An Electromagnet Alternative

User avatar
Tony W
Posts: 69
Joined: Thu Apr 02, 2009 10:42 pm

An Electromagnet Alternative

Postby Tony W » Fri Sep 02, 2016 8:22 pm

Recently, there has been quite a bit of discussion again on this forum about the use of electromagnets, particularly in relation to the operation of AJ couplings. A common thread to this, and previous discussions, has been about the required amount of power needed to operate the coupling effectively. Nigel Cliffe and others have offered sound advice on this matter, but as yet nobody seems to have come up with an alternative solution.

Over the past few months we have concentrated on ironing out a few issues with the running or our layout, Slattocks Junction, and we will shortly be at the stage of installing magnets for AJ operation. Our layout is DCC operated and for some time our electrical guru, Mike Bolton, has expressed concern about the power requirements to operate electromagnets. He has, however, come up with an idea which overcomes the need for electromagnets altogether.
DSC08811 1.jpg
DSC08811 1.jpg (63.04 KiB) Viewed 4118 times

DSC08812 1.jpg
DSC08812 1.jpg (66.69 KiB) Viewed 4118 times

DSC08813 1.jpg
DSC08813 1.jpg (65.84 KiB) Viewed 4118 times

DSC08814 1.jpg
DSC08814 1.jpg (52.58 KiB) Viewed 4118 times

The system uses a 6mm Neodymium disc magnet, which can be obtained easily from various sources, and uses a servo for operation. This configuration has a number of advantages:

1) The servo takes a lot less power to operate than an electromagnet.
2) Having a larger surface area of magnet significantly reduces the chances of the dropper ‘missing’ the magnetic pull.
3) The servo is, of course, capable of being ‘fine-tuned’ which means that the height of the magnet when operated is adjustable.

I’m not going to give sizes of materials as the size of the magnet, which governs the diameter of the other materials, isn’t critical, it’s just that Mike had 6mm in stock and on test so far it seems to work very well. However, if a magnet of too small a diameter is used problems in activating the dropper may be created.

Please note that the tube upon which the magnet is mounted is aluminium as this places less strain on the servo during operation than a heavier metal.

Also, it is not the idea to create a piston so the aluminium tube and the magnet should be a nice sliding fit in the brass collar, which can be covered with a piece of plastikard before ballasting.

Finally, the question has to be asked, “is this a worthwhile alternative to the electromagnet?” I suppose that the answer will inevitably be, “to some, yes, to others, no”.There seem to be clear advantages if used on a DCC layout but I should stress that it hasn’t been tested on a layout yet.

If anybody is interested I hope to have the test rig with me on the Society Stand at Scaleforum at the end of the month.

I was going to post a video of it in operation but I'm afraid my patience has run out completely!!

Tony

nigelcliffe
Posts: 486
Joined: Mon Jul 21, 2008 8:31 am

Re: An Electromagnet Alternative

Postby nigelcliffe » Fri Sep 02, 2016 9:12 pm

I built a servo test rig like the one you show about 6 or 7 years ago, had it with me at the NEC Warley show in 2010. I concluded that there are issues with it, so like any solution, it's horses for courses. For me, I concluded its not worth the hassle of building it.

First current draw. Electromagnets, such as the common Seep unit don't draw that much current, a few hundred milliamps each. They are only used intermittently, and usually one at a time. So, the current draw isn't important. (There are a few poorly designed electromagnets around, such as the "PK" branded ones which do draw stupid amounts of current). I've found the Seep units work fine on 12v DC, so have set things up with a 15v DC supply. Compared to Servos, the current draw is similar in practise as a Servo motor requires a decent power supply to cope with its movement.

Complexity of control. Servos need a control logic board, an operating switch somewhere. Electromagnets just need a switch, though if using a bus-system to a control panel then a control board is probably sensible. Coldfair Green has control panel, and underneath are bus decoder boards (LocoNet based in our case), which are about the same complexity as Servo boards. But our electromagnets can also be operated by local push buttons which means they can still operate with a total electronics failure.

