Booster Question

[Alan Kettlewell]

I'd appreciate if anyone can advise on a rough and ready method to determine if a booster might needed. I'm building this layout featured here:

 

http://www.rmweb.co.uk/community/index.php?/topic/102367-new-layout/

 

I thought, before I continue much further with the build and in particular as I'm in the process of doing the wiring, it would be beneficial to assess now if I should form districts and use a booster, or two?

 

Some statistics, very rough:

 

System: Roco Z21 providing 3.5A

Layout: OO track laid mainly double track in multiple levels around a 30 x 14 ft room. A few hundred feet of track work.

Electronics: 10 x Lenz LS150s handling nearly 60 turnouts - Tortoise motors

4 x Digikeijs feedback modules handling up to 64 feedback inputs (reed switches)

A number of accessory modules to handle the signals - colour light signals for modern image.

Trains: there'll be up to 8 running at any one time. Almost all the locos are new or nearly new so ought to be quite economic on current. Some have sound, some coaches have lights.

 

I think that's it. Just looking for a rough and ready method of working out the max amps needed, also how to measure stall current for the locos, never did know how to do that (grin), please, no in depth studies in electronics or engineering, thanks.

 

Alan

[Mick Bonwick]

I'd appreciate if anyone can advise on a rough and ready method to determine if a booster might needed. I'm building this layout featured here:

 

http://www.rmweb.co.uk/community/index.php?/topic/102367-new-layout/

 

I thought, before I continue much further with the build and in particular as I'm in the process of doing the wiring, it would be beneficial to assess now if I should form districts and use a booster, or two?

 

Some statistics, very rough:

 

System: Roco Z21 providing 3.5A

Layout: OO track laid mainly double track in multiple levels around a 30 x 14 ft room. A few hundred feet of track work.

Electronics: 10 x Lenz LS150s handling nearly 60 turnouts - Tortoise motors

4 x Digikeijs feedback modules handling up to 64 feedback inputs (reed switches)

A number of accessory modules to handle the signals - colour light signals for modern image.

Trains: there'll be up to 8 running at any one time. Almost all the locos are new or nearly new so ought to be quite economic on current. Some have sound, some coaches have lights.

 

I think that's it. Just looking for a rough and ready method of working out the max amps needed, also how to measure stall current for the locos, never did know how to do that (grin), please, no in depth studies in electronics or engineering, thanks.

 

Alan

 

There will be somebody along very shortly to answer your questions in detail, but I think you should plan for at least 2 buses, one for track and one for accessories, probably treated as separate power districts. Whether you do this with another booster remains to be seen . . . . . . . .

[Alan Kettlewell]

Hi Mick,

 

Ah, many thanks, I ought to have mentioned that I've installed a separate bus for the turnouts running off a Lenz TR100 45va transformer. So two buses, track power from the Z21, and turnouts from the transformer. The Digikeijs feedback modules use track power.

 

Cheers

Alan

[Ron Ron Ron]

Alan, this is the first time I've seen your thread  about your new project.

Pretty impressive and a very challenging task, by the looks of it.

 

Firstly, I don't think you are going to power that size of layout from the single 3.5 amp Booster in the Z21, without certain compromises .

 

Secondly, dealing with any shorts and electrical faults could be a nightmare.

 

Thirdly, with the size of layout and number of accessories (point decoders, signals etc) that are to be employed, it's sensible that you are putting these on a separate accessory bus.

 

 

Definitely the layout needs to be divided into a number of Power Districts and Sub-Districts.

One Power District for the accessories and possibly one or two power districts for the tracks, each divided into a number of sub-districts, using circuit breakers, to isolate and identify shorts and to aid fault finding.

The Power Bus design and arrangement needs to reflect that.

 

I'll let others suggest where you should split the layout up into districts.

Factors to consider is where the power will be needed the most, e.g. lots of active locos in a yard/depot/busy station, versus long stretches of running line with only a couple of locos operating etc.

 

Another consideration is the length of the Power Bus wire runs and the maximum distance from the track output of your Z21 (and from any additional Boosters), to the furthest point away via the Power Bus(es) and track.

There's a degree of conjecture over advice given on this matter, but popular advice is to arrange things so that no point should really be more than 30 ft from the track output of any Booster, including the one incorporated in the main system box.

