ABC Braking

[Ray H]

Our club is looking at implementing something akin to the Lenz ABC braking on our O gauge layout where we have some semi automated working.

 

Our initial tests using Zimo decoder equipped locos have had varied results. However, the following refers to tests conducted using the same loco over the same part of the layout and with varied values set in the relevant CVs.

 

Zimo decoders appear to be able to implement a time based stopping facility or a distance based stopping facility.

 

Experimentation with the distance based facility resulted in the loco travelling further to stop (beyond the entry onto the braking section) the slower it was travelling when entering said braking section.

 

The time based facility was similar. The faster the loco was travelling when entering the braking section, the further it travelled within that section. This is less surprising than the results based on distance. 

 

Both of the above were whilst actively (and manually) controlling the loco over the same distance prior to entering the braking section. Fortunately, our requirement means that any of the limited number of locos required to use the braking section(s) will have had the speed set initially, possibly several operating sessions previously and move around the layout as part of an event triggered sequence. On this basis we think/hope that we should be able to achieve a fairly consistent stopping position for each loco.

 

We had also hoped to add these braking sections to the terminal end of sidings and platform tracks. However, our experience from our tests suggest that a loco's stopping point cannot be accurately ascertained when the speed of the loco can be infinitely variable upon entering onto the braking section.

 

Can anyone else related their experiences when using the ABC braking system?

[RAF96]

I have tried the Lenz BM1 module (home made variant) with Hornby R8245 Sapphire decoders but these are time based ABC not distance and stopping is unreliable due to varying input velocity of any train.

 

I was under the impression the ABC capable Lenz decoders were distance stopped and reliably very accurate.

 

You would think it be simple math to break down speed into time and distance and adjust the decelleration rate to suit as usually this value (CV4) can be adjusted on the fly on the main.

Rob

[WIMorrison]

There is a good video here that might be useful

 

[Ray H]

Many thanks for the responses (and the link to the video).

 

We were getting fairly consistent results with the same loco at the same initial speed and with unchanged CV values. Our main problem was that we couldn't repeat that when we altered the speed at the point of entry onto the brake section.

 

I suspect that we could even get several locos to stop at the same place provided that the speed at the point of entry to the brake section remained unchanged on the loco concerned. We couldn't guarantee that the speed would remain unchanged if the loco speed was varied.

 

It is of course possible and may be very likely that we haven't set all the correct CVs that need to be set or we may have confused things by not resetting the CVs to the same base value that we were using at the start of the investigations.

[WIMorrison]

I find that resetting the decoder to default and then working backwards from there is the best way to cure inconsistent or unexplained effects. I know it has saved my frustration a few times going that way

[Nigelcliffe]

My experience of Zimo decoders and Asymmetric DCC Braking (ABC) is they will repeatedly stop in exactly the same place from very different speeds if the constant stopping distance elements are correctly set.  "Exactly" is to the same sleeper or better.     So, I'd be investigating why a different situation isn't achieving this accuracy.

 

One possible confounding factor may be if you are pulling a train with metal wheels.  Each time a metal wheel bridges the gap to the ABC section you reset the DCC to full power, and the braking starts again as the wheel moves over the gap.   There are ways around this, so the train is fully inside a section before the ABC signal is applied. 

 

- Nigel

Edited February 24, 2018 by Nigelcliffe

[Ray H]

It looks like we might be advised to run our tests on distance stopping again using a reset decoder. We changed from time to distance based stopping without resetting the previous values so that may be partly to blame.

 

We were finding variations of as much as six inches (O gauge) in the stopping position.

 

We might also be advised to conduct a separate range of tests using at least a couple of different engines in case that has any bearing on the results.

[Ray H]

We finally found time between converting the (rest of the) layout to DCC to undertake some further tests.

 

We've reset the loco that we'd used previously in case mixing the time & distance options had some impact. This time we concentrated on the distance option and experimented with two separate locos.

 

In both cases we set CV 27 = 1, CV 140 = 11 and CV 141 = 10. We varied the maximum speed we used and each time started the loco from the same place - give or take an inch.

 

We still found that the stopping place varied within an approximate 3 inch range depending on the speed set.

 

Fortunately the stopping position was all but the same each time the maximum used speed was the same.

 

We need the loco(s) to stop reliably at the same place as they are required to stop before a signal at danger and over a sensor that will trigger the movement of a following train.

 

The loco(s) will have the necessary CVs set and then more or less left to run on their own, stopping at any red signals they approach and restarting when the signal changes.

