Heavy Us Freight Downhill Braking

Hi, advise please the correct method of heavy freight downhill braking techinque.
I usually start with dynamic, but after a while if the train is really heavy (loaded coal wagons) the train will accelerate. Then I usually start to add a bit of auto brake (initial reduction - 20%), then the train slows down quickly. So I play with the two brake types all the time what I belive is not really realistic.

 

And do not forget to bail off when using dynamic brakes. BC has to be zero.

 

Long story short - if dynamics up to around B4 are enough then you can just stick with them (some railroads may have rules against this though). Otherwise, apply a minimum application with the auto brake and use the dynamics to fine-tune (while making sure to bail off the loco brakes). If you still need more than half of your dynamics, apply more air brakes and repeat.

 

Thank you Lamplight, always like to learn.
Tomorrow i run another service to try the right way.

 

Also- very important: start with minimum reduction and then add brake pressure gradually as needed, because you don't want to reduce or release the train brake once set, until you're at the bottom of the grade. (in that connection, give the train brake time to actuate all the way back; don't just assume "that's not enough" a second after moving the handle. It takes time for a pressure reduction to propagate down a 1/2 mile or more of brake pipe)

Also: especially on downgrades, a speed limit is not a target. In many cases (especially Sherman Hill), you do NOT want to be running down 1.5% at line speed, because your brakes can't hold the train at that speed. Back when Saluda Mountain was still active, NS took the downhill at all of 8 mph.

 

Realistic? Here is my theory. "If," these unreal trains are like actual trains, then what we have to do to bring them down the hill is realistic. If our unreal trains are not like actual trains, then everything is unrealistic to begin with so why bother worrying about it.
I have never driven an actual train but I have brought a lot of unreal trains down steep grades. I find that twiddling the brakes, sometimes a lot, is unavoidable. This is because you will not find a combination of settings that will last for long. Whatever you do, your speed will in time begin to slow the train too much, or begin speeding up. Going down hill is certainly not time to kick back and read your cell phone messages.
If things seem to be changing so fast you cant ever take your hands off (i.e. out of control) you are no doubt going too fast. Slow down and it will get easier. Wait until you can do everything on Sand Patch before you even try Sherman Hill. Those trains on SH are 11,000 tons and 1-1/2 miles long. One of them is 15,000 tons. Its a different ball game.

 

Fantastic, thank you.
How about hill start? I struggled with it yesterday with a very heavy coal train on SPG. I finally managed to start but snading light was on for minutes.

 

For hill starts apply the loco brake and release the train brake, so the cars behind you are free and it's only the locos holding the train. Then add throttle and slowly release the loco brakes as well.

 

And will the loco brake hold 10000t train without rolling back?

 

It should. After all, the locomotives have enough grip to get the same train started or brake them just by the dynamic brakes (which obviously are loco-only brakes as well). So keeping the train in place should not be a problem either, yeah.

 

Holding a stopped train in place doesn’t require nearly as much force as stopping a moving one

 

And this

 

Can we just appreciate the fact that Matt typed out what we were all probably a bit too lazy to type out on a Saturday resulting in a pretty comprehensive guide to US air brakes?

 

If you want to see Matt demonstrating what he posted above, watch his Sherman Hill streams on YouTube, in which he narrates the process while doing it live. That greatly improved my Sand Patch driving.

 

I'm not sure if TSW does this, but in Run 8 if you apply the train brake while dynamics are engaged, the loco brake does not actually apply. Apparently this is correct and threw me at first, until I queried it on The Depot forum and confirmed actually the case.

 

Depends on the loco

 

Thanks Matt for this detailed explanation, now I have a lot to think about.
I have been practicing now for two days, not an easy job.

 

I hope JustWentSouth dont't mind if I copy here what I have just found in another thread from last year.
It can help understanding this topic (it helped me anyway).

As you can see this is from Trains Magazin.

 

When I went over the crest yesterday after reading Matt's detailed comment, my train started to accelerate. I put dynamic notch 1...2.....3........4......finally 5, but in kept accelerating. No problem I thought, lets put automatic to minimal reduction and decrease dynamic. So I did.....auto to minimum.....dynamic down to 4...3....2....1......off........train just getting slower and slower........oooops I stopped. No worries let's do again......same process.....same result.
It is important to note that the train was only partially downhill when I did this test. But: it was accelerating quickly 3/4 was still downhill and despite dynamic notch 5. As you guys wrote that I should not go over notch 4, I applied auto brake......train stopped.
As soon as the whole train was over the crest all went fine.
So now what? What did I do wrong while I was along the crest?

