Heath Town and other signalling diversions

[lmsforever]

On 28/06/2021 at 08:25, 5BarVT said:

Yes and No: for me it’s more a feeling of sadness - same as when Alun Wyn had to go off and is out of Lions tour.

<Rant on>

And incredulity/frustration at the eejit spectator who caused the first crash.

<Rant off>

The raw emotion when Matthieu van der Poul not only won the stage yesterday but got the yellow jersey too, having calculated that he needed the bonus points from the last climb as well as the stage victory to get yellow, was quite something.

Cycling is a funny sport as it’s individuals who win, but can’t do it on their own.  It took me a while to get used to breakaways that know they have no chance of making it to the finish (usually).  We have the benefit of ITV4 coverage here in the UK and I’ve learnt a lot from the insights of the ex pro cyclists.

Paul

I agree about Le Tour it and the rest of cycle racing on tv via Eurosport is wonderful I was at the Parc De Prance in Paris in 1962 when Tom Simpson finnshed 6th overral that year .i was a racing cyclist during the sixties and seventies when the BBC thought we were an oddity not like now.Cavendish is an rider especially as he now has 34 victories and I think he could win two more this year .This is the toughest tour for quite a while and its great seeing these brilliant men racing thier hearts out.Glad you enjoy it and keep spreading the word we might see the tor back in GB again.  Chris

[5BarVT]

44 minutes ago, lmsforever said:

I agree about Le Tour it and the rest of cycle racing on tv via Eurosport is wonderful I was at the Parc De Prance in Paris in 1962 when Tom Simpson finnshed 6th overral that year .i was a racing cyclist during the sixties and seventies when the BBC thought we were an oddity not like now.Cavendish is an rider especially as he now has 34 victories and I think he could win two more this year .This is the toughest tour for quite a while and its great seeing these brilliant men racing thier hearts out.Glad you enjoy it and keep spreading the word we might see the tor back in GB again.  Chris

Quite agree.  Back then if you saw a cyclist in Lycra you would think “who does he think he is?”, whereas now it just “Oh, there’s someone out for a ride.” [even a MAMIL :-) ].

I’ve only been a utility and recreational cyclist but it’s kept me fit.

Didn’t start watching cycle racing until we went to Lanzarote in about 2000 - daughter had been there for warm weather swimming training  so we had a family holiday there too.  Eurosport was on with the tour, but no commentary, so didn’t pick up any of the races within races, or understand about mountain points, green jersey or the subtleties of getting the same time as the rest of the bunch.

Keeps us well occupied for three weeks in the summer and is a family activity if they are back home then.

Paul.

[St Enodoc]

5 hours ago, 5BarVT said:

Some progress has been made as well as childcare and Tour watching.

 

 

The track has been laid and droppers wired up, apart from two cross board sections to go in lower left.  Both have been purchased and one is stuck down ready for droppers.  The other is mislaid!

More work on power distribution to get DCC to this board and data cabling for the loconet signal.  once that is done it will be complete and a train might just move.

Paul.

So what does all that stuff actually do, Paul?

[Clive Mortimore]

6 hours ago, St Enodoc said:

So what does all that stuff actually do, Paul?

I was thinking the same thing, very tempted to say "It is only two wires" but thought better of it.

[5BarVT]

With two exceptions it’s only two wires . . .

XLR top left carries incoming DCC (and on other boards will also have ‘dirty’ 12V DC).

Stripboard top middle has 8 (100:1 I think) toroidal transformers thus 8 separately detectable DCC feeds.

Each feed leaves to its left pair of WAGO terminals then daisy chained to the right pair, droppers are fed from the terminals.

Back to the strip board - the output from each toroidal is in one pair of a Cat 5 purple ethernet cable (buy the longest available and just cut bits out of it).  As it’s only 7/0.13mm it’s a bit fiddly crimping into the 8way connector!

So 8 o/p = 2x 4pr to the RR Cirkits Watchman board (lower middle) to get the occupancy data back to the computer.

Finally out of the Watchman board is the three wires carrying ‘clean’ 12V and data into the data bus that will run round the layout in the black 3 core flex.

 

It would be so much simpler if I didn’t want track circuits - just incoming DCC and three sets of daisy chained connections running across the board to each set of rails!
 

Paul.

[5BarVT]

15 minutes ago, Clive Mortimore said:

I was thinking the same thing, very tempted to say "It is only two wires" but thought better of it.

With the two exceptions mentioned, it is!

:-)

Paul.

[Clive Mortimore]

23 minutes ago, 5BarVT said:

With two exceptions it’s only two wires . . .