Bear in mind that the Power Bus(es) can follow a circuit, be a loop, radial spurs and any combinations of those.

(see www.wiringfordcc.com )

 

Good luck.

 

Ron

 

 

 

.

[Adrian Wintle]

I think that's it. Just looking for a rough and ready method of working out the max amps needed, also how to measure stall current for the locos, never did know how to do that (grin), please, no in depth studies in electronics or engineering, thanks.

 

Alan

 

Measuring stall current needs an ammeter and a voltmeter as well as a length of track (a loop is good) and a DC power pack. Wire the ammeter into one side of the track feed and wire the voltmeter across the feed (or the rails). Set the loco on the track and turn the DC pack up to 12V. The ammeter will display the nominal normal current draw. Leaving everything the same, push down on the loco until the wheels MOMENTARILY stop turning (stall). At this point the ammeter will display the stall current.

 

 

NOTE: this should be done before the DCC decoder is added to the loco.

 

Personally, I no longer bother testing stall current.

 

Adrian

[Alan Kettlewell]

Measuring stall current needs an ammeter and a voltmeter as well as a length of track (a loop is good) and a DC power pack. Wire the ammeter into one side of the track feed and wire the voltmeter across the feed (or the rails). Set the loco on the track and turn the DC pack up to 12V. The ammeter will display the nominal normal current draw. Leaving everything the same, push down on the loco until the wheels MOMENTARILY stop turning (stall). At this point the ammeter will display the stall current.

 

stall.jpg

 

NOTE: this should be done before the DCC decoder is added to the loco.

 

Personally, I no longer bother testing stall current.

 

Adrian

Hi Adrian,

 

Many thanks for your reply and handy diagramme about measuring stall current. I may just give this a miss though as I've already chipped the locos and don't really fancy taken the chips out to do this. It was really just a way to determine the amps used by my locos to add into the mix with accessories etc so I could get a feel for the overall consumption. On reflection that maybe overly scientific and I think now the best approach might be to make sure there's enough power available even if it means overstating it a bit. However the info and the diagramme will no doubt come in handy at some point eg should I have a suspect loco that seems to be drawing too much.

 

Thanks again.

Alan

[Alan Kettlewell]

Alan, this is the first time I've seen your thread  about your new project.

Pretty impressive and a very challenging task, by the looks of it.

 

Firstly, I don't think you are going to power that size of layout from the single 3.5 amp Booster in the Z21, without certain compromises .

 

Secondly, dealing with any shorts and electrical faults could be a nightmare.

 

Thirdly, with the size of layout and number of accessories (point decoders, signals etc) that are to be employed, it's sensible that you are putting these on a separate accessory bus.

 

 

Definitely the layout needs to be divided into a number of Power Districts and Sub-Districts.

One Power District for the accessories and possibly one or two power districts for the tracks, each divided into a number of sub-districts, using circuit breakers, to isolate and identify shorts and to aid fault finding.

The Power Bus design and arrangement needs to reflect that.

 

I'll let others suggest where you should split the layout up into districts.

Factors to consider is where the power will be needed the most, e.g. lots of active locos in a yard/depot/busy station, versus long stretches of running line with only a couple of locos operating etc.

 

Another consideration is the length of the Power Bus wire runs and the maximum distance from the track output of your Z21 (and from any additional Boosters), to the furthest point away via the Power Bus(es) and track.

There's a degree of conjecture over advice given on this matter, but popular advice is to arrange things so that no point should really be more than 30 ft from the track output of any Booster, including the one incorporated in the main system box.

Bear in mind that the Power Bus(es) can follow a circuit, be a loop, radial spurs and any combinations of those.

(see www.wiringfordcc.com )

 

Good luck.

 

Ron

 

 

 

.

Hi Ron,

 

Many thanks for your reply that's very helpful. As it happens I've already begun some of the bus wiring and at the outset decided to spread the bus(es) radially around the layout. There are three 'arms' (or is it legs?). This lends itself well to the U shape of the layout and so far I've only laid track and buses on one leg of the 'U'. Therefore it is quite easy now to determine where to isolate the track and divide the power bus as I progress with the build. Roco have recently produced a 'Lite' booster at a reasonable price - around £70 - that provides 3 amps. I'll definitely get one of those and maybe two when I've done a few calculations and thought about it a bit more.