 

Practice makes the master

 

Yup. Notice though the corollary to Diagram 4: the air that was in the brake cylinder was vented off to the atmosphere and is lost. It has to be replaced- but refilling the reservoir via the (small) refill groove is very slow. So, if you release brakes and shortly afterwards re-apply, instead of a 10-lb reduction you only have an 8.9-lb reduction, = 22 and a quarter pounds of brake force. And worse with each cycle. And that's just with a very modest 10-pound reduction; a full-service very much makes things worse, and it can take a very long time indeed before your brakes are fully charged again (your locos' air compressors are working at it, but it isn't fast). This is why, once your train brake is set, you don't want to release it again if you can help it.

 

It is true. However, it is difficult to lose the brakes in TSW in this way. I didn't make it - maybe I didn't try hard enough

 

From what I found online, that doesn't seem to be the limiting factor though. An engineer said online that it's about having room to play with. If I'm already in B6 just to hold the train, then I don't have much of a "safety net" of fast applying (dynamic) brakes in case the gradient changes or I misjudged the situation. If I'm only in B4 though, then I can still quickly crank the dynamics way up if the situation calls for it.

 

You might find my TSW2 tutorial playlist handy https://youtube.com/playlist?list=PLWi4sYR9zXqWK563hAInWYVxlLeM8-JMb

Particularly this Dynamic tutorial (It's Sandpatch but the same principles)

Paul

 

Yep wiggle room is always needed. No-one wants to ride that rollercoaster. Here is your worst case scenario spelled out https://www.ntsb.gov/investigations/AccidentReports/Reports/RAR9002.pdf
Essentially

• Overloaded train with estimated instead of actual weights
• No air brake test after re-assembling the consist
• Lack of rest and at least one crew member sick
• Train handed over on the move without completely stopping
• No inspection of the train from the incoming crew
• No hand over from the outgoing crew
• Non functional dynamic brakes on one or more of the locomotives in the consist
• Using all the dynamics that were available not realising they were well beyond their limits
• Not using the service brake until it was way too late which meant it simply couldn't stop and rode the rest of the hill with melting brake blocks
• Basically the train was a run-away before it left

All those things are why you need wiggle room. In my second person training, its always emphasised - if in doubt and there's no communication happening in the cab "big hole" (apply emergency brake) and explain it to your driver after you're safely stopped. Better have a driver mad at you than dead.

Paul

 

Thank you Paul, I have two questions.
- It was advised earlier not to use dynamics over 4 - 5 , then you use Notch 8 (I understand there are 10 in your case). So is that allowed to use higher dynamics?
- Also it was advised not to release automatic brakes until get down the grad what you released. Is that allowed?
Cheers

 

Depends on track layout, train weight, weather, experience, signalling, etc.

Your objective is to maintain speed.

Ther is no golden rule about notches, PSI, etc.

There are some prohibitions: e.g. you cannot brake by setting the reverser in the direction opposite to the direction of movement, e.g. you cannot try to stop a train rolling backwards by setting the reverser to the forward plus power position. You have to use the brakes for this.

 

Wow awesome! Thank you for taking the time to do those tutorials! Bookmarked your playlist

 

How much dynamic to set is situational: the operative term, mentioned above, is "wiggle room" - do you have some extra braking left in reserve, considering the steepness of the grade, train weight and so on? This varies. If I know that I am on the ruling grade- say, down 1.6% on Sand Patch - and I can hold speed with X pounds of air plus dynamics in Notch 7, then fine, because it isn't going to get any steeper, and when it shallows out a bit I can pull way back on the dynamics without having to release the train brake.

 

The 1989 Cajon Pass derailment with a natural gas pipeline rupture. Locomotives there were SD50 SD60 SD40-2 the difference between that one and the game is that the rear locomotives have drivers which is what the banking comm supposed to represent. Had there been multi-player since 2017-2019 the way it's going to work is the front player lead locomotive telling the rear players who are in the rear locomotives applying the brakes when the front locomotive driver does so.

 

El Cajon was a goat-rope for the reasons Paul listed: defective loco, inadequate handoff, sick crewman- but the first was the core reason, compounded by the others. That consist didn't have adequate braking for weight and grade, but they didn't know it. Had they known it (adequate handover, brake test), they would have stopped the train and either added traction or removed cars.