XLR top left carries incoming DCC (and on other boards will also have ‘dirty’ 12V DC).

Stripboard top middle has 8 (100:1 I think) toroidal transformers thus 8 separately detectable DCC feeds.

Each feed leaves to its left pair of WAGO terminals then daisy chained to the right pair, droppers are fed from the terminals.

Back to the strip board - the output from each toroidal is in one pair of a Cat 5 purple ethernet cable (buy the longest available and just cut bits out of it).  As it’s only 7/0.13mm it’s a bit fiddly crimping into the 8way connector!

So 8 o/p = 2x 4pr to the RR Cirkits Watchman board (lower middle) to get the occupancy data back to the computer.

Finally out of the Watchman board is the three wires carrying ‘clean’ 12V and data into the data bus that will run round the layout in the black 3 core flex.

 

It would be so much simpler if I didn’t want track circuits - just incoming DCC and three sets of daisy chained connections running across the board to each set of rails!
 

Paul.

Whizz......that was just above head height.

[Andy Keane]

On 06/05/2021 at 18:00, 5BarVT said:

PRODUCTION LINE

I cleared my workbench last week (cleared, not tidied - it just got moved to a new 'dumping ground' shelf).

Three DCC/Track Circuit interfaces and one power distribution board for the new power box.

Today's task was to make up the first of four MERG District Cut Out Boards (DCOB).  Nice to make with clear instructions and test procedures.  A smaller tip will make the next one easier to solder, so one has been ordered (complete with iron attached . . .).  Now I will be able to swap between fine work and track droppers without having to change tips, just plug in the other iron.

Paul. 

Paul

I am planning on DCC but it seems that you are doing lots of wiring below board that I am as yet not up to speed on. Can you recommend a site or book that summarizes your approach or is this very much your bespoke take on things. I am keen to avoid repeating mistakes that those who have been around this stuff for a long time would see from a mile off.

regards

Andy

[5BarVT]

2 hours ago, Andy Keane said:

Paul

I am planning on DCC but it seems that you are doing lots of wiring below board that I am as yet not up to speed on. Can you recommend a site or book that summarizes your approach or is this very much your bespoke take on things. I am keen to avoid repeating mistakes that those who have been around this stuff for a long time would see from a mile off.

regards

Andy

Short answer is it’s not the DCC causing the complication but my desire for computer control and consequent need for train detection and interface to the computer.

 

So long as your positional feedback is using the Mk1 eyeball you won’t need train detection.  (And for historical accuracy, Helston didn’t have any!)

 

DCC wiring isn’t any more complicated than non DCC wiring if both are done ‘properly’.  A good place to start is Brian Lambert’s website: I found it useful in the early days when I was coming up to speed with DCC.

https://www.brian-lambert.co.uk/

 

Paul.

[Andy Keane]

Thanks - as I cannot bear to watch the football I have retreated to RMweb! I will take a look at the site you suggest. Also as my layout is a large L it spans two rooms and the fiddle yard will not be visibile from the Helston box where I will drive the station, so I do want to automate it as much as possible. Ideally it will detect incoming trains, decouple the loco, run it around and re-form for the next service. As yet I am not sure how best to acheive this but clearly I will need to detect positions etc. I was more inclined to use proximity sensors than track circuits to do this as I will essentially know which circuits I am using but not where the stock is - maybe I will need both. I was hoping to avoid a computer but maybe an Arduino is unavoidable.

regards

Andy

Edited July 12, 2021 by Andy Keane

[Andy Keane]

I have now been and browsed around Brian's site nad found it most interesting. He seems to go to very significant length to avoid accidental shorts on points. Is this really an issue that I should address before I lay my new track - it will be almost all peco code 75 electro frog and I suppose I am naive enough to assume such measures would not be needed on their latest products. But I will certainly follow his instructions for the throat assembly that Marcway are building for the entrance throat.

Also do you have a view on DCC controllers? I had planned to use the Gaugemaster kit but am very open to recommendations as this will be a new way of driving for me.

regards

Andy

[ISW]

On 12/07/2021 at 18:54, Andy Keane said:

I have now been and browsed around Brian's site nad found it most interesting. He seems to go to very significant length to avoid accidental shorts on points. Is this really an issue that I should address before I lay my new track - it will be almost all peco code 75 electro frog and I suppose I am naive enough to assume such measures would not be needed on their latest products. But I will certainly follow his instructions for the throat assembly that Marcway are building for the entrance throat.

