Lego Train 9V Extreme - ready!

[Stefaneris]

I just added four pictures of trains I built with LDD to give you an idea. Since I startet building mostly with scale drawings and the right number of windows etc. I won't realise the trains like this (except for the FLIRT which I already started)

[Haddock51]

Thank you very much! These trains look great! I will get back to you with a separate PM.

Today will be a busy day. Swedish Television will visit me to make a news report about the layout and some trains!

Edited December 5, 2017 by Haddock51

[JopieK]

12 hours ago, Haddock51 said:

Now he has completely converted to 9V and abandonned PF.

When it comes to this kind of layout with gradients, there is no way that PF can cope with these extreme challenges, no matter what kind of train you select.

The main challenge for me with 9V was remote control of the trains. I experimented using DCC, but you need to destroy the motor more or less and it requires a lot of extra work. Apart from that, compared to model trains, LEGO trains are not heavy enough to maintain solid contact to the tracks. With PF at least you can run about eight trains at the same time out of the box. 9V + Mindstorms was a nice way to control at least some trains, but even that is limited. The combination of PF with e.g. Arduino makes it possible for 'more technically inclined AFOL's' to control trains without making too large modifications to the trains and without the need to destroy bricks etc.

[Haddock51]

8 hours ago, JopieK said:

The main challenge for me with 9V was remote control of the trains. I experimented using DCC, but you need to destroy the motor more or less and it requires a lot of extra work. Apart from that, compared to model trains, LEGO trains are not heavy enough to maintain solid contact to the tracks. With PF at least you can run about eight trains at the same time out of the box. 9V + Mindstorms was a nice way to control at least some trains, but even that is limited. The combination of PF with e.g. Arduino makes it possible for 'more technically inclined AFOL's' to control trains without making too large modifications to the trains and without the need to destroy bricks etc.

Interesting comments.

I remember several years back when I started planning this project, there were proposals to proceed with DCC. Given all the requirements and specifications, at the end of the day there was some consensus that DCC might not be the optimal solution in this case.

Referring to my previous discussion with stefaneris, your comment is a good example of different expectations and demand compared with mine. I understand that the option to run many trains simultanously with the help of remote control in your case has been highly prioritized. That obviously results in different solutions.

For me, remote control has not been a top priority in comparison to other important criterias. The most important issues to deal with have been power supply, power- and heat management/control and the possibility to run up to four trains simultaneously (for that you don't need RC on this layout). This is most likely the limit for how many trains you can run simultaneously on this layout because each train requires full attention all the time.

The possibility to run four trains (theoretically with ten engines each) at the same time required substantial modifications of the speed regulators, a stronger transformer - which I have mentioned before - and options to selectively disconnect some power contacts to the rails in order to segment the layout properly.

Other than that, everything else is standard. No modifications of motors or trains - with one exception: the ploughs on the train buffers for locomotives must be grinded in order to avoid contact with tracks when entering and leaving gradients.

I honestly don't understand your comment that Lego trains are not heavy enough to maintain solid contact to the tracks., an issue I have never experienced, not even on 8 percent grades. If you operate long and heavy trains with only a few engines, the 9V engine wheels might start slipping - and you eventually risk to get the engines destroyed. But that is something else.

PS: The only thing I have used from the PF system are lights and extension cables  (front- and tail lights as well as illumination of wagons on some trains: EN, IC and HE). These lights are connected with 9V batteries and switches which provide good and constant lighting. I was never happy with the 9V lights connected to the 9V engines. Why should lighting change when changing speed?

However, and at least for my taste, the PF light system is very clumsy and requires a lot of - not to say too much - space.

 

Edited December 5, 2017 by Haddock51

[zephyr1934]

On 12/4/2017 at 2:18 PM, Haddock51 said:

When it comes to this kind of layout with gradients, there is no way that PF can cope with these extreme challenges, no matter what kind of train you select.