Complexity of build. The servo mount has to be built (or bought), the sliding tube assembly built, connecting rods attached to servo arms. This is a lot more work than drilling a hole and poking the core of an electromagnet through it.

Cost. I think the electromagnet comes out cheaper. But not much in it.

Final issue for me is the field from the rare-earth magnet is still significant, even when they are drawn down the tube. Because I build AJ's onto locos which can be operated from a DCC function key, and those mechanisms also use rare-earth magnets, I've found the "down" servo tube magnets can cause my locos to uncouple (because my in-loco mechanism is very low friction, so any small field will cause it to move). Probably not an issue for most people, but if you ever get a visiting loco from one of the dozen or so people who've bought the AJ mechanism etches from me in the past, their locos might randomly uncouple.


I do rate rare-earth magnets as an uncoupling solution when no power is used, so placing the magnet on a purely mechanically moved arm. For very small layouts, the advantage of no power needed is a significant advantage over any powered solution.



- Nigel

User avatar
Russ Elliott
Posts: 930
Joined: Thu Jun 02, 2011 6:38 pm

Re: An Electromagnet Alternative

Postby Russ Elliott » Fri Sep 02, 2016 9:39 pm

Those Neodymiums are so powerful, I think I'd be worried about ripping out all the undersides of a wagon chassis...

Btw, the SEEPs on Green Street are 24V I think, and have heavy-duty push-buttons. We had to get the tops of the poles as far up as we could (with just a single layer sprinking of ballast) to get them reasonably reliable though.

User avatar
Tony W
Posts: 69
Joined: Thu Apr 02, 2009 10:42 pm

Re: An Electromagnet Alternative

Postby Tony W » Sun Sep 04, 2016 11:20 pm

Nigel
I agree with everything you say (and apologies, I wasn't aware you'd made such a test rig several years ago) but Mikes view is, "since I've already installed the necessary control logic boards for operation of the turnouts it means we already have a power supply, so why go to the trouble of installing an additional power supply and incurring unnecessary expenditure on electromagnets?" Only time will tell if it's the right decision....!

Russ
I hear what you say about the magnets, and in operation on the layout you may well be right, but from the evidence of the test rig it does suggest that this may well be the way to go, for us at least!
Last edited by Tony W on Mon Sep 05, 2016 2:58 pm, edited 1 time in total.

User avatar
grovenor-2685
Posts: 3097
Joined: Sun Jun 29, 2008 8:02 pm

Re: An Electromagnet Alternative

Postby grovenor-2685 » Mon Sep 05, 2016 7:14 am

Tony,
Unfortunately your video is private, you need to make it public on Youtube or provide the sign in details.
Regards

User avatar
Tony W
Posts: 69
Joined: Thu Apr 02, 2009 10:42 pm

Re: An Electromagnet Alternative

Postby Tony W » Mon Sep 05, 2016 1:19 pm

Keith
Is that any better? I don't know why but this video has been a nightmare from start to finish!
Tony

User avatar
grovenor-2685
Posts: 3097
Joined: Sun Jun 29, 2008 8:02 pm

Re: An Electromagnet Alternative

Postby grovenor-2685 » Mon Sep 05, 2016 1:25 pm

Tony W wrote:Keith
Is that any better? I don't know why but this video has been a nightmare from start to finish!
Tony

Fine now, I've edited your post so the video shows up.
Regards

nigelcliffe
Posts: 486
Joined: Mon Jul 21, 2008 8:31 am

Re: An Electromagnet Alternative

Postby nigelcliffe » Mon Sep 05, 2016 2:50 pm

Having viewed the video, I'll add another concern:

Time lag - there is a significant lag from pressing the button to the magnet coming up. This is because the servo is set to operate at a modest speed, which means the servo will be quieter (though not silent) and not jolt the baseboard.
But, with a train propelled over a magnet, the conventional approach to an AJ is to wait until the relevant axlebox is level with a marker known to the operator, and activate the magnet. With the servo board as configured, the operator will need to anticipate the marker by an amount relative to the speed of the train and the speed of the servo movement.