[Ray H]

We have had a few (unrelated) hiccoughs in the interim but have re-started our investigations on how to stop trains reliably, accurately and automatically.

 

Just to recap:-

 

Our circular layout includes a single direction Goods Line that duplicates part of one of the main running tracks. It has three sections that have a colour light signal and sensor at the far end of each section. The lead train is manually triggered to leave the Goods Line and once clear thereof it triggers each of the following two trains to move forward by a section. The new lead train moves forward and the loco eventually covers an LDR based sensor. Under DC operation this cut the power to the (lead) train whilst retaining power to the section it had just left and energising the section in its rear to allow the second train to move forward.

 

The section behind where the (new) second train was standing is now re-energised when the second train stops on its new section and this allows the original lead train to eventually join the rear of the queue as it leaves the main line.

 

We have tended to keep the same three locos on the three goods trains under dc control because each of them had a similar speed for the same controller setting - the Goods line had its own controller.

 

Conversion to DCC should allow us to simplify the operation of the Goods line provided we can get the ABC braking arrangement to reliably stop trains over the sensors that will allow the train behind to move forward.

 

Today's trials were with two non-sound fitted Zimo decoders - previous tests had been with sound fitted locos. Each of today's locos have significantly different motors and both had the Zimo "Active Brake" enabled lest they should ever be replaced on their respective Goods trains. The "Active Brake" was not enabled on the previous locos we used for our tests.

 

The best I could achieve today was a time based stop that depended on the loco's speed on entry to the ABC equipped track section. I did briefly but unsuccessfully try to get the distance based stopping to function.

 

Is distance based stopping compatible with the ABC brake and with the "Active Brake"? 

[Nigelcliffe]

 

...............

 

Today's trials were with two non-sound fitted Zimo decoders - previous tests had been with sound fitted locos. Each of today's locos have significantly different motors and both had the Zimo "Active Brake" enabled lest they should ever be replaced on their respective Goods trains. The "Active Brake" was not enabled on the previous locos we used for our tests.

 

The best I could achieve today was a time based stop that depended on the loco's speed on entry to the ABC equipped track section. I did briefly but unsuccessfully try to get the distance based stopping to function.

 

Is distance based stopping compatible with the ABC brake and with the "Active Brake"? 

 

Should be compatible. 

BUT, if anything connects the brake zone back to the mainline (a single metal wheel rolling over the join can do this), then the brake calculation is reset (loco saw full DCC signal, so braking is cancelled).    So, the "distance based stop" relies on nothing ever bridging the brake zone entry.  

 

Did your tests cover this possibility ?  

[Junctionmad]

I have read the manual and dont you need to calibrate the Zimo Decoders to achieve very accurate stops 

 

ABC is really a kludge 

[Ray H]

Should be compatible. 

BUT, if anything connects the brake zone back to the mainline (a single metal wheel rolling over the join can do this), then the brake calculation is reset (loco saw full DCC signal, so braking is cancelled).    So, the "distance based stop" relies on nothing ever bridging the brake zone entry.  

 

Did your tests cover this possibility ?  

 

All the tests were on a dead end siding. The home made BM1 is all that "joins" the left hand running rail between the "brake zone" and the rest of the layout. The only time we attempt to manually control the train is when we are reversing the loco out of the "brake zone" for the next test.

 

This does exclude the movement of the loco from the main part of the layout into the brake zone where some of the wheels cannot avoid being in contact with the main part of the layout and other wheels with the "brake zone".

 

The brake zone sections on the layout's Goods Line will extend for several inches beyond the signal that we require the train to stop at.

[Nigelcliffe]

All the tests were on a dead end siding. The home made BM1 is all that "joins" the left hand running rail between the "brake zone" and the rest of the layout. The only time we attempt to manually control the train is when we are reversing the loco out of the "brake zone" for the next test.

 

 

That should work. Suggests more faffing with the constant distance settings, because they should work.

 

When I've next got a decoder which has been updated to include the brake key on the bench, I'll do some testing myself.

 

 

This does exclude the movement of the loco from the main part of the layout into the brake zone where some of the wheels cannot avoid being in contact with the main part of the layout and other wheels with the "brake zone".

 

The brake zone sections on the layout's Goods Line will extend for several inches beyond the signal that we require the train to stop at.

I think the issue with the train wheels resetting the brake distance may be better solved by moving the entry to the brake zone back, but have a method to trigger its activation (such as an IR sensor which flips a relay to engage braking), or by a very short (couple of mm) gap in the rail to ensure bridging cannot happen (eg: file a dent, then fill with blob of araldite, allow to set properly, then gently sand flush to rail profile). But, either way, you need the basic constant distance to work before any tests on reliability with a loaded train can be made.