 

Your mention of dividing into further sub districts as an aid to fault detection is interesting too. I'm not up to speed on what circuit breakers are all about or how they'd be wired, so I'll have to do some reading up on that.

 

Thanks very must though for your reply, appreciated.

 

Alan

[Nigelcliffe]

Alan,

 

looking at the scale of your layout,  I'd budget for three boosters (or at least wire it thinking that at least three will be needed).   Run the main Z21 as the central control, and also the bus to the Accessories, then the track (in sections) from each of the boosters.   

 

3A boosters are sensible,  many lower powered boosters are a better solution than one huge 10A arrangement. 

 

Power district breakers go downstream from the boosters, and break the layout into smaller zones.  For example, you might have "storage yard up-direction" and "storage yard down-direction" as two zones fed by different outputs of the breaker, which in turn goes to one booster.    With such an arrangement, a derailment (short circuit) in one power district zone won't affect running elsewhere. 

 

Further zoning might also be sensible, but without actual breakers.  If sections of track are fed through an under-baseboard isolating switches (switching both wires of the DCC bus), then it is possible to turn off sections of track to isolate faults.   This would be only for "really stuck now, something has broken" faults, but its a lot quicker to find a fault when you're down to two turnouts and couple of yards of track, than it is to have "short circuit somewhere, now you find it!" as the message for your entire layout.

 

 

- Nigel

[Oldddudders]

As far as how you choose to break up the layout into power districts, be they separately powered or merely separately protected, you might care to think where inadvertant short circuits are most likely to occur. I think they occur least on the main lines, and most in sidings and depots where points have not been correctly changed during shunting etc. I would suggest separating those two elements of any layout if at all possible.

[mikeg]

Alan, I was advised to add up the size of the run current of all my decoders of locos likely to be run on the layout and divide by 3 as a yard stick as to possible current draw. The fellow who gave me this advise said that although all the locos might be on the track it was unlikely that more than a third would be running, stationary locos only drawing ma each.

 

When I asked about sound he said for every two sound add 1 to the division. His layout was 25 x 12 ft and had 4 main lines and two sheds plus a large freight yard. When I saw it in about 1997ish I had started with DCC but was looking for answers to questions like yours.

 

regards

 

mike g

[RFS]

Just one point about your use of LS150s for your Tortoises.  Yes, they will operate them, but they do so serially.  In other words, if a route needs to be set that requires 3 turnouts to be changed which are on the same LS150, and you've set the switch time to the recommended 3 seconds, then it will take 9 seconds elapsed to set the route.  In model railway terms that's a long time - especially on a busy layout.  I also have Tortoises (about 70) and quickly found LS150s caused serious bottlenecks setting routes on my system that runs under RR&Co automation.  The LS150s had previously been used for Peco PL-10s that were replaced by the Tortoises. I thought I could re-use them but soon found this restriction a major problem.

 

The best accessory decoder for Tortoises is the NCE Switch-8 (MK2 version) as it will operate multiple ports concurrently.  They have 8 ports compared to 6 on the LS150 and cost around £45.  They can easily manage 2 Tortoises on one port for crossovers - one of mine has been running for some time with 15 motors attached.  The second-hand prices of LS150s on Ebay are very good - I found swapping mine for Switch-8s more or less broke even on a port-for-port basis.

 

The Switch-8 MK2 version also allows you to power it with an external DC power supply, thereby relieving the DCC bus of any load apart from signals.

[Phil S]

For a 'Sub-District', the 'circuit breaker' can be as simple as a manually operated switch: one that you only turn off if a fault (short) has occured, and you want to know if it is 'downstream' of the switch .....

In general, when thinking about Power Districts and Sub Districts - DO NOT think in 'analogue' terms of where sections would be - it is not the same.

If you are using live frog pointwork, then they are the most likely LOCATION of a short - when a loco/wagon runs over them the wrong way - hence why Points/Accessories are on a separate bus so that they can still be controlled whilst the track power is removed....

If you use something like a  PS-X 'Intellgent circuit breaker' for each power district, or groups of sub districts, or even each sub-district - the area 'downstream' will autmatically have its power removed in the case of a short ... but then the PSX will try to reset the ciicuit every 2 seconds .... during this time you can either spot the faulty wagon or manually isolate an area .... and if the fault is removed, then you have localised it

 

On my layout I originally planned on 2 x 5A power districts: 'scenic layout/level' and the 'lower level storage' area ... which held lots of trains. Equal.loads.

