DMU mechanical/hydraulic transmission

pepsipowell · Post by **pepsipowell** » Wed Sep 08, 2004 10:21 am

When driving DMUs in the sim it seems that for each notch of the throttle the train accelerates quickly up to a certain speed and then stays at that speed exactly. Move up another notch, and it will accelerate quickly again to another speed, then abruptly stop accelerating. Also, the acceleration seems to be very similar no matter which notch you are in.

Is this how DMUs (in particular Sprinters) actually perform in real life? Or are they more like the performance of diesel-electrics in the sim, and each throttle notch applies more or less power with the maximum speed depending on the gradient etc. and the acceration getting smaller as speed increases?

Jonathan

thenudehamster · Post by **thenudehamster** » Wed Sep 08, 2004 12:43 pm

This is a thorny question, Jonathan.

Much would depend on the type of transmission, what ratios (if any) it uses, what type of shifting it uses, and how the throttle is related to that.
Any diesel-electric system ought to be more or less stepless as the transmission is effectively stepless; however, as US diesel locos show, stepped throttles are nothing unusual. I'd expect them to behave much as you describe.
Mechanical, or hydraulic transmissions are stepped among their ratios, therefore I'd expect a certain amount of stepping in performance; it's similar to the way you drive your car - try taking off in third and see how you get on. Or try staying in first; you'll quickly reach a maximum speed. Even automatics have steps in the ratios, but the fluid drive makes them less definite.

However, in the sim, I've found several older generation dmus with a .eng file lifted straight from a gp38 when the protoype had four gears! Dmu transmission, throttle settings, and .eng file physics in general have been a difficult situation for some time - I kept promising myself I'd look into it, but other things always intervened - and I'm not sure what, if any, real improvements have arisen. Added to this is the fact that I have no real idea what transmission types are used in current generation dmus, and it gets even more difficult. Even my favourite http://www.therailwaycentre.com isn't as helpful as I'd like

If anyone can give me some definitive information on second and later generation dmus in terms of transmission types and ratios, I might try taking another look.

markw · Post by **markw** » Wed Sep 08, 2004 1:07 pm

There is a difference between the typical old-style first generation DMU and the modern Sprinter type. I use the word typical as some heritage DMU's did use hydraulic transmission.

Most first generation DMU's used mechanical gearboxes which worked pretty much as does a car gearbox. The driver would change gear to increase speed. In MSTS the Kiha does have the gear change simulated, and units like the Cravens, the 117, and the Irish BUT units were set up with four manual gears, which you have to change using the "E" key in order to get above about 15mph, top speed in first. In fact, you have to engage first to even move. The only difference between the real life DMU and the MSTS dmu is that you would have to allow the engine revs to die off in real life, or probably destroy the transmission in fairly swift order, whereas in MSTS you don't have to bother with letting the revs die off, so it is probably a bit more like a pre-selector gear box.

Modern sprinters have a hydro-mechanical gear box with an automtic hot shift "gearchange". The unit starts as a diesel-hydraulic, with the diesel driving a torque converter, which is why the engine sounds like it is revving itself to bits, and has a fairly constant note up to about 25mph. At that point, the engine switches to a direct-drive system via a clutch, hence the engine note drop. In effect the engine has two "gears", and in MSTS you would set up the script with two AUTOMATIC gears so that you don't have to press the E key to change gear.

Diesel electric units, such as Voyager or a Thumper, would be set up as per a diesel electric locomotive in MSTS and as Barry rightly says would be stepless. A Voyager on lift off will have the engine run up to what seems like maximum revs and pretty much stay there, there is no "gearchange" engine note drop if I recall, as the current flow to the motors is controlled electronically. By comparison, the Thumper will have noticeable engine note change, but this is to do with driving the unit so as not to put too much current through the motors at start, which would cause a flashover, so you get the driver notching up to say 50% power to get the train rolling, and when the ammeters drop he will then increase power, so as to stay under the red area where the electric motors may be damaged. This is the same for all non-electronic control diesel locos, and something again MSTS does wrong as you can slam the controller to 100% without there being a huge "bang" and everything grinding to a halt.

