47137

It will be important to use feedback in any automation app. because the relays return to state '0' when the unit is powered on, and the power for the tortoises will need to be the same source as the power to the relay unit to be able to maintain point position in the event of power failure (trains will need to be moved clear of any points before power can be restored). It is a shame that they do not have an option for a module with latching relays.

I guess I have fallen on my feet here because I am running the layout (including the relay board) from the UPS I use for the computer in the office.

To be fair, a Tortoise motor stays in its requested state if the power fails. So the problem with stock fouling point blades will happen when the power is restored, all the relay contacts are in the off position, and all the points move to their normal routes. In fairness, I think we could expect some diversity of point positions. I hope so because this will even out the load on the split rail DC supply to the motors. Analytically, during "normal operation" (meaning, running one train at a time and with a purpose, and not trying out every turnout), the largest number of turnouts which need to be reversed for any single movement is three. This is one of the routes from from end to end along the front of the layout. For me, the longest imaginable train can only bridge two of these sets of point blades. However, this argument is very much tailored to my own layout.

Momentary power failures will be a headache, longer-term ones will be easier to deal with. The simplest solution for an exhibition might be to take along a small UPS.

- Richard.

On the same note - I guess app customisation and custom board creation is the only way forward if you wanted to control more than 8 points, using some method of multiplexing via the Bee I/O ports? Or can you see some way to accomplish this without custom boards?

You can save relays by working points in tandem. For example 2a and 2 (trap point) and 7a and 7 on my layout:

If I was pushed, I could work 3 and 4 from one relay, and even 6 with 7 and 7a.

As it stands, I have one spare relay (number .8.) which will work one point on a leftwards extension. If the extension had a run-round loop or a kick-back siding the two points could be worked in tandem. Point x is only for use at exhibitions and doesn't need a motor, and the one point on the fiddle yard is worked by a reed switch tripped by the turntable there. So all in all, I can have 12 points on the whole layout and be happy with 8 relays.

The approach using the relay board doesn't extend very easily. I think for more than 8 relays I would investigate some kind of bus-based system which addresses relays individually. I am not a great fan of bespoke electronic hardware because it makes maintainability and documentation difficult. As it stands, if the Bluetooth module fails one day I can use a USB connection, and if the relay board fails I can pull it out and rewire with 8 SPDT toggle switches.

I forgot to mention, if you want to power the relay board from the layout power supplies, remember to budget for a +5V regulator.

- Richard.

I bought the kit directly from the TinySine web site and the payment was just under £47 including the postage. I was impressed by the level of service - as though I would get the same treatment if I wanted 500 boards instead of just the one. I run AVG anti-virus free under Win 7, which has no problems here although it does detect the rmweb as a source of viruses from time to time!

I haven't worked out whether the board is aimed at hobbyists or the OEM home automation market. It seems to be a "well rounded" sort of thing, it works and I can't think of anything to add without spoiling its universality. It would be nice if you could download the source for the freebie app but I did find something called "Jad v1.5.8e" which has returned 127 files in 27 folders. I installed the Android SDK under XP most of three years ago and it was so long-winded, and my lack of Java knowledge was so overwhelming, I produced a "hello world" app and vowed never again. However, if some of these 127 files can be tweaked - for example, to do add the route clearing functionality I have described - and put back together - solved. I still have the XP box stowed away!

The control protocol does include a command to get relay states, this returns a byte where bit set = relay on. So it is possible for any app (or a Windows .exe for that matter) to initialise itself to match the relay board.

- Richard.

The module has now been in service for two weeks and has performed faultlessly.

When the board powers up, the Bluetooth link is always down and all of the relays are de-energised. So I have arranged the wiring from the board to the point motors so that in these circumstances, each turnout is set to its 'normal' route. Switching a relay 'on' then reverses the turnout.

The Bluetooth link is always initiated from the control device. I am using my Android phone with the TinySine freebie app. If the link is broken (typically by taking the phone out of range) the link drops out (and must be remade manually) but the relay settings remain. During the link creation process, the board sends the state of each relay to the app. It is impossible for the controller and the board to get out of sync with each other.

The only real drawback is the communications from app to board is unidirectional. Having wired the track for DCC, it has occurred to me I could use IR detectors to switch trailing turnouts to avoid a derailment. For this level of sophistication I would be looking towards some kind of bespoke solution to manage the relays and the IR detectors - which would take away all of the simplicity.

I am using seven relays, not six as per my original post. I decided to motorise a turnout for a siding which I had intended to be a local lever. Using the freebie Android app, I memorised the seven point lever numbers straight away. So the idea of a mimic diagram is a bit of a luxury. I do fancy the idea of some kind of route-clearing functionality, whereby setting one turnout returns unrelated turnouts to their 'normal' position. This makes me look towards doing something in Java for the Android platform rather than a bespoke hardware control panel with an Arduino. The only real snag with the phone is losing the link when the phone rings and I wander away from the layout. Perhaps I should buy a small tablet.

The Bluetooth module identifies itself as 'BluetoothBee' with a pre-set password. You need a USB adaptor to change these details and to change the baud rate. I've ordered up one of these boards, partly for engineering satisfaction but also so I can set the ID to match the name of the layout. Probably pure vanity I'm afraid, but it might start discussions at shows if anyone is scanning for Bluetooth devices.