 

As mentioned by Bill solicitr it all comes down to the route. On Sherman I would behave differently and travel MUCH slower. On Sand Patch I'm happy running at (or over when no-one is watching) the speed limit. On Sherman I would not do that because it is so steep. Bill is also very correct that it comes down to route knowledge. If you know for absolute certain that you're holding the train with most of your dynamics and the route does not get any steeper and the weather is dry then go for it. You've still got your air to stop you if needed.

On the air brakes, I differ from the approach Matt takes (which is fine, every task has many right ways of doing things) and is based on many discussions with engineers/drivers over the years. I use the air brake in the saw tooth method. The concept is you use the dynamics for most of your braking effort and in most places along the line dynamics will be enough to control you. As you start to creep up you would use more dynamics because they act quickly then come back down on the dynamics when you are able to. If the dynamics are not enough to control your speed then do a minimum brake application and let that sit for a few minutes so it has a chance to take effect. If that still doesn't control your speed then give it another squeeze and again let it sit for a few minutes. For these applications most people advise to bail off the locomotives to keep the dynamics effective but I've had conflicting advice here - some engineers say bail off always, others say bail off only if the loco goes into wheel slip. It may come down to the localised training and type of route those engineers/drivers run.

With traditional Westinghouse (Wabco) air brakes after that initial set you've only got two or maybe three (if you're lucky) more applications you can make without pumping the train back up if you want your wiggle room. For this reason if the first two applications don't slow you down then you want to make your next application a full equalising application (often called full service) where the brake pipe and brake cylinder come to the same pressure and wait for the train to come to a stop. Once it stops apply your locomotive brake on full (and hope its enough to hold you) and release the train brake. It will take many minutes before the rear of the train is fully pumped up and ready to go. Then you can set off again. If the equalising doesn't stop you, on American Westinghouse you've got one last option - they have an emergency reservoir on each car, when you big hole it / go into emergency that last ditch application will apply the brakes on the cars. Note that the system is its own worst enemy as this emergency air is also used to more quickly release the brakes by helping to pump up the brake pipe.

In the past retainers would have been applied to every carriage/car so that when the air brakes were applied they were physically constrained from completely releasing even with the brake pipe fully pumped back up. I cannot find anywhere that still does this due to modern composite brake shoes - happy to stand corrected - just because I've not found it doesn't mean it doesn't happen.

On the baby railway where I operate we don't have dynamics, we only have the Westinghouse. We brake with one or two applications until the train is fully under control at about half line speed then fully release. We're only dealing with about 150 metric tons of locomotive and carriages though and the brake pipe is fully pumped in two minutes.

breblimator you're absolutely correct on modern locomotives - interestinly using the reverser to brake was acceptable practice on steam engines and on some early railways the only brake you had.

hyperlord you're welcome - I keep doing more, in fact I might do another one for Sherman because the need seems to exist and in places the future Horseshoe route is steeper so the same method can be applied there.

driverwoods#1787 you are correct, the rear locomotive engineer (probably only one since the coupled locomotives are controlled together) would be applying dynamics when told to by the lead engineer by radio but the lead engineer would control the air brake in the entire train including those rear locomotives. The only exception is banking locomotives (sometimes called pushers) who generally are not connected through with the air. In modern times the computers on each locomotive would be talking to each other by radio to manage the descent and the engineer has much more complex options available like DPU fencing where as they crest a hill for example they might have the rear units pushing in Notch 8 while the front units are in dynamics starting to slow the head end. This is critical with long trains to avoid stretching the couplers so much that they break a knuckle and pull apart. Couplers are designed to fail this way because the alternative is to rip the draw bar out of the car. A knuckle can be (and usually is) replaced by the road crew.

Paul

 

As I recall the report on the Cajon derailment, the trailing MU engineer panicked and (on his own) decided to pop the emergencies.

 

With the caveat that increasing the train brake set without releasing costs you nothing- it's releasing the brakes that vents the system. I find on Sherman Hill, given a not-extravagantly-heavy train, I can hold 30 mph with an initial app plus a couple of pounds, plus dynamics. But I still have a lot of air in reserve to add more squeeze if I need it.

 

The first video linked in this topic used a phrase something like (from memory) "put your fence up". Any idea what that meant?! Sorry if I've misphrased it, but the word "fence" was deffo in there.

 

It is about using DPU in the advanced (different power/brake settings for particular locos) way. Not implemented in TSW.

