Shift solinoids?

[propony]

I am having transmission problems that I think are the shift solinoids on the transmission valve body, 104003. My car starts out great, upshifts as it should and doesn't slip. When you come up to a stop, the transmission seems to stay in 3rd gear. Of course pulling the shifter down into 1st gear does nothing to help.
I built a test fixture, 2 LED's fitted to the top of the gov/computer in the engine compartment. Both lights on. Perfect! Shift to 2nd, one light goes out, as it should, hit third gear, still no slipping and no LEDs lit. I should be happy, but the trans stays there, in 3rd gear.
If I snap the throttle microswitch, I can get the lights in the test unit to both come on, but as soon as I pull the trans into "D" The lights go out again, leading me to believe the solinoids are "sticking"?
In my usual throw new parts at it until its fixed fashion, I found that Houston's web site shows the solinoids,104003, are not availible. Now what?
Any help with my latest drama would be apprieciated,
Curt

[Ron]

If a solenoid sticks it will not affect the lights. The lights tap the solenoid wiring and simply tell you whether the governor computer is calling for the solenoid to be energized or not. (Yet it is possible that shorts in the wiring (exhaust is common source) or GC will cause the lights to lie to you.)

[David T]

From your description of the symptoms I am not sure I understand it. If there are no lights lit the shift computer is telling the transmission to be in 3rd. If you see one or two lights lit you are supposed to be in second or first. If you see any lights lit and are in 3rd then you *may* have a problem with a solenoid. Verify what gear you are in with the RPM/MPH chart. As Ron pointed out it is much more common to see a problem with the wiring than a bad solenoid. Remove the wiring harness and verify it is NOT melted going from the diag plug to the shift computer. BTW hitting the WOT switch SHOULD force a downshift out of 3rd. Third gear is a default gear meaning if neither solenoid is energized you get third so you can say it is a "limp home" mode. You can measure the resistance of each solenoid. It should be about 25 ohms each. Dave Swingle had a weird auto trans problem and when he took the solenoids out he found one was plugged up with a blob of silicone gasket sealer. It is rare for the solenoids to fail.
David Teitelbaum

[Ron]

From your description of the symptoms I am not sure I understand it.
+1

I'm wondering if it stays in 3rd after it stops, how does it get back into 1st to shift through to 3rd to begin with?...If it's by switching everything off, 99% you need a governor. If not, what is it that you are doing to get into 1st..?

You said putting it in 1st "does nothing to help"...Does either light come on with shifter in 1st? In 2nd? This is key since either should bypass the GC and directly ground solenoid B and its light (energizing both) no matter what the trans is doing.

You might try this then retest-
Referring to the first section in the chart below, make sure the connections are good at the "L" shaped block. Ensure that the ALL grounds and 12V power coming in are good while apart. (Don't let a volt meter fool you...use an ice pick light or make something that will load them..not LED.)

http://dmctalk.org/attachment.php?attachmentid=953&d=1307760396

[propony]

Thanks,
I checked the harness for any contact with the exhaust, none found, everything is high and tight. I have tried to read resistance from the round connector on top of the GC, using the same contact points my LED are connected to, measuring from the known "hot" post to the solinoid, I got nothing. No resistance. Then I tried to check voltage across same. None, so I tried to a different ground, all three posts are reading 12 volts. This is when the LEDs are not lit. I will try this now that I know how to trip the micro switch to get the LEDs to come back on.
I have replaced the capacitors and added the protective diodes in the boards 4 years ago, and everything has been good.
On occasion I have had to unplug or wiggle the "L" connector so the neutral safety switch will make contact, but I have cleaned all with spray contact cleaner and this problem is gone.
I was under the impression that you couldn't pull the shifter down into first gear to force the valve body to downshift, so am I to read it will?
Thanks again for the help,
Curt

[DMCMW Dave]

My guess is the governor has failed again.