I'd get this approach tested on the layout with some trains under control before a full installation. My personal view is that it's not that good, because it will either lag (infuriating!), or if configured to be quick, will have a Servo buzz as the magnet shoots up. The buzz will be noticeable, and also masks the quiet sound that an AJ usually makes as the dropper hits the sleepers or ballast, thus telling the operator that it has worked.


I am coming to the view that Servos aren't that wonderful for model railway applications (even though I've installed a fair number on layouts), and I'm thinking that either stepper motors or stall motors (Tortoise, etc.) may be better because they are quieter.


- Nigel
Last edited by nigelcliffe on Mon Sep 05, 2016 3:43 pm, edited 1 time in total.

User avatar
Tony W
Posts: 69
Joined: Thu Apr 02, 2009 10:42 pm

Re: An Electromagnet Alternative

Postby Tony W » Mon Sep 05, 2016 2:59 pm

Thanks Keith

Tony

Lindsay G
Posts: 126
Joined: Thu Aug 07, 2008 2:16 am

Re: An Electromagnet Alternative

Postby Lindsay G » Tue Sep 06, 2016 12:23 am

I developed a similar arrangement several years ago which appeared under an inappropriate heading of manual turnout operation so may not have been seen by many. 'Fraid I seem to have spent most on the intervening years on Burntisland 1883 and other distractions so it has yet to be tested to death but seemed to work pretty much like electromagnets from 51L, if anything with a lesser pull so wouldn't part the chassis from a wagon.

I'd also like to use Nigel's DCC AJ arrangement on engines for uncoupling as that doesn't involve an unprototypical reversing of wagons/coaches over a magnet, but if they're affected by magnets retracted even 10mm below the surface then that is disconcerting - something else to be tested in practice. Nigel's arrangement would equally be affected by electromagnets operated too soon which is pretty easy when operating a large layout and the difference between the droppers on either end of a following wagon might only be 30-odd mm apart.

In the video the servo does operate slower than it could, but that might be for demo purposes. It is true that the quicker some move, the noisier they are - but is that a show-stopper? At exhibitions they aren't heard at all. At home I haven't found them off-putting - if they are, just crank up the volume on the engine DCC sound, or Black Sabbath/Holst/Elgar depending on the period modelled.

Lindsay

martin goodall
Posts: 973
Joined: Fri Oct 02, 2009 6:20 pm

Re: An Electromagnet Alternative

Postby martin goodall » Tue Sep 06, 2016 5:00 pm

If I may be forgiven for lowering the high scientific tone of this discussion, I read a suggestion from Iain Rice some time ago for a much simpler alternative to electro-magnets. This is a purely mechanical system whereby a permanent magnet is held on a hinged mounting under the track, and is raised by a lever into the operating position to work the couplings, then lowered again by the lever when uncoupling is completed.

I have never tried it for myself, but this crude and simple method has a certain appeal for someone like me who is a devout techno-phobe, and would run a mile rather than having to have any dealings with servos.

Meanwhile, I have successfully used electro-magnets on my layout for a number of years (plus a few permanent magnets in certain sidings), so the need to look for alternatives has never arisen. Nonetheless, I remain intrigued by Iain Rices's idea.

JFS
Posts: 658
Joined: Wed Dec 16, 2009 3:47 pm

Re: An Electromagnet Alternative

Postby JFS » Tue Sep 06, 2016 8:21 pm

Is it me losing the plot here? I just have this uneasy feeling that the "solutions" are infinitely more complex than the alleged problem!

I can hardly claim to be an expert on AJs - my layout uses them but as it is all passenger stock, it is hardly representative of the typical. That said I have spent a fair bit of effort getting them to work and, I suggest, when they fail to operate, it is "never" the electromagnet at fault!