Also do you have a view on DCC controllers? I had planned to use the Gaugemaster kit but am very open to recommendations as this will be a new way of driving for me.

regards

Andy

Andy,

 

I'm using Peco Code-100 electrofrog points (even the few insulfrog have been 'modified' to work like electrofrog). If I trail a turnout with the switch set the wrong way I'll get a short the moment a loco wheels contact the frog. There are two issues I found:

1. My NCE PowerCab 'trips' (good) but then keeps resetting (not so good). If I left it like that for a long time (I suppose over 5-mins?) it could permanently damage the PowerCab.
2. All the power to the layout is cut off, and all trains come to a grinding halt. Not a problem, but a bit of a pain.

For (1) I simply have to ensure I set the routes correctly using my mimic panels (all controlled by MegaPoints Controllers).

 

For (1) & (2) I split the layout into a number of separate power 'districts', each with it's own power bus. To each power bus I fitted an SCD (short circuit detector) in the form of a MERG Kit #57. Now, when I get a short on a turnout (or from anywhere actually) only that power District loses power and the SCD buzzer sounds. The PowerCab is blissfully unaware of the short, and all trains in the other Districts continue to operate. Once I've cleared the short (usually by me switching the turnout), power is automatically restored by the SCD and the train moves off without any other intervention. Neat.

 

As to controllers, I have the NCE PowerCab and it's associated USB computer connector board. That allows me to also control trains using the free JMRI DecoderPro software on a laptop (and, using their wireless service, my mobile phone & tablet).

 

I hope some of this is of use to yourself.

 

Ian

 

 

 

[Andy Keane]

7 minutes ago, ISW said:

Andy,

 

I'm using Peco Code-100 electrofrog points (even the few insulfrog have been 'modified' to work like electrofrog). If I trail a turnout with the switch set the wrong way I'll get a short the moment a loco wheels contact the frog. There are two issues I found:

1. My NCE PowerCab 'trips' (good) but then keeps resetting (not so good). If I left it like that for a long time (I suppose over 5-mins?) it could permanently damage the PowerCab.
2. All the power to the layout is cut off, and all trains come to a grinding halt. Not a problem, but a bit of a pain.

For (1) I simply have to ensure I set the routes correctly using my mimic panels (all controlled by MegaPoints Controllers).

 

For (1) & (2) I split the layout into a number of separate power 'districts', each with it's own power bus. To each power bus I fitted an SCD (short circuit detector) in the form of a MERG Kit #57. Now, when I get a short on a turnout (or from anywhere actually) only that power District loses power and the SCD buzzer sounds. The PowerCab is blissfully unaware of the short, and all trains in the other Districts continue to operate. Once I've cleared the short (usually by me switching the turnout), power is automatically restored by the SCD and the train moves off without any other intervention. Neat.

 

As to controllers, I have the NCE PowerCab and it's associated USB computer connector board. That allows me to also control trains using the free JMRI DecoderPro software on a laptop (and, using their wireless service, my mobile phone & tablet).

 

I hope some of this is of use to yourself.

 

Ian

 

 

 

Ian, thanks for this. So if I understand correctly you have no shorts when you route correctly but if you don’t you end up tripping the section, and if that is the whole layout it all stops. So you haven’t modified your points, but installed SCD to cope instead?
regards

Andy

[ISW]

4 minutes ago, Andy Keane said:

Ian, thanks for this. So if I understand correctly you have no shorts when you route correctly but if you don’t you end up tripping the section, and if that is the whole layout it all stops. So you haven’t modified your points, but installed SCD to cope instead?
regards

Andy

Andy,

Basically yes. All I would add is that having several power Districts (I have 8) means that a short in that District only affects that District (& no others), and the SCD does all the detection, warning, and is self-clearing once the short is removed.

Ian

[5BarVT]

PROGRESS AND LE TOUR

Well that’s Le Tour over.  Pleased that Cav equalled Eddie Merckx’s record.  Made for some exciting sprint finishes.

 

There has been progress on the construction front too.  Board 1 is complete apart from adding parts that I have but can’t find. :-(   Part 1 of static testing revealed some faults that have been rectified (better QC inspection would have helped).  Part 2 revealed something more that will take a bit of finding (it involves a long cable).

 

Photos may be added, or in a following post.

 

Paul.

Edited July 18, 2021 by 5BarVT
Deleted iPad inserted words that changed the tone of Cav comment.

[5BarVT]

FAULT FINDING

18 hours ago, 5BarVT said:

Part 2 revealed something more that will take a bit of finding (it involves a long cable).

