Advice on E-links & boosters

[LuLuJo]

I would welcome some advice on the use of power boosters with elinks and railmaster pro edition?

My layout Windermere & Hincaster Junction is a fairly large layout approx 24ft X13ft. It has a double track circuit with a scenic fiddle yard Hincaster is one side and elements of Carnforth on the other. There is a double track branch to a scale model of Windermere station. There are around 75 locos and 120 points and signals. I have an elink controlling power bus A which is split into 3 separate power sections for track power . Bus B has another elink for the points and signals with 4 separate sections. Both elinks have 4 amp power supplies. All the sections on both track and accessory buss can be isolated with switches.

I have been concerned for some time that loading the system up was becoming very slow and unreliable and the Accessory bus always had to be switched on in sequence a section at a time to charge the point decoders slowly.

So I invested in a new PC running windows 11 and transferred the software as advised and set it up with all the recommendations about the ini file etc. The railmaster system is the latest version 1.745

It initially worked okay until I tried to add power boosters. I added 3 tam valley power boosters one to each track section on Track Bus A. These have 4 amp supplies. The boosters were set up for track power. If I try to start the railmaster system with the power boosters powered up the elink will not initialise the elink.

Windows deice manager shows both the elinks there on Com 3 & 4 and supposedly working correctly.

I also have a hornby power booster which had worked on the track bus A in the past and tried adding that to power Bus B. That will not initialise either. If I take it out and try to start the system with all the Power bus B sections turned off and then switch them on one at a time it might initialise.

As an aside I have a voltmeter on each power bus . I know this is not an accurate reading but it gives an indication of what is going on. On switching on with no section live on Bus A reads 13.6volts. Without the boosters it used to decline to 12 volts if all 3 sections were live. With the boosters in the system it stays on 13.6volts. Bus B reads 13.6 volts with one section switched on if another is added the voltage drops then recovers. Adding the 3 & 4th in sequence drops the voltage and it recovers to 12.8 and stays there

So any advice as to what is going on. It looks like I must only apply power to the boosters once railmaster is loaded and the elinks have initialised but I dont understand why? I am not sure if the Hornby power booster is suitable for use on an accessory bus as it does not seem to work and the Tam Valley boosters do have a separate setting for such use.

I appreciate that I may be pushing the system to its limits but would welcome any advice and comment

Mike

[96RAF]

The Hornby booster is more accurately described as a DCC signal booster and isolated district power supply as that is all it does.

The isolated section is self powered and the booster module listens to the DCC signal from the controller and passes it to the isolated section.

Bus B should behave exactly as Bus A does. There should be no power drop on either.

Have you tried powering up the isolated section boosters first and the controlled section last. That way there should be no inrush from the controller in the first section to the next in the daisy chain.

[Fishmanoz]

Wow, that’s some setup Mike. You have enough power there to run a small village (I exaggerate!).

I think you know the reason for sequential startup on your accessory bus - it’s the inrush current to the CDUs in the accessory controllers. And the advice when this exceeds the overload state of your controller is exactly as you are doing - sequential startup.

Such a situation should not occur on your track bus. There should be no significant current draw at startup. That should only happen when you power up locos. Am I wrong and you do have something drawing power when you start? Or do you perhaps have some fault condition causing it?

It seems even more mysterious that wiring in the boosters has made things worse. By rights, all the power to each section should be coming from each booster power supply. All the controller should be contributing to these sections is low current DCC control signals, even with 3 of them. Do the Tam Valley boosters say what this input power requirement is? Surely only milliamperes, not amps? Do you have them wired correctly so this is the case?

[LuLuJo]

Thanks for the replies.

I have made a little progress but still have not much idea what is going on.

If you follow the following start up procedure exactly the system will initialise and run:

1. Power up the elinks
2. Switch on the computer then load up railmaster
3. Power up the boosters
4. Switch on the power sections on both bus A then B in sequence.

At least this means that i can run the system but it does not feel very robust.

Powering up the boosters first or at the same time as the elinks does not work.

My volmeters show a no load output of 13.6 on both buses. The voltage on Bus B in use with all sections active and which has no booster drops by 1.6 volts to 12. The voltage on Bus A starts at 13.6 or 12.8 with the boosters off then rises to 13.6 once they are switched on and all sections are active.

I would like to put a booster onto bus B as i feel that the 125 points and signals are pushing the capacity of the system but putting my hornby one into the system shuts the elink down. The yellow light does come if that indicates anything helpful. Sadly i cant get any more Tam valley boosters .

[LuLuJo]

A further update. I have acquired a dcc concepts alpha meter which allows for a bit more accurate voltages and also shows current draw.

With no load the elinks outputs 15.6 volts according to the alphameter. This compares ith 13.6 on my cheap meters of ebay.