[David T]

You actually have to remove the wires and run your hands over every inch and visually inspect the wires. Once I had a car and everything looked fine till I took the shift computer (and it's wiring harness) out. If the wiring gets melted it can short out the power going to the solenoids and that can blow the output transistors. The most common failure on the automatic is the shift computer itself. The 2 large caps blow and there are cold solder joints. Eventually every shift computer needs work. This failure mode does not sound like yours though. That's why I suggested examining the wiring harness. Unusual symptoms generally sound like a melted wiring harness. Remove the 3 brass hex nuts in the round diag plug and the wires under them. Measure Ohms directly at the wires. These wires go to the solenoids. If you do not see 25 ohms (or thereabouts) there is something wrong between the ends of the wires and the solenoids.
David Teitelbaum
David Teitelbaum

[Ron]

Thanks,
I checked the harness for any contact with the exhaust, none found, everything is high and tight. I have tried to read resistance from the round connector on top of the GC, using the same contact points my LED are connected to, measuring from the known "hot" post to the solenoid, I got nothing. No resistance. Then I tried to check voltage across same. None, so I tried to a different ground, all three posts are reading 12 volts.

That is normal...I think I see what you are missing. Look at those three posts as one long wire starting at Post C). It runs down to the trans and branches. Each branch coils around forming Solenoids A and B then returns to Posts D and B, respectively. So you were virtually measuring around on the same wire (which is always hot with key on). Note the "normal" mode for the trans is 3rd (as when pulling the fuse). With the key on, the solenoids are always hot and the GC grounds them to energize them (1st gear) and drops ground (via Post E) to Solenoid A for 2nd, and Solenoid B for 3rd, to up-shift.


I was under the impression that you couldn't pull the shifter down into first gear to force the valve body to downshift, so am I to read it will?(Forget the valve body here.)
When you select 2nd, you bypass the GC and ground Solenoid B directly turning it on [Think, "NO 3rd allowed!"], "forcing" a down-shift. But when you select 1st, the GC decides if its appropriate [Think, "No higher than 1st, please?"], "requesting" a down shift. So No, not for 1st, just 2nd... This is part of why I asked about the lights when stopped and 'stuck in 3rd' then moving the selector (*ahem?).


On occasion I have had to unplug or wiggle the "L" connector so the neutral safety switch will make contact, but I have cleaned all with spray contact cleaner and this problem is gone. There are two connections there for the ground side of the start inhibit relay. (Except for the switch (S1), the neutral safety circuit/wiring is independent of the GC.)
As said above, there are several ground sources coming in through that connector, and if they can not handle the load the GC will act up (and may damage itself...Similar for power supply coming in.) I can see how a bad ground(s) can cause the symptoms here. Same for section(s) in the GC...the odds of this are much lower, but if it is weak or not getting what it needs, a power off may allow it to work a few times...
So far I agree with Dave -- 90% it's a bad governor...(*I can help you narrow the odds down better, if you answer my questions... ;-)

[Ron]

You actually have to remove the wires and run your hands over every inch and visually inspect the wires. Once I had a car and everything looked fine till I took the shift computer (and it's wiring harness) out. If the wiring gets melted it can short out the power going to the solenoids and that can blow the output transistors. The most common failure on the automatic is the shift computer itself. The 2 large caps blow and there are cold solder joints. Eventually every shift computer needs work. This failure mode does not sound like yours though. That's why I suggested examining the wiring harness. Unusual symptoms generally sound like a melted wiring harness. Remove the 3 brass hex nuts in the round diag plug and the wires under them. Measure Ohms directly at the wires. These wires go to the solenoids. If you do not see 25 ohms (or thereabouts) there is something wrong between the ends of the wires and the solenoids.
David Teitelbaum
David TeitelbaumI totally agree.
What bugs me is that it works until he stops... bad grounds can work magic and week/bad component(s) can act "OK" until it's in a certain situation that requires power off/cool off/etc. Grasping at hopes$...probably, but maybe an easy/cheap fix...:yesss:

[propony]

Ok...
What if it is a "missing" ground? If I have 12v at all points on the test connector, the coils in the solenoids are recieving power(+), but if I am looking at the schematic correctly, the solenoids always have (+), but the ground is what drops out/off...am I correct?
The reason I think this way is looking at the downshift request circuit, and the fact that when I click the microswitch, I can make the LEDs light up. I think the Microswitch provides a momentary ground that closes the solenoid, downshifting the transmission. Is that what is happening?
If that is what is happening, would it be possible to "backfeed" a ground to the solenoid from a new ground source through the wires I have the LEDs connected to? I am curious if this is a redneck paddle shifter waiting to happen?
Has anyone tried to backfeed a ground to a solenoid from the test port to try to activate the solenoid, and light the LED? Isn't the transistor waiting for a ground so it can do the switching?
I am starting to believe I am suffering from a cold solder situation. Does this sound logical?
Thanks again for the help,
Curt

[Ron]

Ok...
What if it is a "missing" ground? If I have 12v at all points on the test connector,...
Then you are all over the problem -- E is a ground.
(I'm hoping you meant the others only...;-)


...the coils in the solenoids are recieving power(+), but if I am looking at the schematic correctly, the solenoids always have (+), but the ground is what drops out/off...am I correct?
Yes.



