Mastering Brake Compensation

Hi Brian

I hope I've understood your problem - that is to get the loco to stop in the same place each time.

Stop Markers are inherently unreliable.  That's why I went to dcc uncoupling of my box cars, as I can't get the loco to stop in the same place twice.  It seems to me that the problem relates to the fact that TC "guesses" the stopping distance, based on a timer.  The software takes the information it's gained in the profiling into account, but it's still a timer - not a linear measuring device.

The vagueries which will throw this off are exacerbated in my case by the fact that I'm using a sound decoder which has an internal timer matching the sound sequence to the motor parameters, as well.  When I add in the age and infirmity of the mechanism of the loco, TC is really working in the dark - or at least the tilight zone. 

Some users have resorted to using IRDOTs and other sensing devices to try to build accuracy into the results, but I understand that it's a bit complicated to set up and very inflexible - ie., it only works in the spot where the IRDOT is located.

It's possible to get Brake Markers to bring the loco to a smooth stop.  John's had some success with this, but it might be worth experimenting with changing the start/ramp combos.  So long as they add up to the Stop Marker value, it doesn't matter how they are divided up.

I hope that this has been of some help.

Would you like me to set up a test sequence on my plank, Brian? 

[user=449]brianpr1[/user] wrote:

I have to admit that I have not bothered to use the brake compensation test. It seems uncertain in its effect and I don't think it offers practical help with my stopping problems.

 

I guess we will have to agree to disagree Brian 

In my experience, insisting that my all my locos stop within .25" of a given mark on a test bed saves a lot of time when they are asked to stop in a variety of places on the real layout. 

I am intrigued by the concept of having the brake distance slightly less than the stop distance so that the loco crawls at threshold speed for a couple of inches before the stop marker activates. Based on the manual I had always thought they had to match…….when I re read the manual there is a suggestion this doesnt have to be the case……although all the illustrations have matching distances:roll:……………….I guess I should test it out……………..I do fiddle with ramp distances to ensure consistent stopping 

Kind Regards

Brian,

I thought the same as you do now about brake compensation but after experimenting I have found that it does improve stopping control……. it just takes a lot of time..

John,

I was going to post to say that having the brake distance slightly shorter than the stop distance hadn't created any  problems with the RR&Co software  and that I use it to replicate that short distance where locos crawl before engaging passenger coaches at the terminal or on run-around - but I see from your last post you are going to experiment with it. The only issues I have found are

• the loco speed profiling must be spot on - otherwise the motor stalls and the loco ends up a few centimetres from the carriage
• it doesn't work for really slow running locos - unless you want to replicate a scale speed of a tortoise

 

[user=540]gdaysydney[/user] wrote:

John,

I was going to post to say that having the brake distance slightly shorter than the stop distance hadn't created any  problems with the RR&Co software  and that I use it to replicate that short distance where locos crawl before engaging passenger coaches at the terminal or on run-around - but I see from your last post you are going to experiment with it. The only issues I have found are

• the loco speed profiling must be spot on - otherwise the motor stalls and the loco ends up a few centimetres from the carriage
• it doesn't work for really slow running locos - unless you want to replicate a scale speed of a tortoise

[/*]
I already get that problem on some locos where the throttle is still showing movement but the loco has stopped short :sad:

I find it quite difficult sometimes to establish the "best" threshold speed in order to achieve realism and reliabilty……….and of course the threshold varies from loco to loco…….which is probably at the root of most stopping problems……particularly where long ramps are involved. 

[user=449]brianpr1[/user] wrote:

John,
I think it's a matter of horses for courses:  Because my layout is an out and back I need my express locos to stop accurately at the uncoupler in platform 2 when going forward, and at five rear of train markers when going backwards. Nowhere else really matters and stopping position in the storage loops is not critical. It is easier for me to do the tweaking of brake comp at the actual markers.  Where you need to stop at markers around the layout it makes more sense to fix brake comp at a standard test position.

