Successful Use of Electromagnets?

Discussion in 'FAQs' started by Gary S., Jan 15, 2007.

  1. Gary S.

    Gary S. Senior Member

    It worries me that only one person has reported the successful use of the electromagnets:confused:
  2. Nomad

    Nomad Active Member

    Gary. if you do a google, there's a lot of bad stories about them. That's what turned me off.

  3. Gary S.

    Gary S. Senior Member

    Loren, I'm determined to make these suckers work on my layout. I think the double coil system is the way to go. I am building a mock-up piece of my layout shelfing, and will use it to experiment, both with the electromagnets and with a mechanical turn-out control for my switches.

    My current thinking about the electromagnets is to buy 2 of the # 307 electromagnets, and buy 1 of the #810 O scale electromagnets (which comes with 2 coils). The #307 is about $8 and the #810 is $11, for a total of 27 dollars. This will make 2 uncouplers with double coils and are strong enough to be mounted under the track without cutting through the ties.

    Also, I have developed a control system for the uncoupler using a lighted pushbutton (momentary) and a relay/time delay so that pushing the button will make the magnet energize for 5 or so seconds then turn off. I probably mentioned this already.

    I'll take some pics in the next few days of my experiments.
  4. Nomad

    Nomad Active Member

    You go, Gary:) . If your as tenacious with these as you were with track planning and operation, they will definately turn out right.:D

  5. Gary S.

    Gary S. Senior Member

    I'm going to bump this to the top one more time to see if anyone else has successfully used Kadee electromagnets.

    I am still experimenting to figure out what is the best system for the magnets, plus my "one-shot" timing circuit for the pushbutton worked flawlessly. My workbench is out of control though!

    Attached Files:

  6. Nomad

    Nomad Active Member

    Looks like Frankentrain's lab:D

  7. Nomad

    Nomad Active Member

    Hey Gary:wave: How you doing with the uncouplers?

  8. Gary S.

    Gary S. Senior Member

    Coming along nicely. I've spent the last week gathering supplies and putting components together. I've got some lighted pushbuttons and installed LEDs in them, built my relay modules and time delay circuit, got my power supply into a cabinet, and have 5 of the dual coil electromagnets put together. I don't have the camera this weekend, but will get some photos next week.

    On the electromagnets, I am using a 24V DC supply. The two coils are connected in series so each is getting 12 volts with a current draw of 1.5 amps or so. I also added extra metal sideplates which help "focus" the magnetic field to where it is needed. A 12 volt supply could also be used by connecting the 2 coils in parallel, but be aware that the 12 volt supply would have to be capable of putting out 3 amps.

    The O scale Kadee electromagnets come with two coils and four sideplates for around $11. This seems to be the most economical way to get the coils. I combined the parts out of these with the HO scale electromagnets I already had, and I used some sheet metal to make the extra sideplates. I also made some mounting brackets out of aluminum sheet so that I can easily mount the magnets... my layout base is 1/4" birch plywood with 2" of blue foam on top.

    As mentioned, I'll take some photos next week.
  9. Nomad

    Nomad Active Member

    Right on:thumb: , glad to hear there working. Where are you finding them at thay price?

  10. shortliner

    shortliner Member

    A single coil one ,with thicker steel sideplates might solve the problem - the couplers need to be moved further to the side and it sounds like they aren't moving far enough
  11. tetters

    tetters Rail Spiking Fool!

    A tutorial is definately in the works.

    I need step by step photos, parts list, and idiot proof instructions. I need all the help I can get.

  12. Gary S.

    Gary S. Senior Member

    I did a strength test with all manner of single coil and 2 coil configurations with various number of sideplates, and ended up with the 2 coil version as my choice.

    Here is the history: Probably 6 or 8 months ago, I had bought a kadee #307 which is the "through-the-ties" version of the uncoupler. I did not even realize they made a Kadee #309 which is the "under-the-ties" version. It is my understanding that Kadee has discontinued the #307. The instructions with the "through-the-ties" #307 stated the coil should be used with a 12 volt source. So I assumed that 12 volts was about the max that the coil could withstand. Now I have discovered that the #309 under-track model uses the same coil but they tell you to use 16 to 18 vdc. In the kadee 309, they include a full-wave bridge rectifier and a filter capacitor so you can use AC and convert it to DC for the uncoupler.

