What is the mechanism by which DPUs communicate?

[Highland37]

https://www.youtube.com/watch?v=In3zXS4dKQA

In the above video (amazing!) there is a distance between the units. How does the information get from the driver to the DPUs on what to do?

 

[Groningen]

https://en.wikipedia.org/wiki/Multiple-unit_train_control

 

[jopsuk]

They use a radio system called Locotrol

 

[Groningen]

You hope that the message by radio is somewhat encrypted and can not be hacked.

 

[ac6000cw]

You hope that the message by radio is somewhat encrypted and can not be hacked.

You'd better ask GE then

Incidentally, if the description on YouTube is correct (125 car coal train), it seems to have a lot of power in relation to the weight - about 34,000 hp (8 modern locos) for about 16,000 tonnes of train. Almost at the hp per tonne level of a reasonably 'hot' intermodal train.

--- old post above --- --- new post below ---

There is a brief description of how Locotrol is used (from the drivers perspective) in the answer to the second question here - http://trn.trains.com/railroads/ask-trains/2010/02/ask-trains-from-february-2010

If you have several locos in a DPU group e.g. the four mid-train Montana Rail Link SD70ACe in the video, one acts as the Locotrol slave and controls the others in the group via the normal AAR MU connections. The system allows for up to 5 DPU slaves per train - the most I've ever seen is three, on an intermodal train in Canada that seemed to go on forever...

 

[Taunton]

There have been a number of different systems over time. The first ones typically used a separate unit, not a diesel, to control the mid-train units, very often a stripped-out former B-unit from an old streamliner diesel loco set. This allowed the remote control supplier to provide their own vehicle without needing to get into loco operation, it controlled the rest of the locos attached by its existing multiple unit connections.

Radio became the accepted control system but some of the earlier versions used strain gauges in the couplers to detect whether to push or not, as the early radio attempts had their own problems such as loss of communication through tunnels.

Initially they were used as pushers at the back in traditional manner, but gradually moved to a mid-train position, and more recently it's becoming common to have a further one on the rear end as well. The ability of couplers to handle the strain through the train was what limited operations with the locos just at the front, especially where the train would be partly on the uphill and part on the downhill on undulating routes, and the old big loco formations, albeit with units of lesser power than nowadays, seem no more. I can recall Union Pacific running with maybe eight (yes) locos on the front in the 1970s.

 

[Highland37]

Many thanks everyone. Very interesting.

 

[ac6000cw]

as the early radio attempts had their own problems such as loss of communication through tunnels.

They still do have those sort of problems sometimes, but the modern 'slaves' just stay in the current state (pushing or dynamic braking) until either the radio link comes back, or (eventually) a timeout occurs and they slowly throttle down. I assume if they are braking they continue with that (as it's probably the safest thing to do). I think if they detect a brake application (in the 'loss of radio' situation) they throttle down immediately.

and the old big loco formations, albeit with units of lesser power than nowadays, seem no more. I can recall Union Pacific running with maybe eight (yes) locos on the front in the 1970s.

Not entirely disappeared...there are eight on the front of this BNSF train I saw a year ago, climbing up to the Tehachapi 'loop' - https://youtu.be/hVjog6nJITc?t=439 (although I strongly suspect some of them were not powering the train - it's hard to tell when it's basically a wall of noise going past). The next day I saw 6-7 on the front of another BNSF mixed freight climbing the pass.

In 11 days in California, most of the DPU setups I saw were rear-only, a few mid-train only, and one very long BNSF intermodal had mid+rear (a 3+2+2 setup - https://youtu.be/ynK5qiSFF5U?t=589 ).

 

[Highland37]

You'd better ask GE then

Incidentally, if the description on YouTube is correct (125 car coal train), it seems to have a lot of power in relation to the weight - about 34,000 hp (8 modern locos) for about 16,000 tonnes of train. Almost at the hp per tonne level of a reasonably 'hot' intermodal train.

You think so? They seem to be working pretty hard on that climb and not exactly flying up it. Impressive none the less.

It makes me sort of want a massive quarry to open near Inverness and aggregate to head south through the Slochd with 8x Class 60 or 66s

 

[ac6000cw]

You think so? They seem to be working pretty hard on that climb and not exactly flying up it. Impressive none the less.

It kinda depends on how fast you want/need to climb the hill in relation to track capacity. Mullan Pass is nominally a 2.2% gradient, and it's climbing it reasonably quickly for a coal train, as you would expect with eight modern AC-drive locos hauling it (the SD70MAC on the front is oldest, the first AC-drive production locos in the US). On lines which predominately carry coal trains it's always been a 'biggest possible trains hauled by the minimum number of locos' situation to minimise the cost - it's OK to breast the summit at 5 mph (when the weather is bad) in that situation because the other trains on the line won't be doing any better. I'd guess that six SD70ACe/ES44AC locos could handle the climb but they might be too slow.

