The International Simutrans Forum

 

Author Topic: how accurate are late game DMU/EMU tractive effort specifications?  (Read 5510 times)

0 Members and 1 Guest are viewing this topic.

Offline ӔO

  • Devotees (Inactive)
  • *
  • Posts: 2345
  • Hopefully helpful
  • Languages: en, jp
To me, quite a lot of them seem high, even accounting for the number of drive wheels and their lighter weight.

for example

Baseline, locomotives with known figures:

Diesel
1960: BR class 55: 2460kW / 222kN
1962: BR class 47: 1922kW / 255kN
1976: BR class 56: 2420kW / 275kN
1989: BR class 60: 2300kW / 474kN
1998: BR class 66: 2385kW / 409kN

Electric
1959: BR class 81: 2390kW / 222kN
1965: BR class 86: 3010kW / 267kN
1973: BR class 87: 3730kW / 258kN


DMU/EMU for comparison per power unit

DMU
1958: BR class 205: 650kW / 75kN
1963: BR class 123: 343kW / 83kN
1984: BR class 150: 212kW / 91kN
1990: BR class 158: 260kW / 100kN
1998: BR class 170: 315kW / 150kN

EMU
1962: BR class 309: 846kW / 135kN
1981: BR class 317: 764kW / 142kN
1988: BR class 321: 996kW / 145kN
1999: BR class 357: 560kW / 165kN


From 1950 to 1988, DMU/EMU TE seem slightly high
From 1988 onward, they seem excessively high, especially diesel.

A typical class 309 10 car configuration will have 2538kW / 409kN / 453t, compared to class 81 with ten mk 1 coaches is 2390kW / 222kN / 473t. I think 409kN is probably slightly high, with somewhere around 300kN to 350kN being more realistic.


The real problem is later in the game.

with an 8 car configuration of class 357: 3360kW / 990kN / 358t
with an 8 car configuration of class 170: 2520kW / 1200kN / 388t
or, basically, the DMU is out performing the EMU, which does not seem right, and even the EMU TE seems fairly high.

Offline Junna

  • Devotee
  • *
  • Posts: 1081
Re: how accurate are late game DMU/EMU tractive effort specifications?
« Reply #1 on: July 22, 2013, 03:13:23 PM »
I assume those are where possible the real numbers; for DMU's, those are often specified per engine, necessitating a multiplication which might not always give an accurate end result; but they are indeed almost universally above those of a locomotive. The 170 is only run in three-car sets as far as I know, so this is just an accumulated TE.

Online jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18382
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: how accurate are late game DMU/EMU tractive effort specifications?
« Reply #2 on: July 22, 2013, 03:40:36 PM »
It is actually very difficult to discover tractive effort information about multiple units for some unaccountable reason. There is information about the tractive effort of the Class 444 here, on which I have based extrapolated guesses for other units, but, on re-reading the page, it is not clear as to whether it means that the whole 5 car unit has a tractive effort of 200kN or that each vehicle has that tractive effort.

If anyone has any idea where I can find information on this subject, even approximate information, that I can extrapolate/interpolate, I should happily recalibrate these.

Offline ӔO

  • Devotees (Inactive)
  • *
  • Posts: 2345
  • Hopefully helpful
  • Languages: en, jp
Re: how accurate are late game DMU/EMU tractive effort specifications?
« Reply #3 on: July 22, 2013, 03:58:28 PM »
one 5 car set of BR Class 444 with 1500kW / 200kN / 227t would still give it twice the TE/ton of a class 81 locomotive with mk1 coaches.

750kW / 100kN per motor car sounds about right.
« Last Edit: July 22, 2013, 04:03:29 PM by ӔO »

Online jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18382
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: how accurate are late game DMU/EMU tractive effort specifications?
« Reply #4 on: July 22, 2013, 04:02:13 PM »
Can anyone think of a way of discovering (or deducing) what sorts of tractive effort that earlier DMUs and EMUs (and later DMUs) had?

Offline ӔO

  • Devotees (Inactive)
  • *
  • Posts: 2345
  • Hopefully helpful
  • Languages: en, jp
Re: how accurate are late game DMU/EMU tractive effort specifications?
« Reply #5 on: July 22, 2013, 04:23:52 PM »
There are a few things I would consider

- Drive wheels per car
obviously, Bo-Bo would have better traction than 1A-1A or Bo-B

- Power regulator type
Around 1988, a new power regulator was designed, which allowed great improvements in TE. Most evident with class 60
Before that, electric power regulation to the wheels was lousy, resulting in poor power or wheel spin.

- Maximum speed / gearing
If the train is designed for higher speed, it usually sacrifices TE for power. (i.e. class 43, class 91)
Conversely, slower trains may sacrifice power for TE. (i.e. class 60)

---

Class 43 has 1678kW / 80kN, which is 21kW/1kN or 420kW/20kN/axle.
Class 180, 220, 221 has 560kW, are newer, but have 2-B configuration.
Therefore, as a guess, maybe somewhere between 21kN to 25kN per driving axle?