    Well, I already had a good 12 volt DC supply and a 24 volt DC supply, and with my assumption of a max 12 volts on the coil, I began experimenting. With only 12 volts on the coil, it just wasn't strong enough no matter how much metal i added to the sides. So I doubled up two uncouplers... 2 coils and 2 cores with a longer bolt, and double sideplates. The coils were connected in parallel at 12 volts. This double coil arrangement draws around 3 amps, so you need a pretty beefy power supply.

    The kadee electromagnets are only for momentary use, they will get hot if you leave a continuous current on them. I started thinking about a "one-shot" circuit with a momentary pushbutton and quickly came up with one using an on-delay timer and a relay. Well, I already had some relays with 24vdc coils, so that is what prompted me to put the 2 electromagnet coils in series so they could work on 24 volts from my 24 volt supply. I have already built 5 of these things and have the parts on the way for 5 more.

    In retrospect, a single coil with extra sideframes and operated at 18 volts dc will probably get the job done. I'm sticking with what I have because of the power supply and relays that were already in my possession.

    One other thought: I purchased a bunch of LEDs for my lighted pushbuttons and turn-out indicators, and some resistors from "JAMECO" which is an on-line electronics supply. Prices were good, and the stuff showed up on my doorstep a mere TWO DAYS:) after I placed the order! I ordered on-line some pushbuttons and other electrical stuff from a place called "Factorymation"... got the stuff in TWO DAYS:) ! Also, I ordered some more uncouplers from Internet Trains and it has been 5 days and I have not even been notified that the stuff has been shipped yet:cry: .
  13. Gary S.

    Gary S. Senior Member

    The LHS had the O scale model for $12, I bought what they had, and ordered some more from Internet Trains for $11.
  14. BigJim

    BigJim Member

    The trick with any electromagnet is to focus as much of the flux where you want it. Flux = AMP turns > smaller wire = more turns & less amps > bigger wire = less turns and more amps.

    Since the coil size determines the AMP turns per volt the other variable is volts. The plates are used to direct the flow of the flux. For maximum flux density you want the easiest path everywhere but the "gap", the narrowest possible gap and the gap as close as possible to the required action.

    Since your two coils spread the plates and widen the gap you are losing a little density.

    Looking at the pics I would try the following:

    1 Coil at higher voltage. With your time delay (EXCELLENT IDEA!!!) and very low duty cycle you shouldn't have a problem with heat. Just remember the wattage (and flux) goes up with the square of the voltage. 12 volts = 18 watts. 18 volts = 40.5 watts. 24 volts = 72 watts. So one coil at 18 volts has more flux (and a smaller gap) than two coils at 24 volts. If Kadee uses the coil a 18 volts with a "manual thumb" delay for duty cycle then I would think the much lower duty cycle your time delay offers should allow the use of 24 volts with four times the density. Just don't hit it to often during testing.
    Larger diameter bolt. Since you don't have a picture of the coils without the plates I can't see the inside diameter. Use a larger diameter (and shortest) bolt if it will fit through the coil. Drill the plates as required. DON'T USE A STAINLESS BOLT. Multiple or thicker plates also help.

    From the picture the plates seem to span about 8 ties. Nice for having a longer uncoupling zone - bad for flux density. Unless you need a long area try cutting the plates to about 4 or 5 ties wide to focus the flux.

    Notch the top of the plates so the come up even with the top of the ties. You could still cover these with ballast.

    If I were you I might try getting some flat steel strap, about 1 > 1.5 inches wide and up to 1/4 thick, a drill a hole for a large bolt and round the bottom to match the coil. Notch the top to match the ties. Use one coil with voltage as high as heat will allow.

    Let us know.
  15. BigJim

    BigJim Member

    One more thing. Gary, I can tell from your work bench and equipment that I don't need to tell you this (and much of the previous information) but for others with less experience > since all of the power supplies we use are isolated ground there is no problem using a 6 volt and 12 volt in series to get 18 volts. Just tie the positive of one to the negative of the other and the other two connections to the coil.
  16. Gary S.

