Atomic Systems

How are you supposed to use the dynamic brakes on the default (US) locos in Railworks? Applying dynamic braking similar to in MSTS does not give any results before you apply it 100%, and then you get insane wheelslip and a marked speed drop after only a few seconds. Just don't seem right to me... It is more effectively to fully apply airbrakes, and then after a while release them, and keep on doing so (something which in the real world would deplete the air supply after a few times), and that just does not seem realistic... Is there something I've overlooked here, or are the dynamic braking as it is now practically useless?

Move the lever a bit and then wait... and wait a bit, and a little more, and all of a sudden you'll get braking in the face. I may not know how US stuff is set up, but I'm really not convinced it's like *that*.

Yes there is a fair bit lag before the dynamic brake kicks in after applying it. Note as the train slows the dynamic brake gets less effective until it has no effect at all at low speeds, after which you need to apply the air brake to stop the train completely. If you are going down a gradient then only a small amount of dynamic brake is required to maintain a steady speed (when it eventually kicks in). More dynamic brake is required to be applied the slower you are moving - it does take some practice to know how much to apply for the speed you are travelling and the gradient your are decending. This is how it works in Railworks, does it mimic how a dynamic brake works in a real locomotive? Probably not but I think the basic fundamental operation is similar to real life.

As for the air brake running out of air, it depends on what type of air brake fitted. If it is a triple valve system (more modern) I think you are looking at it the wrong way round, air brakes use compressed air to charge the brake reservoirs of each wagon / carriage. If the main brake pipe loses pressure then the compressed air stored in each wagons reservoir is used to apply the wagon's brake. So if you ran out of compressed air the brakes would actually apply and you would not be able to release the train brakes until the brake pipe is charged with air again. I'm not an expert on these things so of I'm wrong someone will correct me! I think the main problem of using the air brake instead of the dynamic break is that you will wear out the brakes very quickly due to having them applied for long periods as you descend gradients. The dynamic brake uses the locomotives traction motors to slow down the train, so no brakes are being used.

There are three major types of air brake in use today:

Triple Valve - the oldest type still in use. This rapidly and fully releases the brakes if the pipe pressure rises only slightly. Used as a backup system on some old EMUs, and as a primary system on American freight trains. Because a very long brake pipe can take a long time to recharge after use, it is possible to "run out of air" on this type and be left with no braking ability for an extended period of time.

Distributor Valve - used throughout the UK and Europe as a primary brake for passenger and freight trains, probably also used in America for passenger trains. Brake cylinder pressure is proportional to the pipe reduction, regardless of whether applying or releasing. Because the brake pipe must be fully recharged to release the brake, it is not possible to "run out of air" in normal service.

Electro-Pneumatic Brake - offers very fast brake response, for both apply and release, and is therefore used on most recent DMU and EMU types. Old (pre-1970s) types were not fail-safe and thus required a conventional air brake as a backup. The HST (IC125) uses a variant of EP brake where the conventional air brake pipe is controlled by EP valves from both ends of the train simultaneously, doubling brake response speed compared to a conventional air brake.

This ignores ye olde vacuum brake and numerous minor variants of the above air brake types.

It does appear that dynamic brakes on large diesel locos have a certain amount of setup time. The typical instructions for a US freight loco are to advance the control to the Setup position, wait until the ammeter shows a reading (several seconds), then advance the control further until desired retardation is achieved. However, I do not know what the correct amount of braking for these is.

The dynamic brakes on the Class 86 are much easier to use, as they are integrated into the normal train braking system (an advantage not available with triple valves). There is still a lag of a couple of seconds until they activate, but this time is covered by the loco's tread brakes, which are also used if the dynamic brake fails for some other reason.

Many EMUs also have dynamic braking which has no lag. The PEP-derived units repurpose the notching resistors for the job, and it is easy to hear the traction motors working to retard the train, just the same as when accelerating. More modern units can actually regenerate into the rail or even the wire, using GTO or IGBT inverters (though this ability is often not turned on), but again with no lag and controlled by the ordinary brake handle. The Helsinki Metro units apparently regenerate into on-board batteries, which can also be used for emergency propulsion.

Hm, so the system has a configuration time? I guess it is somewhat "like that" then. I'm also guessing that like the powertrain with contactors that you can't go from zero control straight to some large setting and expect it to work. That's quite easy to set up, but does require a bit of scripting.