 

 

 

[Ray H]

Thanks Nigel.

 

There is a 2-3mm wide physical gap between the start of the brake zone and the end of the rest of the layout. I will make sure a wheel can't touch both rails simultaneously. 

 

I am slightly confused with a generality.

 

I (think that I) understand the concept of the ABC "system" in that you need a section of track - the brake zone - that is normally only linked t the "rest" of the layout's plain DCC network by the BM1 or equivalent. The train - with the loco's decoder appropriately set-up - will stop within the brake zone. The train will restart when the BM1 is over-ridden by making a direct link between the brake zone track and the rest of the layout (and the train won't stop if that over-ride is in-situ when the loco enters the brake zone). That's probably not the best way to describe it but I hope you get my drift.

 

I get the impression from the Zimo manual that there are ways of bringing the train to a stand within a pre-determined distance but it isn't clear whether that also requires a BM1 or some other means of triggering the Stop command. The decoder manual only references the CV27 setting(s) in association with the ABC "system".

[WIMorrison]

This page might give you some ideas

 

http://www.aandhmodels.co.uk/abc-braking---automation-without-a-computer-1242-c.asp

 

I am not sure that Zimo support what you are trying to do , I know it works with Lenz, but that won’t help you

Edited July 1, 2018 by WIMorrison

[Nigelcliffe]

Thanks Nigel.

 

There is a 2-3mm wide physical gap between the start of the brake zone and the end of the rest of the layout. I will make sure a wheel can't touch both rails simultaneously. 

 

I am slightly confused with a generality.

 

I (think that I) understand the concept of the ABC "system" in that you need a section of track - the brake zone - that is normally only linked t the "rest" of the layout's plain DCC network by the BM1 or equivalent. The train - with the loco's decoder appropriately set-up - will stop within the brake zone. The train will restart when the BM1 is over-ridden by making a direct link between the brake zone track and the rest of the layout (and the train won't stop if that over-ride is in-situ when the loco enters the brake zone). That's probably not the best way to describe it but I hope you get my drift.

 

That's pretty much it. 

The brake module puts a bias into the DCC signal, so the voltage seen on one-half of the DCC waveform is reduced by about 3v, but the other half of the waveform is normal (or reduced by 0.7v).    An ABC capable decoder can detect this bias in the DCC signal and do things with it.  Usually that "do things" is "slow down and stop, but if reversing behave normally".      

When the brake module is bypassed (either by a control switch, or a wheel bridging a rail gap), the DCC signal is normal full volts on both halves, so the ABC capable decoder just runs normally. 

 

I get the impression from the Zimo manual that there are ways of bringing the train to a stand within a pre-determined distance but it isn't clear whether that also requires a BM1 or some other means of triggering the Stop command. The decoder manual only references the CV27 setting(s) in association with the ABC "system".

 

Zimo's constant braking distance can be set to be triggered by external events, or to be the "cab braking"  (ie. throttle instructions will also have constant braking applied to them).   The latter is weird if you want to actually drive the trains by a human, but I can just about see the point if driven by an automated system. 

 

Zimo support at least three external event types for triggering braking:   brake on DC rather than DCC (also called Marklin braking),  brake using Asymmetric DCC (Lenz ABC), and Zimo's own HLU signalling.    The latter can get really expensive in specific Zimo hardware.   I think Zimo may also support the "half speed" trigger which is possible with one of the other Lenz BM modules (that alternates the DCC signal on one rail as "full - reduced - full - reduced").  

 

But, Zimo do NOT support the auto-reversing/shuttle feature found in Lenz (and thread implies also in Hornby Sapphire) decoders. (unless Zimo have put the auto-shuttle into a recent firmware update which I've not spotted).

 

Relevant Zimo CV's

CV27 - determines if ABC is on or not, and which directions of rail apply.  Can be values 0 (off),1 (Lenz forward convention),2 (reversed convention),3 (stop in both directions of travel).  Probably want 1 (or 2 if loco has its track wires crossed). 

CV134 - sensitivity of decoder to ABC signal (ie. if loco fails to stop, can be tweaked) - read manual on how to change it.

CV142 - higher speed additional sensitivity.

 

CV140 - selection of braking method. 0 (no braking),  1 (HLU or ABC will cause braking), 2 (cab = throttle brakes, probably not wanted), 3 (HLU/ABC and cab/throttle, probably not wanted).  Add 10 to these values if different deceleration slope method required.  (using the +10 will mean a slow loco will decelerate very gently to the stop point, whereas without the +10, the slow loco continues into the stop zone until it's calculation of deceleration following CV4 value will bring it to the correct stopping place). 

