Dodgy physics

[MikeK]

I think we've also had a dialog about stretching and bunching. When this speed increase happened, it was, BAM, instantanious. From 40 to 50 in a split second. And would setting the dynamic to just 20 do that much braking? Throttle was at idle going downhill. Like I said I'm gonna run it again. No independent, just train brakes at first, then when they say test the dynamic, I'll take it slow.

That sounds exactly like what I described in the first post. I coupled to some stationary cars at 1 mph, and the next thing I know I am accelerating past 10 mph.

[MikeK]

No idea if the independent brake is propogated ( it certainly wouldn't be for a UK diesel, so I have doubts ). Might be able to script that also.

Someone please correct me if I am wrong, but I was under the impression that any changes to the independent brake would only be propagated to the locos that are part of the set at the head end. In other words, it doesn't apply to distributed power.

The reason was that changes to the independent brake setting happen via one of the 3 outer hoses that are connected together then multiple locos are coupled to each other.

So for distributed power in the rest of the train, they get throttle, train brake and dynamic brake settings only, and the engineer has the option of making them mirror the lead locos, or controlling them independently of the lead locos.

For all the locos in the head end set, they mirror all the lead loco settings, including throttle, train brake, independent brake and dynamic brake.

[TrainMaster1]

From Arizona Chris post:

I think we've also had a dialog about stretching and bunching. When this speed increase happened, it was, BAM, instantanious. From 40 to 50 in a split second. And would setting the dynamic to just 20 do that much braking? Throttle was at idle going downhill. Like I said I'm gonna run it again. No independent, just train brakes at first, then when they say test the dynamic, I'll take it slow.

Your goal on a down hill is to keep the train bunched and not running in and out like an accordion. Good way to derail and/or break a knuckle. You do that by controlling the speed with automatic application (never the independent) and dynamics.

What I am confused about is setting dynamic to 20? Dynamic on most locos has eight notches because it mirrors the power (throttle) handle. Dynamic 2 is 25% of full force of whatever this particular loco delivers. Each notch deliver 12 1/2% of the dynamic force available.

The throttle has to be set to zero as no engine on the planet can deliver both power and dynamics from the same traction motor. Throttle to idle (making sure your amp meter is at dead zero or your would burn the motors into paperweights), then into dynamic set up. This basically changes the traction motors into resistors to create drag and slow the train. Early locos had a selector button and a single handle to go from power to DB. After the mid 1960s or so there are two separate handles on a control stand-one for power and one for DB. The power handle must be at idle and amp meter at zero that is when set up happens. Then notch out your DB to control the train. Personally once I set air on a downgrade it stays on until either I am off the grade or I know I have a long enough stretch before the next downgrade.

If not I will cut the dynamic back to zero , leave the air on and pull against the application for the short distance to keep the train moving until the next downgrade. Then throttle goes off and back to DB for the next section of the hill.

The best way to run downgrade when you have DPU or helpers is called running the picket fence. No sim however has this capability yet. This is where the helper engines say may be in Run 6 while the leaders are in Run 4 to keep the train bunched. The two crews or DPU controlled from the leader can apply power and brakes to keep the train under control far better than the same number of engines on the head end. Same thing for going up grade. Trailing unit bunches the train because you are pulling only one car as opposed to the whole train when you start...leaders start to notch out and get the first few cars on the roll before you even think of releasing the air. You want that forward momentum or you get to see the world in reverse.

We had that happen in a operating session on the SP Cascades where dispatcher stopped a crew in siding at Oakridge. Upon getting clearance, engineer did not bunch train, lost adhesion and panicked releasing the air as well. Train ran beautifully in reverse until grade flattened and air was restored. Thankfully no trains were behind him for a few miles and he recovered.

On BNSF Scenic Sub crew went hurtling backwards at 50 mph plus and derailed at bottom of hill at CP Scenic East also trapping another train in the siding there. That would happen exactly that way in 1:1 scale too. Nice to see that the sim in that case had it right.

I am enjoying the daylights out of this thread. (In case you had not noticed.)

Nick

[Kali]

The best way to run downgrade when you have DPU or helpers is called running the picket fence. No sim however has this capability yet. This is where the helper engines say may be in Run 6 while the leaders are in Run 4 to keep the train bunched. The two crews or DPU controlled from the leader can apply power and brakes to keep the train under control far better than the same number of engines on the head end. Same thing for going up grade. Trailing unit bunches the train because you are pulling only one car as opposed to the whole train when you start...leaders start to notch out and get the first few cars on the roll before you even think of releasing the air. You want that forward momentum or you get to see the world in reverse.
Nick

What's interesting about apparently being able to change the dynamic brakes in script is... if I had enough buttons and levers in the head unit I could give you independent control over every engine's dynamic brakes, although that of course needs all the stock to be passing messages around properly as we don't have radio control in the sim. Obviously the entire consist behaving as one means I can't do anything about the auto brake. If I could work out how to gain control over rear engine throttles then obviously the same thing is possible for power, but at present I have no idea how the game is doing that and if it's controllable by scripting. Someone would need to build extra cab bits too obviously.