25kN/axle would be the same as class 444, although class 444 has four driving axles per motor car versus only two for 180.
« Last Edit: July 22, 2013, 04:46:31 PM by ӔO »

Online jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18382
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: how accurate are late game DMU/EMU tractive effort specifications?
« Reply #6 on: July 22, 2013, 04:54:53 PM »
Presumably the tractive effort per driving axle would be considerably different between a 1940s suburban EMU, a 1950s rural DMU and a 2000s inter-city DMU or EMU? I don't think that we can extrapolate from modern inter-city D/EMUs to older suburban/rural units very well, can we?

The Class 60 I know used creep control, which cuts off power for a very short period just as the wheels began to slip - I don't know whether this was ever used on anything other than freight locomotives, though. Was it this to which you were referring?

Offline ӔO

  • Devotees (Inactive)
  • *
  • Posts: 2345
  • Hopefully helpful
  • Languages: en, jp
Re: how accurate are late game DMU/EMU tractive effort specifications?
« Reply #7 on: July 22, 2013, 05:36:01 PM »
on modern DMU/EMU, I don't know if they have creep control, but they do have far superior power control over the wheels thanks to pulse width modulation. It's like having more increments of power you can apply. As an example, although I don't know the exact numbers, PWM can do 5% increments of applied power, where as before, the controller could only do 10% to 15% increments of applied power, which allowed the driver of the trains to keep their schedules, even in adverse weather.


Timeline wise, I think you can break it down into...
1930 to 1960, early electric and diesels
1960 to 1988, improvement in power and TE (10 to 15% improvement in TE)
1988 and beyond, improvement in TE and power efficiency (15% to 30% improvement in TE)

Online jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18382
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: how accurate are late game DMU/EMU tractive effort specifications?
« Reply #8 on: July 22, 2013, 05:48:00 PM »
Ahh - so you suggest that I back-extrapolate from the modern units? Is there any particular source for those figures?

Offline ӔO

  • Devotees (Inactive)
  • *
  • Posts: 2345
  • Hopefully helpful
  • Languages: en, jp
Re: how accurate are late game DMU/EMU tractive effort specifications?
« Reply #9 on: July 22, 2013, 07:39:36 PM »
Ah, sorry, I should clarify that those numbers are guesses.

There is a length article here, which is oddly only in japanese, about power regulators in trains.
http://translate.google.ca/translate?client=firefox-a&hl=en&ie=UTF8&oe=UTF8&u=http://ja.wikipedia.org/wiki/%25E9%259B%25BB%25E6%25B0%2597%25E8%25BB%258A%25E3%2581%25AE%25E9%2580%259F%25E5%25BA%25A6%25E5%2588%25B6%25E5%25BE%25A1

Basically, the introduction of new power regulators allows more power to be applied to the wheels instead of being lost along the way.
Or effective horsepower vs. brake horsepower.
« Last Edit: July 22, 2013, 07:45:14 PM by ӔO »

Online jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18382
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: how accurate are late game DMU/EMU tractive effort specifications?
« Reply #10 on: July 22, 2013, 07:59:22 PM »
I appreciate that they are guesses - in the absence of any information, informed guesses are better than nothing. Did you mean to extrapolate backwards with your percentages from the one or two known data of modern units, or had you something else in mind?

Offline ӔO

  • Devotees (Inactive)
  • *
  • Posts: 2345
  • Hopefully helpful
  • Languages: en, jp
Re: how accurate are late game DMU/EMU tractive effort specifications?
« Reply #11 on: July 22, 2013, 08:27:28 PM »
Yes, I think it would be best to work with the class 444 specs and go backwards from there.

25kN/axle is probably the higher limit for DMU/EMU, beyond that, there is probably not enough axle load to keep the wheels from slipping.

I do know that 187.5kW (250hp)/axle is at the high end for DMU/EMU. Going higher than 200kW/axle can lead to wheel spin, unless the gearing is changed for higher speeds (over 200km/h), which would lower TE as a result.
« Last Edit: July 22, 2013, 08:35:22 PM by ӔO »

Online jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18382
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: how accurate are late game DMU/EMU tractive effort specifications?
« Reply #12 on: July 22, 2013, 11:04:05 PM »
I am in the process of calibrating now, and have found some interesting snippets of information. There is some discussion on this forum about the tractive effort of some modern EMUs; and this web page suggests that the BR Class 205 DEMU had a tractive effort of 56kN despite two powered axles (presumably because all the equipment was rather heavy).

Edit 1: One difficulty is that it is hard to find out how many axles are powered on early DMUs. If anyone can find this information, I should be most grateful.

Edit 2: I have now pushed the changes to Github, and also uploaded a spreadsheet giving approximations based on your figures above. I am not sure whether this needs further adjustment in light of the information that I have turned up in the first edit.
« Last Edit: July 22, 2013, 11:29:47 PM by jamespetts »