Memory wire

[Rhys]

I am just wondering if any members on this board have had practical experience with using memory wire to actuate points etc on a layout. I've been reading up about it and there doesn't seem to be much mentioned on the internet. I'm interested to hear about how well it works over long periods of time, and some first hand accounts would be very useful.

[Rhys]

Memory wire is very temperature sensitive: a setup installed on your warm attic in summer will cease when you take it to a sparsely heated hall for a show in late november.... :-\ Personally, I wouldn't use memory wire for anything vital to operations, like points, signals and other items. It's perfectly OK for animation (a guard raising his flag on a level crossing, etc) but if it needs to change position reliably in all environments, use a motor or solenoid to power it.

I was planning to conduct some temperature tests to see just what sort of movement I was getting under varying temperatures, and factoring that in to the movement. As I'm a scientist I find the concept a bit to sexy to pass up (maybe I should get out a bit more?). With a bit of forethought it should be OK (and I'm using it to work N gauge hand laid points, so I think it should be possible to get enough movement regardless of the air temp.

Being based on the other side of the world, we don't tend to get the temperature extremes that you would see in Europe. Having said that, at one train display last winter my 5 minute araldite was taking 15 minutes to set.

[Donw]

Where temperature can affect the operation you need to build in compensation. You can see examples of point rodding where compensators are used so that half of the rodding is in tension and the other half in compression so any temperature changes cancel out. This would very much complicate using memory wire. Personally I would go for tortoise point motors the stall mechanism effectively absorbs the temperature changes.

Donw

[Nickey Line]

Hi, you may be interested in this...

 

My link

 

...and this...

 

My link

 

The memory wire itself is shielded from drafts and therefore far less susceptible to temperature changes.

I've been using these units for some months now and the only problem I've encountered is if the wire comes into contact with the plastic.

This can result in the plastic melting thereby destroying the unit!

Care needs to be taken in constructing the units to ensure this doesn't happen...

[bertiedog]

I experimented with Memory Wire a lot, and used it for points, perfect operation, as long as the layout is in a stable temperature all the time, and I do not think it's at all suitable for exhibition layouts, a bit un-predictable for this use for points, but OK for signals.

 

But at home it works well, summer and winter, but not in extremes or sudden changes. It is a tidy system, small, but has a bigger footprint under the board than say a tortoise or coil motors.

 

Once set up carefully, the lifetime is as good as any electronic gear, the metal lasts, but the power used must come from a properly regulated source, no over voltage etc.

 

Stephen.

[Nickey Line]

I experimented with Memory Wire a lot, and used it for points, perfect operation, as long as the layout is in a stable temperature all the time, and I do not think it's at all suitable for exhibition layouts, a bit un-predictable for this use for points, but OK for signals.

 

But at home it works well, summer and winter, but not in extremes or sudden changes. It is a tidy system, small, but has a bigger footprint under the board than say a tortoise or coil motors.

 

Once set up carefully, the lifetime is as good as any electronic gear, the metal lasts, but the power used must come from a properly regulated source, no over voltage etc.

 

Stephen.

 

Bertiedog, I suggest you take a look at Pinkmouse's very neat memory wire turnout motor (link in earlier post). Big footprint? I think not!

Neither is it susceptible to fluctuating temperatures. My version takes up approx. 30mm x 70mm x10mm. Compare that to a Tortoise!

[bertiedog]

I can't see how the temperature differences can be disregarded, they are acceptable as I said at home in a relatively stable environment, which may not be the case in exhibition halls.

 

The experiments I did where extensive, for a point motor maker, to find out the extremes they would work under, and they are more sensitive than other types of motor..

 

I rigged strain gauges between the blades and the switch rails, which read the actual force on closure, and it read like a thermometer, as the temperature varied, so did the force.

 

But it maintained contact despite these differences, but I found the critical thing was to set the mid position at the temperature you expect to use them under.

 

Now, in comparison , ordinary motors do not alter with temperature at all......so any changes the Memory wire has are a disadvantage, but can be controlled, designed out, or suppressed to a point where it is under control.

 

The best method of suppression is a temperature sensor feeding a compensating supply, complete stability results. This worked a treat, but adds complexity.

 

One the size, the ones shown are very small, about 1/2 the size I used, but I found a larger unit was more powerful, with leverage to multiply the power from the longer wire. This could handle two changeover switches as well as the points very reliably. Narrow, but long, and longer than a twin solenoid.

 

This was why I used the expression foot print, rather than saying larger or smaller, the Tortoise has a small footprint but is massive in volume.

[Pinkmouse]

Certainly. But does the unit have switches build-in? The Mighty Mole (as delivered by the Proto:87 Stores) has one, the Tortoise has 2 and the Bemo/Herkat unit have 4, whereas the Lemaco/Lematec offers 2-4 switches. And the Bemo/Herkat unit aren't that much bigger then your memory-wire unit

 

But are they as much fun?

