aleopardstail

to be fair MS Paint isn't exactly known for circuit drawing but it was to hand, yes current transformer. have had a few other ideas on the detection circuit side of this which I'll be experimenting with which is likely to cut out the ADC module

managed to sort the PCB issue out, went with a wet chemical transfer method that worked pretty much spot on 8:3 mix of Ethyl Alcohol and Acetone, pattern printed on a bit of a magazine (thin glossy paper), pattern placed over the cleaned board and the chemical mix drizzled over the back. left for a few seconds (long enough to recap the bottle really), then pressed down through a bit of kitchen roll, came out as above, nit perfect round the edges but the important bits worked. when the weather cleared up Saturday I had a go at etching it came out very nicely, better than my accuracy with the drill at least *hangs head in shame*, will get there though. for scale with a bit of N gauge track will be ordering some bits later this week to assemble them, planning to try the HW-201 infra red proximity module boards but also a custom circuit as I want to add a bit of hysteresis into the response to save having to debounce the output before using it. have also made a (small) start with some scenery, nothing fixed in place and done mostly to get a handle on the scale and size of it Metcalfeville strikes again, picked up some paints for the exposed edges to experiment with as well

Still working with these beasties, added the ability to read/write to and from an EEPROM IC, its not an optimised library (it writes byte by byte and not in pages for example) but it does for what I want, its also set to 8k EEPROM only. Still not gotten round to playing with PIO though

today has seen some initial experimentation with preparing the sensor PCBs, not quite successful enough to etch, but learnt a fair bit, will have another go tomorrow second attempt, better than the first but didn't quite transfer correctly. rectangles are pads for 0805 components, then circular pads for connecting wires. the four pads in a row fit between N gauge sleepers nicely

It would (wood?) have been a lot heavier if not for the good denizens of this site providing some much needed advice on what would (ply?) actually work without being overkill. its 9mm ply, it could probably have been 6mm and still worked. have not tried to move it too far but on and off the desk it sits on isn't too bad, basically kneel down, slide it forwards and can get a hand on the centre of the "X" which makes a very good handle. lesson so far though: it needs handles on the sides just to make stablishing it easier, also a bit more thought about where the electrickery bits will go would have been useful - specifically how to access them during this build stage - wondering on cutting out some of the side at one end to make an access point - eventually covered by a removable metal plate that can have things like USB ports and a power connector added. there is quite a bit more to go on here as well (needs at least a low backscene on three sides and a facia on the front, indeed one end likely needs to be lower as well - but this needed to protect the scenic bits) so a few concerns about weight here, in retrospect actually going 4'x4' or 4'x3' but in two halves would have been a lot better, the design seems made for it really and would have made working on it a lot easier. still the point was to be a learning exercise. Before more track goes down (and I do actually have the rest of the track) I need to get some experiments with printed circuit board manufacturing completed

space for an Arduino Pro Mini? would need an opto-isolator circuit and a power regulator & rectifier circuit to go with it but reasonably cheap and can drive pretty much any number of outputs if you are willing to add output shift registers or extra port multipliers - can do a decent number on its own though. can also get the chips on their own (inc. the SMD ones if you can work with them, these are pretty small)

Bit more progress, I combined the C/MRI & RS485 code and hardware with my previous servo controller, added it all up and got this gork awful mess however while ugly, and in dire need to be fitted to stripboard and nailed on the underside of the board eventually it works, sensors still need some calibration but they work. can now drive what there is of the layout from the computer. the next step really is work on the infrared position sensors and to add some position feedback for the turnouts. there is another PCA9685 board on order which will become the early stages of some colour light signals, probably oversized mock up ones initially to get used to how they work, the layout isn't in a suitable state to add actual scale ones as yet which would be way too fragile

cheers, got the current sensors working in the end needs a slight tweak to cope with noise from a nearby servo driver but it works, next to build the IR sensors. managed to get JMRI to see multiple physical blocks as a logical section

glad it was useful to someone have adapted the controller its running on slightly with an RS485 interface, it now speaks a slightly tweaked by of C/MRI code that also runs on the RP2040 board and can drive the points from JMRI. I'm amazed it worked really, but work it does