An issue with a heavy slow train going down a really long hill with the brakes on for a while would be heat buildup, rheostatic brakes are much better at dumping energy given they have actual radiators.

Hiya, looking at the blueprints, there is a programmable delay to the dynamics. You can set the delay to whatever time you want to.

As far as I'm aware the more realistic operation of dynamic brakes on a US-style diesel (we use some of the core technology as the basis for Australian locomotives that usually end up more efficient and capable than what they're based on!) is to move the dynamic braking handle from off to a "setup" position first. Then after a few seconds the dynamic brakes become available and the position can be varied at will.

Newer locos with more computer control technology are likely to be very different I'm guessing. With ECP starting to become a reality in Australia on some freight services there's probably some kind of automatic hill-descend mode available on the newest DownerEDI and UGL locomotives used for coal transport in the eastern states.

The HST braking - to me that sounds a little like a hybrid of ECP and the distributor system where you've got one distributor operating from each end of the train. Am I on the right track?

Trainguy76 wrote:Hiya, looking at the blueprints, there is a programmable delay to the dynamics. You can set the delay to whatever time you want to.

Yeah, but if you push the lever full open, after the delay you have 100% brakes. This isn't quite the same thing as waiting for bits of machinery to go "chunk".

HSTs have twin control air pipes and distributors on each vehicle; controlling each end just means the air pipe pressure equalizes faster.

There was a thread about the dynamic breaks on trainsim.com week or so ago, and I think there is some suspicion that the wrong units are being used again (like with the German electric locos). In the blueprint, there is a MaxForce entry for dynamic brakes and it seems the values entered are in the ballpark for the force in kN, but for the engine MaxForce I think it has been established that the value should be in klbf (though the developer docs say kN, presumably wrongly), so the max braking force from the dynamics is likely to be 4.4 time too large.

The force certainly seems to be far to strong as in the game you only need something 5% setting to hold speed down a steep hill, and get wheelslip above about 10 or 15%.

In addition I think (but I'm not totally sure) in real life there is a 'set up' position of the handle which doesn't apply the dynamic brakes but sets up ready for use without the delay. As far as I can tell there is no such position in RW - once the brakes are off you need to wait for the set up time again, and at any positive % position there is some retardation force - its a bit difficult to tell mind you as the whole useful range of the control is so small due to the MaxForce problem.

For a DE, maxforce in ( engine part of ) the blueprint seems to do nothing but change the scaling of the ammeter...

I think you can log acceleration in script, so it would be easy enough to set up a brake test & then see what values do actually make a difference.

Kariban wrote:For a DE, maxforce in ( engine part of ) the blueprint seems to do nothing but change the scaling of the ammeter...

I think you can log acceleration in script, so it would be easy enough to set up a brake test & then see what values do actually make a difference.

Here's the thread I was referring to: http://www.trainsim.com/vbts/showthread ... mic-Brakes. Unfortunately I haven't had time to tweak around myself, but the feedback there was that reducing MaxForce for the dynamic brakes did make them more realistic (though they were reducing by more than the factor of 4.4).

I forgot to say that MaxForce for engines is in lb/ft (whyyyy).

I think it requires some proper scientific method to be sure; there are tons of variables that look like they ought to do good things, and only end up changing something visual. Fortunately we can do that!

There is a BrakeEffortVsSpeed graph... presence of graphs in kuju physics usually complicates matters :S

This "lag" because the the dynamic braking, and resistor contactors and systems need time to switch to braking mode from traction mode. As i see US locos have a dynamic brake notch called "set-up" which should "tell" the loco to switch to braking mode, but don't brake, until the driver moves the handle further. Of course in rw this isn't works, the set-up position means a simple 10% dynamic braking(and it isn't a notch actually in game), but fortunately at least the "lag" is implemented in game.

I would not be surprised if MaxForce of dynamic brakes would be klbf instead of kN.

For the MaxForce of the whole vehicle, I found AI acceleration match that value, if interpreted as klbf (and Mass in tons). So why not could it be klbf? It would be consistent, after all. Not with documentation, but with the other usage in the same file.

All of the default locos really need going over to fix silly things like this. Even if there were no visual changes, it would make them behave better.