I don't know the situation on Slattocks, but I did note in the other thread that the writer mentioned that he was referring to HINGED AJs. In my (limited!) experience on a previous layout, I simply could not get this design to uncouple regardless of the power of the magnet. The reason? There is no sideways flexibility in the hinge, thus when the dropper is drawn down by the magnet, the shank is unable to pass round the tail of the opposing coupling. Thus, either it simply hangs up on the opposing tail, or, worse, it drags that coupling down as well. Either results in a failure to uncouple. Putting it another way, the flexibility in the shank of the original AJ design allows the sideways deflection essential to successful uncoupling. The magic of the AJ concept lies exactly in here - when the the dropper rises after passing over the magnet, the shank returns on the "wrong side" of the opposing tail preventing re-coupling. Hence sideways flexibility is intrinsic to the principle. I am looking forward to hearing from those who have overcome this issue with a hinge design exactly how this is achieved.

I was also very surprised by Nigel's comments above that the AJ dropper "hits the sleeper or ballast". In a properly set-up coupling it most certainly does not - the shank of the coupling is restrained by the axle before the dropper touches anything - otherwise the vehicle will be derailed when uncoupling on the move which is the whole point of the concept after all.

I am slightly concerned that the whole world seems to be trying to invent "a better AJ" whereas I would venture to suggest that the original design (whilst not without its inherent weaknesses) was extremely well thought out and almost impossibly to better.

My experience of AJs was transformed by reading Chris Pendlenton's article in MRJ 223. Therein, he answers every question and provides plenty of examples of how to solve some tricky problems. In particular, the traditional problem of them getting out of alignment is neatly solved by the "forward pulling post" idea - utterly brilliant! I did as The Man said and it worked for me!

For what it is worth, I buy my electromagnets from Wizzard Models (because I am lazy - sorry - busy) and power them from a 12v transformer salvaged from a low-voltage halogen lighting system, rectified with a bridge rectifier and with snubber diodes across the coils. Hardly rocket science even for a steam engineer like me!

Best wishes,

Lindsay G
Posts: 126
Joined: Thu Aug 07, 2008 2:16 am

Re: An Electromagnet Alternative

Postby Lindsay G » Tue Sep 06, 2016 11:56 pm

Hi JFS,

This thread isn't about the AJ coupling itself but the means by which it is made to work (albeit that there was passing reference to Nigel's coil operated AJ's fitted to DCC engines).

Traditionally this has been via an electromagnet. Alternative solutions to the electromagnet have formed the basis of the thread, all involving magnets that are moved upwards when uncoupling is required either by servo (or possibly stepper or other motors) or manually (wire in tube or whatever). Some, perhaps the vast majority, may be perfectly happy with using electromagnets for AJ activation. Those involved with Slattocks had their reasons for wanting to move away from electromagnets. I don't know why Ian R advocated an alternative approach, perhaps it was akin to mine.

I was initially following the conventional route and wired my own layout for electromagnets but hadn't yet bought them. I was a bit concerned about having to use 24 strand wire (I was told it was needed) and the fact that buttons shouldn't be pressed too long and that there are arcing issues. You can then add to that the cost of buying a good number of commercially available units as there was no way I was going to spend the time on winding my own items. When I came across servos for signal operation and introduced to them physically, I started to think of other uses for these cheap but powerful and reliable items such as pointwork or elsewhere, and it was then that I came up with the idea of AJ operation. Hardly a Eureka moment, merely an alternative to the accepted norm.

There must always be room for questioning the norm, or else we never progress beyond where we are - like P4, and EM, stemming from 00. Some ideas will reposition the norm (CSB's notably in the last few years), others will not. 'Tis all a matter of turning the theory into practice and proving things or otherwise. Let's hope that theoretical debate on AJ activation doesn't take over and run to 13 pages as is happening elsewhere on the Forum.

Lindsay

P.S. As a wee warning, an electromagnet was fitted to the new engine works boards on Burntisland 1883 (by me!), the button being on a horizontal surface. During a coffee break (we have a few of them), there was a smell of burning and it turned out something had been laid or fallen onto the button keeping the electromagnet permanently on - luckily beyond the smell and a highly discoloured 51L unit there was no more damage done - but under different circumstances there might have been. A few lessons were learnt there.