It makes me sort of want a massive quarry to open near Inverness and aggregate to head south through the Slochd with 8x Class 60 or 66s

Aggregate Industries (ex-Foster Yeoman) have a 'super quarry' near Oban, but it was designed to use marine transport only :cry: (and 'freight train interference' would become a popular excuse for late running passenger trains if you tried your idea )

 

[Taunton]

I recall an informed article on Distributed Power (probably sufficiently long ago that they were called Mid-Train Pushers) which described one of the key aspects was ensuring the Centre of Propulsion was always ahead of the remote locos. That is the point where the couplers are slack, the cars ahead are being pulled by the lead locos, the ones after that are being pushed by the remote locos. This point can move forward and back, but must not get to the remote locos to the extent that they are being pulled, even fractionally, by the lead locos, as that was found to lead to broken couplers too easily.

That was OK with only remote locos mid-train, you just ensure they are sufficiently far back to handle the changes in gradient etc that the train is on, but with further locos on the rear as well it must be more difficult, as there are presumably two such Centres.

A 125-car train, even with an average of 6 inches of slack in each coupler, is going to have 60 feet of slack to pull in as the couplers slack and tighten, which can give quite a thump. I've certainly seen a Canadian train stop on an upgrade and the rear vehicles recoil more than a complete car-length, which must be a severe strain.

 

[Highland37]

Aggregate Industries (ex-Foster Yeoman) have a 'super quarry' near Oban, but it was designed to use marine transport only :cry: (and 'freight train interference' would become a popular excuse for late running passenger trains if you tried your idea )

There is a very good reason for that. It's on the other side of Loch Linnhe from Oban and has no road or rail access anywhere near it. You are thinking of Glensanda Quarry.

I wonder how the same number of 66s would get on, on that Montana train?

 

[ac6000cw]

I wonder how the same number of 66s would get on, on that Montana train?

With around 740 kN of continuous tractive effort at about 15 mph for an ES44AC/SD70ACe versus 260 kN for a 66, if you fitted the 66s with AAR couplers (twice as strong as ours) they could probably get the train moving on the flat, but they wouldn't make it very far up the 2.2% gradient...

A bigger issue is that the 66s don't have dynamic (rheostatic) braking, so the Americans might not even allow them to take the train down the slope on the other side of the summit for safety reasons. Note how slowly this UP coal train is starting its descent of Cajon Pass on one of the 2.2% grade tracks - https://youtu.be/yP8Dh-h0x80?t=257 (the track nearest the camera is the original 3% grade route, which that train would not be allowed to use).

And besides which, the 66s are just not LOUD enough to be serious heavy-haul diesels

--- old post above --- --- new post below ---

If you are interested in 'Mountain Railroading', the April 2004 issue of Trains magazine - http://trn.trains.com/issues/2004/april-2004 - is an excellent read, full of interesting stuff from cover to cover. (My copy is very well thumbed!).

The cover story of this issue - http://trn.trains.com/issues/2006/november-2006 - is about how GE, EMD and CSX got together to develop the current generation of heavy, high tractive effort diesels (up to 160 kN per axle, 890 kN per loco).

You have to register on the website to view it, but there is a short cab-ride video showing the use of Locotrol here - http://trn.trains.com/issues/2010/~/link.aspx?_id=575A9E0BB7BA4EB0BAC646F0FF60412A&_z=z

 

[Highland37]

I have wondered if the 66 body shell could handle the forces involved also.

This is an amazing video: https://www.youtube.com/watch?v=PCOD92ux0T0

Another long one I'm afraid, so if you want to get straight to the action skip to 6:30. This video shows the old Iron Ridge Tunnel on Mullan Pass, and the new cut that replaced it. Not the best quality either, as most of it was filmed at max zoom in hazy weather. Coal train was 125 cars plus 8 locos, 548 axles, @ 11mph. Strange to see DC power leading a coal train, especially with the many AC locos showing up on the much lighter stack trains going over here recently.

Thanks for your input. I would like to go to Montana to ski and watch this stuff.

 

[ac6000cw]

I have wondered if the 66 body shell could handle the forces involved also.

I don't know, but I suspect if you distributed the locos along the train properly it would be OK. Incidentally, 'distributed power' isn't a new idea, it was used to some extent in the steam era to get a train up a difficult climb if necessary (but of course it needed one crew per loco, using whistle signals and experience to control it all). The usual alternative was to break a train into sections at the bottom, take them up one by one, then reassemble the train at the summit - takes a long time, but only needs one crew.

Thanks for your input. I would like to go to Montana to ski and watch this stuff.

I've never been to Montana either (it's on the list). The photos in this Trains blog - http://cs.trains.com/trn/b/staff/ar...ut-west-part-1-montana-rail-link-at-last.aspx - are lovely, especially the last one. To quote the blog "Once the westbound cleared, the eastbound struggled to the summit and dove into the tunnel beneath me. With a volcanic exhaust and deep voice of power, the train muscled its way across the pass."

But in the mountains the trains can be infrequent, so expect to have to be patient (take a good book - and an umbrella, the scenery is green for a reason). Even on busy mountain climbs like Tehachapi, there can be lulls lasting several hours if you are unlucky.