Built a cable tester - a plug and socket joined to a choc block with a loose wire that can short out DCC. (Then the district cut out squeaks.)

Cable is fine.  Socket on the first board (which feeds the second) had DC cable wired to DCC pins and vice versa.

18 hours ago, 5BarVT said:

(better QC inspection would have helped).

There seems to be a theme here.

 

More seriously, I’m glad I found it here as at the same fault on the other socket and testing with a loco could have blown up the command station as the loco ran across the board joint.  (Known in my trade as a wrong side failure.)  Static testing will precede dynamic testing for definite in future.

 

Paul.

[ISW]

4 hours ago, 5BarVT said:

FAULT FINDING

Built a cable tester - a plug and socket joined to a choc block with a loose wire that can short out DCC. (Then the district cut out squeaks.)

Cable is fine.  Socket on the first board (which feeds the second) had DC cable wired to DCC pins and vice versa.

There seems to be a theme here.

 

More seriously, I’m glad I found it here as at the same fault on the other socket and testing with a loco could have blown up the command station as the loco ran across the board joint.  (Known in my trade as a wrong side failure.)  Static testing will precede dynamic testing for definite in future.

 

Paul.

Paul,

 

One of the benefits of building a layout in disconnectable / separate baseboards is that they can be individually tested. I tested all my baseboards with a multimeter to ensure correct polarity / continuity on each bus / frog polarity at turnouts. The same process was followed as each baseboard was connected to the previous one. It's tedious and boring, but much better that trying to find a fault 'somewhere' later on!

 

Ian

[5BarVT]

MORE STATIC TESTING AND SOME DYNAMIC TESTING (ISH)

Having fixed the errant plug coupler I took my Olfa Silver knife and plonked on each track section.  Cut out squeaked each time.  So I placed a test loco on the track.  IT MOVES!  I’m confident that both boards are wired correctly for power to the track.

 

Now onto indications back over loconet.  NOTHING.

 

Right.  Laptop brought through to train room, use a new 6c cable. Still nothing.  Try it in a different ‘RJ12’ socket - success(ish) - I can see the speed commands but still nothing back from the new boards.  Eventually I thought I might check the data cable connection at the system end.  Ah! Might help if that was plugged in. All of a sudden, the indications started working . . . 
Not QC this time, just being an eejit!

I was still puzzled why loconet indications to laptop were not working over a 4c cable as I was sure I had done it before.  Lots of faffing with different cables and connectors in case one I hadn’t used before was faulty, but no combination worked. Finally tracked it down: Digitrax put rail sync on the extra 2 wires (a copy of the DCC signal which their boosters use) and although I don’t need the signal, the computer interface uses it for power to that side of the opto isolation.  I hadn’t powered up my test card (in the same box) because I wasn’t using it and it would be a faff to bring the PSU through too (but lot less faff than I’d already done) but I now know it needs to be on to power the interface.

 

Now onto some real testing results.  One track section doesn’t indicate at all.  Could be a poor crimp (7/0.13 is horrible to crimp because it’s so thin) or poor soldering on my home made toroidal transformer board.  Some swapping over of parts to see if the fault moves with them will be next.

One track is very sensitive, going occupied whenever any of the other tracks on that board show occupied, or constantly going clear/occupied/clear repeatedly.  I had wondered if my transformer boards might generate interference between the sections due to proximity of wiring.  That will be more of a pain to sort if true, but, I would expect it to affect more that one section if that was the mechanism.

Watch this space.

 

Laughing emojis will be assumed to be aimed at my incompetence!  Sympathy will be assumed to be related to the possible interference issue.

 

Have fun,

Paul.

Edited July 20, 2021 by 5BarVT
Spooling

[ISW]

10 hours ago, 5BarVT said:

Could be a poor crimp (7/0.13 is horrible to crimp because it’s so thin) or poor soldering on my home made toroidal transformer board.

Paul,

 

I'm not surprised. I had similar problems crimping 7/0.2 wires. I eventually found that if I push the wire a bit further into the plug, the crimping was much better as the insulation sleeve crimped on the outer crimp bit (strain relief), with some insulation and core wires crimped on the inner crimp bit (electrical). I used the smallest part of my crimping tool to ensure good contact. I then had to use pliers to 'reshape' the squashed plug into a more 'circular' shape so that t would fit in the plug sleeve. After a few hundred, I was getting quite good at it! By my reckoning, I've used ~2000 crimps on the layout.

 

10 hours ago, 5BarVT said:

I had wondered if my transformer boards might generate interference between the sections due to proximity of wiring.

