Getting motor and gear box to run smoothly in chassis

[Captain Kernow]

I've built quite a few chassis kits over the years, and virtually all of them have ended up running smoothly enough for my satisfaction, often after a bit of tweaking and bad language, but in a few cases, pretty much straight away. Many of these have been in OO, some in P4. Some have been built rigid, some compensated. That's fine, and it's great when something comes together and works right first time. You learn things along the way and try to put that knowledge to good use next time.

 

Now, however, I am virtually at the end of my tether with a chassis that I am building for a BR Standard Class 2-6-2T, and I've got to the point where I have run out of ideas. I simply don't know what is wrong.

 

The chassis is a modified Comet 76XXX kit, which is now intended for the 82XXX. It is being built to OO, with Romford type wheels, and rigid.

 

The basic chassis block went together fine, using a relatively good quality chassis jig (as it turns out). I'm satisfied that the chassis is built square, and that the wheels are OK on the axles. There is no rocking when the chassis, with wheels, is put on a piece of flat glass.

 

I'm also happy that the rods are fitted OK and are not binding.

 

I've run the motor in (Mashima 1420) , and then a bit more with the gearbox attached. There were no problems assembling the gearbox (a High Level 'High Flier' 54:1). Unfortunately there wasn't room for a flywheel. The motor and gearbox run smoothly enough, with a nice, steady rotation when an axle is inserted (but not mounted in the chassis at this stage) and the drive gear tightened up (grub screw type).

 

I've also had the motor/gearbox and centre axle running nice and smoothly in the chassis, with the drive (on centre axle) in place. The leading and trailing wheels have also run smoothly, when rods fitted, one minute, but then the next minute, I'm getting some resistence - for no apparent reason, enough to produce a faltering gait to the loco (this is with the motor directly wired to lightweight wander leads from the controller, so it's not the pick ups either).

 

The gearbox is 8.7mm wide, and the chassis (between the frames, being OO) is approx 11mm.

 

The gearbox is held in place at the rear (non drive end) by a piece of insulated wire soldered to the motor terminal at one end and a chassis spacer at the other - a good soldered joint, yet giving the motor some flex when running.

 

However, I'm wondering if there is some kind of 'law of physics' going on here, that governs the relationship between the motor/gearbox and the driven axle. Should there be any play in the driven axle? Should the motor (at the non-driven end) be held in place more securely - is any kind of flex there a bad thing and enough to introduce seemingly random poor running?

 

The 'High Flier' is quite a tall gearbox, but it needs to clear the top of the chassis, valve gear, motion plate etc.

 

Anyone got any good ideas out there, please, because I'm fresh out!

 

Thanks.

[micklner]

I am sure that High Level do not have any flex in the geared axle. I would fit washers to remove any space and try again.

 

good luck

[Theakerr]

I put together a comet 0-6-0 chassis using their jig and I thought it was square, but I had similar problems to those you are describing.  After much effort it turned out that two axles were square, the centre and back end, but the third was not square as measured using a square and an extended rod.  I had to melt out the bearing and file the chassis to move the bearing and get a correct alignment.  It is still not perfect but it isn't too bad.  FYI, I also assembled a second chassis kit, but this time I checked the bearing hole position.  The two front holes did not align perfectly.  This time I corrected by filing out both the front and back axle holes a little, rounding out the driven axle a little oversize and assembled accordingly.  Works OK.  

[Miss Prism]

The gearbox is held in place at the rear (non drive end) by a piece of insulated wire soldered to the motor terminal at one end and a chassis spacer at the other - a good soldered joint, yet giving the motor some flex when running.

 

I cannot picture what is going on here, what direction or axis of 'flexing' is being allowed or intended?

 

And what kind of play do you think there should be in the driven axle?

[edcayton]

Surely if the driven axle is running OK (can you try running the loco with no rods to check?) then the problem is somewhere in the rods/quartering/parallellity of the axles.

 

Have you tried running it with just one side's rods on? Are the crank pins square to the wheels?

 

Ed-no brain left

[robert17649]

Surely if the driven axle is running OK (can you try running the loco with no rods to check?) then the problem is somewhere in the rods/quartering/parallellity of the axles.

 

Have you tried running it with just one side's rods on? Are the crank pins square to the wheels?

 

Ed-no brain left

Ihad a similar problem which turned out to be down to a slightly bent crank pin, cured by changing to a new set of pins and loctite.  It may be no help but worth a try

[sjp23480]

I put together a comet 0-6-0 chassis using their jig and I thought it was square, but I had similar problems to those you are describing.  After much effort it turned out that two axles were square, the centre and back end, but the third was not square as measured using a square and an extended rod.  I had to melt out the bearing and file the chassis to move the bearing and get a correct alignment.  It is still not perfect but it isn't too bad.  FYI, I also assembled a second chassis kit, but this time I checked the bearing hole position.  The two front holes did not align perfectly.  This time I corrected by filing out both the front and back axle holes a little, rounding out the driven axle a little oversize and assembled accordingly.  Works OK.  