 

It might be terminology, but releasing puts air into the system - it releases from the brake cylinders but its going into the train brake pipe and auxiliary reservoirs on each car.

Normally when you apply it will be in increments governed by the feed valve, most of the time this is 7 pounds.

The rear helper doing it old style could pull their emergency valve without touching the brake stand and that would apply the emergency brakes from the rear travelling forwards. The lead locomotive would start to fight that application using its compressor, that would get awfully messy much like a break-away.

hoagy old mate breblimator explained that well. In a massive train you might even have multiple sets of DPU doing different things although how one person keeps track of that in their head amazes me.

Paul

 

Thanks for the explanations, much appreciated.

 

My bit of advice in this excellent thread: keeping a train stretched out over the crest requires that you don’t hit the crest at track speed and stay on minimum throttle as long as possible, ideally until the rear is over the crest. Then, you can slowly bunch up the train and begin braking with the dynamics.

On SPG, I find myself using the saw tooth approach with the air because the minimum application is 9-10 pounds on the CSX stock in TSW; this will bring most consists to a halt going down the grade.

For the SD70Ace, the minimum application is a more realistic 6-7 pounds of air which really allows you to dial in a speed. That’s why I was disappointed to see the SD70 mostly disappear from the SPG timetable. Hopefully, the locos for Horseshoe Curve will have the lower minimum set!

 

Except in the C40, which gets it right.

Anyway, going down the grade eastbound, if with loaded coal or grain hoppers it takes minimum air (permanently) plus dynamics to hold. OTOH, empty, autorack and intermodal trains can be held with dynamics alone, unless all you have is two SD40s.

 

For two SD-40s Eastbound trips try to make them light which means copying 1878 metric tons RRO petroleum train separation that their horsepower per ton handle without stalling at Start then apply the method described here.

 

Admittedly, I have been driving for over a year using the dynamic break as primary and tapping the air break (auto-brake | train brake) as required. It always worked. However I was running the dynamic at 7 or 8. Though I never got in any real trouble, I was always concerned that I had no headroom if needed. But I kept to what I knew.
In the early days, I experimented with setting the air brake as primary but abandoned it because I could not imagine that any friction brake could or should be left on for an extended time. I'm no expert in Train brake pads but running down Sand Patch or Sherman where they would need to stay engaged for 10+ miles seems a bit wrong to me. Does one expect to change brake pads at the end of each service? Not to mention the heat.
I would expect they would get hot and quit working. I have not noticed any consideration for that in the physics model. Its like they are magic in the Sim. Last forever, work forever. I have had some conditioning to believe since neither my truck nor my motorcycle can go down long hills using a friction brake. Its a wreck waiting to happen. Trains are different. (I guess)
So, when I'm wrong, I say so. I have started using a minimum application of air brakes and then adjusting with the dynamic breaks. Now I am going down hill with dynamic brake settings of like 2 or 3 instead of 7 or 8. It is a whole different experience and I love it. (got me some headroom)
"I was wrong OK can I go now"
But I would like to hear more about whether the pads have to be cooled down etc. or whether this "new way" is actually real world viable. I would think they would be less effective the hotter they got requiring the engineer to increase the application - and eventually they would just stop working ( within 10 miles for sure)

 

The video linked by breblimator in post #2 was done by an actual engineer, so yes, this is the real world practice. Requiring engineers to have a minimum air brake application on steep gradients is nothing out of the ordinary - there’s a similar rule in Germany above a certain gradient.

As far as wear and tear goes: Obviously, it’s hard on the brakes. Seeing how the air brakes are not applied very hard (if everything goes right), the wear and tear is probably manageable though.

 

True. Although IRL a mile-long consist would also have locos at the rear and probably also in the middle.

 

Brake blocks are replaced very frequently. Even our baby 110 ton train sets on the heritage railway I work on has the brake blocks replaced roughly monthly, more often in summer when we run longer and heavier consists due to increased passenger numbers. Modern composite blocks last somewhat longer than old fashioned cast iron. Wheels will dissipate a lot of heat as long as you stay below the point where the grease in the bearings catches fire then all good

If we had proper DPU in game then not all the dynamic braking would be at the front, even now the braking is a simulation of having a second crew in another set of locos at the back isn't too bad.

It is important to keep the train either stretched or compressed - either should be ok from a physics perspective both in game and in reality. Having them bounce back and forth would set up standing waves which would likely at least brake couplers if not worse.