Nice looking plan, it shows just what you can do with a little extra depth over the 24" many N gauge designs seem constrained by, I went to 2'6", but what you have got there into 3'x4' is amazing. should be excellent to see running too

a bit more progress have now hooked up both the long sidings, have also nailed the current detection so it works and works reliably basic circuit, but it works, will pick up a 10k resistor, which is sensitive enough for me. Also had a play with JMRI & C/MRI, adapted the Arduino to speak "C/MRI", resulting in: yes, not that impressive I admit, not sure why the block at the top is black, anyway this is reading the seven block occupancy detectors, showing the locomotives sitting in the two white blocks, and it updates nicely as they shuffle around. minor problem, the servo controller I have for the points isn't an Arduino, nor does it talk DCC.. Its a Raspberry Pi Pico RP2040 board... which doesn't speak C/MRI. Hmm.. so into the C/MRI arduino library, which actually isn't using anything special in the hardware so I had a go at a modification... the first prototype of a Raspberry Pi Pico RP2040 MCU communicating with JMRI over the C/MRI protocol, ok again not that impressive, it just controls one LED - however the fact it worked at all amazed me, next up is bi-directional stuff but then critically, controlling the PCA9685 board to start driving servos, and eventually to take over the world. the code is very much a bodge currently, I want to clean it up a bit so its a compatible version of the main library not a hacked separate file - to be honest its nothing special, the only bits that needed changing relate to the serial port comms code

Dunno who this project is torturing more, myself of both viewers.. regardless! progress! the sensors are working reasonably well, at least well enough to fit them and worry about making them work better with a bit of signal processing later.. so this afternoon I removed the cat, tidied the board up, put the stock on the shelf, after removing the cat again (seriously I'm buying a water pistol) and flipped to board on its end - oh the benefits of a portable board - as and when a larger layout is built, it *is* being made in sections. and while on its end I added the current transformers and generally tidied up the dropper wires then added the DCC bus wires - not fixed as they will end up cut to add more sections (hence the slack) its a mess, none of it is labelled, what could possibly go wrong? then some very temporary wires from the sensors were added - these will eventually terminate "somewhere" on the underside, but for now go to the top in one corner servo controller not yet installed as I want to do a bit more with it first. board flipped back round, no cat related trouble as the furry terrorists had taken the hint its worked! no short circuits or wires the wrong way round - test train went round quite merrily - though the ancient traction tyres on the Class 86 are long age hardened so three coaches becomes an issue with the fact the board isn't quite level on the shelf. Glad I didn't add a deliberate gradient really. ok, thats the DCC wiring to the blocks, what of the sensors? well the monitoring block is to the right, so far five blocks, only four monitored as I need to assemble another board, but these are the results so far: you can see the train starting in the first column, vanishing into the unmonitored block, then reappearing in the second column and finally the 4th column. issue so far is the spurious readings when no train is present - cut the power and they all show nothing, a train stationary and it reads around 0.9-1.3, with then other blocks then around 0.01-0.05.. the sensors appear to be picking up a lot of noise or stray magnetic fields - I suspect its electrical noise in the wires though - may have to have a look at some filtering techniques. but, hey a train can run, so progress

have a (basic) understanding of real world signalling and how multi aspect & blocks works, only real reason for multiple blocks or point detectors is to replicate the drivers ability to react to a signal - this will be a loop with four signalling blocks and four aspect (because if over complicating something dive right in I say), so we have either tail chaising or tail chasing with an intermediate point - that rare thing "common sense" suggests while each signal has a stopping point the actual speed before that will indeed depend on what the preceding signal was - so a single train will run at whatever line speed is here but have two and both will run slower as they will be under a caution aspect of some sort all the time. aware there are other systems, but the Yorkshire in me is cheap.. may end up with something more dedicated anyway as none of the signal logic will be massively complicated and there won't be a huge amount of stock to keep track of. every rabbit hole seems to lead to more and more rabbit holes currently using DCC++EX to drive things, pondering changing that at some point to run on something slightly more powerful but thats a while off yet

it is almost certainly overkill, the goal though is to be able to manage trains that lack calibrated speeds (or even roughly calibrated), its a tiny layout but really experimenting with stuff that could work on a larger one. operating goal is to know a block is occupied so as not to allow anything else in with the main "block" sensors (which could be several sub blocks into a larger virtual one), then use the higher accuracy ones to trigger actions within the block (e.g. trigger a "stop" action such that a train has a bit of space to gradually stop (using 6" here) to then be a few inches clear of the signal. currently not used JMRI much other than for programming decoders, aware some sort of custom logic could do this without too much trouble, just wondered how hard it would likely be.

