Ultrasonic cleaning to restore the YH-409 HVAC sensor and park brake vacuum switch

[Hillbillycat]

When I had tried if sealing up the brass housing works (before I had found my plunger) I had tucked a small pice of neoprene inside the hollow stem. Sealed up perfectly but did not work, of course.
I do however think the plunger needs to be free for motion. If it only should stick out why in the world did they make it a costlier, moveable two-piece design instead of just a needle end on the brass housing like a carb needle?

My sensor´s completey reassembled now so I can´t try that for I fear the bent end of the spring won´t stand another straightening. It will sure break off.
Maybe you can try that Lutrova? Could use a piece of wax instead of the neoprene, too. This is just for testing so the wax won´t hurt and could be melted out with a hot needle.

And yes, the "drifting" sounds like the plunger not seating correctly. I do think that the ultrasonic cleaning will help here since mine was prefectly clean inside. No debris or dirt tracks at all. Have not seen a non-cleaned disassembled though, so I can´t thell if they´re clean or dirty inside normally.
BUT from how dust loaded the bimetal was (and is on my donor car one) I really think that the exact dust adheres and builds up inside the brass housing. And that can prevent the plunger from fully seating into the valve body - making it stick/bind/ not move correctly.

[Lutrova]

I don't know why Ford would've designed the plunger as a separate piece from the rest of the brass, but I also can't think of a reason it can't be fixed in the extended position. I filled the stem behind the plunger with a piece of basswood, which is just what was at hand. It should keep the plunger from falling into the stem without restricting its ability to rotate, whenever that may be needed. In bench testing without the spring attached I was seeing a drop of one inHg per nine seconds. And that's drawing vacuum from the supply port, relying entirely on the plunger to hold vacuum.

With everything reassembled I'm currently seeing a loss of about 1 inHg/sec at room temperature with the temp bias set to full heat. Again, same test method. Any other temp position will drop vacuum rapidly, but at least build a little with a finger held over the servo vac line. Just what this will mean for servo operation in practice, I don't know. I probably could've stood to bend the bimetal some more. The spring adjuster is turned all the way in, so there's no room for adjustment if it generally biases too much toward cold. But now I'll have to install it in the car and see what happens.

[gadget73]

unless maybe the seperate piece was a way to make up for sloppy manufacturing, basically letting the floppy bit self-center in the seat ? That seems unlikely though.

[sly]

Also, the failures I mentioned before are actually the tension spring, not the bi-metal.

[Lutrova]

Well I finally got around to installing the ATC sensor and going for a spin. I can't say whether it's totally fixed, but I've definitely got more operation now than I did before. Moving the temp lever from cold to hot triggers a vacuum release from the temp servo. I didn't run the engine too long - and I never shorted the TBL when I removed it - but I did feel a bit of heat with temp set to hot on Hi-Lo. Fan speed also appears to be at least somewhat responsive when set to Auto and moving the temp lever around.

I'm probably not going to test drive more until I replace my wheel bearings. The AC compressor or clutch might also be acting up. So a conclusive result may still be a few weeks off. But since the starting point was full cold and no response to input, I think this could already be at least a partial success.

Knowing what we do now, anyway, I can't imagine the sensor failing in ways that haven't already been discussed. If the spring isn't broken, the bimetal has enough oomph to resist its pull, and the plunger remains extended far enough to seal the valve, the sensor ought to work - at least to some degree.

Thanks, Hillbillycat, for taking the plunge on this and figuring things out. If you can replicate the results I've found, we may finally have a workable solution for these sensors.

[Hillbillycat]

Hey, congrats on your success!
I dis-arched (is that even a word?) my bimetal a bit through the opening for I had overdone it before. Can´t disassemble or I´ll brake the bent spring end. Now I get beautiful vacuum modilation on my bench with the plenum hooked up but a no-work condition in the car. I still think that the in car temp is too hot these days to get a proper funtion. Or can you choose for heat now during the summer? Fixed all vacuum leaks in the car by now.

[Lutrova]

Well it's pretty much going to be 100 or more for the next couple months, so I'm probably not a good candidate to verify ATC heat in the summer. What little testing I've done so far suggests that it might be possible. I set the controls to Hi/Lo and temp to 85/full heat, and I got the auto fan to cut to low and could hear the temp door servo lose vacuum. So the system seems to be at least somewhat responsive. When running in Panel AC, I have gotten the car to eventually reduce fan speed when it reaches whatever it believes to be my desired temperature. And bumping the temp selector down a few degrees ramps the fan back up. Obviously the ATC sensor working and the sensor being properly calibrated are two different things, but I don't care enough to try to get it dialed in to actual temperatures. It's good enough for me if the car will hold a given temperature, and if I feel cold or hot I can nudge the selector left or right accordingly.

If the bimetal was too arched, couldn't the adjuster at the end of the spring pull it back? I would imagine if you were really patient you could unhook the spring without having to bend it. Just use a small pair of pliers and a lot of deep breathing. Or maybe make friends with a laparoscopic surgeon.

[Lutrova]

It's been a little cooler here lately, so I've been able to confirm full operation on my ATC sensor. I feel pretty strongly now that the plunger falling out of position and failing to seal the valve is the chief failure point for these sensors. So long as the plunger is still present when disassembled, I think they can be reliably repaired.

Hillbillycat, if your system still isn't working and you don't feel the sensor will survive another surgery, I can find another one in the junkyards around here and ship it to you.

[Hillbillycat]

Hi, thank you for the kind offer.

