Simple Magnet Pole Tester

[ruddycrazy]

G'day Guy's,

            Talking with Bruce this morning on irc I mentioned my simple neo magnet pole tester and he asked for details so I decided to let everyone know about it too.

All thats needed for the circuit is a UGN3503 hall effect sensor, a green led( or any colour) a 330 ohm resistor, a 0.1uf decoupling cap and a 5 volt battery supply. For the battery I just used a 4.8 volt nimh rechargeable battery.

I made the prototype on a breadboard and ever since it's done the job so I haven't worried about making a pcb for it

As the UGN5303 supply pin is at 1/2 the supply voltage when there is no guass present the led is 1/2 lit

When a north pole is passed by the sensor the led will go completely off if the magnet is strong enough.

When a south pole is passed by the sensor the led will be fully lit.

I reckon this a quick easy way to ensure the magnet poles before I do my motor conversions. The last thing you need is getting 1 magnet wrong. This little circuit will ensure you know which side the pole is on everytime and a simple texta mark can be used for the reference.

Cheers Bryan

we did a little editing and now the schematic and the part numbers are correct in the story make things easyer for people following.
Kurt

[ruddycrazy]

Doh in the schematic the led is shown the wrong way around.

[Flux]

If the IC sources current then the LED is correct.

If you think the LED is reversed then the chip must sink current and the LED must be supplied from the positive line ( probably this is the case so the connection is wrong as well )

Flux

[Flux]

I can't find anything on your chip but UGN3175 is open collector current sinking and I suspect yours is the same.

The led should be fed from the positive line and needs reversing.

If you have a lot of magnets to deal with especially if using more than one magnet per pole then it is a good idea to build something like this.

Flux

[Janne]

Hello,

Are you sure you didn't mean the "UGN3503"? When i last needed a hall effect sensor for a simple job, that's what i got since it was the cheapest

Anyways, if it was the 3503, it should be able to both sink and source current. And with "idle" output voltage of 2.5V indeed a green led should just lit up.

[Flux]

Yes UGN3503 makes more sense.

It will work as shown but the led may glow with no field. With one polarity it will go to full brightness and with the other polarity it will go out. With no field the thing is biassed at half supply volts, it is a linear device not a switcher like the UGN3175.

Flux

[ruddycrazy]

Hi Janne, Yes you are right I made a big typo on the chip it should of read UGN3503, that what ya get when it takes about a minute to make a schematic.

This simple unit will come into it's own with my 4kw conversion as i'll have 98 magnets to install.

Cheers Bryan

[bj]

  Thanks Bryan, have made a few of those mistakes, so I think I will build

one today.

[RP]

If you want to mess with it, you can delete the picture and upload a new corrected one with the exact same name.

[dsmith1427]

Is this a valid low tech version... Take a compas (the kind that has a needle marked north and south used by backpackers, hunters, etc) and hold it to the pole of the magnet.  If the north pole needle points to the magnet, then it is the magnetic south pole of the magnet.  If the south pole needle points to the magent then it is the magnetic north pole of the magnet.  Can I do this or is it not valid?  Thanks.

Thanks.  

Don

[ruddycrazy]

I tried that method with my 1.5kw motor conversion. The 2x1x1/2" neo's destroyed the compass thats why I made this circuit. As the UGN3503 is designed as a linear sensor and the sensor will go to zero output when a strong north pole magnet comes within range it's a sure fire way to ensure you get the poles correct.

The same sensor is used to find the gauss in magnets and I have to re-find the circuit I saw on the net as it would be good to test each magnet for gauss and match them in motor conversions.

When I get my cnc finished one of the first jobs will be routing a pcb for this project and use some coin cells for the power. Then the tool be be nice and small so it can get into nitty gritty places if needed.

Cheers Bryan

[RP]

Oh, Oh, Oh!  

You could put it in a keyfob shell with a momentary push button to turn it on

[Flux]

The compass needle is the age old way of doing this but it has never been easy. It takes a skilled person to do it without reversing the magnetism of the tiny needle.

Usually it sticks on its pivot and gets reversed magnetised. In extreme cases if you get too close you will pull the thing off its pivot.

I use a marked neo magnet and just feel for attraction or repulsion but again this is not something I would recommend to anyone else. It is safe if you know what you are doing. I also have a flux meter which does effectively ehat Brian's circuit does and that can be easier when you have lots of small magnets in one pole. For single magnets on axials it's not that necessary.

Flux

[bob golding]

this circuit would make a good tacho with a bit of modification. change the led for a small coil and put a reed relay in it then you could use a bicycle computer in the same way as inspeed use for there anenometer. i tried using one as a tacho but you need to get the reed relay in just the right position,and keep it there. i know you can buy hall effect sensors to do this but this looks cheaper and easier. need to experiment to get the current though the coil right. if you used 3 core cable,old telephone cable? you could power it from the ground, which would be better.

bob golding

[ghurd]

Neat idea.

Not sure how long a reed relay would last.

Could maybe mount the chip (or one of the others mentioned) on the tower, LED being the LED in a optoisolator at ground level, output to the tacho?

Can extend a name brand bike speedo cable a long way.

I do not know how far, but more than the longest I ever extended one.

G-