However, changing from ZTC as controller, to Roco caused me to change to the 3-3.5A limit of your Roco/Fleischmann Z21, and then with 'boosters' for more districts. I have  a RRampmeter for each of them ... it is shows maybe 1/4A quiescent current, but upto 1 Amp where I have fully illuminated trains parked. A running loco is about 1/4-1/2A nowadays. Sound and lighting adding to the load.

I have re-used Roco Amplifiers '764 as boosters, as they are the same internally, however they do not pass Railcom data: For the Z21 they have now announced Single or Twin Boosters - which support Railcom - so you may wish to look at them before choosing others.

My Storage level is still amply covered with 1 booster, as only 1 or 2 trains will be MOVING at a time, and my scenic area is divided into 'Left, Centre, Right  Power Districts - each with PSXs,  and then sub-divided further into eg freight yard, platform tracks using manual switches, .... on a 'geographical' basis which minimises interboard wiirng too.

[Alan Kettlewell]

Alan, I was advised to add up the size of the run current of all my decoders of locos likely to be run on the layout and divide by 3 as a yard stick as to possible current draw. The fellow who gave me this advise said that although all the locos might be on the track it was unlikely that more than a third would be running, stationary locos only drawing ma each.

 

When I asked about sound he said for every two sound add 1 to the division. His layout was 25 x 12 ft and had 4 main lines and two sheds plus a large freight yard. When I saw it in about 1997ish I had started with DCC but was looking for answers to questions like yours.

 

regards

 

mike g

Hi Mike,

 

I like this back of a fag packet method. I'll have around 22 locos, perhaps half a dozen with sound so that supports what others are suggesting that I'll need my Z21 at 3.5A plus a couple of 3A boosters. Great that, many thanks.

Alan

[Alan Kettlewell]

Thanks to everyone for your extremely comprehensive and welcome replies. I think I have a good steer now as to what I need to do regarding boosters, power districts and sub district division for easier fault finding. I just need to get some sheets of paper out and start scribbling.

 

Thanks RFS, appreciate your 'points' (sorry) about LS150s and Tortoise motors. However I have 9 of these left over from previous layouts and I've already fitted two banks of 4 of them handling some of the storage yards. Anyway, I don't mind the wait while they all move into position when a route is set as it allows a little bit of calm between all the excitement of automated train running. I notice that with the sound locos, when in the yard and selected by Train Controller for the next run, and instructed to start engine sounds, it takes a while to start up and Rev up prior to setting off so this kind of fits in with the time scale of things as routes are being set up ready for the schedule to run. It's all good fn and thanks again.

 

Alan

[Alan Kettlewell]

I spent some time this morning thinking about power districts.  I'm looking at buying 2 x 3A boosters - the Roco Dual Booster at 229 Euros from MBL in Germany looks good.  So, making 3 power districts at 3.5A + 3A + 3A.  I copied and renamed my Train Controller files and used these copied files to draft out potential power districts shown in the screenshots below.  In essence I've put all the 3 storage yards in one district, the main station with east and west spirals in another and the country section on the other leg of the layout in the third district.  I think that more or less evenly separates the flow of traffic and potential current draw.  Next, to work out some sub districts that I'll put through some on-off switches for ease of shorts detection.  So far so good, many thanks for all the advice.

[Nigelcliffe]

Alan,  my main concern is the black area.  There are a lot of spirals on the black, and I suspect spirals can have high current loading (loco working up hill, on a curve).   I'd look to move one spiral to another zone, or arrange the wiring so the black area can be split into two should that prove necessary in the future. 

 

- Nigel

[RFS]

Alan Kettlewell, on 13 Dec 2015 - 19:31, said:Alan Kettlewell, on 13 Dec 2015 - 19:31, said:Alan Kettlewell, on 13 Dec 2015 - 19:31, said:

Thanks to everyone for your extremely comprehensive and welcome replies. I think I have a good steer now as to what I need to do regarding boosters, power districts and sub district division for easier fault finding. I just need to get some sheets of paper out and start scribbling.