[kipford]

We built 20 units for one of our club layouts. We ripped them out and changed to Tortoise! Why? Temperature stability in our club room, but more important we were getting corrosion and hence poor contact between the wire and posts. I would never use them now, Servos are the best route.

 

Dave

[Pinkmouse]

... but more important we were getting corrosion and hence poor contact between the wire and posts.

 

So you didn't crimp or solder the power feeds to the memory wire? I'm not surprised you had a bad contact then. Would you just twist two bits of wire together and expect to get a reliable connection?

[bertiedog]

It was always my understanding that crimping, or a screwed connection is advisable with all common types of memory wire. The alloy used does not solder too well, and if a flux works with it, any traces will corrode the wire.

 

I always connected with a tiny copper sleeve, inserted into an electrical connector block, and the copper sleeve was crushed on to the wire, very firm contact and no chance of corrosion.

 

I found the wire some what brittle, and if a screw bore direct on to the wire it could sever it easily, the copper tube protected it completely.

 

Stephen

[Donw]

If you want to play with memory wire goahead. But don't be surprised if it gives you some hassle. If you just want your points to work try the tortoise worth every penny in my book.

Donw

[Pete the Elaner]

What advantages do you think memory wire would have over solenoids or motors?

 

I believe Dagworth has used memory wire for raising/lowering of pantographs on AC electrics.

[kipford]

kipford, on 09 May 2010 - 09:52 , said:

 

... but more important we were getting corrosion and hence poor contact between the wire and posts.

 

 

 

So you didn't crimp or solder the power feeds to the memory wire? I'm not surprised you had a bad contact then. Would you just twist two bits of wire together and expect to get a reliable connection?

 

No they were all crimped!

 

Dave

[beeman]

Interesting to hear other members findings on memory wire as i have recently been 'actuating ' Ratio signals using this material. I made a number of types of mechanism before setling on Mk.4 variety. The signal is mounted on a wooden plug that fits into a similar sized baseboard hole so that the signal/actuator drops into the baseboard from above. I found that any friction in the actuator had a detrimental effect on its operation in my early type design. I think my final model is the possible ultimate way of using this, for signals at any rate. The actuator is mounted on an umbrella spoke. I have a crimp on the looped upper end of the memory wire along with a phos/bronze wire, which is eventually soldered to the signal operating wire for adjustment. The M/W bottom end is clamped by an 8BA bolt, the bolt was soldered to a small piece of stripboard, epoxied to the umbrella spoke. The 'load' spring is attached to the spoke with a small hook soldered to the spoke, made from a brass pin. The M/W loop hooked onto the spring ,the P/B wire passing thro the spring for attaching to the signal. I use a purchased 3.3v PS unit for operation and found some of the signals will operate on around 300 mA. Each actuator has a series resistance, made from wire off the roll, to adjust its operation. The M/W is 100mm long, this gives a movement of around 3mm. My most comprehensive is a LNER double bracket with 3 operating arms. Didnt bother with the distants, pleased to get the 3 homes working due to the fact that the Ratio signals are somewhat fiddley for a 70 year old. I have made a total of 8 posts with a total of 15 operating arms all fitted in the same way such that they can be 'unplugged' out of the baseboard after disconnection below should maintenance be nescessary. Please no comments from the 'rivet counters' about making them 'bounce', I am well pleased with their realistic effect and operation. I will attempt to do a Dwg of the method I concocted If required, having a spot of bother with the PC at present. Beeman

[2ManySpams]

Memory wire is very temperature sensitive: a setup installed on your warm attic in summer will cease when you take it to a sparsely heated hall for a show in late november.... :-\ Personally, I wouldn't use memory wire for anything vital to operations, like points, signals and other items. It's perfectly OK for animation (a guard raising his flag on a level crossing, etc) but if it needs to change position reliably in all environments, use a motor or solenoid to power it.

 

The signals on Treneglos were operated by memory wire but we too found that the differences in temperature during the day in an exhibition hall may ongoing recalibration through the day a requirement. I'd seriously consider using tortoise or similar slow action point motors next time round.

[beeman]

Perhaps a bit late in the day for comments, but for reference to any potential users. What about the old reliable Omega loops,you can forget your electronic compensation and variable temperatures if you engineer the mechanism to accept these variations.With both signals and points it is possible to fit 'stops' to the operating mechanism with Omega loops to take up any excess movement which will cater for any temperature variation. It can depend on the ingenuity of the maker to achieve. Ive been making solenoid operating uncoupling units of late,using the coils and armatures off of small relays, but whilst making/fitting found certain limitations with the initial design and modifying to model 4 before being satisfied. Says one thing, perseverance is rewarded eventually. Beeman