The reason I think this way is looking at the downshift request circuit, and the fact that when I click the microswitch, I can make the LEDs light up. I think the Microswitch provides a momentary ground that closes the solenoid, downshifting the transmission. Is that what is happening?
Yes, indirectly through the GC. via a '"request" line' different than the one earlier.
The microswitch has its own ground source and its own path to A and on to the board along the green harness wire...Note it taps the power feed circuit behind a 1.2 K resistor at a point between D2 & D3. Look at it as your place in the circuity to contribute to pulling the circuit low.
[I'm not going deeper, no mater how many questions you don't answer :devil:]


If that is what is happening, would it be possible to "backfeed" a ground to the solenoid from a new ground source through the wires I have the LEDs connected to?
Yes.


I am curious if this is a redneck paddle shifter waiting to happen?
Yes!


Has anyone tried to backfeed a ground to a solenoid from the test port to try to activate the solenoid, and light the LED?
Yes.
It works...Both. Be careful! Grounding A only is undefined. Only the selector switch grounds B directly. All others are by the governor or by request through the governor. And you certainly don't want to shift it at the wrong RPM!


Isn't the transistor waiting for a ground so it can do the switching?
Yes.
From the governor for A, and, either the GC and/or the Selector switch for B.



I am starting to believe I am suffering from a cold solder situation. Does this sound logical?
Yes.
Especially if you confirm the power and grounds are good, and, the GC itself isn't toast or weak...but it looks like we will never know LOL

[propony]

Sorry, I guess I missed the question...oops! I am not getting the same results twice. went out today, everything worked perfect. But yesterday I did pull the shifter down into second, and then to first, and got no lights to glow. Until I built the test LEDs, I could only give "seat of the pants" report, and I concluded it was in 3rd by pulling the fuse with no change in gear. Pull out of my driveway I could feel it shift twice, never felt it downshift at the stop sign, not sure I have ever felt it downshift when slowing to a stop in the past. When I pulled out into traffic, I could tell it must have been in third, and tried to pull it down manually, and got no attempt at downshifting from the trans. I moved to the shoulder and contiued to pickup speed gradually, as not to slip the trans clutches.
Is there a switch on the shifter, or in the linkage?
Thanks again for all the help, only a few more weeks of driving before the snow flies here, and it would be nice to find a solution, or plan of attack before I put it up for the winter (and pull it apart!)
Curt

[Ron]

NP!
Well, if ya have to, since you have been in it before it will be a lot easier this go...
FYI- The clutch pack(s) don't like "limp mode" (I don't go along with that idea 100% since limp mode should be 2nd not 3rd, IMHO). The one pack is a little bigger than a fist..not much there...(guess how I know), I'd park it actually, if possible.

[propony]

Are there anypoints outside the gov assembly that could provide some type of additional ground? I know there is a lot of plastic on it. If I could provide a "cheater" ground on the outside, is there any chance it could help the board?
Also, how does the shifter tell the trans to downshift? I am going to look at more diagrams. Am I looking for another microswitch?
Curt

[Ron]

Are there anypoints outside the gov assembly that could provide some type of additional ground? I know there is a lot of plastic on it. If I could provide a "cheater" ground on the outside, is there any chance it could help the board?
CurtAll you can really do is make sure the grounds in the chart are good...after that, ground is ground...not like adding a stronger battery...


Also, how does the shifter tell the trans to downshift? I am going to look at more diagrams. Am I looking for another microswitch?

When you select 2nd, you bypass the GC and ground Solenoid B directly turning it on [Think, "NO 3rd allowed!"], "forcing" a down-shift.
There is a cam that presses one of two two way switches (S2). It connects B to the ground coming in on the L shaped plug pin 3. This grounds solenoid B(2nd). For 1st it's too complicated to explain here...basically the other switch closes and grounds the circuitry at a point that 'helps' tell the opamps to control the base of the transistor which actually grounds the solenoid(s)....
All of the circuits are covered here (http://dmctalk.org/showthread.php?350-Automatic-Transmission-Governor) (I thought you knew that).