    Gary S. Senior Member

    Great food for thought. I had considered notching the steel plates so they would come up through the ties. But that seemed like a lot of work... still it is a viable option.

    As for the core diameter, Kadee supplies a circular piece of steel that fits nicely in the coil. It is drilled for a 10/24 bolt. I am using those.

    The idea of using some 1/4" flat bar for the side plates is a good one, and is probably what I should have done.

    Do I dare put 24 volts on a single coil? It would probably be okay as long as it isn't kept on for too long. Right now, I figure that 3 to 4 seconds would be a long enough time to uncouple a car. Center the couplers over the magnet, push the button, back the train a tad to give some slack, then pull forward, then back again for the delayed action so the car can be pushed back to where it needs to be spotted.

    Back to the drawing board? I'll have to put 24 volts on a single coil and see what happens, see how quickly it gets overly hot.
  17. Torpedo

    Torpedo Member

    You could try a "smoke test" to see how long a coil would hold up under 24vdc. It might be worth sacrificing one unit to establish a baseline absolute maximum on time. Besides, sacrificial testing can be quite entertaining. :cool: :D
  18. BigJim

    BigJim Member


    If you don't have one I would recommend you (and everyone else) get an angle head grinder. I see has a cheapo 4.5" one on sale for $9.99.:thumb: Might not be the quality you want for daily use but should be OK for the odd job. Haven't tried recently but with most of their tools you could get a two or three year replacement warranty for a few bucks. Wait - you are into model trains - money must not be an issue - get the Porter Cable one. Only $129.00.:)

    That and some cut-off disks will do your notches in a couple of minutes and cut your strap and round the corners and.... You hack saw will get very little use after you get one of these.

    P.S. You will not believe what you can do with the "flap-sanding" disks (the flaps are on the side - not around the edge). Wood, steel whatever. The diamond wheels do a nice job on tile, granite, concrete & rock, etc.

    BTW Harbor freight has a retail store in Houston to save on shipping.
  19. Gary S.

    Gary S. Senior Member

    The "smoke test" is definitely a cool procedure. I did that with an LED the other day. I had a 1/4 watt potentiometer, 0-5k and was playing around with LED brightness vs milliamps by turning the pot up and down. I wasn't paying attention and turned the pot all the way to zero ohms, and the result was a frying noise and smoke coming out of the potentiometer, and the LED turned all black and had some smoke oozing out of the bottom around the leads.

    I just got through putting 24 volts on a single coil. I put 4 seconds of power on the coil, about 5 times in one minute. No harm done, but the windings were obviously hot, almost to the point of being hot enough to burn my fingers.

    Next, I did a strength test, comparing one coil at 24 volts versus 2 coils in series at 24 volts. I did the test by seeing which coil set-up could lift a piece of steel from the greatest height. The 2 coil magnet could lift the steel from about a 20% greater height.

    Some of my thoughts (which could be wrong): The strength of an electromagnetic is directly proportional to the Amp-turns, not necessarily the wattage. I say this because a certain percentage of the wattage applied to the coil is being used up as heat. The single coil and the double coil have the same Amp-turns--> the single coil has half the turns of the double coil but double the amps. The double coil has twice the turns of the single coil but half the amps. So the amp-turns are equal. I can't explain the 20% difference in strength, the double coil being stronger than the single coil. Some kind of magnetic saturation, or, with the single coil, much of the wattage is being used as heat?

    With the single coil, we get 24 volts x 3 amps = 72 watts

    With the double coil, we get 24 volts x 1.5 amps = 36 watts.

    The double coil at only 36 watts is stronger than the single coil at 72 watts.
  20. Gary S.

    Gary S. Senior Member

    I have access to a 4.5" grinder, but didn't think about using a cut-off disk to make the notches.


    I can't explain it, but I have always been a "manual" kind of guy. Hack saw, files, pliers, vice grips. I just recently got a Dremel and I am finding it very useful and thinking why the heck didn't I get this sooner?

    I'll have to look up the address and check them out.

Share This Page