Mike, I meant "no idea if it's propogated in the sim", and it isn't. I let a consist run away with just the lead engine's brakes on and then tried scripting it like the dynamics, and it ran away just as fast.

[GaryG]

Visualize going downgrade and you start by applying the trainbrakes minimum. You end up with braking applied on locos as well as cars. With loco brakes more efficient, you should bunch slack at the front end somewhat, no problem so far. Too bad we don't have bailoff so we could keep the locos rolling freely and keep the slack stretched.

Now you decide to start up the Dynamics. As soon as you do that, the loco brakes are released and the locos will roll freely for the ten seconds allowed for the dynamic setup. What happens now? You had bunched (compressed) slack which will now start to be pulled out by the free rolling locos so speed should increase until the DB's start up. If you let them start with a high setting you're going to slam all the released slack back in again so again we have an almost instantaneous speed change again along with another kick in the pants once it bunches again.

I would think that a smoother operation would be to start at the top of a grade with DB's at minimum then a minimum brake application whitch should let most of the train stay stretched. Now, as needed, add DB's until the maximum recommended setting for the grade then if more braking is needed, slowly drop the trainline pressure to keep the speed checked. This method, I assume would prevent hard slack runin and runout. Slack will still runin as the DB setting is increased but it should occur slowly and eventually over the length of the train.

Of course, I have no actual experience but I feel this method should give a smooth ride for you and the train.

GaryG

[arizonachris]

OK, this Thread is getting very interesting. You're right, Nick. I ran several different scenarios to test brakes, along with the one I ran last night. What I found is that locomotives vary greatly. The SD70ACe is so far removed from the F7 it's not funny. SD40-2 is way different, too. But I'm getting the hang of it. Train brakes to just where cylinder pressure applies, keep the pipe pressure up as much as you can, increase or decrease throttle to stay within speed. But, that keeps the consist stretched, yes? And that's not what we want? I would think that would be better than having it all bunched against the lead engine, as any helpers don't help at all with braking that I can see.

I reran that brake test scenario, kinda boring but still gave me some insight. Was running perfect. Writing down % of brake handle vs. pipe PSI vs. cylinder PSI. LOL, "DERAIL" WTF? Why? OK, start over. Running fine. 25 ~ 37% throttle once up to speed, 31% on the train brakes. That's 31% graduated self lap limited. 83PSI in the pipe, 61PSI on the cylinders. Pretty much didn't touch the brakes, just modulated the throttle all the way down.

Now, I realized too late I was gonna overrun the ending point. 100% train brakes, zero throttle, hit the Dynamic, I watched the second engine bunch right into the lead and "DERAIL" . Wasn't paying attention to my F3 HUD and distance to the next destination. Pfffttttt. Third time's a charm, right?

The F7 seems so different, if you give it 6% train brakes, it's the same pipe and cylinder pressure as 50%! Something's not right there. And the SD70ACe has quite a few bugs itself. Anyways. Time for a snack, then maybe another scenario or two!

[Kali]

At present in game, I don't think the engine's brakes *are* better than the stock - at least it depends on the stock. I'm pretty sure DB don't bail any brakes off the engine either in-game. I'm in the wrong place to check atm, but the last car I looked at weighed 91tonnes ( was set to permanently full ) and had a brakeforce of 70% of that, and the latest SD40 uses 59% of it's weight for brakeforce. I don't remember what a SD40 weighs, but the end result is not far enough ahead of the car. I have tended to try and bunch the consist up initially with the independent brake and then the dynamics, and finally a bit of autobrake if it feels like it's running away. Problem is the autobrake tends to stretch the rear back out somewhat so getting off it again is a bit fraught.

I would set a car or two up right if I could get some idea of how much they really have - I appear to have lost the test report I had, annoyingly.

Chris would you try the same thing with the SD40 I posted a couple back? the brakes on that are rather slower ( a lot slower ), but the default pressures are ok and you can actually get a decent minimum application. And the dynamics work properly.

[arizonachris]

Chris would you try the same thing with the SD40 I posted a couple back? the brakes on that are rather slower ( a lot slower ), but the default pressures are ok and you can actually get a decent minimum application. And the dynamics work properly.

I'm not the sharpest tool in the shed as far as any kind of editing. As long as you tell me where and how to install it, I'd be glad to test it out. Also what you are looking for, and what route.

[Kali]

It's a zip file with a RWP, just install it like normal and you'll have an extra black SD40. Just try that ( or 3 of them or whatever ) in the same scenario to see how you get on compared to the others.

[Dan]

I don't know if it would be possible but it would be nice if someone were able to write some prototypical driver/engineer training scenarios, with appropriate messages about the correct way to drive.

[Kali]

Should be possible, the HSC scenarios walk you through doing various evolutions.