So there - a working wireless remote control system for a layout, independent of the track controller technology, with off-the-shelf hardware, no programming and even no soldering to the board!

- Richard.

Richard

 

do you have link on ebay for this board. I am unable to find any near that price, even from china

 

Alan

Hi Alan,

This is the kit I bought:

http://www.tinyosshop.com/index.php?route=product/product&path=141_144&product_id=371

The kit contains the relay board, the Bluetooth module (which you have to plug into the relay board), a mains power supply for the +5V to drive the board and a Bluetooth dongle to plug into a USB port. The power supply is auto-ranging for the mains voltage but has US prongs so you need a shaver adaptor for the UK. So I first tried out the module in the bathroom, before I found a plug adaptor in Poundland :-)

I went for the budget USD $4 delivery service and the kit arrived in six days, packed in a sturdy box. The value was written on the package so no customs duty to pay.

Sorry for this rather tardy response, I simply didn't notice this post 6 minutes before the post after it.

- Richard.

The S stock is surface stock which is virtually the same size as main line stock.

Not really - the Met loading gauge is more generous than BR, and S stock is wider than main line stock. If these are true scale models, they will foul some "surface" model station platforms. Perhaps someone can post details of the routes used for delivery of the trains over Network Rail.

- Richard.

You mean you told "the bank manager" you wanted to buy a set???  Mistake No.1.

Experience suggests it is best to just do things, and then ask for forgiveness afterwards

- Richard.

Anyone ordered the S Stock yet? It's cash with order I believe.

Yep.

Last week I sorted out my contract to the end of the year and today the car passed its MoT. This release couldn't have come at a better (or worse) time, I cannot possibly run it at home but it will fit on the club layout.

- Richard.

Some ivy on the viewing side of the tank steelwork could soften the appearance, and also possibly block up more of the view.

- Richard.

Looks good to me, and also suitable for light railways and secondary lines where the track seems to be on top of the ballast rather than restrained by it.

- Richard.

I am thumbing through the "Railway Magazine Illustrated" for 1975 (volume 121) ...

Jan page 2 has the APT-E beside the motorway at Mill Hill, undated but implied in 1974

Jan page 3 has some details of the P train

Feb page 104 has tilt testing using APT non-articulated trailer bogies under a former Hasting line demu

Mar page 153 has the APT-E passing St Albans City

May page 236 has a P train trailer body shell undergoing strength and stiffness tests at Derby

Nov page 528 has the E train and the prototype HST at Swindon.

Hope this is of interest and sorry if this is a repeat of an earlier posting. There may be others but "APT" doesn't appear in the index for the year!

- Richard.

 

Rapido function in a certain way and that is to determine an absolute production quantity based on pre-orders and carry no inventory. In this case however Rapido are producing for a commissioning client in the form of Locomotion. The orders placed up to the deadline then give Locomotion the necessary information to determine the quantity which will be manufactured. They could choose to have more manufactured to cover any replacements necessary (and I'd support them in actually have a small quantity on a shelf or on the website at release date if it meant suppressing inflated opportunistic resales on ebay at the point of release) or drastically they could choose, if they wished,  to manufacture less if they chose to factor that some buyers who've placed a deposit may not be able to pay the balance.

 

I agree with this, but the Locomotion web site does state, "Limited Edition - produced strictly to order".

The claim on the web site may well be true, but if it includes unlimited extra orders placed by Locomotion themselves, it rings pretty empty to me.

On the bright side, it didn't influence my decision to buy at all - I am really looking forward to these models.

- Richard.

I am not quite sure what you are trying to say here.

 

Given that there is only one seller of the APT-E (Locomotion), it is really quite simple.  At whatever deadline Rapido and Locomotion have setttled on Locomotion will tell Rapido that they need x of each catalog number, at which point Rapido will go off and produce x of each item (plus a small number of extras for warranty issues).

 

If you are trying to imply that Locomotion is going to destroy they reputation they have built up by playing games with order fulfillment, then I suggest you take it up with Locomotion because baseless speculation on this forum is inappropriate.

I was trying to inspire some discussion but tapping it all out on a smartphone became too much and I probably didn't write enough.

1. To my mind, the APT-E model is a speculative venture by Rapido Trains. At the beginning of the project they ran a risk of insufficient interest, and having decided to go ahead there was a risk of under-supply or over-supply.

2. Investment by punters before the deadline was a speculative investment. You might have got a train, or your money back, or perhaps even a partial refund or no refund if something went badly wrong.

3. The removal of the deadline makes under-supply seem unlikely.

4. The prior limit of two pre-orders per customer stopped unscrupulous dealers/investors making bulk pre-orders and not fulfilling them, leaving a huge oversupply - good. But, the new limit of four per customer seems to be tailored exactly to small, if not unofficial, dealers.

5. In the event of over supply, the dealers may end up selling off the surplus stock at a discount. So this makes a new risk to the original punters (my item 2) - they may find themselves having paid over the odds by paying early.

