Large Scale Central

Davenport Surges?

General Discussion
#1

As reported on MLS:
http://www.mylargescale.com/Community/Forums/tabid/56/forumid/17/postid/56931/view/topic/tpage/2/Default.aspx

Quote:
. . . 12/30/2008 8:15 AM Quote ReplyAlert Just received and put into service my first (last?) Davenport. I was initially impressed with the construction and detail. I wired it for battery operation and put it to work. I was stunned when I found out it has the same flaw (with a single power truck) as do the two truck Shays! This flaw is - when going down hill (2-3% grade) the power train binds up releases, binds up release … etc. I won’t go into this as the Shay problem has been beat to death here on MLS and if you are interested I am sure you can locate a lot of it in the archives. For its size the little sucker is a good puller on flat ground or up hill. Best, Ted . . . .
and:

http://www.girr.org/girr/tips/tips10/davenport_surging.mov

from Our Buddy George!..here is his work page on the Davenport:

http://www.girr.org/girr/tips/tips10/davenport_tips.html

anyone else suffering the same with grades? I think I just saved $150 (for now)

cale

#2

Wow, that’s pretty bad surging in George’s video. I wonder how it’ll behave with a pile of cars behind it?

#3

I just double-checked mine, and noticed no surging at all on grades up to 8%. I tried running both ways, with and without a load, running on as little as 1.5 volts (a single AA battery). I’m not disputing others’ experiences, just offering that they may not be representative of the whole. Why one does and one doesn’t, I can only chalk up to the vagaries of mechanical design.

If it’s surging, try putting a little extra lube on the worm to see if that helps. There was plenty on mine, but I know from trying to smooth out other surging locos that a bit of extra slipperiness does help, if it doesn’t eliminate the problem.

Later,

K

#4

If Bachmann had made this engine 3 or 4 years ago, I’d be interested, but today I’ve so many HLW mack varients, I’m just left a little cool to this model. be interesting to see whats causing the surging, that would be an absolute deal breaker on my 4% grades

#5

look again at the site for a 2nd video where the surging has been, at least for now, fixed

http://www.girr.org/girr/tips/tips10/davenport_tips.html

BTW, when it was surging, a load behind it didn’t seem to do much

  • gws
#6

Gentlemen;

About ten years ago, I got a call from TOC complaining about surging with one of my drives. A bit later I was visiting relatives in the Seattle area and had arranged to visit Dave to witness the surging for myself. It was vicious!!, flipping logs off of a couple of the cars on an 80 ft. long 4% downgrade. A real problem. Went back to Vegas and started experimenting. Tried flywheels, no help. Investigated other drives that didn’t surge, seems hey all had double-thread worms on the motor shaft. So tried it in one of mine, but had to come up with means to get back to the original gear ratio.

The single thread worm on a direct drive has a gear ratio of the number of teeth on the driven gear; i.e., single thread worm on 30 tooth worm gear is 30 to 1. A double thread worm on a 30 tooth worm gear is 15:1. To get the gear ratio back to 30:1, ideally a 15 tooth spur gear on the same shaft as the worm gear would turn the axle gear at 30:1. I wasn’t able to buy 15 tooth spur gears; either 14 or 16, the 16 tooth worked better for me. I had tried 12 tooth and 20 tooth, both had drawbacks.

By the way, with this gear drive, I haven’t suffered a single gear failure in the last five years (one exception where everything failed, in a commercial application).

To check the worm on the little diesel, look at the end of the worm on the motor and count the number of ramps, a single thread will only have one, a double thread will have two (directly opposite from the other). There also triple, quad and more thread worms for very special applications.

Barry - BBT

#7

I’ve found loads behind a locomotive tend to exacerbate the surging problems. Essentially, the load behind the locomotive is pushing the teeth on the gears up against the worm opposite to how the worm is trying to push the gears, setting up something of a see-saw game as things tend to catch here and there. That’s where the copious amounts of grease come in–reducing friction (if not drag) so the gears have less tendency to catch.

On my dad’s railroad, he’s got a fairly long 5%+ grade, which tends to cause surging on a fair number of locos as they bring their trains down. The most common solution is to introduce a bit of drag on the train, so you don’t have the entire consist bearing down on the locomotive. I’ve often thought about ways to rig some kind of adjustable brake mechanism for the cabooses for just this reason.