CV141 - the distance to travel to a stop. 

 

CV157 - allocating a Function Key to the "MAN" command.  Disable ABC (and HLU) braking of loco whilst the function key is on. 

 

 

When I've used it on Zimo, I have locos which stop at the same sleeper each time. 

 

 

 

- Nigel

[Junctionmad]

I did some similar tests a while ago and Im trying to recall exactly what we found 

 

Firstly decoders can take some time to reliably detect the Asymmetric DCC signal , and if the pickup is mediocre or the track is bad , Then some decoders can take 1-2 seconds, Zimo has a CV That controls the recognition time, but as you make the recognition faster , you introduce some uncertainty into exactly when the decoder will recognise the Asymmetric Signal. This uncertainty translates into " jitter " in the stopping distance and is worse as the loco is going faster , hence sometimes programming in a long recognition time is better.

 

Secondly the issue of exactly when the loco is " inside " the ABC section can be variable , depending on how many wheels are picking up and exactly how, you could find that the exact point the decoder decides to activate ABC is a variable event , This is not helped by the fact that  any full DCC that then gets transmitted ( because say that wheel suddenly conducts ) , then resets the ABC system . Again this results in a variable stopping distance 

 

In the second case ( i.e. pickup and shorting  issues ) The effect is more pronounced the slower the loco is going through the transition section 

 

The Zimo decoder is capable of millimetre stops , but all this assumes that everything else is perfect , which it rarely is 

 

This is why I said ABC is a kluge , The system gets even more kludgy when you add a train of metal wheels, DCC coach lighting , etc as all these can confuse the exact point at which the decoder implements ( or keeps implementing ) ABC 

 

 

The reality is the only way to reliably trigger ABC is to detect the train in the section absolutely reliably ( light beams , magnets , current sensing  , and then switch in the ABC signal ( in reality a few diodes ) , you may need to have very long ABC sections to do this reliably as in reality , the whole train needs to be inside the braking section , before  the ABC signal is triggered.  One way to do this is to have TOTI sensing ( train on track ) that then switches the whole section under the complete train to ABC.

 

We have abandoned ABC and broadcast Speed zero looks  like a better process because it cant be interrupted by poor DCC signalling 

but this method does not have any off the shelf products ( to my knowledge ) and is affected by decoder compatibility issues in some cases 

 

My own view is that if you want the simplicity of ABC, then you will have to except that you will get a degreee of unreliability in the exact stopping position due to a number of factors , not necessarily the decoder either . Anyway the prototype didnt stop in exactly the same place either !

 

The sad fact , is that " local Automation " , i.e. where you dont have a whole layout computer controlled , is very tricky to do with DCC , The system has huge flaws that were never addressed by NMRA ,particularly the issue of bidirectional communications.

 

by the way the problem is exacerbated as the models get larger and the wheelbase essentially grows , i.e. O Gauge 

 

dave

Edited July 1, 2018 by Junctionmad

[Ray H]

Thanks for the above.

 

I've managed to get a copy onto paper.

 

I'm not back at the club until later in the week and erroneously forgot to bring one of our home-made BM1s home but I can cobble something together so that I can experiment with settings on my OO layout beforehand.

[Crosland]

Thanks Nigel.

 

There is a 2-3mm wide physical gap between the start of the brake zone and the end of the rest of the layout. I will make sure a wheel can't touch both rails simultaneously. 

 

 

That's not enough if a wheel can drop into the gap. A smaller gap packed with plasticard filed to the rail profile is often recommended.

 

Andrew

[Junctionmad]

That's not enough if a wheel can drop into the gap. A smaller gap packed with plasticard filed to the rail profile is often recommended.

 

Andrew

 

and remember this is O Gauge !

[Ray H]

I believe that I understand the concept of a wheel being in contact with the two adjacent rail ends at the same time.

 

Is this any different from an item of motive power with pick-ups on several wheel sets having (front) wheel sets on the braking zone track and (rear) wheel sets on the rest of the layout power supply.

 

Presumably the brake zone activity only comes into play once all electrically connected wheel sets are within the brake zone.

 

On reflection this may cause us some fun (so to speak) because one of the locos likely to regularly run along the goods line only has pick-ups on the tender wheel sets.

[Crosland]

I believe that I understand the concept of a wheel being in contact with the two adjacent rail ends at the same time.