JFS
Posts: 658
Joined: Wed Dec 16, 2009 3:47 pm

Re: An Electromagnet Alternative

Postby JFS » Wed Sep 07, 2016 6:26 am

Lindsay G wrote:Hi JFS,

This thread isn't about the AJ coupling itself but the means by which it is made to work (albeit that there was passing reference to Nigel's coil operated AJ's fitted to DCC engines).



Hello Lindsay,

Fair point, but I am still trying to understand why it takes so much "power" to operate other people's AJs. I argue that there must be a reason for the couplings being "stiff" and if that is uncovered and fixed then the problem vanishes. You have failed to convince me that the potential issues with with electro magnets are any less than those which might arise from servos.

Perhaps Tony can be persuaded to amplify exactly what problems were foreseen on Slattocks?

Nothing wrong with challenging the norm, but my fear is that beginners might rush to complexity when there may be some very simple reasons why their electro magnets fail to work their couplings. Equally, anyone challenging the norm is under an onus to demonstrate that are not inventing a more complex wheel!

And these solutions are complex - and, in the event of problems arising with the servos, by no means easy to fix in the heat of an exhibition.

Best wishes,

User avatar
Tony W
Posts: 69
Joined: Thu Apr 02, 2009 10:42 pm

Re: An Electromagnet Alternative

Postby Tony W » Tue Sep 13, 2016 10:30 pm

Hi JFS
There are absolutely no issues with Slattocks (at least, non relevant to this thread :D), it's just that Mike (our electrical guru) didn't fancy installing another power supply when there was already one in situ which could be utilised. We don't have a problem with electromagnets, and have used them for many years, but given Mikes reasoning it seems logical that, at this stage, we should should try out what he has proposed. He is concerned, however, that we don't have the situation Lindsay described on Burntisland. I don't (at this stage) accept what has been written in this thread about the problems of delay since the hand uncoupling in the video is done at a far greater speed than uncoupling should be done on a layout. From the experience of those who have operated servos in the heat of an exhibition I'm not aware that they give rise to any more problems than electromagnets do.

I suppose, at the end of the day, my reason for posting this alternative was because people seem to be getting so bogged down with all sorts of issues with electromagnets in the other thread that it seemed appropriate to offer an alternative.

With the Alex Jackson coupling never take anything as 'definitive' - it is always capable of being improved (except, perhaps the hook, itself). I've said this on many occasions but what worries me is that people sometimes take things written as gospel when in fact what is really being written is the latest thoughts of people seeking practical alternatives.

The coupling and its operation can take many forms. What suits one may not suit another. For example, Nigel Cliffe say's he isn't a great lover of servos and identifies stall motors such as tortoise as potentially a better option. On Slattocks we went down the tortoise route for turnout operation and eventually gave up on them as they were no end of trouble so moved to servos. We still have them in the fiddle yard and they can still give us problems from time to time. You've said you prefer the fix coupling to the hinged yet I find the latter to be far superior - its each to his, or her, own and you use what suits you best - there's no 'wrong' solution.

Apologies to Lindsay - I know this thread is about magnets not the coupling itself but since JFS has raised the issue about the coupling not clearing the tail of the opposing coupling in this thread I'll briefly answer it here. JFS - what you have described has got nothing to do with the power of the magnets, or sideways flexibility of the hinged coupling. Since I've been there, done that, and got the teeshirt (didn't I use that phrase in the book?) I'll stick my neck out and say I'll guarantee your problem will be a burr on the tail of the coupling, and/or a counter-weight which is too light. My advice would be to use a stone to deburr the coupling tail and to stick a bit of blu-tack on the counter-weight to see if that cures the problem. I'm sure it will. I had exactly the same problem all those years ago when Norman Whitnall was teaching me the rudimentaries of the coupling - he soon put me right.

Hope that helps - I'll shut up now!

Tony


Return to “Couplings”

Who is online

Users browsing this forum: No registered users and 1 guest