9v is to die for, but with some of the third party devices that are now available I think it wouldn't be hard to tackle those grades with PF. The PF XL motor is far more powerful than any of the train motors. The hard part with PF is getting all of the motors on the same speed at the same time. Difficult to do with the standard IR, but the SBrick allows you to do that. So you could have distributed motors all synchronized to respond to the same speed command. But you would need a $50 SBrick and perhaps a $50 battery for each cluster of motors (I didn't say it would be cheap), and it would be a lot harder to hide than the low profile 9v motors. You could probably even get fancy with the user interface to have different sets of motors run at different speeds to get past a certain spot in the layout. In other words, never say never.

But none of that takes take anything away from the fact that you managed to do the impossible with 9v.

 

 

 

17 hours ago, Haddock51 said:

I honestly don't understand your comment that Lego trains are not heavy enough to maintain solid contact to the tracks., an issue I have never experienced, not even on 8 percent grades. If you operate long and heavy trains with only a few engines, the 9V engine wheels might start slipping - and you eventually risk to get the engines destroyed. But that is something else.

I think what he was getting at is that the lego trains are so light weight that if you are pulling a heavy train from the front (all motors in the locomotive) the cars can derail on the tight curves simply because the force pulls them laterally across the tracks more than longitudinally along the tracks. I've had that problem when attempting to pull more than 50 cars around standard lego curves. Given how weak the motors are and how much friction there is already in the lego bogies, you can't really add a lot of weight to the cars like you would in an HO layout. Initially I was surprised that you were not having that problem, but as you discussed earlier in this thread, you've avoided it by putting the motors under the cars as well as in the locomotive.

[Haddock51]

6 hours ago, zephyr1934 said:

. Initially I was surprised that you were not having that problem, but as you discussed earlier in this thread, you've avoided it by putting the motors under the cars as well as in the locomotive.

There are no motors in/on the Dm3.

My decision to mount your rods on all three cars of the Dm3 meant that I had to sacrifice the idea of having 9V engines on the locomotive.The same goes for the three Da locomotives on the timber train. 

This was the first time I surrendered when it comes to one of my favourite principles - functionality always precedes design... Despite the problems that this decision lead to, I never repented it. The Dm3 is indeed a jewel in my train collection.

I was obviously aware of the fact that this would result in a tremendous breakpad (more than 1 kg) in the very front of this train, but after many hours of testing, I finally managed to solve this issue.

PS:  Yesterday, Swedish Telivision visited me to make a reportage about my layout! It was sent the same evening on local news at 6.30 pm - best prime time! (unfortunately without english subtitles...)

Edited December 6, 2017 by Haddock51

[Haddock51]

8 hours ago, zephyr1934 said:

I think what he was getting at is that the lego trains are so light weight that if you are pulling a heavy train from the front (all motors in the locomotive) the cars can derail on the tight curves simply because the force pulls them laterally across the tracks more than longitudinally along the tracks.

Sometimes, I am still experiencing derailing issues with the EN - another heavy breakpad - and its tender . Even though it carries a 9V battery, it apparently is still too light and often pushed off the track when passing points and/or s-curves uphill, something that I unfortunately can't do much about it. Again, the trick is to find and apply the appropriate speed when passing these critical parts of the layout.

Edited December 6, 2017 by Haddock51

[Haddock51]

19 hours ago, zephyr1934 said:

 The PF XL motor is far more powerful than any of the train motors.

From what I have seen myself when watching trains/locomotives with PF motors, I am still convinced that one of the main disadvantages with PF compared with 9V are the inbuilt mechanical frictions with all these parts to transmit that engine power to the wheels. Would be interesting to know the real efficiency of PF engines after discounting for these frictions. And then you have the constant issue of decreasing battery load ....

To my opinion, the key and remaining advantage with 9V engines is the fact that power supply is stable and engine power is transmitted directly to the engine-wheels.

I would simply hate the idea to watch my trains stop somewhere halfway up to the top because of power  shortage....

Speaking about long trains in grades, I don't believe in the idea of mounting a super strong motor in the very front of such trains. As you wrote yourself, you need to spread engine power. And for that there is no need for engines like a PF XL.