 

Thanks RFS, appreciate your 'points' (sorry) about LS150s and Tortoise motors. However I have 9 of these left over from previous layouts and I've already fitted two banks of 4 of them handling some of the storage yards. Anyway, I don't mind the wait while they all move into position when a route is set as it allows a little bit of calm between all the excitement of automated train running. I notice that with the sound locos, when in the yard and selected by Train Controller for the next run, and instructed to start engine sounds, it takes a while to start up and Rev up prior to setting off so this kind of fits in with the time scale of things as routes are being set up ready for the schedule to run. It's all good fn and thanks again.

 

Alan

 

Hi Alan - appreciate what you've decided to do but I would suggest doing some tests because I still think you may have an issue here. Although each LS150 will execute change requests consecutively, Traincontroller (TC) will assume all turnout changes happen concurrently.  In railroad.ini the parameter "turnoutinterval" is set, by default, to 0 which means concurrent operation.  So in the example of 3 turnouts needing to set for a route that are on the same LS150, it will actually take 9 seconds to complete, but TC will assume all is done after 3 seconds.  Hence a train could be started and potentially reach the turnouts before the last one has switched. If nothing else, the sound of the turnouts still changing when the train moves off can be distracting.

 

You can set turnoutinterval to a non-zero value (in milliseconds).  You would need to set this to 3 seconds to ensure TC waits for all turnouts to complete. This means it always will take 9 seconds to set the route even if the turnouts are all on separate LS150s.  And TC also sends a change command even if the turnout is already in the correct position. This is where you may have a problem as while this route is being set, no other turnout changes can take place.

 

I did have a similar problem earlier with my LS150s when I had Peco solenoids. The problem here was that all the LS150s were powered from the same TR100 transformer, which at 3 amps didn't have the power to reliably switch two motors at the same time.  This could happen if both motors were on different LS150s which would then be sharing the output of the TR100. I therefore set the turnoutinterval to 1 second, but found that was too high and in busy situations delays built up resulting in some moving trains being signal-checked. (I can do a passable imitation of Clapham Junction when there are grandsons to be entertained!).  Setting this to 500ms mostly resolved the problem. But you will require much longer to set the Tortoises if they're operated consecutively. I have one route on my layout at the station throat that requires 8 turnouts to be set which would have been impossible with LS150s. And in your layout diagram, a train travelling from "stn plat 5 east" to "block 69" will require 9 turnouts to be changed.

[Phil S]

When wiring, take the precaution of always cutting both rails - whether for separating into Power districts (an absolute Must). or sub districts - as you cannot always predict which rail which short, and it also makes fault-finding easier - total isolation when both are switched out. EVEN though many prtective devices will only isilate 1 rail themselves ... also make sure they are all the same way around.

 

I recommend the use of LEDS (with series resistor and protective diodes) on the TRACK side of your islating switches ...

in this way, the LED(s) will light when a coach bridges a gap, even though that section is not being powered directy !.

 

The diagram shows my switch a

 

rrangement - which has had minor adaptions/ reallocations since 2007

 

(Diagram attached)
 

 

    The 'hex nut' row in each case is a centre-off toggle DPDT switch  allowing each destination to be powered from EITHER of 2 sources. - Both for fault-finding and power economy...

If I am not running much, then the single Amplifier/Central unit  output is sufficent to run all!

The TOP row is LEDs fed from TRACK sections (Power Districts are 'Barnstaple', Left and Right (and previously 'Central'),  and Exeter/Terminii which is a sub-district split of the Storage Power Disitrict, as having ALL of the storage level as one section was too large for quick fault-finding!  where 'Exeter' are my through loops, and physically away from the Terminal storage platforms.

Each of these have PSX protection - as separated by the red lines (4 PSX. 4 RRampmenters)) 

Further 'sub-sub-districts in the form of switches or Wago Level Cable Connectors are used wherever it is felt useful to be able to isolate further.

 

The middle row is basically the dcc accessory distribution.   I use Rocoline track, with points set to Live Frog. Many points have integrated digital motors, others are via LS150s, or TrainTech Dcc signals...  Hornby Dublo Semaphores might be the highest individual power consumers !  (They now have Bill Bedfor LSWR arms and LED lighting)

 

And the 3rd row is 12-16Vac distribution, primarily  for the LS150 decoders from either a spare ZTC transformer, or an Ikea 12V 'lamp transfomer' will do, as these are NOT Pecp motors but <100mA.