Chris the zip was in a post near the bottom of the previous page in the thread btw.

[arizonachris]

Yeah, got it, Kali. Won't have time until later after breakfast and NASCAR.

[TrainMaster1]

Okay let's clear up a couple of items.

#1 In the real world Independent brakes only apply on the first locomotive. Some games may have figured out how to get them to go to all the locos. That is not prototype. So if you want to run right...first loco only.

#2 You also should know what a prototype consist looks like before you create scenario or try to run a train in free roam. All to often I have seen guys try to run non-prototypical on weight and length. There is a reason why 1:1 has rules for TOB (Tons per Operating Brake), length restrictions, weight restrictions, power axle restrictions and mnay more. If there is nothing good on TV sometime (which is pretty much always), pick up any roads Special Systems Instructions and get ready 8-10 hours of uninterrupted sleep. Sometimes it seems like restrictions have restrictions. You do not need to all of them just the biggies (weight, length, grade rating by horsepower are three to know)

#3 Many roads require you to stop and do a brake test at the top of a grade. We do this in our operating sessions when required by rule book. This allows you to be in control and set up your run to the bottom properly. On a downhill, set and hold a minimum application. Grabbing more air is playing a losing game. You will almost always grab too much and then it becomes a release....no wait...apply....and before you know it you have zero in the reservoir and you are regretting not listening to mom and wearing clean underwear for when they find you at the bottom of the hill. If a minimum application and full DB still cannot hold your train back this is because your consist and train are either not prototypical (i.e too heavy and long) or you have made a release somewhere you should not have along the way, or the equipment you are using is faulty. I have seen that more than once where we have been asked to BETA test items long before they see the light of day. Some guys get it right out of the box and others are completely clueless. My particular favorite item was an Alco C-628 with a desk top controller! Just proving that you should always wear a helmet when riding your bike....

#4 Make friends with your Dynamics. They work when used in conjunction with a solid minimum reduction. If you have done your brake test at the top of the hill you train is now under control and you can get underway. You should be in DB set up once the train starts moving. As you get into the 8-12 mph range make your minimum reduction and leave it alone until either the grades really ends or you know the route well enough to know you have a nice long flat area between downgrade #1 and #2 so you can have full air in the reservoir and you have time to get into set-up.

If you have a short stretch of flat ground before the second grade starts, hold the minimum reduction, leave DB and return to the power handle. You can safely pull against an application for short stretches. Once the next down grade starts get back into set up and starts notching out your DB again. If you hold you application it should allow the train to run in naturally behind you. Yes you will get smacked in the butt just like in the real world but that is controllable if your know it is coming. Do not panic at this point. Let the train settle which it should unless again your train would never see the light of day on this line or the equipment comes with lousy physics.

That should give you a realistic ride down the hill and get you home in one piece. So remember Independent is always on the first loco only and does not go to other engines in the consist. A minimum application held down the hill minimizes slack action and should be held until you are clear (last car) of the major grade area. DB can be adjusted from set up to 100% as needed and in the real world works off of an 8 notch handle in all but desk top engines which have a 0-100 graduated and combined handle. Also before you apply DB, when you come off of power...your amp meter should read zero. Anything else and you are frying the motors if you try and grab DB at that point. The set-up is so that the motors can transition from supply power to preventing it by going into resistor mode.

Hope that helps get you home in one piece and enjoying your ride a lot more...

This is just one many new things you will learn if you get involved in our upcoming RW operating sessions. You will learn a ton of things that will make increase your enjoyment big time. If you are interested in learning more about prototype operations come listen our next session on 10/29 and then sign up for our Intro Events where we cover everything you need to know to run according to the book. Send me a PM and I will get you set up (there's that phrase again) with the info you need.

Nick

[Chacal]

The F7 seems so different, if you give it 6% train brakes, it's the same pipe and cylinder pressure as 50%! Something's not right there.

The F7 uses 24RL brake equipment which is NOT self-lapping.
It behaves very differently from the newer 26L equipment.
This might explain the difference, if the sim models the brakes correctly.

You have to make a 6% reduction, then move the handle to "lap", and then it will stay at 6%.
Then if you want a 50% reduction, you apply more brakes until you get 50%, then move the handle to "lap" again.

[MikeK]

I am enjoying the daylights out of this thread. (In case you had not noticed.)

It's good that so much prototype information is surfacing, but honestly if this thread has taught me anything it is how truly unrealistic RW is underneath the pretty graphics.

Do the RW team really not know how bad the sim is for modelling long north american trains, or do they just not care? I find it incredible that can say the "simulation" is more realistic now that they have some clouds and splashes of water when the brakes don't even work properly.

The team working on this product is clearly very, very talented and capable of a lot of incredible work, they just don't seem to have the same priorities as we do.

When I look at the "F5" debug values in OpenRails, it is amazing how much closer to reality that sim is. If only we could take the physics from that game and transplant it into RW then we would be getting somewhere!