Jump to content

Problem Stopping Locos with RailMaster


Recommended Posts

Hi All,
I have been an Elite user more or less since it was released and was keen to give RailMaster a go. From what I can see from other posts some people find it works really well but I seem to have a problem controlling locos.

When I drop

the throttle controls to Zero, the locos don’t slow down, often they accellerate! Only pressing Stop will stop a loco. Currently this problem seems to affect all my locos. They all work fine and generally run noticeably smother when controlled directly from

the Elite itself.

I also have a problem with one loco where the throttle drops to zero a second after the mouse leaves it.

I have followed the advice in the Manual regarding erratic control and tinkered with the Baud Rate but it hasn’t made any

difference.

Has anyone got any ideas?
Link to comment
Share on other sites

Hi GPMark,
I have had a very similar problem with RailMaster, my locos all start fine when I click the shunt button or the cruise button but when I click the stop button nothing happens and I have to click 2 or 3 times before the throttle indicator

goes to zero and the loco reacts.
This has caused a few tragic accidents on my layout and my previously excellent safty record is now shot to hell.
I have also tried to change the baud rate to no avail.
Any ideas to solve this will be appreciated.
Link to comment
Share on other sites

All locos behave normally when controlled from the Elite. This is just a problem when controlling from RailMaster. Locos seem to behave fine when I use the Shunt, Cruise and Stop buttons. It’s just the slider that causes problems.
Link to comment
Share on other sites

Hi GPMark,

Have you tried stopping the speed slider just short of the Zero position and bring the train to a crawl, then pressing the loco's stop button?

It is possible that there is a bug in the software which needs correcting.

There

will be a sub-routine that tests for the slider position, then reduces the loco's speed. When the slider reaches the Zero position this routine should invoke the Stop sub-routine. What is probably happening is that the slider subroutine is not invoking the

stop routine, but jumping back to full speed ahead. By moving the slider just short of the Zero position and clicking on the loco's stop button you are invoking the stop routine and the terminating the slider routine.

Try this and see if it works.

Does

anyone else have this problem?
Link to comment
Share on other sites



Ive have the same and some times the Elite locks up and will not power up after a power cycling. I have to press the menu buttom on power up to regain control. Not only

does the loco not stop but am having problems with the F2 sound not stopping and I have played around with the macros and nother just more F2 sound and then will not stop.
Link to comment
Share on other sites

Hi GPMark,

All users of Railmaster need to list their problems (in detail) here, so that we advise Hornby of the problems we are having.

I am pleased that the temporary solution works, but as you say they should not happen. Unfortunately,

software developers, tend to be sloppy with their programing, partially because they do not findout exactly what is needed by talking to users.

They rely on users to debug their programmes, big or small, they all suffer from this problem. One of the

rasons for this is that they test each sub-routine to see if it works, but they do not test all possible combinations of subroutines. The number of lines of code needed to carry out a simple task, does not help debugging.



Link to comment
Share on other sites

I believe the problems you are all having is linked to the speed of the USB bus on your PC and the amount of data which can be transmitted to the Elite (as the Elite is using a virtual RS232 driver through a USB port).

RailMaster sends real-time

data as you slide the slider, to the USB port, which is then translated by te Elite driver to a serial signal. Some USB ports, particularly version 1.1 and running on slower CPUs may not be able to hand the data traffic, which is why there is a "Send real

time data" checkbox in RailMaster's settings. Un-checking this will cause RailMaster not to send data as the slider is moving, but only when the mouse ro finger is released. This eliminates the problem of buffering data to the USB port.

You could try

un-checking the "Send real time data" box in settings, or try swapping your USB cable to another port, perhaps USB version 2.0 and see what happens.

I know the developers built RailMaster on the back of talking to users directly, but of course cannot

cate for every combination of CPU, driver, USB, Windows operating system and so on, which is why they put the Help Request system in the software.
Link to comment
Share on other sites

Hi LMS Tim
Thanks for the advice. Un-checking the Send Real time speed data option has resolved the problem. The stuck throttles no longer mysteriously return to zero a second after you let go and the others do all stop when you drag the slider back

to zero. Speed changes aren’t exactly smooth but for the time being I can live with that in the hope that a future upgrade fixes the issue.

As far as I am aware all by USB ports are USB 2. I’ve only had the PC a couple of years.

Do we know if

the developers read these forums?
Link to comment
Share on other sites

I believe the developers do read these forums now and then. What I do know is that they take all and any comments and advice seriously, so maybe you want to put together exact details of what happened in your case, and details of your PC and set-up and

send details to Hornby.
Link to comment
Share on other sites

Guys,
I suspect that the usb channel is not the issue.

Google hasn't led me to an effective baud rate for virtual serial ports over usb. let's assume that 9600 b/s or approx 1000 bytes/sec is the lowest imaginable rate over usb.

an xpressnet

message is around 10 bytes. this means that 100 messages per second could be sent over even a very slow connection.

I would expect that railmaster would put some sort of limiting on the rate at which it sends messages but even if it didn't. 100%

throttle to 0% - 128 changes could be sent in 1.3 seconds.

It is interesting that turning off the real-time data option has resolved (or perhaps reduced the frequency) of the issue.

I would suspect the computers on either end of the link

rather than the link capability itself.

Greg
Link to comment
Share on other sites

Greg

Just to clarify, Xpressnet actually works at a bit rate of 62,500 BAUD. It is nothing like the USB connection, which to the Elite works most effectively at 19,200 BAUD.

The theoretical maximum amuont of data which can be sent does not

take into account the fact that RailMaster must obviously poll the Elite frequently to check settings of all locos and so on. This is shown when you use the Elite control knob the RailMaster slider for that loco goes up and down.

Remember that the Elite

is a very low powered CPU compared with a PC and that the bottleneck is the USB connection to the Elite, going through a serial driver, itself.
Link to comment
Share on other sites

xpressnet does work at 62.5kbaud with specific real-time requirements. this is the rate that (hopefully) applies when the select is used as an xpressnet throttle.

I have never found a definitive statement but i think that the baud rate on virtual

serial ports is largely ignored. as far as xpressnet is concerned I think the speed on the link to the pc can be anything. in tests I have done with the elite at different rates the performance of the elite has not changed in any way I could detect.

if

by "bottleneck is the USB connection to the Elite" you mean the actual hw/sw inside the elite then I think that this could well be the case.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
  • Create New...