 I was very careful to place my MegaPoints network cable away from the power bus cables for that very reason. It does seem to work, as I've not experienced any such electrical interference, yet.

 

Ian

[5BarVT]

39 minutes ago, ISW said:

I'm not surprised. I had similar problems crimping 7/0.2 wires. I eventually found that if I push the wire a bit further into the plug, the crimping was much better as the insulation sleeve crimped on the outer crimp bit (strain relief), with some insulation and core wires crimped on the inner crimp bit (electrical). I used the smallest part of my crimping tool to ensure good contact. I then had to use pliers to 'reshape' the squashed plug into a more 'circular' shape so that t would fit in the plug sleeve. After a few hundred, I was getting quite good at it! By my reckoning, I've used ~2000 crimps on the layout.

I’d started doing something similar to that for my 7/0.13 fold the conductor back over the insulation and crimp both together.  This particular cable predates that.  I haven’t done that with 7/0.2 yet but I do squash down the conductor part of the crimp to be sure of contact.

 

39 minutes ago, ISW said:

I was very careful to place my MegaPoints network cable away from the power bus cables for that very reason. It does seem to work, as I've not experienced any such electrical interference, yet.

Ditto, data down one edge of the board, DCC power kept to the other as far as possible.

 

On 19/07/2021 at 20:11, ISW said:

One of the benefits of building a layout in disconnectable / separate baseboards is that they can be individually tested. I tested all my baseboards with a multimeter to ensure correct polarity / continuity on each bus / frog polarity at turnouts. The same process was followed as each baseboard was connected to the previous one. It's tedious and boring, but much better that trying to find a fault 'somewhere' later on!

100% with you there.  All was well on the individual board, it was only when joined up that the problem was identified.  It is so much easier working on fault finding without having to crawl about underneath everything.

 

On a three line whip for an early walk before the temperatures rise, then house cleaning and tidying in preparation for grandchildren over the weekend.  (Possibly the first time the youngest has been away without parents.)  More fault finding has to wait!

 

Paul.

Edited July 21, 2021 by 5BarVT
Tryping

[ISW]

1 hour ago, 5BarVT said:

I’d started doing something similar to that for my 7/0.13 fold the conductor back over the insulation and crimp both together.  This particular cable predates that.  I haven’t done that with 7/0.2 yet but I do squash down the conductor part of the crimp to be sure of contact.

Paul,

I wondered about that method, but 2 things concerned me; I couldn't be sure to have the wire the 'right way up' to ensure good contact inside the crimp, and the risk of 'stray stands' causing shorts (or worse, intermittent faults). Given that my crimped terminals fit in 'dupont' style connectors (at 0.1 inch spacing) a stray strand was best avoided.

Ian

[5BarVT]

47 minutes ago, ISW said:

Paul,

I wondered about that method, but 2 things concerned me; I couldn't be sure to have the wire the 'right way up' to ensure good contact inside the crimp, and the risk of 'stray stands' causing shorts (or worse, intermittent faults). Given that my crimped terminals fit in 'dupont' style connectors (at 0.1 inch spacing) a stray strand was best avoided.

Ian

I think I have avoided that, I kept the stripped length short so that it was still within the electrical part of the crimp.

But I will check!

Paul.

[Clive Mortimore]

We all have our own way of testing our layouts.

 

Mine is a good running loco and when it says "I ain't gonna budge" do some visual checks, like the right wires going to the right place, solder joints have been made and are good uns, and the loco hasn't picked up anything, like track pins (it do happen). I then check the loco hasn't died for some reason, that can happen. It is then the multimeter comes out. Only to find what seemed the right wires had been labeled up wrong, I swapped the wires and labels and off went the loco. The only other problem I had with the wiring of Sheff Ex was the GM500s not functioning correctly when trying to fire two Peco point motors at the same time. I bypassed the resistors and that was that cured.

 

I have had more problems due to my own fault when wiring up the changes to the tracks in the diesel loco yard than the main layout.

[St Enodoc]

17 minutes ago, Clive Mortimore said:

I have had more problems due to my own fault when wiring up the changes to the tracks in the diesel loco yard than the main layout.

Too many wires, old fruit.

[5BarVT]

IT’S INTERFERENCE

But not quite how I imagined.

Each board physically separate but joined electrically, all indications work fine; joined together, the interference resumes. Sheet of A4 between the boards, works fine, so a rail /rails is/are touching across the joint and causing run round paths.

Solution, grind back to ensure a gap, starting with the most suspect.

That will be tonight’s task.

 

Paul.