I remain to be corrected if my fading memory of "The Ladybird Book of How It Works: The Locomotive" and various other books is faulty!

lateagain · Post by **lateagain** » Wed Sep 08, 2004 2:06 pm

Interesting post guys.

IMHO the Cravens that came with Mid East UK had by far the best "simulation" of the older DMU's. As a commuter on 117s into Paddington for a number of years these "felt" and sounded right. As stated the only bit missing was the dying of the revs between each gear. On 117s you'd also have to get the sound of the sliding windows almost rattling out of the slam doors to really capture it perfectly

I've travelled on the turbos occasionally since (having taken out a mortgage on the single to Ealing Broadway

) and cant say I heard them at all, and beside being impressed at how smooth they were in comparrison to the old stock I couldn't help but wonder how they improved all this and yet fitted such awful seats.

Geoff

jbilton · Post by **jbilton** » Wed Sep 08, 2004 3:33 pm

Yeah even diesel-electric transmissions do have steps...in that the traction motors have different stages of field weakening....these are not controlled by the driver...but by the control gear(very various ways of doing this)...and this alters the load on the diesel engine and hence the speed and tone.
The driver obviously contols the throttle setting....(some locos also have control mechanism between though)....which sets the rate of acceleration.
This is where some drivers got themselves a bad name for causing flashovers....although some classes were more prone to it than others.
One of my favourites was the deltics...which when driven with a passion the driver would know his steed and thrash it to within a mph or 2 or the field weakening....but without tripping......and you'd be able to hear the engine going through its acceleration curve all the way from 0 to 100mph without a let up.....fantastic.

phill70 · Post by **phill70** » Wed Sep 08, 2004 9:43 pm

jbilton wrote:Yeah even diesel-electric transmissions do have steps...in that the traction motors have different stages of field weakening....these are not controlled by the driver...but by the control gear(very various ways of doing this)...and this alters the load on the diesel engine and hence the speed and tone.

HST's dont have field divert, but they are the only ones that i know that dont have it.

Zackybong1 · Post by **Zackybong1** » Wed Sep 08, 2004 10:30 pm

That brilliant sound package for the sprinters from Railwaves is great to simulate the change from Major to Minor in MSTS. Well worth the fiver or so I paid for it.

jbilton · Post by **jbilton** » Thu Sep 09, 2004 12:19 am

phill70 wrote:
jbilton wrote:Yeah even diesel-electric transmissions do have steps...in that the traction motors have different stages of field weakening....these are not controlled by the driver...but by the control gear(very various ways of doing this)...and this alters the load on the diesel engine and hence the speed and tone.
HST's dont have field divert, but they are the only ones that i know that dont have it.

Hi
Very true....as all modern diesel electric locos 59,60,66 etc....but you do still get a noticeable change in engine tone as the various control systems change the electrical load on the alternator.Strangely enough some of them make even more pronounced auditable changes than before.

"The control of the voltage pulses and the frequency has to be matched with the motor speed. The changes which occur during this process produce a set of characteristic buzzing noises which sound like the "gear changing" of a road vehicle and which can clearly be heard "

Very interesting site....if you;ve into it.

http://www.trainweb.org/railwaytechnica ... l#ACMotors

pepsipowell · Post by **pepsipowell** » Thu Sep 09, 2004 11:31 am

Well, I'm trying to set up the transmission for Sprinters, but after some trying it seems MSTS doesn't like it if the direct drive gear is also the highest (which is effectivly the case for the fluid coupling in Sprinters), giving error messages (cannot read .eng file etc.) when loading.

When I did get it working by adding extra gears, the acceleration was terrible, managing only 3 mph accelerating at full throttle all the way along the platform at Aberystwyth. It took ages to get up to higher speeds, but even by 50 mph the acceleration had barely changed at all. Also, again the acceleration was independant of the throttle setting, as long as you were above a certain speed for each notch. When I tried climbing the bank out of Westbury towards Warminster, it stopped very quickly.