The key question is this: Do all your engines stop accurately at all your markers? That is what is giving me most trouble.  By standardizing brake ramp length and position and making endless test runs I can get closer than 1/4" to the reference marker and 1/2" to each other rear marker in platform 2 with all of my engines, but so far I have not checked what happens in platform 1.

I guess the honest answer to the key question is……most of the time:roll: but certainly sufficient to give me the confidence to keep plugging away

I meant to mention this before Brian……..most of my current TC development is very similar to yours……based on out and back. I am running 2 and 3 car sets from either the branch or the storage yard into 3 bays at the main station.

I have compromised in the sense that like you (?) I use different platform faces……..Bay 1 for the 2 car sets and Bay 2 for the 3 car sets.

I also use the same loco types, for the 2 car sets 0-6-0 T (57xx Panniers) and 2-6-2 T (45xx) for the GWR 3 car set.  One of the many challenges I have yet to face is dealing with the LMS component where I have 3 entirely different locos….Fairburn 2-6-4T, Ivatt 2-6-2T and Ivatt 2-6-0 Mogul which can run tender first………………

The other problem is the layout is still very much under construction so I havent run a train for the last 2 weeks as I have been focussing on track painting, ballasting and platform construction.I still have to run the autotrain,2 car and 3 car schedules all at the same time together……….I suspect you are way ahead of me there.

Finally I should perhaps have mentioned earlier the only really critical marker for me is stopping the incoming loco over the kadee magnet………I always draw the uncoupled loco 6" +- clear of the carriages…….. so as long as the new loco couples within 3-4" I am fine……I think varying the ramp as you and Dave have done may help here 

 Regards     

 

 

 

 

[user=434]John Dew[/user]

Finally I should perhaps have mentioned earlier the only really critical marker for me is stopping the incoming loco over the kadee magnet………I always draw the uncoupled loco 6" +- clear of the carriages…….. so as long as the new loco couples within 3-4" I am fine……I think varying the ramp as you and Dave have done may help here

That and increasing the length of the magnet area by using the Sol super magnet approach

Good Point Dave........I meant to mention that for greater certainty I was toying with the idea of doubling up on the magnet.

Up to now I haven't bothered entering detail in the "weight" section on the general tab of the engine setup.

I've not done this previously as the measure is in tons and I have assumed that the weight of the original would not have any meaningful relationship to the weight of the scale model. 

Anyone have any ideas as to how it works or the impact it may have on braking/acceleration?  

I'm just about to set up the speed profile for the "city of truro"so I'll look up the weight on the WWW and input it and see if I can see any difference between the locos performance with and without the weight measure.

OK first problem - is it the weight with a full load of coal?

I assume it is with a full load in the tender - According to Wiki its 55 tons or 92 with the tender ( so I'm assuming that an empty tender would not weigh 37 tons):???:

I guess.

I enter an approximation of the loco weight whenever I set a loco up but in all honesty I dont really know what effect it has

[user=434]John Dew[/user] wrote:

I guess.

I enter an approximation of the loco weight whenever I set a loco up but in all honesty I dont really know what effect it has

Hi John,

Have I got the time wrong or are you on the computer very early ( or very late)?

I'm still experimenting ….. had a few issues with setting the top speed in the decoder:oops:  The loco is fitted with a bachmann 21 pin decoder … and as I have discovered they don't program like most others the max setting for CV5 is 63 !!!   Needless to say when I set it first up I didn't read the table of CV values and set CV5 at 120 ( my starting default value as I never run locos at more than 70mph scale speed….  testing it for max speed the loco shot of the track and noise dived onto the floor !!!  Fortunately there was no damage.

Meanwhile , on more "weighty"matters  whilst twiddling my thumbs during the speed profiling process I came across this in the RR&Co forum

http://www.freiwald.com/forum/viewtopic.php?f=1&t=15971&p=98740&hilit=weight#p98740

Jurgen seems to be saying that the weight may be used in some future software release - I think :roll:

updated later ..  well as far as I can tell adding the weight details does nothing at all in regard to braking

Hi Dave

I think you are right…….I read that thread originally but the weight issue kind of passed me by……….