Wiring is white 'figure of 8' heavy speaker flex marked by a black stripe for polarity.

Each Roco Amplifer/Booster is powered by a 18Vdc SMPS (4x)

Different colour-sleeved '6A flex' is used for the track and accessory bus distributions: Black, White, Gold and Grey being readily available.

 

There is also a 12Vdc distribution used for lower-level layout wiring with LED strips, and streetlighting / building lighting.  Currently running at about 3A of a 4A SMPS.(*Red&Black' speaker figure of 8)

There are also various Phone-Boxes, Street Lights etc which are powered from adjacent  'track'/ accessory / 16Vac busses to act as a visible diagnostic that they are on.

 

My programming track is an integral part of the layout, and thereore has a switched connection from either a SPROG or the Roco system.

Hope this goves you, and others, some ideas about dcc wiring arrangements with built-in diagnostics.

[Ron Ron Ron]

I spent some time this morning thinking about power districts.  I'm looking at buying 2 x 3A boosters - the Roco Dual Booster at 229 Euros from MBL in Germany looks good.  .....

 

I personally think you're wise to use Boosters that will allow RailCom, to match that from the  Z21's own Booster output, even if you're not planning to use RailCom initially. 

So that's a good choice.

Unusually, the price of the 10807 Z21 Dual Booster is listed the same everywhere at €229; even from the well known German discounters.

 

 

.

[Alan Kettlewell]

Hi Alan - appreciate what you've decided to do but I would suggest doing some tests because I still think you may have an issue here. Although each LS150 will execute change requests consecutively, Traincontroller (TC) will assume all turnout changes happen concurrently. In railroad.ini the parameter "turnoutinterval" is set, by default, to 0 which means concurrent operation. So in the example of 3 turnouts needing to set for a route that are on the same LS150, it will actually take 9 seconds to complete, but TC will assume all is done after 3 seconds. Hence a train could be started and potentially reach the turnouts before the last one has switched. If nothing else, the sound of the turnouts still changing when the train moves off can be distracting.

 

You can set turnoutinterval to a non-zero value (in milliseconds). You would need to set this to 3 seconds to ensure TC waits for all turnouts to complete. This means it always will take 9 seconds to set the route even if the turnouts are all on separate LS150s. And TC also sends a change command even if the turnout is already in the correct position. This is where you may have a problem as while this route is being set, no other turnout changes can take place.

 

I did have a similar problem earlier with my LS150s when I had Peco solenoids. The problem here was that all the LS150s were powered from the same TR100 transformer, which at 3 amps didn't have the power to reliably switch two motors at the same time. This could happen if both motors were on different LS150s which would then be sharing the output of the TR100. I therefore set the turnoutinterval to 1 second, but found that was too high and in busy situations delays built up resulting in some moving trains being signal-checked. (I can do a passable imitation of Clapham Junction when there are grandsons to be entertained!). Setting this to 500ms mostly resolved the problem. But you will require much longer to set the Tortoises if they're operated consecutively. I have one route on my layout at the station throat that requires 8 turnouts to be set which would have been impossible with LS150s. And in your layout diagram, a train travelling from "stn plat 5 east" to "block 69" will require 9 turnouts to be changed.

Hmmm ... Thanks for this, you make a good case and might be right. I'll set up some tests as soon as I have enough feedback detectors in place to run some simple schedules between my Level 1 storage loops and the station. I too previously used Peco motors but sacked them this time due to their unreliability. So, this is my first go with Tortoise motors. Any way, thanks for your input, I'll let you know how it goes.

Cheers

Alan

[Alan Kettlewell]

Blimey Phil, my eyes glazed over reading about your super diagnostic wiring method. Very much a Rolls Royce system you have there methinks - all kudos to you for the work put in on that set up and brilliant to have the know-how to do all that. A bit over my head I'm afraid but thanks for posting it all. I'm sure though that I can perhaps take at least some of those ideas and put them into use on mine. I'll certainly make sure I isolate both sides of each track for my districts and run through DPDT switches for my sub-districts. I was thinking of having just DPDT switches on the front of my layout with led lights to show the districts as powered, on if power, off if not, so I can see where a short might occur. Too simple?