None of this says anything bad about Rapido trains, but I do think Locomotion Models have muddied the waters by changing the rules part way through the game. A simple statement from Locomotion Models would help a lot.

- Richard.

Hi all

 

Just a quick reminder that the order book for the APT-E and Trailing cars closes on Thursday. Visit our website www.locomotionmodels.com/ to see our brilliant countdown clock and get your order placed in time!

- Richard.

It wasn't really a deadline at all! I suspect, the NRM are wanting to use unfulfilled preorders to build up their own retail stock.

The extension of the deadline sounds like bad news to me. It is as though the production run has been decided but not, as yet, fulfilled by orders. New buyers who are keen for a model can place new orders and choose to pay the full amount. The suppliers are free to say, "we will honour paid-up orders as our top priority, and we will refund the £50 deposit paid on orders we cannot fulfil". I don't like this at all.

- Richard.

Locomotion Models are now showing a maximum of four per customer on their Web site.

- Richard.

Probably easier said than done.

Have you tried soldering to very thin flexible plastic printed wiring assembly?

 

Keith

Yes I am showing my age and it is a long time since I had a keyboard in pieces. My current thinking is I do want some tactile feedback through mechanical lever switches, so I ought to use the freebie Android app for testing, and the H&M lever frame / Arduino, as a project in its own right, as the goal to aim for.

Right now, the track is wired but relies on Croc clip leads to set the frog polarity. Usability will take a leap forward when I can install the point motors and wire up the frogs properly. I could start this right now if I wasn't sidetracked into arranging a narrow gauge feeder. Then again, it is a hobby!

- Richard.

If they are going at high prices on Ebay at the time, I will be happy with the 2 sets I have managed to buy.

Seems a shame as by limiting them to 2 per customer it makes it an expensive way to buy them if peoples only choice in future would be to obtain them from Ebay sellers who are just in it to make money.

Just remember if people have not pre-ordered them before the deadline it might be that they have not got the money at the time or they might want to see the fully finished product before they part with any cash, its a shame if 30th of april is the deadline no more models made as you could easily get many future orders, otherwise its expensive tooling that might never be used again.

My understanding is the limit of two per customer is for orders with the £50 deposit. No limit if you pay outright now.

Richard.

The Heljan socket is too long and it is in the wrong place too! The socket should be short enough to let the 'fingers' on the coupler emerge and lock around the back of the socket.

The specification for the size and location of the sockets is NEM 362, "Aufnahme für austauschbare Kupplungsköpfe" - German text only as far as I know, but at least it's easy to understand the illustrations.

- Richard.

nem362-d.pdf

My only remaining issue is the supplied couplings. They are way too short. On other stock, the tension lock coupling bar is generally on the same line as the buffer faces, whereas if the Class 14 couplings are pushed home, they are too far back, and don't swivel through the slot in the buffer beam.

 

Has anyone sourced viable replacement couplings, or found a reliable solution please? I suppose I could just glue the existing couplings into the sockets and hope the glue is stronger than the weight of any trains it hauls?!?

My model has Kadee #20 couplers fitted (I use Kadees on all my trains now), these are long enough to put the rear of the knuckle in front of the buffers. So (and I know this is far from ideal for your situation), if you could attach a wagon with a similar Kadee on one end and a tension lock on the other, this would work.

- Richard.

This was last week. We sat on the bank of the Chelmer and Blackwater Navigation with our picnic (this was on a landing stage near Heybridge Basin), throwing bits of bread at the ducks, and seeing none of them would take a scrap of interest. After a while, it dawned on us we had duck sandwiches.

- Richard.

I must say this APT-E is a brilliant release, to me it remains perhaps the most futuristic train ever produced in Britain and the most stylish train made in the post steam era. I remember how as a very small boy it represented a much brighter future.

There was something special about the APT-E. I ordered a 6-car analogue set some months ago, fair enough I suppose but having seen the "cybermen" video I've now ordered a 4-car DCC+sound set too. Neither train is remotely suitable for my own layout. I am telling myself, I shall run the analogue set on the club layout, and I shall run the DCC set on the new club layout, and when both me and the club settle on DCC I shall sell on the analogue set. But the vibes I am getting tell me I shall assemble a seven-car set, for whatever layout is next, and keep all of both trains as a heirloom. I have declared a personal moratorium on motive power purchases for the rest of 2015.

- Richard.

I pursued the career and soon discovered the opinions of the politicians and the bean counters do override ideas like good design. But having enjoyed reading the accounts in the last two posts, I am now asking myself, "to follow apt-e, why not build one, four or five car, 25kV set, to refine the technologies?". It seems like the p-train was rather a shot in the dark, as though someone was desperate to see a train in commercial service. Perhaps someone in Government was wanting to wrap up the R&D activity and jump straight into a finished product? Going from one experimental set to five prototypes seems a bit of a leap of faith.

- Richard.

Yes I'd like an official view too. The gas turbine might have been expensive to run, but it would allow operation on a wider range of tracks. The apt-p has always struck me as a peculiar beast; a set of three prototypes tested using customers, which doesn't fit easily into the idea of design experimentation, or design development, or a revenue earning train.

- Richard.