Hall effect door switches

[AdmiralSenn]

Hey guys. I remember reading somewhere that someone had replaced the door plunger switches with hall effect sensors and attached magnets to the doors to activate them.

I'm thinking of doing this (even with vacuum caps on my plungers, they still sometimes don't turn off) and I was wondering where I should go to get the proper component. I guess I could just graze through Radio Shack and see what they have but I'd rather not have to replace the sensor in a year if I grab the wrong one or get a cheaply made one.

Anyone do this, and feel like sharing? Thanks!

[Elvis]

a short piece of rubber hose does the job...

[AdmiralSenn]

Well, I have larger (I think they're 5/16") vacuum caps over the original plunger tips, and they turn off the lights consistently. The problem is that because the switches are longer with the caps on, the shafts get twisted and don't come back out every time when the door opens again!

The door moves in enough of an arc that the switches take the downward force of the door before they take the inward force - one of them stays angled down whenever the door opens and doesn't turn on at all.

I'd be happier with a no-contact "works every time" system. I know people have talked about it before, and I think someone's already done it somewhere.

[Elvis]

I've seen that, too.
Took the switches apart and filed down the plastic surface to be smooth again.
A little lubricant can't hurt either.

I've done this to two or three cars in the last 10 years.

[Farrar]

Why not just put a small rubber pad on the part of the door which contacts the switch? There's a metal plate there already. Just add some thickness at that spot and your switches should open the circuit properly.

[Bitsyncmaster]

Why not just put a small rubber pad on the part of the door which contacts the switch? There's a metal plate there already. Just add some thickness at that spot and your switches should open the circuit properly.

You may be better off not using any rubber so the parts will slide with less effort.

I like the idea of a no contact switch but that would take some playing with to see if it would work. Not sure how sensitive a magnetic switch would be.

[DMC5180]

Ed Uding sells these longer switches. No need for a rubber tip and they slide easy against the contact pad on the door.

http://www.delorean.eu/catalog/product_info.php?cPath=20_141&products_id=65049&osCsid=6d464e071a84130a4f1c4d3c83b3104a

[jangell]

I replaced the mechanical door switches with magnetic ones. I can't remember if the switches need to be normally opened or normally closed (easy enough to test), but I got the wrong ones and had to wire up a transistor to invert the signal.

I used magnetic switches I got online that are meant to be used for security systems to tell if a door is open/window. I think they're just reed switches. I stuck neodymium (rare earth) magnets to the metal plate on the door. I found that alignment was important, so I closed the door and stuck a marker through the hole the switch goes through to mount the plate.

I imagine you could use a hall effect sensor, but I don't know if you'll need extra hardware for that, or if you can just get a hall effect switch, or how much load the hall effect sensor can talk.

Someone also pointed out that the magnetic switch I used is only viable if you use LEDs for your interior/door lights, as the current draw from the old bulbs is sufficient to overload the switch.

There was also a $165 kit that was available at one time, but that site no longer seems to be available. My solution is about much less expensive.

Here's my original post on DMCNews: http://autos.groups.yahoo.com/group/dmcnews/message/71215

Not too long ago I found that my door lights were staying on when the
doors were closed. Repeated closing of the doors finally took their
toll: the switches still worked, but the plastic that they were
mounted in had collapsed and bent, so the switch no longer pointed at
the door and thus didn't completely depress. I tried using shims,
but that didn't help with switch's aim.

I figured that even with the shims, this would eventually happen
again, so I decided to go with a cheap, contactless solution that
required no permanent modifications to the car and no new wiring:
magnetic switches.