On Bus B (the points) adding the first section drops the reading to 15.4V and current draw is 0.43A. Adding the second section produces 14.97v 1.00A, third 14.33V 2.02A and fourth 14.05V 2.35A.

Changing the order produces the same final outcome. I wonder what the maximum output of an elink with a 4 amp supply is?

On Bus A the track adding the first section (Hincaster) reduces voltage to 14V and current draw is 0.94A. The second section (Carnforth) produces 13.93V 0.99A Running a sound operated garratt with 30 wagons produces little change. I suspect the load from the track sections may come from the frog juicers as I have seen a reference to a largish layout having a similar load. The voltage output from the elink is stable however many track sections are live suggesting that the boosters are taking the load off the track elink.

I am now pretty happy that the track side of this is okay. I just need to find a way to add a booster(s) to the accessory bus.

An aside. I setting up the system i checked that the elinks were present and working using device manager. The default baud rate for the driver in Windows was set up at 9600 so I changed it to match that on railmaster 115200. It seems to me that the elink now initialises faster. I have not seem it suggested that this is worth checking but I may well have missed it

[Fishmanoz]

It seems you have transient currents in the system somewhere on startup but, now you have a startup sequence that works, then it’s no longer broke - so don’t try to fix it.

That’s even more the case with your B bus. You could have a thousand points there and no problem if you only throw one at a time (I exaggerate). And you can 100 LED lights/signals and be drawing less than an amp.

And if I can say, you’ve taken a very conservative path on boosters and buses. I’m want to say that, until your layout takes up both car spaces in a double garage, you likely don’t need a booster. Even then, big layout is a proxy for running lots of things at once. It’s the lots of things that draw the current, not the big.

[P-Henny]

The default baud rate for the driver in Windows was set up at 9600 so I changed it to match that on railmaster 115200. It seems to me that the elink now initialises faster. I have not seem it suggested that this is worth checking but I may well have missed it

 

 

I think you will find that this is documented in the separate Windows Driver Installation.pdf manual that RM installs. It is also mentioned in the main RM 1.74 Manual on Page 16 (just checked).

[LuLuJo]

My point about baud rates is that I had missed the need to change it in windows it at some point in my moves over the years from XP to Windows 7 then 10 and now 11. The manual is very out of date and the excellent advice in this forum about transitions also talks only about the changes in railmaster. So I suspect I may not be alone.

If its wrongly set the system still works but apparently more slowly

[LuLuJo]

"That’s even more the case with your B bus. You could have a thousand points there and no problem if you only throw one at a time (I exaggerate). And you can 100 LED lights/signals and be drawing less than an amp."

Is that really right? I have 125 points and a continuous load of 2.4 amps. The load is proportional to the number of decoders in each section.

One of my reasons for pursuing this is the hope/possibility that the layout may expand again significantly and its already beyond what would normally be in a double size garage. So I want to find a way of reducing the load on the elink.

I will ask around to try to find someone who has done it

[Fishmanoz]

To be more specific to get a better estimate:

• how many lights/signals do you have that are continuous LED/non-LED?
• what type of accessory decoders and point motors are you using? Solenoids have zero standing current, everything else will have some so there is a practical limit.

But 2.4 Amps on a circuit with 3-4 Amps available - no problem for now.

[P-Henny]

In addition to Fishman's questions above. Are you using Electrofrog points with live switched frogs. If so, how exactly are they wired and what product is being used to perform the frog switching. I'm asking this question because I have an idea regarding why the Hornby booster shuts down the B eLink, but an answer to this question will indicate to me if my idea has any traction in it or not.

EDIT: Last minute thought. Since adding the Hornby booster causes the B eLink to shut down, could the booster actually be faulty. Has it been tested on the A eLink to confirm it works on that one?

[Dukedog.]

You dont say what cable sizes you are using, if you where to increase the cable size to the first section, then decrease at the end of that section and so to all the other sections to the end of the power bus as the current demarnd drops after each section the voltage will stay more stable. with low voltages always best to go bigger than you think with cable sizes as the current draw goes up the voltage goes down. look to increase the first one or two sections to 4.0mmsq or even 6.0mmsq this would reduce the risistance and keep the voltage up.

[96RAF]

@P-H

With regard to the Hornby ‘booster’ please see here http://www.halton96th.org.uk/page18.html

@LLJ

Which wiring method did you use to connect your Hornby ‘booster’ as the Xpressnet method has been found to be unreliable and that a direct district to district signal connection works best.

[P-Henny]

LLJ has stated that both A & B are eLinks and since the eLink doesn't have either a Boost or Xpressnet port, then he is most likely using the direct district to district connection method, as per Fig 4 in the booster manual.

[Fishmanoz]

Hi Dukedog, I’m not sure that you’ve understood booster operation. You seem to infer that there is a power connection between sections? This is not the case, each is powered separately, the connection is only for signal transfer.

