Electromagnet for reed switch (battery powered)

[kael]

@Collin55, so I'll get 50h then out of 2xAA, like you calculated it?

How many cycles do you want? The reed switch has vey limited life.

I thought it was the other way! Think about all the reed swicth used at bike electronic speedo's.
Here **broken link removed** I've read this:
•Because the Reed Switch has no wearing parts, the contacts can switch low level signals well in the billions of operations.

 

[kael]

I find it intresting to see how this simple setup: a reed switch that needs to be triggered by an electro magnet where both components are next to each other, can have different opinions how to realize.

@Cachehiker:
OK now I get how you do the calcualtion, one more thing;
If I do a more correct calculation of the wirelength with the, on each layer, changing circumference I get this;
Turns PI 3/8" 36AWG inches
200 3,1415 0,375 0,005 238,75
200 3,1415 0,38 0,005 241,90
200 3,1415 0,385 0,005 245,04
200 3,1415 0,39 0,005 248,18
200 3,1415 0,395 0,005 251,32
200 3,1415 0,4 0,005 254,46
200 3,1415 0,405 0,005 257,60
200 3,1415 0,41 0,005 260,74
200 3,1415 0,415 0,005 263,89
200 3,1415 0,42 0,005 267,03
200 3,1415 0,425 0,005 270,17
200 3,1415 0,43 0,005 273,31
Total: 2799,08

While calculating it in 1 pass has an about 14.5% greater value for the total wire length.
2400 3,1415 0,375 0,06 3279,73

The manufactor of the ready made small electro magnet, Kuhse Gto 18 (1,4W 12VDC) says;

this application is not the standard application of these solenoids. Normally they are used to hold a plate our something else made of steel. That is why the magnetic filed goes not more than a few millimetres away from the holding surface. Especially this little type of holding solenoid has only 2-3 millimetres magnetic field.

detailed PDF specs:
https://www.kuhse.de/fileadmin/download/datenblatt_de/Katalog_Haftmagnete_de_uk06.pdf

So it still looks like a challenge to find a suitable electro magnet to trigger the reed.

 

[cachehiker]

While calculating it in 1 pass has an about 14.5% greater value for the total wire length.

I expected more than that. When wound perfectly with bare wire, the distance from layer to layer will actually come in a hair under 5 mils. My numbers are a little fast and loose but I have to account for the nature of hand wound windings and the thickness of the lacquer. One isn't likely to get 2400 turns in 12 perfectly wound layers. I would expect a 13th and possibly even part way into the 14th. It also provides a conservative estimate of the amount of wire needed. It's much easier to stop winding after 2800 inches or unwind the last 480 inches than it is to splice in an additional 480 inches.

I'm interested to see that. According to the classic equation for magnetic field strength, B, in the center of the solenoid: B=kμ[SUB]o[/SUB]nl where k is the relative permeability of the core material; equal to 1 for air, and 5000 for iron. n is the number of turns. l is the current in amperes. μ[SUB]o[/SUB] is the permeability of space.

I oversimplified that statement more than I should have. A ferrous core can be used to reduce the reluctance in a magnetic circuit, and it can therefore be used to extend the range of a solenoid driving a reed switch. It's easy to assume that it's going to improve things by a factor of 5000 but, unless you start getting elaborate with the shape of the core, the path length of the field lines through the air limits its effectiveness to a tiny fraction of that. It's the field strength inside the reeds and not the core that matters.

I'll have to see how slow things are a work today. Maybe I'll get lucky and find a couple of already drawn diagrams online.

 

[kael]

I did some further testing with the coil of a relais, AZ530-09-02
On my first test I didn't remove the attached switch part, so all the flux went through this metal part instead the of the reed I was holding against it. So now with the switch removed from the relais, the flux searches is way through the reed and is closing alright.
The resistance of the coil is 192 Ohm & taking about 40mA on 9V.
Suppose the wire is AWG36 (5mil): 1000/441,8x192 = 434.5 Ft on the coil, if there are 20 layers and 150 turns on the about 1/4" diam by 0.8" coil then I have:
20x150=3000 total turns * 40mA = 120Ampturns
It looks I can start working with this coil.

 

[ccurtis]

A ferrous core can be used to reduce the reluctance in a magnetic circuit, and it can therefore be used to extend the range of a solenoid driving a reed switch. It's easy to assume that it's going to improve things by a factor of 5000 but, unless you start getting elaborate with the shape of the core, the path length of the field lines through the air limits its effectiveness to a tiny fraction of that. It's the field strength inside the reeds and not the core that matters.

Granted that the field acting on the reed switch will not increase by 5000, but better than no increase I would think. Agreed that shaping the core, say into a U and then the ends bent inward slightly, can significantly guide the magnetic field from the inside of the coil to the outside where it will be most effective. Something else, I think, would be to make the length of the coil shorter compared to its diameter, the windings layered more on top of each other instead of spread out. With a short coil, a greater portion of the field exists around the outside of the coil than with a longer coil.

 

[kael]

When selecting a relais for using the coil with my reed switch, does a high sensitive relais coil (i.e. 9V 44mA 203 ohm) give more AT than a standard one (i.e. 9V 59mA 153 ohm)?
There's probably more length in a sensitivity coil (=more AT) than a standard coil (more Ohm) but mA is lower (= less AT)?

 

[alec_t]

Not easy to say. I guess it depends on what current/voltage the contacts have to switch and hence the armature springiness that the magnetic force has to overcome. The specified coil dissipation is probably a good guide as to its magnetising force; more mW = more pull.