And somewhere along the line, you start to ask yourself if you really want to proceed with all the necessary modifications to meet your requirements ...

PS:  Maybe I am wrong, but then somebody has to prove  that the combination of long trains and PF - in a track environment with 8 percent grades - really works. Meantime, my impression is that this entire discussion about PF - including third party devices - in this kind of environment is highly speculative, not to say just an illusion.

 

Edited December 6, 2017 by Haddock51

[Haddock51]

12 hours ago, zephyr1934 said:

You could probably even get fancy with the user interface to have different sets of motors run at different speeds to get past a certain spot in the layout. In other words, never say never.

Maybe my own imagination is not big enough.for this fancy idea.

To me, the concept of running different sets of motors at different speeds with a train like the iron ore train is wishful thinking.

Referring to my scheme with the red marked problem sections, I just can't imagine how you can manage to change speed of 13 motors - probably within seconds - when passing these difficult sections.

Yes I know, never say never ...

Edited December 6, 2017 by Haddock51

[Edde]

It would be interesting to make a similar layout as this but with 12V, and test the possible gradients.

[Haddock51]

2 hours ago, Edde said:

It would be interesting to make a similar layout as this but with 12V, and test the possible gradients.

Unfortunately, I don't know anything about 12V. Hopefully somebody else with good knowledge about and experience with this train product line can give you more advice/help.

W.r.t. testing, in this video you can see - and hopefully get some inspiration from - my test layout with a climbing spiral and test gradients back in 2014:

 

 

Good luck!

Edited December 6, 2017 by Haddock51

[Steamdemon]

Guessing wheel slip was an issue during the early testing stages.

[Haddock51]

14 hours ago, Steamdemon said:

Guessing wheel slip was an issue during the early testing stages.

Yes it was, mainly due to the lack and distribution of engine power. I had calculated that I needed 8 percent gradients to get the trains from floor to ceiling within reasonable space. So the grades were a given from the very beginning.

Edited December 7, 2017 by Haddock51

[Haddock51]

22 hours ago, Edde said:

It would be interesting to make a similar layout as this but with 12V, and test the possible gradients.

Coming back to testing grades, let me share with you the "hump" approach (this might not be applicable to 12V).

When building layouts with gradients - and particularly with 8 percent gradients - you are faced with a problem that eventually can cause derailing: a sharp edge between the upper part of the gradient and the flat level..

In order to "disarm" this potential derailing pitfall, I have built humps, resulting in three minor track edges instead of one big at each edge:

 

As you can see on this picture, I have put 1x2 plates under the linkage of the first and second straight on each side of the edge. In addition, I used a grinder to grind all three track edges.

With this "hump", you get a rather smooth transition fram gradient to flat level (and vice versa), something you can also observe in the iron ore train video.

This also provides safer "landing zones" for 3 wheel constructions, e.g. on EN, Dm3 and Da locomotives.

Edited December 7, 2017 by Haddock51

[baard]

On ‎01‎.‎12‎.‎2017 at 5:22 PM, Haddock51 said:

Very few indeed. The last weekend in October, we spent ten hours on filming and taking pictures. Prior to that weekend, I had spent many, many hours on testing - and modifying - the iron ore train, primarily speed,  number and sequence of engines (a reduction of length and/or weight has never been on the agenda).

The room dimensions are 5.5 x 7.3 m. And approx. 20 square meters are still available ....

Envious...., hmm, must clear my Lego room from all the boxes, but then I have to clear the attic first ...... etc. etc.

[Edde]

On 12/7/2017 at 6:17 PM, Haddock51 said:

Coming back to testing grades, let me share with you the "hump" approach (this might not be applicable to 12V).

When building layouts with gradients - and particularly with 8 percent gradients - you are faced with a problem that eventually can cause derailing: a sharp edge between the upper part of the gradient and the flat level..

In order to "disarm" this potential derailing pitfall, I have built humps, resulting in three minor track edges instead of one big at each edge:

 

As you can see on this picture, I have put 1x2 plates under the linkage of the first and second straight on each side of the edge. In addition, I used a grinder to grind all three track edges.