Parts:
- Three cylindrical magnetic switches (replaces existing switches):
$2 to $5 each
- Three rubber grommets (to mount the switches in the old holes):
$0.50 each
- Six blade-style connectors and crimp tool (to connect the switches
to the existing wiring)
- Three rare earth/neodymium magnets (mounted on the doors to trip
the switches): $20 for a bunch of them
- Black Paint (if your switches/magnets aren't black): $3.00 for a can


Magnetic Switches:
I went online and found a pair of magnetic switches, which are
commonly used in security applications. They're pretty simple: the
switch changes from on to off depending on how near it is to a
magnet. The ones I got are small cylinders, about 1/4" or 3/8" or so
in diameter and an inch or so long. I went with cylinders (also
called "recessed") because they would fit through the hole where the
old switches were mounted. I don't have a link to the ones I
actually got, but here are some quick Google matches; they all
basically work the same way:

http://www.google.com/products?q=recessed+magnetic
+switch&hl=en&um=1&sa=X&oi=froogle&ct=title

Magnetic switches are listed has having a 3/8" or 1" or what not
"gap", which is how close the magnet needs to be to the switch before
it changes states. I think this mostly refers to the strength of the
magnet it comes with; if you want to try to use the magnets it ships
with, you'll want to get switches with the largest gap you can find.

The only other issue is to make sure you get a switch that is
normally in a state that will ensure the lights are off when the door
is closed. I managed to get the wrong ones, so I had to use a
transistor to invert the state. I remember thinking that I needed
normally closed switches (ie: completes the circuit when the magnet
is not near the switch), but they operated like normally open
switches when I they arrived. If in doubt, you could probably just
get one of each kind -- they're only a few dollars a piece, and it'll
certainly be easier (and use less wires) than the transistor method I
wound up using.


Switch Mounting:
You want a switch that is smaller than the hole that the old switch's
wires go through, since that's where you'll want to install it. To
mount mine, I took the switch to the hardware store and found a
rubber grommet that it fit snugly into. The grommet itself mounted
securely into the old switch hole with no tools or glue. I also
painted my originally light grey switches black so they wouldn't
stand out. The final installation is nearly flush with the trim and
looks surprisingly professional.

After mounting, you just crimp on a couple of blade-style connectors
and plug the two wires from the old switch into the new switch, and
the lights on the doors should go on.


Magnets:
To get the lights to turn off, you need to mount a magnet on the
door. The switches usually come with magnets, but I found them too
weak for what I needed; that part of the door is surprisingly far
away from the switch when closed. I wound up going to ThinkGeek.com
and ordered a $20 collection of rare earth (neodymium) magnets. This
gives you a bunch of magnets to try, and spares in case you lose one,
and they have a much greater range than normal magnets:

http://www.thinkgeek.com/geektoys/science/770f/


Magnet Mounting:
I didn't need to glue the magnet to the door or drill any holes, as
there was already a convenient metal plate there that old switches
pushed against, and I could just stick the magnet to that. The
magnetic switches I got were somewhat finicky about where the magnet
was: as long as it was inline with the switch it worked, but if it
was off to the side too much it wouldn't.

To figure out the proper placement of the magnets, I removed the door
switch, closed the door from the inside, and marked the door plate
through the switch hole with a chalk pencil. I then just placed the
second longest cylindrical magnet from the set to the marked location
on the door plate; the magnet is strong enough that the door lights
go off when the door closed at the point that you can feel the
striker pin pressing against the latches, but the latches aren't yet
engaged, which is pretty much what I wanted.

The magnets haven't fallen off the door yet, but it might not be a
bad idea to use a dab of super glue just in case. [note: I've never needed to do this, although sometimes they may slip and need to be re-aligned] I've also been
meaning to paint the magnets black, since the small chrome cylinders
on the door stand out a little.


The total cost of this project was about $40, which includes the
switches, grommets, paint, shipping, a bunch of extra magnets, and
the transistors. The magnets were easily the most expensive part,
but that's because I got far more than I needed. You might be able
to find switches that ship with stronger magnets and not need to get
the extra set from ThinkGeek or elsewhere, or just buy smaller
quantities of stronger magnets, or you could mount your switch so
that it sticks out more and thus is closer to a weaker magnet when
the door is closed. You probably already have paint, blade
connectors and a crimp tool, so no need to add those to the price.