a bit of progress, first up four points went down, with servos attached and from below simple, not terribly original but it works, the driver at that point looked thus: its crude but it works, its since gained an EEPROM chip which has been tested but isn't yet used - it will store the default settings and power on positions when I get it coded. The screen and keypad are likely to end up on wires mounted elsewhere to plug in as this will end up on stripboard under the layout. after this the track to support these two electrical blocks went in to experiment with current monitoring the current sensors need more work but the droppers are in place. that was over the last few days, today a bit more went down the two storage sidings went in first, the gap in the sleepers is where the droppers are, each siding being its own electrical block. then two further blocks were added completing the outer loop, again droppers in place, though these have yet to be electrically tested the current monitoring can be seen bottom left, it works, almost, the 10 bit ADC on the Arduino Nano not being quite enough, picks up moving locomotives easily and will pick up two stationary ones but can't see a single one. have some 16 bit ADC boards arriving later this week which should nail it. it is possible there will be a further electrical block added, for signalling it will be a sub block but may as well hook it and monitor it individually. Jobs up for the next few weeks are creating the Infrared point sensors, which means making some small custom circuit boards to hold the 0805 SMD devices (LED & phototransistor), once they are assembled, tested, working and installed this loop can be fully glued down. prior to that it will be getting powered up to be tested

Currently building a small test layout, and working on train detection. The plan is to have current sensing for "block" detection, this is semi-working and should be nailed with some better resolution ADC boards shortly (the 10 bit in the arduino not quite managing it, figure a 16 bit ADS1115 should be fine in comparator mode). This will say "something is in this block" and I gather JMRI can work with that. It was then planned to have an IR sensor (LED & phototransistor) looking up to provide a "train is at this exact point", to trigger changing speed or halting completely. I don't yet have these, I'm familiar with the technology though and should be able to assemble small surface mount sensor heads to fit between N gauge sleepers. however, I'm wondering, can JMRI manage a "virtual block" that has actual sub blocks? e.g. what I'm wanting is for the signalling logic to see a train in any or all of 1a,1b & 1c as "occupying block 1", but picking up entering block 1c as the "ahh, now slow for the signal" point in the same was an IR sensor at point "x" would do so the train can slow to stop roughly as point "y"? in practice some blocks would probably just be two detection zones, a main bit and a signal approach bit, others would be three especially if running in both directions Q: how viable is this? Q (bonus!): which approach is generally considered "better"?

the actual servo control is managed by the PCA board, the code to drive is isn't huge so should run fine on arduino as well - the Pico is better for running a screen with a full frame buffer. note this is the RP2040 Pi Pico microcontroller, a £3.60 device, not the full on raspberry Pi computer (I've not tried but I suspect the code would run on a Pi4 without much alteration though) mostly went with the Pi Pico as a learning exercise to be honest, they work and seem to work well

now I actually have the beginnings of a layout, here is a picture of the somewhat homebrew servo controller as it currently is The green board is the Pi Pico RP2040 microcontroller, black button is a reset button, which resets just the MCU, need to adapt to cut the power to the display and keypad to reset them as will (eventually will "reset" by actually cutting the power) the PCA9685 is the blue board, then there is a 0.96" OLED and a touchpad for user input. It needs, but does not yet have, an EEPROM chip adding to be able to save settings. The software is very much in "test" mode currently but useful for making adjustments when installing servos. also as yet no way to talk to the outside world, intended to provide a DCC interface to actually control the servos but also an RS485 interface to a C/MRI connection to a PC or other device so this can provide actual feedback. Intended to have position switches on the points feeding back into this, probably via SN74HC165 shift registers, and likely some LEDs via SN74HC595 shift registers to show the "actual" position (driven by software though not directly by the switches themselves) so far it works