Alan

[Alan Kettlewell]

Alan,  my main concern is the black area.  There are a lot of spirals on the black, and I suspect spirals can have high current loading (loco working up hill, on a curve).   I'd look to move one spiral to another zone, or arrange the wiring so the black area can be split into two should that prove necessary in the future. 

 

- Nigel

Hi Nigel,

 

Yes I agree, good advice thanks. I'll rethink that a bit.

Cheers

Alan

[Phil S]

" I was thinking of having just DPDT switches on the front of my layout with led lights to show the districts as powered, on if power, off if not, so I can see where a short might occur. Too simple?"

Alan

Phil: My background has been teaching maintenance on (broadcast) equipment for xx years... so a bias toward easy fault-finding is 'natural' 8-)

       It starts with simply having positive indications such as having LEDs which light when things are okay,   A fault lamp won't light if there is no power to light it !, and being able to follow the path as the signal (track power etc) progresses to its destination. 

 

Illuminated buffer stops are a simple example - either modern off-the-shelf, no wiring, or a led, diode and resistor .... I use small 1.8mm 'red'  LEDs which face sidways, and look like a lamp mounted on a square sighting board ... painted over to leave just the small lens for the red light to show.  (Protective diode eg 1N4148 in series to protect against reverse voltage breakdown of LEDs at maybe 5Vreversed   1/3penny extra when bought in bulk.)

 

I have changed to WAGO (lever operated connecting blocks in many places - now avaiable form Screwfix) for interconnecting wires because they can be disconnected without having to find a screwdriver.  (In the garden, I smother them with vaseline to weather protect them). Their disadvantage is being slightly bulkier, and needing 2 each time... one for each wire (but in 2-5way types)

 

My ZTC 511 had an inbuilt current display.  Earlier When I started with Zero-1 on its release, I soon learnt to add a moving-iron (ac) ammeter to monitor what was happening.  (and a 'scope helped too 8-) )

My ZTC once showed a 3 A current with no trains running ... I traced it to a failed switch in a Fleishmann point, which melted inside as a result....   tracing it being simply an ability to isolate sections until you find the fault 'disappear'... and then maybe a multimeter to spot the are with the lowest resistance.  In the Garden my Massoth wireless handsets show me the current from the base sation.

Roco MultiCentrale Pro and Z21 monitor current ... others too.

 

With the central panel's section switches; because they were wired for a choice of 2 inputs in each case, that gave an indication, if the source power was at fault, and also gave the layout 'fault tolerance' because another power supply could be tried (IF one was confident that using it would not cause 'catastrophic failure, of course).

The Rrampmeter and LED on the SMPS also showed power supply faults themselves ...  so the LED by the switch is more useful showing the TRACK side, than the source side. .... especially as it can then show 'bridged' gaps.

 

The PSX's I use are alongside the Central Box - so that their LEDs can be seen without crawling under a board - they have 2 LEDs showing their operation ... and one flashes in a retry situation.

 

I could have made the LED indications better, by using bi-directional 2-colour LEDs .... so that the colour would show dc on track (eg testing with a battery) or a combo colour for AC (AS IN DCC).

 

I do work from the prinicipe that FAULTS WILL ALWAYS OCCUR ...  and disbelieve those who say their layout NEVER has a problem ... therefore I want to find it as easily as possible......

(I have, regretably discovered, that I cannot 'logically fault find' when being continually asked by  a young helper what is wrong, before I have located the problem, and why am I testing that thing...

During maintenance courses, a 'real fault' was always inconvenient,and 'best found'/most easily found, after the students had gone home for the evening, because it was invariably NOT in a part of the machine being covered either at that time, or previously and therefore disrupted the schedule.8-(

 

Incidentally, I have just been testing the helix (3-3.5% r=511mm),  for a new layout I am building - and the single car Roco RS1, and Piko 2-car DMUs, both (silent) dcc fitted, ran up the incline beautifully with a 9V battery held to the rails .... I believe in simple tests with 9V batteries to check both analogue and digital locos where possible !! 

[Alan Kettlewell]

Hi Phil,

 

Apologies I haven't replied to your post sooner.  Your exploits and set up are very interesting and you have certainly made fault detection into an art form.  However, for my layout, I'm going to have a much simpler set up as I don't have the wealth of knowledge for anything more elaborate.  