I put cork padded stops on both ends of my test track because no matter what I do the engine almost always overruns the run off sections.

I have had a couple of chips where I have had CV5 almost as low as that…….Lenz have virtually priced themselves out of the market for me………..which is a shame because I find them far more stable and easy to programme than any other. I am going to give Bachmann 2 function a go in my new Panniers

ps I think we are +- 18 hours behind you……………I wrote that email about 9.30pm Monday and this one at 7.30 am Tuesday

Kind Regards 

[user=434]John Dew[/user] wrote:

Hi Dave

I think you are right…….I read that thread originally but the weight issue kind of passed me by……….

I put cork padded stops on both ends of my test track because no matter what I do the engine almost always overruns the run off sections.

I have had a couple of chips where I have had CV5 almost as low as that…….Lenz have virtually priced themselves out of the market for me………..which is a shame because I find them far more stable and easy to programme than any other. I am going to give Bachmann 2 function a go in my new Panniers

ps I think we are +- 18 hours behind you……………I wrote that email about 9.30pm Monday and this one at 7.30 am Tuesday

Kind Regards 

John I thought you were going to have a play with DCC Concepts decoders? I have got one at the mo and it's performing quite well and it profiled nice as well.

I have made some progress with brake profiling but no two engines stop the same on the same stop markers. Mind you my uncoupling is unreliable  so I am going to give Dingham couplers ago operated by an electromagnet in the engine.

[user=450]wogga[/user] wrote:

John I thought you were going to have a play with DCC Concepts decoders? I have got one at the mo and it's performing quite well and it profiled nice as well.

I got a 5 pack with the stay alive function…….I fried one but that was my fault………they are very good …..2 locos that previously were right offs ( a 14xx and would you believe a Lima Railcar) now carry out simple scheduling duties.

I ordered another pack and didnt hear anything…..I posted an "anyone at home" here and Ron (Sol) kindly phoned them for me…….apparently Richard was away and his email had gone down…….I waited another week or so still didnt hear anything so I ordered the Bachmanns from Hattons……at a great price (12 GBP)………if they work that is :roll:……………..cancelled the DCC Concepts order and by return got a charming apology from Richard……….I still have locos to chip so I will try them again.

Cheers

[user=450]wogga[/user] wrote:

John I thought you were going to have a play with DCC Concepts decoders? I have got one at the mo and it's performing quite well and it profiled nice as well.

 

Pete,

When did you purchase the DCC decoder? I bought a pack of five with the saty alive feature and had a real problem trying to achieve a smooth profile - they all had a speed step that was well out of line with the others.  Somewhere I have a posting from another forum that came up with a work around but its currently in the two hard basket.

Strangely though they work fine if used manually ( although that could be because the locos I put them into never run at more than 40mph scale speed.:???:

Iansa had the problem with the mis-steps in the speed table.  He reset the BEMF and that fixed it.

[user=434]John Dew[/user] wrote:

I got a 5 pack with the stay alive function…….I fried one but that was my fault………they are very good …..2 locos that previously were right offs ( a 14xx and would you believe a Lima Railcar) now carry out simple scheduling duties.

 

John,  how did you get the stay alive capicitor into the 14xx.

I have found the problem with the "stay alive" in  steam locos is that the very locos you need it for ( small wheel base) are the ones where space is a premium.   That said where I have managed to fit them they work well ( apart from the profiling issue - which maybe confined to the first batch.

Re Bachmann decoders  ( accourding to railroad programmer they are ESU decoders)-  I have found that the ones they sell now work fine as long as you don't want to set up speed steps.  I have now profiled about 20 locos in RR&Co and have found the best results are obtained by setting CV2 and CV 5 , performing the speed profiling and then adjusting the acceleration and deceleration  CV's  before finally  finally adjusting the brake compensation back in RR&Co.