Everyone needs a shed, and two are better than one, these arrived earlier the exterior looks decent, needs a brush taking to it to weather it and stop it looking like and unpainted plastic kits, the inside is less impressive, it will do for now, will be given some lighting but I can see making my own to have a bit more internal detail at some point - I want some offices along the side facing the camera as well, or portacabins at least. that said they look the part. also starting to prepare for some experimentation around mounting servos and sensors, so the test layout has gained its own test layout, a hellishly impressive one spring removed from the turnout, waiting on the piano wire arriving and will have a go at mounting a point servo, can attach bits of track with power feeds via current sensors but leave them unsecured pending the IR sensors going in at a later stage. for the main layout the turnouts can be installed prior to the sensors being finalised so thats the plan, I just want an experiment for the mounting first.

this may well the the solution to go for then, ahh well only need to do it twice - three packs making two pits for the length. ho hum

Hi I've got a pair of inspection pits for Dingleberry, the kits come with short lengths of rail, how are you meant to join them to each other? no space for a fishplate, I wonder if the idea is to power from both ends but I've got three lengths so still need to power the central length. I could remove sleepers from flexi track but seems to make sense to use what it comes with if possible. soldering from the top? any ideas?

foam to clad the board in arrived today, so cleared all the track off and dusted a bit, then stuck it down, leading to two interesting discoveries 1. copydex stinks every bit as much as I remember 2. moggies stay out of the room, this is useful information lilac colour and not bright pink, which is a sort of bonus printed the track plan off and stuck that down, line in red is where an additional siding may end up, storage road for the parcels depot. rest of the track is now here, though not all the rail joiners yet, however once the glue has all dried I can start marking out the electrickery bits and drilling holes for the wire droppers and point servo motor actuators, plus the planned location for the infra-red position sensors - the parts to actually make the sensor heads may be late May though - I can at least get the track assembled and droppers added in the mean time, just can't really fix it down or wire it up properly

Flooring didn't arrive, rescheduled for tomorrow apparently, can live with it. However some wire to make droppers from did turn up so could hardly resist soldering a couple of power feeds and seeing if it worked First powered movements at Dingleberry a few issues with some of the points have been discovered, e.g. none of the sidings get power, suspect the second hand points are not switching right, should be solved with a bit of cleaning, destined for individual power feeds anyway so not too bothered. I do suspect the economic viability of the two trains shown is somewhat questionable though

more progress, some XPS floor covering is due tomorrow which will be the sound insulation, in a lovely shade of pink so there may be a bit of painting to be done soon. made a start cutting flexi track to length, none of this is powered up yet as the electrical schematic has yet to be finalised. two train storage sidings along what will be the front of the layout, each able to take four Mk1 coaches and a locomotive, which is more than I'd normally be running here - expecting closer to two or three coaches to fit within the planned signalling blocks (which are unrealistically short but this layout is as much about testing design ideas as testing stock, get it right and will be able to have a pair of smaller trains on each loop chasing each other through the signals. station area has gained the inner loop road able to manage push through running and the flexi for the two main loops is cut but not yet installed, can also see the start of the parcels depot area, though the siding will be replaced with flexi to get the length right - three Mk1 coaches on each road and since an extra 17th turnout has been found I'm considering a further siding alongside here, the curved road on the middle right of the picture being the headshunt for this area as well as the loop platform road probably not "safe" working practice but its a 4' long board, so shoot me. its all just sitting there for now, to be removed when the XPS goes down and the glue sets. Also had a number of IR sensors arrive, the ones using the 5mm IR LED & Phototransistor. Quick experiment showed they will pick up N gauge stock reasonably looking up through the sleepers, however the plan is to remove the LED & Phototransistor and replace with short wires to a small circuit board holding a pair of surface mount devices that will actually fit between the N gauge sleepers and likely work a lot better. all good fun. Plan is to get the outer of the two loops properly installed first as electrically this is the most simple - needs four power feeds (with occupancy current transducers) and four IR sensors, likely a couple of weeks before the stars align and thigs can be glued down properly while I fiddle with the various bits to go between the track and the board it is however likely the track will be added and not glued for some testing so should be able to make things moooooove

board now built, rigid as a rigid thing, my thanks for the advice please not that no accurate measurements were harmed, or indeed involved, in the construction of this project