Later,

K

#8

the way that I see this, it is a tug-of-war between gravity and drag. The number of leads on the worm is a secondary player that masks the impact. if gravity wins, the loco will be continually drawn downhill building speed within the backlash range of the gearing. When the gear train binds up against a single lead worm, the loco virtually stops until the worm can unwind the gear train again.

if drag wins, the drive train must work to pull the train downhill and the backlash never comes into play because it is taken out all the time.

The Davenport, and Railtruck, both use those roller power pickups that provide essentially no drag. They free roll through their backlash region easily. The force of gravity is an acceleration. It takes a little time to build up speed so when the acceleration due to gravity is halted by using up all the backlash, it takes a little time to build up speed again. In the meantime, the motor has driven the loco to the other end of the backlash and the loco is then set to free roll ahead of the worm again.

The double leaded worm allows the loco to actually drive the motor instead of just banging up against a stop like the single leaded worm provides. This allows the loco to burn it’s potential energy by steadily rolling downhill without bouncing to a stop.

  • gws
#9

…and, if the gear ratio is too high, say like the 1:14.5 on the K-27, a double start worm will tend to let the engine freewheel and accelerate downhill from a cruise speed on any reasonable kind of grade.

#10

Tony, this is true. I’ll also bet that the K doesn’t surge but once at the top of a grade as the train, and gear train, transition from being stretched to bunched.

When a train is on the flat, there is no component of acceleration due to gravity either fore or against the direction of travel. Therefore, drag wins and the train, and gear train, stay stretched out. No surge.

When the train is going uphill, the acceleration due to gravity ADDS to drag, same result. The train stays stretched out, no surge

Only when the train is going down hill and gravity overcomes drag will it surge. And it will only do that when the gear train can bind. In any gear mechanism where the motor can be driven by rolling the loco on the track, the tendency to roll downhill is absorbed as the motor, in effect, becomes a smooth dynamic brake. In this case, the train is bunched up, but it stays that way until it gets to the bottom of the hill and stretches out again.

  • gws
#11

I will add a bit more.

When the mesh is exactly correct and the gears are well matched one other feature develops: the loco can be pushed with a finger and the wheels will roll (like rolling down the hill freely).

My observation was that the double thread worm allowed the momentum generated by the gravity exerted on the down grade to work together with the motor which smoothed everything out. The force was being transferred directly to the motor via the double thread worm. The angle of the threads on the worm can be made more aggressive to enhance this ability; see the worm in the K-27. The K-27 doesn’t surge.

Barry - BBT

#12

The K-27 certainly does not surge downhill but it can run away downhill such that speed output must be reduced.

#13
TonyWalsham said:
The K-27 certainly does not surge downhill but it can run away downhill such that speed output must be reduced.
I really don't want to turn this into another K-27 thread, but your response begs a question...

Wouldn’t this ‘feature’ only be detrimental to roundy-round running with grades? I would think that in operations mode, where the engineer is always with the train, that this reaction to grades would actually be adding some realism. The operator needs to anticipate downgrade and reduce speed on approach, and the grade as needed.

Unfortunately, there are no brakes (without DCC back EMF) so I suppose it’s possible it might run out of control. I have a steep grade with a sharp curve and a 3 ft. drop at the edge of the curve., I’d want to be sure the K could get down this hill and safely around the curve before I could consider one.

To get back on track -

I think the Davenport is a neat little loco. It would be perfect for a yard shifter on my layout with R1 yards indoors. My indoor yard is nearly dead flat, so this surging issue wouldn’t probably bother me unless I took it out on the main.

#14

Brake Van -

How about a small battery powered braking mechanism in a caboose. A mercury switch (or environmentally friendly equivilant) would sense a down hill slope and switch on an electromagnet to slow the wheels down (or launch a boat anchor on a chain). You wouldn’t want to run it backwards, though, as uphill climbs would be difficult.

Another thought… Just have a functioning brake wheel on the caboose and adjust it manually as you go… Prototypical for some time periods.

Sincerely,

Joe Satnik