in another battery temporarily, that would tell
the tale!

Just a thought. Wouldn't it be a kicker if that was the problem after
all your agony to date?

Good luck.

 

Yes. These transmissions are electronically shifted. The only "true"
hydraulic gears are 2nd and Reverse.

If the parking brake is properly adjusted, it should hold the vehicle
in place if it is at idle and in gear. There is also the option of
putting the vehicle up on jackstands....

Hope this helped.....