Author Topic: magnetic reed switch issue  (Read 2237 times)

0 Members and 1 Guest are viewing this topic.

Offline cncjerryTopic starter

  • Supporter
  • ****
  • Posts: 1269
magnetic reed switch issue
« on: January 24, 2018, 12:45:48 am »
Hello again. 

I built a vertical "bug" style CW key (W.R. Smith designed) and it works fine, or did, but during a recent move the dot magnetic reed switch was broken.  Major bummer as I had it dialed in pretty well.  It was making well over 100 dots in a row.  I was using a double pole miniature magnetic reed switch and closing the normally open contact.  I had another of the same type of reed switch that I installed but the issue was that once made, the connection wouldn't break unless I moved the reed far away from the magnet.  Obviously this wasn't going to work for CW so I order a bunch of single pole, "little Fuse" brand subminiature switches thinking that if I had to try a couple to find one that was faster on break than the others I would have them.

The first switch I tested seemed to work great until I took a phone call and I guess while it was loose in the mounting tube it must have been in the magnetic field too long as it started doing the same thing.  Once a dot was made, it wouldn't break.  The make break pendulum distance is only about maybe 1/32nd inch when working, maybe even less.  So for the heck of it, I reversed the switch and now it works fine.  This doesn't make sense since if the switch became magnetized, if that was the issue, it shouldn't matter from what side it gets closed to then break.

So is a constant magnetic field causing the switch to get magnetized so that once made it doesn't break?  I can see where the pull-in would need to be greater than the holding magnetism, but when I had the switch reversed, I had to move it about an inch away from the magnet in order to get it to break.   Whereas when it works correctly, it makes and breaks within 1/32 or less.

Anyone have experience with these small magnetic reed switches?

Thanks!

Jerry
 

Offline duak

  • Super Contributor
  • ***
  • Posts: 1038
  • Country: ca
Re: magnetic reed switch issue
« Reply #1 on: January 24, 2018, 02:30:48 am »
I had a problem  recently with a pneumatic cylinder that uses reed switches to detect a permanent magnet built into the piston.  The switches slide in grooves that are part of the cylinder extrusion and allow for some leeway in setting length of stroke but cannot easily allow for how near or far the switch is radially spaced from the magnet.  One of switches disintegrated and the other's wires were crunchy so I replaced both.  Both worked well for a few cycles then one started to stick.  After a few more cycles the other switch started to stick.  This didn't make sense.  In the end, I taped both switches on the surface of the cylinder about 5 mm further away than if they were in their grooves.  It seems that the switches were too sensitive or the magnet was too strong.  I'm ruling out contact welding as the reed switches are speced to drive a PLC and this one has been working for years.

The manufacturer/supplier wasn't helpful.  I found a few documents on line and one mentioned sticking if the switch is subjected to a large magnetic field.

I tested the switch I removed and found that if I used a much stronger magnet I could get it to stick.  I also found that the orientation had some effect but the original installation only allowed for one way.  I also found that a piece of sheet steel layed on top the the switch reduced sensitivity as it diverts magnetic flux around it.

For what it's worth,
 

Offline cncjerryTopic starter

  • Supporter
  • ****
  • Posts: 1269
Re: magnetic reed switch issue
« Reply #2 on: January 24, 2018, 04:44:40 am »
thanks for the comments.  My switch is installed inside a bar of brass that was drilled 1/8" so I'm not sure about the flux distribution.  Maybe I need a flux capacitor? :palm:

I tried spinning the switch within the tube so the orientation would be changing but once it started sticking, it stayed that way.  But the same switch when installed 180 degrees lengthwise no longer stuck.  The magnet cuts a short arc lengthwise with the switch.  Also, another switch of the same type works from either direction.  It's clearly a flux issue.  I wonder if I were to demag the switch if it would help?  They only cost $1.49 per so I can afford to waste a dozen or so.

Once I get the switch set again and adjusted correctly, the key is really fun to use.  I spent a couple hundred hours machining it.  I nickel plated most of the hardware and powder coated the base.  I have to tear down the back plate and power coat it and then practice.  The bug action takes more practice than an iambic paddle.

Thanks again.

Jerry
 

Offline mjs

  • Regular Contributor
  • *
  • Posts: 117
  • Country: fi
Re: magnetic reed switch issue
« Reply #3 on: January 24, 2018, 12:19:05 pm »
Do you happen to have any filtering capacitance that is short circuited by the switch ? Reed switches can get welded together with quite small caps, I've had problems even with 100nF/3.3V. The short circuit current peak was 4-8A! Adding a 100 ohm resistor in series helped and we had no failures after that.
 

Offline cncjerryTopic starter

  • Supporter
  • ****
  • Posts: 1269
Re: magnetic reed switch issue
« Reply #4 on: January 24, 2018, 04:32:53 pm »
the reed switch is switching an FT-897 transceiver directly.  It's low voltage and current.  The switch I have in there now works perfectly.  The only thing I can think of is that either 1) the prior switch got magnetized with the rare earth magnet I am using; or 2) maybe the switch got tacky from a broken seal.  I'm going to try to demagnetize it with a tape head demagnetizer to see if it changes.
 

Offline Leo Bodnar

  • Frequent Contributor
  • **
  • Posts: 801
  • Country: gb
Re: magnetic reed switch issue
« Reply #5 on: January 25, 2018, 02:19:30 pm »
I had very bad experience with COTO reed switches recently.
A lot of failures - like 2-5% on a few thousands sample set. We played with temp profiles to no avail then started mounting them manually and this did not fix the problem.  Then I have noticed that some had cracked glass body while still on unhandled tape.
During this time COTO reed switch relay failed in my LeCroy scope.
This completely wiped out my trust in reeds.
Just say "no"
Leo

P.S. The failures were anything from no contact at all to static 100-400 Ohms resistance to stickiness.  They were subject to 3.3V, 1-2mA  and no inductance in current path.
« Last Edit: January 25, 2018, 02:25:24 pm by Leo Bodnar »
 

Offline babysitter

  • Frequent Contributor
  • **
  • Posts: 893
  • Country: de
  • pushing silicon at work
Re: magnetic reed switch issue
« Reply #6 on: January 25, 2018, 08:35:43 pm »
Burned from reed switches here, too:

Tried a NC reed switch to keep a battery-powered device off when its in a transport box with a magnet.
Bad experience:
Each switch needed the magnet to be held in a individual position. After some playing with the magnets, the reed switch turned into a NO device.
My magic coworker was soon able to select NO or NC operation just by the right movements of the magnet.
NVE GMR sensor and FET were later used to do the trick.

BR
I'm not a feature, I'm a bug! ARC DG3HDA
 

Offline cncjerryTopic starter

  • Supporter
  • ****
  • Posts: 1269
Re: magnetic reed switch issue
« Reply #7 on: January 26, 2018, 06:07:45 am »
New switches arrived and all work as expected.  So I put the "sticky" switch in and that one now works again.  :-//

I found this one brand, other that Little Fuse, and it really works well.  My dot count went way up and the dots were more adjustable for length.

I still can't figure out what caused that switch to stick like that, what the heck. I'm thinking of changing the dash contact to a magnet as well now that the dot is working better.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf