Railmaster Start Up - Point Setting Speed Vs Decoder Recharge Time

[walkingthedog]

When I had powered points on my layout the CDU used to be powered by the controllers AC supply, every time I changed a point the locos would slow down for a few seconds, so I gave it its own supply and all was well.

Maybe some things are still best

 

the 'old' way.

[Graskie]

I was just recently thinking that myself, fishy. Thanks for confirming in your final paragraph what I now know I should do when and if I eventually get round to DCC operation of my points, especially with regard to my Graskie Junction which would often

 

need to be route set.

[walkingthedog]

Bit old fashion though isn't it Graskie?

[St1ngr4y]

The original question was about recharge time during the startup process. I don't know what sequence used by RM to set each point to it's startup position, but if RM were to fire all of the port 1's first, then the port 2's, etc., then the decoders would

 

have longer time to recharge their cdu's. This would only benefit layouts with more than one accessory decoder, of course - the more decoders, the larger the benefit.

[DRB49]

Would the 4 Amp power supply be any better? I have the luxury of owning the Elite and so I'm using that PSU with the E-link. I have not experienced any problems so far and have a couple double cross-over points.

I have just made the transition from

 

switches to the Hornby accessory decoders, so I'm interested in a successful solution. I wouldn't personally want a solution using relays.

If you are going to pursue the automation option on Railmaster, you could easily program a second pulse to each point

 

and in a delayed sequence, which would almost guarantee correct operation.

The only option is perhaps a booster.

This is my first ever contribution, so please don't be too quick to rubbish it.

[St1ngr4y]

RM does fire each point twice, and if you have two points on the same address e.g. for a crossover, it seems to send three commands. However, it only does this if you use the red/green buttons on the layout diagram. If you fire a point from within a program,

 

it only appears to send one command.

[DRB49]

Ok, sorry I hadn't realised. I haven't properly "played" with the programming much yet. The priority last weekend was locating the last two R8247s in World and then rewiring the points. I guess the work around within programming would be to repeat the

 

point command about 1 second later, perhaps for the opposing point. It's not clean but it's a lot better than adding relays.

[Pete172]

DRB49 said:

Ok, sorry I hadn't realised. I haven't properly "played" with the programming much yet. The priority last weekend was locating the last two R8247s in World and then rewiring the points. I guess the work around within programming

would be to repeat the point command about 1 second later, perhaps for the opposing point. It's not clean but it's a lot better than adding relays.

Hi, as far as I can see the only option is to increase the time between point firing, to allow the

CDU to recharge enough. The Elite power supply is 4 amp, with 3 amps going to the track, so that should be more than enough to re charge the capacitors as quickly as is possible. Now, if Hornby had designed the R8247 with capacitors on each output port, then

there would be no problem !
I'm going to my layout now to try to change the pause time up to 3 seconds, will let you know how I get on....probably break it!
All good fun ..
P.

[Fishmanoz]

If you go back and read the entire thread from page 1, you can see that Pete172's problem, using Elite and so 4Amp supply, is that only the port 1 points fire on startup. The solution found to work by others is to set the point firing delay to 3 seconds,

 

giving the R8247 time to recharge between throws. Except for Gregd99 who is using a non-Hornby 6 port decoder which uses as separate supply to fire the points, so no such delay needed.

[Fishmanoz]

And the thing about a CDU is that it produces a high burst of current for a short period and then recharges at a lower current for a longer period. Also, the Elite only allocates 1Amp to accessories leaving 3 for locos. So if a number of decoders are all

 

trying to recharge at once on startup after each has fired port 1 in rapid succession, the recharge time for each is liable to be longer then normal. It would be interesting to know exactly what the recharge current is to see how many can be recharged at once

 

before this becomes a factor.

[Pete172]

Hi,

Well I managed it !,.. I had to open Notepad as an administrator, then use it to navigate to the file, make my change for the point timer-from 0.75 to 3.00 secs, then it allowed me to save the change.

Now, when I start up Railmaster, my points

 

change direction to their default position. Yes, it takes longer, but at least it works now.

After this change my observations are :

1. When pressing the green/red buttons, the points seem to take the full 3 secs before I can fire another, before it

 

was faster - on manual pressing.

2. I can now hear a distinct 1st & 2nd pulse to the points, with a long (3 sec ?) pause between them ! is this what would be expected ?

3. At start up, my point setting are strange, (I have only 2 decoders installed

 

at the moment)the port firing sequence is 5,6,1,2,3,4,7,8. Very odd, as obviously 567&8 are in decoder No2. I might try going back to the design and removing the point buttons, saving, then re-installing them - any thoughts ?

 

Anyway, at least it is

 

working now, it seems to have taken many nights to get to this stage. Lots of fine tuning and 3 more decoders to fit..

My main remaining problem is that Railmaster keeps stopping working !(using Vista), but that's another story.

Thanks for all your help.

 

P.

[Gregd99]

DRB49 said:
Would the 4 Amp power supply be any better?

A CDU is designed to limit the current that flows into it thereby avoiding overloading effects on the feeding supply. the lower the current limit the longer it takes to recharge

the CDU. Somewhere on the forum are some calcs that i did showing recharge times.

I would "guess" that the limit in the 8247 is around 100mA. This would mean that you could have a select with a 1A supply able to run a couple of locos and power two 8247

recharges at the same time. Note that the current draw form the 8247/CDU is only (mainly) during the recharge cycle. At other times the draw will be very low.

Would the 4A supply help? i don't think so. the current limit is inside the 8247 and it has

no idea whether the supply is 1A or 4A. It will be designed to work with the 1A supply.

This is my first ever contribution, so please don't be too quick to rubbish it.

Welcome aboard. Questions shouldn't ever be rubbished. I think

that most people look to see that someone has though about their question and have tried a few tings first so that they can ask a good question.

The more thought out the question ... the better the assistance can be.

[96RAF]

Pete172 said:
3. At start up, my point setting are strange, (I have only 2 decoders installed at the moment)the port firing sequence is 5,6,1,2,3,4,7,8. Very odd, as obviously 567&8 are in decoder No2. I might try going back to the design and

removing the point buttons, saving, then re-installing them - any thoughts ?
P.

I wonder if RM initialises the points in the order they were added to the plan.
Testing on my setup (which is still on the demo version restrictions pending

receipt of my eLink-RM combo) the points fire as I laid them on plan 1,4,5,7 (using 2 ports on each of 2 x R8247s).
It may be possible in this case to amend the order in the track plan file (wherever that is and if accessible to edit) so that each R8247

gets a chance to recharge - although as stated earlier charging in parallel may not actually be any more beneficial than charging in serial.
Rob

[DRB49]

The idea of suggesting the 4A PSU over the 1A is that with its lower internal impedance, the initial charging of the capacitor inside the accessory decoder would be marginally quicker.

Another suggestion (hard hat firmly in place) - solder another capacitor

 

in parallel with the one on the accessory decoder.

[Gregd99]

DRB49 said:

The idea of suggesting the 4A PSU over the 1A is that with its lower internal impedance, the initial charging of the capacitor inside the accessory decoder would be marginally quicker.

Gosh that's a thought provoking

answer (for me at any rate).

My forte is micro-controllers rather than regulated power supplies but is it necessarily true that a higher capacity supply has a lower internal resistance?

[Fishmanoz]

Sounds like an attractive proposition but it will only make a difference if the charging current drawn by the decoder is more than 1Amp, and that is clearly not the case. Given the voltage is the same, internal resistance is nothing more than a proxy for

 

the current capacity of the supply.

[Pete172]

DRB49 said:

solder another capacitor in parallel with the one on the accessory decoder.

By soldering 2 capacitors in parallel...would that not provide a very beefy kick to the points motor and cause some extra load on the component

(thyristor ?)that is switching the capacitors.
Surely the answer is to have a capacitor for each output port...

P.

[Pete172]

Hi all,

I'm still trying to understand what the figure 15 represents in the "Delay" box. This can be found when pressing the "Accessory Modules" button on Railmaster top tool bar.

This option may or may not be useful to me when setting up my point

 

decoders .

I have examined page 41 of the Railmaster PDF, but don't see any reference to it ?

 

Thanks,

Peter..

[Fishmanoz]

Pete172 said:

DRB49 said:

solder another capacitor in parallel with the one on the accessory decoder.
By soldering 2 capacitors in parallel...would that not provide a very beefy kick to the points motor and cause some extra load

on the component (thyristor ?)that is switching the capacitors.
Surely the answer is to have a capacitor for each output port...

P.

Sorry, I missed this suggestion from DRB when I was reading the thread previously. There is no way you

can add to the charge storage in the decoder without considering the complete design of the unit. Simply adding a capacitor sounds to me like a good way to have the internals of your decoder looking blacker and more melted then they did before, and to have

the corrosive internal contents of the capacitors now being on the outside corroding everything else. Or put another way - don't.

And given the original problem was recharge time, not sufficient energy to fire a point, doubling the capacitance will

double the recharge time, making the problem worse not better. The trade off might be that you have enough energy stored to be able to fire 2 points before it is fully discharged, but only if the whole circuit is designed to do that.

Yes it would only

fire points in rapid succession if sufficient energy was stored to do this, and one capacitor per port would be a way of achieving it. But that is a complete redesign of the unit.

The bottom line - if you are going to use 8247s, you are going to have

to live with their design limitation of only being able to fire a point every 3 seconds or so. If you need to throw points more often than that, you'll need to look for an alternative.

[St1ngr4y]

Pete172 said:

Hi all,
I'm still trying to understand what the figure 15 represents in the "Delay" box. This can be found when pressing the "Accessory Modules" button on Railmaster top tool bar.
This option may or may not be useful

to me when setting up my point decoders .
I have examined page 41 of the Railmaster PDF, but don't see any reference to it ?

Thanks,
Peter..

Hi Peter,
I think the 15 second delay refers to the gaps between reading data from and writing

data to the decoder.
Ray

[HornbyRailMasterSupport]

The Hornby R8247 accessory decoder, when used with RailMaster, does not require recharging after every 3 seconds or so per point fired. The "Points Timer=0.75" setting within the RailMaster.ini file is set to the optimum firing time when firing Hornby

 

point motors consecutively, that is three quarters of a second per point motor.

 

It then takes between 2 and 4 seconds to recharge the CDU within the decoder, depending on the amount of power available to the unit.

 

Some point motors, e.g. motor

 

driver (like seep) require more power, for longer, to fire therefore you should set the "Points Timer=" setting to a higher value, e.g. 2 seconds and use higher values until all four points motors linked to an R8247 fire correctly.

 

Also bear in mind

 

that if you have two point motors linked to the same output port on the accessory decoder (to fire two sets of points in a crossover, for example), this will require slightly more power, however the 0.75s timing covers this if you have doubled up one port

 

on the module (i.e. you have five point motors in total).

 

RailMaster fires points twice to ensure that any sticky or lazy points are indeed fired. The second firing requires far less power than the first.

 

The delay within the accessory decoder

 

programming window is the number of seconds needed to program the module correctly. RailMaster supports several popular accessory decoders and the optimum delay for each is built in. RailMaster also covers things like programming the R8247 three times automatically

 

in order to allow the CDU to charge up.

 

We hope this clarifies things.

 

[Pete172]

HornbyRailMasterSupport said:

The delay within the accessory decoder programming window is the number of seconds needed to program the module correctly. RailMaster supports several popular accessory decoders and the optimum delay for each

is built in. RailMaster also covers things like programming the R8247 three times automatically in order to allow the CDU to charge up.

Hi , Many thanks for your reply.
Now that I know what the "15" refers to, I won't need to touch it. Should

this information be inserted in the Railmaster PDF instructions ? Anyway my moto is, if it ain't broke don't fix it, so I will leave it well alone.
Many thanks,

[Pete172]

RAFHAAA96th said:

the points fire as I laid them on plan 1,4,5,7 (using 2 ports on each of 2 x R8247s).
It may be possible in this case to amend the order in the track plan file (

Hi,
I managed to sort out my point firing

sequence on start up.
I went back to the track plan, deleted the point buttons, and re-inserted them in logical order, so now they all start up in sequence 1,2,3,4,5,etc.
Interestingly, if you want a particular firing sequence, I think they may be able

to fire in the order that they were placed in the plan - but I would need to re-check that to be sure...

[Gregd99]

HornbyRailMasterSupport said:

The Hornby R8247 accessory decoder, when used with RailMaster, does not require recharging after every 3 seconds or so per point fired. The "Points Timer=0.75" setting within the RailMaster.ini file is set to

the optimum firing time when firing Hornby point motors consecutively, that is three quarters of a second per point motor.

It then takes between 2 and 4 seconds to recharge the CDU within the decoder, depending on the amount of power available to the

unit.

Could you please clarify that this means
1. a pulse of duration 0.75S is applied to the point motor
2. after the pulse is removed the CDU needs between 2 and 4 seconds to recharge to the level where another motor can be operated.

Thanks

[HornbyRailMasterSupport]

Not a pulse of 0.75 seconds, but a pause after the point is fired. You can set the pulse time in the accessory decoder programming window separately, as long as your accessory decoder supports pulse firing. The Hornby R8247 does however the older R8216

 

does not.

 

We are not sure what you mean by "After the pulse is removed". Hornby accessory decoders will happily fire Hornby point motors with a gap of 0.75 seconds between firing. This timing only applies to RailMaster setting your points and signals

 

on start-up and also rapid firing of points in succession during normal use.

 

As mentioned previously an accessory decoder module (Hornby) then requires between 2 and 4 seconds to recharge based on a) the number of any other locos/accessory decoders

 

being powered and b) how drawn the current was on the accessory decoder in question (e.g. how many points may have been paired to a single output and what kind of point motors are used). This is normally nearer 2 seconds for a satisfactory charge to fire one

 

point again.

 

In normal operation, once you have set, say, four points for a single accessory decoder, it is unlikely that you will want to set all four again within a few seconds thus you should not need to worry about recharge times.

 

To confirm

 

the order of firing points/signals on start-up, this is done in the order in which these items were added to the track plan. If you are concerned about the firing sequence (you should not be if the "Points Timer=" setting is specified correctly in the RailMaster.ini

 

file for the motors/decoders you are using) then you can set the order by removing your point indicators from the track plan and re-entering them in the order you want them to fire.