And it’s this that makes the startup sequencing problems more of a puzzle.

[Dukedog.]

Yes my bad sorry miss read, but as the voltage drops in the feed bus with each addition of each booster points to the feed supply to the booster dipping so these are the cables that need to be increased in size.

[P-Henny]

I may be wrong, but I think that one or both of you (Fishman / Dukedog) have misread the OP posts.

My reading of the posts is that currently it is only BUS A (eLink) that has Boosters in circuit and that BUS A voltage is predominantly stable.

LLJ says he wants to modify BUS B (eLink B) to include a Booster which currently doesn't have one. What he does say is that BUS B is made up of more than one section which can be switched in and out of circuit using switches. It is the sequential adding of each section to the SAME BUS B that is affecting the voltage and causing the current to increase proportionally. It is the sequential current increases observations that is driving the desire to add a Booster to BUS B.

Maybe LLJ will confirm or not this understanding.

[Dukedog.]

yep agree, maybe a simple drawing would help?

[LuLuJo]

Thanks for the comments and questions. Hopefully answers to the questions:

The set up is as described by P-Henry

"My reading of the posts is that currently it is only BUS A (eLink) that has Boosters in circuit and that BUS A voltage is predominantly stable.

LLJ says he wants to modify BUS B (eLink B) to include a Booster which currently doesn't have one. What he does say is that BUS B is made up of more than one section which can be switched in and out of circuit using switches. It is the sequential adding of each section to the SAME BUS B that is affecting the voltage and causing the current to increase proportionally. It is the sequential current increases observations that is driving the desire to add a Booster to BUS B."

Added Comment and answers

The sequential switching on of the sections on Bus B is necessary because there are quite a number of ADSX8 decoders which have 16 capacitors each and will trip the elink due to current draw if not switched in a slow sequence. I switch each one on watch the voltage drop wait a few seconds for it to stabilise then switch on the next

The frog juicers are Gaugemaster DCC80's fitted to all the points

I tried wiring the Hornby booster as set out in the manual using normal wires

The booster had previously worked on the track bus.

The point and servo decoders are an eclectic mix of ADSX 8 & 4, ADFX8, Traintech 4, hattons combined solenoid & decoder, megapoints servo controllers, Traintech SC2?(dapol semaphore)

There are 120 points all solenoids with decoders and the 20 signals are all but 1 semaphores and mainly servos but some old hornby dublo solenoid

The 3 tam valley boosters which are now in use on Bus A and I am happy with have a jumper which needs to be changed if its being used as an accessory booster. This made me wonder if the hornby booster is more suited to being used to power the track.

As a general comment having an instant display of from the 2 voltage meters in the A & B circuits is extremely useful. They cost £8 and I think I would have given up and taken up stamp collecting without them!

[Dukedog.]

You have answered your own question here.

The sequential switching on of the sections on Bus B is necessary because there are quite a number of ADSX8 decoders which have 16 capacitors each and will trip the elink due to current draw if not switched in a slow sequence. I switch each one on watch the voltage drop wait a few seconds for it to stabilise then switch on the next.

It is the current inrush from the ADSX8 pulling the feed voltage down.

They work ok in the track as minimal current draw.

[LuLuJo]

So you are saying that the hornby booster is not usable on an accessory bus ? That may be true but its not due to current inrush.

The point is if I put the booster in the system is it will not allow the elink to initialise before there is any current draw at all from the bus because the bus is isolated at that point

[P-Henny]

Totally relying on memory here, so the content detail of this post is the best I can remember.

Several years ago I recall there was a long debate on this forum (thread now long gone as the images in it have been lost in the forum platform transfer) where eLink BUS B was being used to power multiple DCC Concepts ADS8SX and was affected by inrush current issues.

In that lost thread, it was suggested that the OP installed Hornby R8239 Boosters (two) to split his ADS8SX decoders into smaller groups (maximum of 3 units per group if I recall correctly as 4 units immediately tripped the short circuit protection). The OP went away and did this, then reported back that the suggested solution was successful.

Personally, I would try swapping the two eLinks over to see if eLink B still functions correctly with the TamValley Boosters on BUS A. And also does the R8239 still affect eLink A when used on BUS B.

[96RAF]

I trust you get such comprehensive answers to your FB request for help. Beware of Sharons.

[LuLuJo]

Thanks P-Henry. That seems like a reasonable suggestion to swap the elinks. I will give it a try when I can then feedback the results.

Its interesting that I also first hit trouble when I got to 3 ADS8SX decoders and had to start loading them up in sequence and then splitting the bus. The earlier FX versions are much less of an issue.

But also very useful to know that someone has used the R8239 boosters successfully on Bus B.

Mike

[LuLuJo]

I thought these days you mainly had to beware of Karens?