 

I'll be splitting my layout electrics into 3 power districts and all I intend to do is connect an LED across the track power in each district eg District 1 - call it Blue District with a blue LED, Red District with a red LED and so on.  I'll mount these on a small panel board on the front of my layout.so I can immediately see which district has a short or fault if the LED goes out - probably couldn't get much simpler.    

 

I've no prior electronic knowledge to call on so I've been reading up on LEDs, resistors and simple circuits.  I've spent just a few pounds on a some 5mm LEDs in four colours with some mounting bezels and resistors.  Working out the which resistors to use has proved to be very interesting - and a tad confusing at times - but I think I've just about grasped it now.  Here are my workings to calculate the appropriate resistors:

 

Supplied track Voltage:  18V   (@3.5amps)

Forward Voltage across the LED: 3.2V

Difference being 18 - 3.2 = 14.8V

Forward Current across LED: 20 MAmps

Therefore Resistance required is 14.8/0.020 = 740 ohms.  So I'll use 820 ohm resistors.

 

The standard 0.25W resistors I first bought get too hot so the track power is clearly too much for these.  Further reading up and calculations show the power across the resistor: 18V x 0.020amps = 0.36W.  So I've ordered in some 0.5W @ 820 ohms.  It's a good job these small components are very cheap to buy - but quite fun experimenting!

 

I'm thinking, after re-reading your post above, that I should provide further protection with a diode to prevent the flow going the wrong way - more reading up required ..

 

Perhaps you could give these calculations a check over?

 

Anyway, it's all interesting, good fun and opens up another world of skills to tack on to our modelling exploits - although I want to be careful this doesn't lead into a major digression away from building my layout - so easily done!

 

Thanks

Alan

[Phil S]

Happy New Year - Christmas gets in the way of many activities !  I'm just returning to working on my new Skandi Portable layout.

You appear to have grasped the prinicples:  3 Power Districts is probably appropriate - if chosen on a 'modular'/geographical basis (mine was, remember, originally based on 2 x 5A ZTC outputs, and then adapted to match the 4 x 3.2A of the Rocos ... but with the benefit of Rampmeters on each output to actually see the current being taken

 

(the ZTC had a current display built in AND SO DOES THE Z21 and MultiCentrale Pro before it. Some other Central Units do too - eg our Massoth Dimax used for the Garden LGB).

MOST of the time our current demand is well below 1A per district ... but allow for more LED coach lighting and Sound in the years to come... we are up to more than 1A in the storage area when filled with Mumtiple Units, all lit internally.

 

However - I do not use Peco point motors, and my Analogue Roco point motors are powered from Lenz LS150s with their own 16Vac power  supply.  Digital  Roco point motors only take about 100mA - so neither present any great load on a power disitrict.   The Accessory DCC bus is normally fed from the (Z21/RocoAmp) 'Master', and not via any PSX or other protection. - so it can still operate points and accessories when track power is cut-off.   (Peco motros would be best powered from CDU-based decodwers)

 

Power Monitoring LEDS: I would suggest PAIRS of LEDS - 1 for each polarity: BOTH on shows ac (dcc), but only 1 when dc applied (eg when doing a continuity test - or to show any reversal (in dc)

I would give each of the LEDs their own Resistor and series protective diode....

 

Many may consider the 1/3p extra cost of the diode excessive protection ... but it allows you to use eg 12V flashing LEDs etc which have only 0.5V reverse capability, and SOME LEDS may fail at -5V.

I would also aim for 10mA  or less (small m for milli  , M=MEGA 8-) )  as modern LEDS are much brighter than older types  .. so double or quadruple your Resistor value, which then brings down their power requirement considerably

                                                          P=I^2 R  or P = V^2 / R    [V=IR, P=IV]   

ALSO - and only applicable when ac(dcc): the LED is only on 50% of the time, so the power rating required can be dropped further - in theory.   

{ I have a small LED Tester from  'KEMO' Germany, available from many UK suppliers like Maplins - with which to check LEDs with differing currents 1mA- 50mA}

 

And adapting from my method .... planned redundancy .... use the same type of plugs/sockets to connect the Power Districts so that ANY of the Sources (Z21 or booster o/p) can feed ANY or ALL  of the Power Districts ... and during low power requirement sessions, just use the Z21 sent to all.

 

rgds

phil