Ideas to resolve uneven running

[jwealleans]

I built a Crownline J19 at the end of last year, which has a Mashima 1624 and Branchlines Multibox as the power unit. Under exhibition use, although it runs smoothly enough, it slows and speeds up, almost as if the controller is being turned up and down as it runs. It isn't regular enough to suggest a tight spot - it may go several feet between slows, or only a couple of revolutions. Testing shows that the chassis without the body runs as smoothly as you'd want.

 

There is no apparent contact between wheels and mechanism and the body; we've ruled out intermittent contact between motor contacts and the body and also shorts from the pickups. With the grub screw backed off the chassis rolls smoothly in both directions. What seems to be true is that the phenomenon gets worse the more the mechanism is under load.

 

I'm casting about for ideas now and would appreciate suggestions. I am ordering up the same motor for another J19 I am building and will probably swap that in to see if that cures it, but if anyone else has any ideas please pitch them in.

 

Thanks.

[DLT]

Weird!  

When you fit the body to the chassis is the wiring getting trapped anywhere? Could there be a bad solder joints between pickups and motor, or possibly an internal break in the wires?  

I have know the brush screws to come loose on Mashima motors, you could try tightening them?

Otherwise I would suggest that the motor is faulty.

Puzzled, Dave.

[34theletterbetweenB&D]

... Testing shows that the chassis without the body runs as smoothly as you'd want...

I take it this means that with the body removed the chassis does not exhibit the spontaneous speed variation?

 

That points at a mechanical effect, and it has to be one that builds up as the loco runs to vary the speed.

 

If the body is very heavy, it could just be a tight motor bearing heating under the greater load: binds, slows the motor until the bearing cools, speeds up, cycle repeats.

 

Inevitably the body slightly deforms the chassis, and this can be passed into the driveline and create a slight bind. What can then happen is that the mechanism then 'winds up' tight somewhere (because machined shafts and bearings have a slight screw thread on them); and thus slows. Does the spontaneous slowing occur both forward and reverse? The cure is typically a little more slack 'somewhere' in the driveline, although I did once cure this effect with much stronger pick ups bearing on the wheelbacks to positively prevent the driven axle getting thrown over enough to bind the inside wheelhub against the frame.

[jwealleans]

Thank you for your thoughts - chassis twist isn't something I'd thought of. I have ordered up another motor and we'll see what that brings. I'll update the thread in due course.

[PGC]

Hi Jonathan

 

The J19 has a boiler and firebox of max 4' 9" diameter, which equates to 19mm diameter in 4mm scale. With a 16mm motor running along the axis of the boiler, that means there's only a max of 1.5mm air space between the outside of the motor and the boiler/firebox before the thickness of the boiler material is taken in to consideration, so you may in fact only have a 1mm air gap.

 

Is it possible that with this lack of air space around the motor that the motor is overheating a little? I've had a similar problem before. Personally, I'd use the same gearbox but use a Mashima 1020 or 1024 motor.

 

Of course, if you do decide to try a smaller motor you'll possibly have to change the gearbox as well unless you can just change the worm as the 10 series motors have a 1.5mm shaft whereas the 16 series have a 2mm shaft.

 

While writing, would you be prepared to pass on your thoughts about building the J19 kit? I'm contemplating buying the PDK kit which I understand is the same set of castings and etchings so would be interested in any comments you make about ease of build etc.

 

Phil

[jwealleans]

Thanks for that, Phil. I see your point, but the motor stands upright in the firebox and at an angle, so there is rather more space around it. In addition, in exhibition use it only gets one circuit of the layout before parking back up for 7-10 minutes, so I'm inclined to discount that as a cause. This is it:

 

 

The build is illustrated on my workbench thread here starting towards the bottom of the page. It was a second hand kit and some parts were missing or had been replaced with scratchbuilt components which complicated the build unnecessarily. I'm building another one at the moment and that has gone much more smoothly. The current one is a Crownline issue, but they seem to be exactly the same. Whether PDK roll the boilers I don't know - Crownline didn't - and I've found a couple of gotchas in the instructions (they don't tell you to put the securing nuts inside the tender body before you close it up) but if you think a bit about what you're doing there won't be a problem. I previously built their J17 and it went together almost by itself. I'd recommend them.

[jwealleans]

I think we may be on to something with this.  Running improved with the body screws slacked and I think there's a slight twist to the body which  - as 34C pointed out above - then gets transmitted to the chassis.  It ran much better on a load test last night, so I shall put some strategic packing in place and then we'll see how it behaves at Folkestone.

 

Interestingly I am finishing off another of these at the moment and that has also ended up with a slight twist in the same axis.  Either I'm making the same mistake consistently or there is something in the kit which encourages this slight error.  The first one had a lot of replacement scratchmade parts, but this is a completely 'out of the box' kit.

 

I have a J20 to do from the same stable, so I'll take care when that one is going together to try to see when/if the same thing happens.

[DLT]

Distortion sounds like a very likely explanation.  Can you you support the model upside-down and look to see what happens to the corners of the chassis when you tighten the screws?

If this is the problem then there are three ways to get round it.  One is to correct the fit of the chassis to the body, so that no distortion takes place.  Sounds easy, but can require quite a bit of trial and error filing and/or packing.

 

Second, as it stands you effectively have four points of contact with the body, one at each corner of the chassis.  Can you file away either the chassis corners or the body at one end, so that the contact point is at the screw, effectively giving you THREE points of contact instead of four?  In that situation tightening the screws can't twist the chassis.

 

Third method is get rid of ONE of the screws and replace it with some sort of hook arrangement to hold that end of the chassis in place, but fairly loosely.  Then you can screw up the other end with causing any distortion to the chassis.  The photos below show the body/chassis fit of the S.E.Finecast E1/R kit, using this method.

 

Apologies, I can't upload the photos at the moment, but they are towards the bottom of page-1 of my Workbench Thread

http://www.rmweb.co.uk/community/index.php?/topic/2359-dlts-southern-locos-adams-a12-jubilee/

 

Hope all that sounds clear!

Cheers, Dave.

[Jol Wilkinson]

Jonathan,

 

I would strongly endorse what DLT says about body fixing. I found it a real problem when I modelled OO, but a little less so with P4. I think that hornblock mounted axles are more tolerant of chassis twist.

 

However, I now build locos with one rigid and one "sloppy" fixing. I've incorporated that into the etched loco kits I've designed. They have one fixing screw and either a tongue and slot fixing at the drawbar end or the front coupling hook through the buffer beam and chassis to fix the front end. That leaves just enough "slop" to avoid the body distorting the chassis.

 

Jol

[jwealleans]

Thanks both for your suggestions. I will try to find time to get the loco onto the bench this week and decide what to do. A small amount of packing may be enough, otherwise I'll simply leave one screw slightly loose and use Tacky Wax to prevent it working its way out.

 

I'm a bit peeved with myself for letting the body develop a twist as I always build them on a piece of 2 x 1 cut to length so I can tell if they are out and I didn't notice anything while I was putting it together. The chassis is jig built so I'm sure that's OK - in any case, it runs well so the problem can't have been there.

[Horsetan]

.... Whether PDK roll the boilers I don't know - Crownline didn't - and I've found a couple of gotchas in the instructions....

 

In my experience, PDK do roll the boilers (if they're not resin mouldings), but occasionally not very well.

 

As far as instruction "gotchas" are concerned, you are going to pick out a few if you build for P4 or intend to deviate from strict instructions anyway.