The one from the parts car seems to work. Although I get the feeling I don´t get full heat. AC is finally rebuilt, charged and working since last Friday. What a joy.
The one I had disassembled doesn´t work properly, though.
If you can grab one from a JY, I´d be pleased to have another spare. Maybe it´s working better then the one I´m using right now.

Do you think the repair is rather to fix (glue) the plunger than to re-arch the bimetal?

[Lutrova]

Do you think the repair is rather to fix (glue) the plunger than to re-arch the bimetal?

I think so. I've become more skeptical of the idea that the bi-metal loses tension over time. It's two different metals joined to each other, whose layering ought to provide some rigidity. And it seemed fairly resistant to bending when I tried doing so by hand. It also strikes me as no less strong than the metal spring acting against it, and yet these sensors all tend to fail in one direction - cold. Of course if the bimetal does lose its arch over time then bending it back would seem to be the fix. But knowing what we do now about the innards of these things, I'd probably start by addressing the plunger before trying to adjust the bimetal.

A full cold failure can also be explained by a permanently out-of-position or missing plunger. While one whose temperature wanders would also point toward a plunger position issue. I can't think of any problem that would come from fixing the plunger in its outermost position, while any number of problems seem to arise from it sitting anywhere else. My personal method was to jam a piece of basswood behind it, which holds its position without limiting its ability to rotate if for some reason the diaphragm and plastic valve don't meet head on. Would it cause problems if the plunger were glued in place instead? I don't know.

Another question I still have is why Ford designed the plunger as two pieces? Maybe it's as simple as that's what one engineer came up with. It doesn't appear to be a very good decision, but it is what it is. Then again, my sample size for these sensors is one, and it was broken. I wouldn't be surprised if there is some sort of glue holding the plunger on working sensors. There was glue holding the calibration on the spring gear. If proper assembly relied on the assembler to apply enough glue to fix the plunger then that could explain why some sensors are still working while others (most?) are not.

Now that it seems these things might be repairable, I'll start grabbing them from cars I run across.

[Hillbillycat]

Finally fixed! ATC Sensor fully working again!

Hi, I fixed my spare sensor!

Caution! Glueing the plunger in the outermost position will not work. The plunger protrudes too far and the valve will not seal.
Jamming a piece of wood behind the plunger like Lutrova did however gives enough movement so it can be seated back enough untill the valve closes.

Here´s a how to do:


To repair the sensor you need to disassemble the unit. There are two tricky parts:
First to unhook the spring. Initially I did bend back the end at the adjuster wheel, but later found out that the spring can be unhooked from the brass valve stem with needle nose pliers through the side opening.
Next tricky part is NOT TO LOOSE the plunger that it´s all about when the spring gets unhooked. I lost mine again but found it back. Working on a clean table or better some sort of large container or pan will prevent the loss in case he plunger falls out. Clean the whole unit the best you can. Ultrasonic bath get´s everyting out of the small passages. But carb cleaner and compressed air will do good, too.

Next to the actual repair step:
Make sure the plunger is inserted with the round end first. Then a small piece of wood must be jammed into the hollow stem, pushing the plunger fully forward. The wood piece needs to be short enough to clear the hole for the spring.
Put the valve/diaphragm unit back into the sensor body and gently press on the brass valve stem. This way the plunger gets backed up and the valve fully closes. You can see this by the diaphragm fully seating in its recess of the sensor body. Install the star washer and hook back the bimetall and reassemble the whole sensor.
With the plunger being jammed in position there´s no fear of loosing the vital plunger during further reassembly.

So we can finally conclude that these sensors fail due to plunger issues. Maybe dirt accumulation in this tiny passages, maybe these plungers had some foam support when new that got disintegrated and vacuumed up. I don´t know. But we can sure say these sensors can be restored by jamming the plunger in place.

[DerekTheGreat]

Good information. Next one which comes across my driveway will get this treatment if needed.

[Hillbillycat]

UPDATE - UPDATE! Plunger fail prooved!

I just swapped sensors from my Town Car and disassembled the non-working one this minute.

And now it´s officially prooved:
The plunger fails!

Behind the plunger is a small plastic rod, triangular shaped. And behind this plastic wedge is a piece of foam/neoprene. I wasn´t able to really tell.

So the actual cause of fail is the foam disintegrating, preventing the plunger to extrude far enough to seal the tiny pilot hole in the sensor valve body.

The plunger and plastic piece (that nobody ever noticed before, or had been broken on some, too) only falls out during disassembly because the foam is missing. The foam keeps everything thight and in place so it can be handled easily during production.

[Lutrova]

Well done.

It's still only a sample size of two, but I think plunger/needle valve failure has got to be the primary issue these sensors have. As you found, it seems the plunger needs to be free to rotate independently of the rest of the brass piece. Maybe this is because the spring plus gravity doesn't always pull in a straight line away from the valve?

I'd be interested in seeing whether any of these sensors actually lose their plungers prior to disassembly. The hook of the spring ought to keep it in place even after the original foam retainer disintegrates. I'm also skeptical that any other parts in these things is prone to failure. The plastic valve assembly and enclosure have no moving parts. The bimetal and spring both seem pretty robust. And I don't think the rubber diaphragm is exposed to the elements enough for it to deteriorate.

Then again, the Ford engineers probably didn't expect the plunger retainer to deteriorate either. It's good you actually caught sight of the retainer. Now we know exactly how these things fail.

[DerekTheGreat]

Yah, great information. They were voodoo magic before.