 

Is this any different from an item of motive power with pick-ups on several wheel sets having (front) wheel sets on the braking zone track and (rear) wheel sets on the rest of the layout power supply.

 

Presumably the brake zone activity only comes into play once all electrically connected wheel sets are within the brake zone.

 

On reflection this may cause us some fun (so to speak) because one of the locos likely to regularly run along the goods line only has pick-ups on the tender wheel sets.

 

You have the idea.

 

With a loco at the front you only need to worry about the length of the loco that includes pickups. A loco with only tender pickup is not really any different to a tender loco with all wheel pickup. Once the tender is in the section you are good.

 

It's more of a problem if the loco is propelling the train, the whole train has to be in the section before everything is isolated and the braking works as expected. The braking section will need to be longer or the braking more harsh.

 

There's also an issue if you have, say, carriage lighting, with pickup on more than one axle. Each such coach will bridge the gap.

 

Does ABC support slowing to a crawl and then stopping. The final stop can perhaps be more accurate.

[Ray H]

Luckily all trains will be pulled not pushed and all locos will run loco rather than tender first. However, we may use a Heljan Class 37 on one of the three regular train workings.

 

I did see a loco slowing to a crawl in our earlier tests but even then we couldn't repeatedly get an accurate stopping point even with the speed step the same when the loco being tested entered the brake zone. Admittedly, the inaccuracy wasn't too bad and bordered on the acceptable but we really need to get reliable results to get the (at signal) sensor covered without actually passing the signal. We'd sooner not have to re-position either the sensor or the signal.

 

The layout's Goods Line has been out of commission for a while whilst we've concentrated on re-wiring the rest of the layout and fitting decoders to the greater majority of locos. That point has now been reached which is why I've bumped the thread again and have done the more recent tests. I'm at the club tomorrow and will experiment with Nigel's suggested CVs and see how I get on.

[Nigelcliffe]

 

Does ABC support slowing to a crawl and then stopping. The final stop can perhaps be more accurate.

 

That's down to the decoder makers.   I know how Lenz and Zimo have their behaviour/algorithms, don't know about ESU. 

 

If constant distance is turned off, loco just decelerates following CV4 (and any additional CV's influencing deceleration) until loco comes to halt.  Clearly will stop at a different distance depending on speed of loco.

 

With constant distance turned on:

 

Lenz - decoder decelerates following CV4.  If the "constant distance" comes up before loco has reached "stop", loco performs an instant emergency stop.  If "constant distance" comes up well after loco will reach "stop", the loco will inch along the track at speed-step-1 until it gets to the distance. 

 

Zimo - two options:  (a) loco continues to run at current speed, until it calculates that from this speed, and following CV4 (etc), it will come to a stop at the correct distance.   (b) loco computes distance and speed, and comes up with a deceleration rate which will bring it to a stop at the correct distance.   In the latter, the rate of deceleration will be variable depending on loco speed on entering the stop zone, in the former case the loco will run past the start of the stop-section until it calculates the appropriate time to begin deceleration.   

 

 

 

Some decoders will support "slow approach" ABC signals.  The decoder will slow to its set "slow approach" speed on that signal.   But a device to create that DCC track signal is rather more than six diodes.  

[Ray H]

An update from today.

 

The Goods Line has been "converted" to DCC operation today. I have used a BM1 that was home-made by a colleague. I'm not 100% certain that it is either wired or installed correctly which is why I've brought one home to check over tomorrow morning (if I run out of time tonight).

 

All today's tests have been conducted with a single sound equipped  (ESU) decoder loco that has not been tweaked since its receipt save for changing the loco number from 3 to agree with that on the body!

 

The loco stopped dead each time it entered the Brake Zone and it restarted when the "BM1" was bypassed.

 

The Goods Line is sub divided into three sections with c/o relay contacts used to set each brake zone to either "brake" or "restart" and on that basis I have used a single BM1 with an input from the DCC bus and an output to the relevant switch contacts on each of the the three relays.

 

I'm currently minded to believe that the BM1 is working as designed and that the ESU decoder's default CV values (possibly tweaked by the sound file provider) is causing the loco to stop when it crosses into the brake zone. My reasoning is that either way round the BM1 shouldn't cause a loco to stop dead - in the correct direction it should either slow on continue at its current speed, whilst in the other direction the loco simply ignores the BM1.

 

Anyhow, I shall do some further testing tomorrow and report back in due course - that may be in a little while as I shan't have access to my computer for several days after Friday's club visit.

Edited July 4, 2018 by Ray H