Driving reed switches.

[Optikon]

Anyone know of of a low cost idea for driving a reed switch? (The switch only, not an integrated switch+coil called a reed relay)

I'm thinking about some standard SMT coil like thingy. Maybe an air core inductor with 1000's of turns of very fine wire...

I plan on placing reed switches on one PCB and driving them from another physically close PCB. I briefly entertained the idea of making a coil inside the PCB but this is not a low cost approach and I can't get huge numbers of turns to meet the Ampere-Turn spec.

Any ideas? Think low cost! Anyone know of a standard inductor part with lots of turns that will possibly create a strong enough field?

The MFG's also offers the standard "test coils" but I've no idea about their cost but it would certainly work. Looking for novel/wacky ideas here that are low cost..

Thanks!

 

[williB]

how about winding your own.. cant get much cheaper than that..

 

[Roff]

I would think that you would need a lot of ampere-turns to switch a reed switch when the reed is external to the coil. When it is internal, you get basically all the flux linkages passing through the reed. When it is external... :cry:
Since you seem to be an opto guy, I'm guessing you have already considered and rejected optically operated switches (e.g., phototransistors).

 

[williB]

all ya gotta do is drill a hole in (something round & flat ).
apply hot glue to the outside (around yer reed switch.)
and wind the inside with 30 gage wire (or smaller) .

 

[Roff]

all ya gotta do is drill a hole in (something round & flat ).
apply hot glue to the outside (around yer reed switch.)
and wind the inside with 30 gage wire (or smaller) .

Willi, you need to read the question a little more carefully.

 

[Klaus]

If you use an inductor with a soft iron core then your idea might work over a short distance. You will have to experiment which way to orient the reed switch relative to the inductor to get best sensitivity. I think you can forget about using an air cored inductor, 1000's of turns or not, if the reed SW is *outside* the inductor.
You could find small iron cored inductors by salvaging a solenoid from a junked video player or similar or dismantle a small relay to get the coil & core.

Reed switches can be quite sensitive if you place them *internal* to the inductor. For example, I made a simple one layer coil of about 15 turns that fits over a reed switch. The coil carries the current to my fridge, so its a reasonable size wire gauge. The reed switch turns on a battery powered hour meter when the fridge compressor is actually running.
Klaus

Anyone know of of a low cost idea for driving a reed switch? (The switch only, not an integrated switch+coil called a reed relay)

I'm thinking about some standard SMT coil like thingy. Maybe an air core inductor with 1000's of turns of very fine wire...

I plan on placing reed switches on one PCB and driving them from another physically close PCB. I briefly entertained the idea of making a coil inside the PCB but this is not a low cost approach and I can't get huge numbers of turns to meet the Ampere-Turn spec.

Any ideas? Think low cost! Anyone know of a standard inductor part with lots of turns that will possibly create a strong enough field?

The MFG's also offers the standard "test coils" but I've no idea about their cost but it would certainly work. Looking for novel/wacky ideas here that are low cost..

Thanks!

 

[Optikon]

Thanks for the thoughts!

I guess I didn't mention this is not a one off thing.. I'm looking for something repeatable & manufacturable which is why there is emphasis on low cost.

Willi, I understand your idea but do not think that labor operations can possibly be lower cost than an off the shelf part that is placed my a machine. I apologize for not stating that.. but I did think about your idea of winding my own (well it would be a custom built part)

Ron H,

I agree externally I would need lots of ampere turns but these reed switches are spec'd at 10-15 AT (which isnt alot imo) _but_ this is at a specified distance so some experimentation is required to see how much I have to boost it for my increased distance.

Also I'm not an opto guy! But I did consider opto-coupled devices.

Klaus, thanks for the info about the iron cored inductor.. I guess air cored is out (I was thinking more turns would be available with this type)

A little more elabortation may be in order,

I'm looking to get extreme capacitive isolation from relay coil driver to the contacts. I can accomplish less than 0.1 pF with an external coil. Incidently, relays are offered with this amount of isolation but they are expensive. This is why I couldn't swing an opto-fet/tran type of device.

Other details: switch can be low voltage (no 100V+ requirement)
must carry 1.0A or better (tough for hi isolation opto devices)
dont' care how long it takes to switch or bounce or any similar antics.
I need life of switch to be > 500e6 cycles. I'll be switching gently (no hot switching)

Sounds like a coil with sufficient strength is the answer.. but how to do it low cost? I've thought about wacky things like using a dummy (but strong) relay as my off-the-shelf pre-built coil driver.. or inductors or somehow permanent magnet but I'm not sure how to move it around in a low cost way..due to reliability, a low cost motor is out.

Thanks for the thoughts guys! If there are other considerations, please share them I think I'm going to go iron core inductor hunting..

 

[Roff]

Is there no source of DC on the switch board? What kind of load are you switching? Is the current AC or DC?

 

[Optikon]

Is there no source of DC on the switch board? What kind of load are you switching? Is the current AC or DC?

There is a source of DC onboard. The system is multiboard so I have th e opportunity to put coil drive (whatever that is..) on another board and this provides an isolation advantage. However, that is not required provided I can get the isolation I'm after. For this problem, I am willing to make any rail necessary and many watts are available for drive.

Load being switched is DC (1A max. It is strictly speaking, time varying but very slow- sub 0.1 Hz, essentially DC) I'm charging a large capacitance with constant current from ~4V to 9V.

I need to isolate from a high frequency AC common mode signal (full of trash into 10MHz) on those switches. With reed switches at ~ 0.5pF across contacts and <0.1pF to coils, I'm there. If I can't get this isolation in a low cost way, I'm willing to junk the idea in favor of an entirely different approach I have rolling around in my head... but not quite there yet!

I only have a block diagram available on paper for this whole mess and didnt want to unload all this with my orignal post..but will if you have some clever ideas to work out.

Thanks for the help!

 

[Roff]

Well, if you want to post the block diagram, sometimes it helps to get it looked at by someone who has a different perspective.

 

[JimB]

Remove the coil from a large relay, place the reed switch across the magnetic circuit of the relay coil.
You could use a thin copper foil between the relay and the switch to minimise capacitive coupling.

JimB

 

[Optikon]

Well, if you want to post the block diagram, sometimes it helps to get it looked at by someone who has a different perspective.

Ron H,

I'll "electronify" it and get back to you with it..

Thanks for all the help!!!

Jim, I think what you describe works just fine for switching. Unfortunately, I'm not in a position to have relay disassembly an option for volume manufacturing. Thanks!

 

[JimB]

[ Unfortunately, I'm not in a position to have relay disassembly an option for volume manufacturing. Thanks!

In that case why not approch a "coil winding company" and see if they can make something in the quantities you require at a price your project can afford?

JimB