With this "hump", you get a rather smooth transition fram gradient to flat level (and vice versa), something you can also observe in the iron ore train video.

This also provides safer "landing zones" for 3 wheel constructions, e.g. on EN, Dm3 and Da locomotives.

Good solution. I've thought about the possible "hump" than can occur when making slope before. I'll look into making simple gradient tests today for 12V. If I do any special findings I might make a new thread about it. From my own experience, the system is very strong and powerful, so it might be possible to achieve gradients higher than the one's you've managed for 9V.

[Toastie]

When going to the extreme - and I have the impression this thread id fostering new powerful and current hungry solutions - you should also pay careful attention to the power delivery system for the tracks. As @Haddock51 has pointed out before, he distributes the power feeds very carefully and rather densely over the entire layout. Switch points as part of the entire circuit need particular care; I believe it is advisable to mount pairs power feeds not that far way from the switches in the main power consuming directions. The internal mechanism is sometimes a litte oxidized and this constitutes rapidly a "hot spot" as the heat production roughly scales with the square of the current. I have melted down a switch point upon accidentally producing a short. The LEGO regulators take (partly) care of that but when you go to custom power supplies (my layout is permanently powered by a 12 V / 10 A supply as I use 9V train motors for both, as power pickups and motor, which is operated by PF or RCX) for operation at high amperage, you should address this issue carefully. The same holds true for the original train track power feeds; when these are not having good contact (low resistance) to the metal rails, same thing may happen (and has happened to me - the feeds may actually deform to an extent that they don't have any contact at all to the track). If you don't mind using a soldering iron and don't have that many original power feeds at your disposal, I'd solder a custom power wire directly to the metal rail (on the outside). Works nicely, safes a lot of money (and trouble ...).

Best whishes

Thorsten      

[Haddock51]

3 hours ago, Edde said:

From my own experience, the system is very strong and powerful, so it might be possible to achieve gradients higher than the one's you've managed for 9V.

Testing gradients back in 2013/14 showed that 9V engines on short and light trains could manage grades of 10 percent or even somewhat more.

For me it was important to find the right gradient so that all trains - without exceptions - could make it through the planned layout. Back in those days when I used to run trains with up to ten engines on the test gradients, there were problems with wheel slipping with 10 percent grades. Therefore, I decided to go for 8 procent to be on the safe side.

Edited December 9, 2017 by Haddock51

[Edde]

14 minutes ago, Haddock51 said:

Testing gradients back in 2013/14 showed that 9V engines on short and light trains could manage grades of 10 percent or even somewhat more.

For me it was important to find the right gradient so that all trains - without exceptions - could make it through the planned layout. Back in those days when I used to run trains with up to ten engines on the test gradients, there were problems with wheel slipping with 10 percent grades. Therefore, I decided to go for 8 procent to be on the safe side.

I see, my problem lies in the lack of 12V wagons being constructed atm and the lack of space. Hopefully I can test a little during the holidays.

[Edde]

12V test 

I did a very simple and crude test with a 7760 locomotive. The end section raises by about 5 bricks per section of track. I think 7 could be possible, but I don't have enough bricks lying around. In the future I'll try to test with a plank and get a longer track. The only non-lego parts are the rubber tires, however they should be identical to lego's ones. It can go a little faster as I didn't have to turn the dial the whole way (which is why I believe steeper slopes are possible) but I didn't want to risk it falling down at the end. Of course, this is just a small train without any wagons, but it's something. 

Edit: assuming the slope is just a triangle with the hypotenuse of 16 studs and height of 5 bricks (6 studs), it would be a 40% gradient.

Edited December 9, 2017 by Edde
Additional information.

[Haddock51]

1 hour ago, Edde said:

12V test 

I did a very simple and crude test with a 7760 locomotive. The end section raises by about 5 bricks per section of track. I think 7 could be possible, but I don't have enough bricks lying around. In the future I'll try to test with a plank and get a longer track. The only non-lego parts are the rubber tires, however they should be identical to lego's ones. It can go a little faster as I didn't have to turn the dial the whole way (which is why I believe steeper slopes are possible) but I didn't want to risk it falling down at the end. Of course, this is just a small train without any wagons, but it's something. 