Overall, the magnets seem to work quite well, and I haven't had any
problems with my door lights since I installed them.


Tom Neiland replied with a few important notes:http://autos.groups.yahoo.com/group/dmcnews/message/71230


If I may add a few thoughts. The "proximity switches" come in 3
types, the most common is "form A" which is closed circiut in the
presence of a magnet (and most common because they are used in alarm
systems). Then there is "form B" that is open ciruit in the presence
of a magnet (good for door switchs since they draw no current when the
door is closed). "Form C" has both open an closed circuits (3-wire SPDT).
I built a similar circuit for my car and wanted to share a few
things I learned. With normal bulbs, the D's door lights pull about 1
AMP each side. It is abut 1/10 that or 100 mA with LEDs. The typical
magnetic switchs is rated for 500 mA for form A and 250 mA for form B.
So, it is probably good that you used the transistor. I used the form
B switch with a relay so I could do the function of both of the
driver's side switches with one magnetic switch. The down-side of the
relay is that you have to wire it to a power source under the dash
(only used when the door is open). It would have probably been simpler
to use a transistor like you!
I too, really like the results. I never have to worry about the
switch alignment again. There is no contact between the magnet and the
switch.

-- Joe

[Farrar]

Ed Uding sells these longer switches. No need for a rubber tip and they slide easy against the contact pad on the door.

Wow. I bought a few of these from Sherco for $2.25 each.

Farrar

[DMC5180]

Wow. I bought a few of these from Sherco for $2.25 each.

Farrar

Yeah they are over priced for what they are. I bought a small number of similar style switches 2 years ago for a couple bucks apiece too (came from the UK). They are different from Ed's. I actually like the appearance of his better. Do you have a link for your Sherco switches?

Never mind I found them. Did you use the open style or the one with the rubber cover. Did you need to shorten them or still use an extension cap?

[Farrar]

Do you have a link for your Sherco switches?

Toward the bottom of this page (http://sherco-auto.com/switches.htm). There's also a rubber-covered version of the same switch for the same price, but I preferred the look of the chrome mount.

[Bitsyncmaster]

Just had an idea. Use a mercury switch mounted in the door. Will have to think about it a while. You could have it break/make the door lights without adding any wires but the dome delay unit may need a wire from the mercury switch.

You really want a passive switch (reed switch or mercury) so you don't have to keep it powered and add battery drain.

[jangell]

With a mercury switch, would you have to worry about the lights flashing on briefly when you hit a bump? Probably more likely if you have LEDs instead of the slower incandescents. And assuming the mercury maintains contact long enough for that to even be an issue.

A short timer could work around the flashing issue, although that adds complexity. The only other obvious difference would be that the door would have to be more open than normal before the lights come on, although how far probably depends on the orientation of the switch.

Neat idea though.

-- Joe

[Bitsyncmaster]

With a mercury switch, would you have to worry about the lights flashing on briefly when you hit a bump? Probably more likely if you have LEDs instead of the slower incandescents. And assuming the mercury maintains contact long enough for that to even be an issue.

A short timer could work around the flashing issue, although that adds complexity. The only other obvious difference would be that the door would have to be more open than normal before the lights come on, although how far probably depends on the orientation of the switch.

Neat idea though.

-- Joe

Good thought. Since the door will swing quite alot, the switch angle can be extreme. I don't think a bump in the road will upset my dome delay timer.

Edit:
It looks like all you need to do is wire the mercury switch to the door lights and ground. Then leave the original door switch disconnected.
Bad thing is you will not get a door ajar warning.

[AdmiralSenn]

Looks like McMaster-Carr has everything I need. I think in my next few paychecks I'll go ahead and experiment with this.

I guess I don't really need a Hall effect sensor if a magnetic switch will work reliably. Seems like it simplifies things quite a bit this way.

[Bitsyncmaster]

I'm going to try one door with the mercury tilt switch. I'm pretty sure this is going to work. Only consern is going over bumps may flash the dome lights. You would not see it with incandesent bulbs but LEDs at night you would.