As well as this, I seem to remember many problems encountered by people running in reverse (which will definately feature with these trains in my route when done). I think I'll stick to diesel electric in the eng files for the time being, even though it is wrong it seems to give a reasonable performance in the sim. If anyone has more to add/solutions etc. , please
do continue posting, it's been very useful and I've learned a lot so far!

Jonathan

MartinH · Post by **MartinH** » Sun Sep 12, 2004 9:31 am

If you want a reasonably realistic diesel mechanical eng file, take a look at the freeware Cravens 105 or the 3DTrainStuff 101 that comes with the Cambrian. The 101 is slightly better in that the 105 is underpowered (the tractive force associated with each gear ratio is too low). Unlike the GP38, both units have also been changed to vacuum brakes as they should be in real life. The 108 on the Swanage route also uses the same eng file.

Martin

thenudehamster · Post by **thenudehamster** » Sun Sep 12, 2004 3:50 pm

The 105 (if that's the one from the MidEast CD) was deliberately underpowered by the author so as to 'give the impression of a clapped out' unit after some twenty years of service by the period in which Mid East is sort of set.

It shouldn't be too difficult to upgrade it to a more reasonabl power/TE level.

baldwin · Post by **baldwin** » Sun Sep 12, 2004 4:53 pm

In order for MSTS to accept any form of gearing the correct parameters must be in place in the eng file. I have just knocked up a two speed hydraulic for a 158, which while still a bit rough round the edges, works.

Here is the section from the eng file

DieselEngineType( hydraulic )

GearBoxNumberOfGears( 2 )
GearBoxDirectDriveGear( 2 )
GearBoxMaxSpeedForGears( 25 95 )
GearBoxMaxTractiveForceForGears( 25000N 10000N )
GearBoxOverspeedPercentageForFailure( 170 )
GearBoxBackLoadForce( 4500 )
GearBoxCoastingForce( 2000 )
GearBoxUpGearProportion( 0.85 )
GearBoxDownGearProportion( 0.35 )
GearBoxOperation( automatic )
GearBoxEngineBraking( direct_drive )

As said, this still needs tuning, but works.

Mervyn

jimmyladd · Post by **jimmyladd** » Sun Sep 12, 2004 5:16 pm

looks interesting! I wonder why nobody has ever done hydraulic transmission on the UK hydraulic diesel models?

micksasse · Post by **micksasse** » Mon Sep 13, 2004 3:17 pm

Going back to the original posting, even the best of the .eng files for the 1st generation DMUs suffer from the limitations of MSTS itself: what Jonathan says is correct: regardless of the gradient or other conditions, they'll accelerate rapidly to a given 'barrier' speed, and then just bounce off that speed until you change up, whereupon the same process is repeated (a bit like in my car at 9000 rpm...). It's coarse and a bit messy, and above all you can shift up a gear without 'taking your foot off the gas', i.e. without throttling down to zero. Another, err, unusual bit (apart from the explosive performance in reverse - MSTS speed entries in .eng files not recognising 'negative' speeds?) is... have you ever tried just putting your class 101, 105 or 117 straight into 4th gear from a standstill and driving off? You'll have no problem and it will accelerate just as quickly up to its top speed!

Again, one of the limitations of MSTS.

Jim, I suspect people haven't bothered with 'hydraulic transmission' for diesel locos because MSTS doesn't seem to let you make them behave realistically - I also think I'm right in saying that MSTS doesn't support re-counters.

I'd love to be proved wrong on this: it would be particularly interesting to have the entirely different set-ups between the 'geared' Mekhydro transmission and the torque-convertor setup used by Voith (I think that was the big difference between classes 42 and 43, i.e. BR and NBL Warships?).

baldwin · Post by **baldwin** » Mon Sep 13, 2004 7:16 pm

Mick

I am not too familiar with the differences between the Mekhydro and Voith transmissions. How did each type work?

Mervyn