Edit: assuming the slope is just a triangle with the hypotenuse of 16 studs and height of 5 bricks (6 studs), it would be a 40% gradient.

That's very steep indeed! Amazing!

Next time, you have to show us a video with locomotive and some (heavy) wagons. That might be a different story ...

Edited December 9, 2017 by Haddock51

[Edde]

Just now, Haddock51 said:

That's very steep indeed! Amazing!

Next time, you have to show us a video with locomotive and some (heavy) wagons. That might be a different story ...

Of course. The wagons for the 12V system has more friction than the 9V system, so the difference between a single locomotive and a train should be bigger than that of 9V. However as i said I'd want to test that with a plank and do a longer slope.

[Haddock51]

Just now, Edde said:

Of course. The wagons for the 12V system has more friction than the 9V system, so the difference between a single locomotive and a train should be bigger than that of 9V. However as i said I'd want to test that with a plank and do a longer slope.

This is of course a balance between space and grades. The more space you have, the longer ramps/planks and the lower grades you can use. In addition, you have to decide how high up you ultimately want to go.

[Edde]

Just now, Haddock51 said:

This is of course a balance between space and grades. The more space you have, the longer ramps/planks and the lower grades you can use. In addition, you have to decide how high up you ultimately want to go.

Yup, you're lucky to be able to use a whole room for your trains, my house is filled with lego basically everywhere.

[Haddock51]

7 hours ago, Toastie said:

When going to the extreme - and I have the impression this thread id fostering new powerful and current hungry solutions - you should also pay careful attention to the power delivery system for the tracks. As @Haddock51 has pointed out before, he distributes the power feeds very carefully and rather densely over the entire layout. Switch points as part of the entire circuit need particular care; I believe it is advisable to mount pairs power feeds not that far way from the switches in the main power consuming directions. The internal mechanism is sometimes a litte oxidized and this constitutes rapidly a "hot spot" as the heat production roughly scales with the square of the current. I have melted down a switch point upon accidentally producing a short. The LEGO regulators take (partly) care of that but when you go to custom power supplies (my layout is permanently powered by a 12 V / 10 A supply as I use 9V train motors for both, as power pickups and motor, which is operated by PF or RCX) for operation at high amperage, you should address this issue carefully. The same holds true for the original train track power feeds; when these are not having good contact (low resistance) to the metal rails, same thing may happen (and has happened to me - the feeds may actually deform to an extent that they don't have any contact at all to the track). If you don't mind using a soldering iron and don't have that many original power feeds at your disposal, I'd solder a custom power wire directly to the metal rail (on the outside). Works nicely, safes a lot of money (and trouble ...).

Best whishes

Thorsten      

Interesting comments.

I basically agree with you - with the exception of soldering directly to the metal rail, something I commented in one of my previous replies. However this is a separate discussion.

I guess this is the right time for bringing up a mysterious experience that I still fail to understand and find a logical explanation to.

As Toastie correctly pointed out, the distribution of power feeds is very important. There are indeed substantial losses of power where you have sections with several points in a row.

Being aware of this problem, I therefore decided to mount power feeds on each of the four yards and main line on level 50 according to the following scheme:

The idea was to assure appropriate power supply and to selectively turn on/turn off each of these power feeds with the help of switches on the box to the right of the speed controllers (in combination with adequate points settings).

However, this did not work (all power connections all the way to the cable terminal have been carefully checked).

And this is what happened: even with turned off power feed - and closed switches for the relevant yard - neighbouring yard(s) would still be on power! (The same mystery occured on the siding on level 175 with an additional power feed halfway into the siding).

After several hours of testing, I ended up with all these extra power feeds being taken away. Now, everything works properly and power on yards and siding is shut down with the help of points. However, as a consequence, sensible power losses occur, particularly on yards # 4 and # 5.

Can somebody explain this mystery?

 

Edited December 9, 2017 by Haddock51