This is worth checking as I had a similar issue with an old A3 chassis. Everything looked square but on closer examination the axle holes were not aligned.

 

Steve

[ozzyo]

Are you running it on a rolling road or in space. As both of these can show this type of running but when you put it on the track it runs OK. I think that this happens because the locos own weight helps to keep all the wheels in sync.

[twiggy1969]

are your rods one peice as this would pick up any side ways flex with motor and gear box

as the high flyer is high as to say more lightly to be able to flex in any direction  which in turn is transfered to the axel then the wheels then rods

 

just a thought

 

mark

[dymstocklr]

Could it be the crankpin nuts 'tightening' as it runs?

Bill

[Captain Kernow]

I cannot picture what is going on here, what direction or axis of 'flexing' is being allowed or intended?

 

And what kind of play do you think there should be in the driven axle?

 

Hi Miss P - hopefully this diagram will make the gist of my query clear - 

 

 

There had been approx 30 thou of lateral movement on the driven (centre) axle, which allowed the gearbox and motor assembly to 'flex' slightly up above.

[Captain Kernow]

I put together a comet 0-6-0 chassis using their jig and I thought it was square, but I had similar problems to those you are describing.  After much effort it turned out that two axles were square, the centre and back end, but the third was not square as measured using a square and an extended rod.  I had to melt out the bearing and file the chassis to move the bearing and get a correct alignment.  It is still not perfect but it isn't too bad.  FYI, I also assembled a second chassis kit, but this time I checked the bearing hole position.  The two front holes did not align perfectly.  This time I corrected by filing out both the front and back axle holes a little, rounding out the driven axle a little oversize and assembled accordingly.  Works OK.  

Interesting to note your experiences, thanks.

 

In this case, the chassis was assembled using a Hobby Holidays jig, which should ensure that the axle holes on both side frames are precisely aligned, and this appears to have happened OK, judging by visual checking now that the chassis unit is assembled.

 

Similarly, the rods were done using the jig (even though this strictly shouldn't be necessary). The rods are articulated via the centre crank pin.

[Captain Kernow]

Are you running it on a rolling road or in space. As both of these can show this type of running but when you put it on the track it runs OK. I think that this happens because the locos own weight helps to keep all the wheels in sync.

This is very interesting. The latest problems happened when testing the chassis 'in the air'.

 

However, I must stress that I have previously had the chassis running nice and sweetly, both 'in the air' and on the track with the loco body on. The subsequent testing is to ensure that, as each section of valve gear is added, the chassis still runs nicely.

 

What I was getting, then, was unexpected and unpredictable faltering in the running, with leads soldered directly to the motor, without having taken the motor or gear box out, yet just a few minutes earlier the same chassis was running OK. The only thing that had changed was that I had handled the chassis by means of holding the motor gearbox assembly, and I'm wondering if I've done something to upset it's 'equilibrium'.

[Captain Kernow]

are your rods one peice as this would pick up any side ways flex with motor and gear box

as the high flyer is high as to say more lightly to be able to flex in any direction  which in turn is transfered to the axel then the wheels then rods

 

Hi Mark - the rods are in two pieces.

 

One other thing I had already done, given the long side frames of the High Flier gearbox, was to solder on a piece of scrap nickel silver towards the bottom, not foul of the drive gear, to give the whole thing a bit more rigidity. This hasn't affected the running, as I'd already tested it OK.

 

Considering how motors are fixed in RTR locos, I'm now thinking of building in maximum rigidity fore and aft, by means of a rear brass bracket and bolting the gearbox on one side to one of the side frames (there's room inside of the body cavity for the motor to be consequentially slightly off-set).

 

 

[Gruffalo]

Thanks for the info on this thread all. I have just ordered a rolling road so that I can "settle" all my stock as I change to finer wheel standards ready for code 75 and to bed in two or three chassis kits. 

Just one thought, are the pick-ups having any influence?

[Captain Kernow]

Thanks for the info on this thread all. I have just ordered a rolling road so that I can "settle" all my stock as I change to finer wheel standards ready for code 75 and to bed in two or three chassis kits. 

Just one thought, are the pick-ups having any influence?

The pick ups aren't affecting the running in this particular instance, as far as I can ascertain from my testing so far, as much of the problem has been encountered with the chassis directly connected to flexible, lightweight 'wander leads'.

 

I had previously given the chassis some running in on a rolling road, but my personal preference is to run the loco in on a circle of track.

 

Good luck with your chassis building!

[Miss Prism]

Chassis diagram_irf.jpg

 

Ok. What was confusing me was your terminology.

 

On your lower diagram, the lateral 30 thou sideplay ('slop' is ambiguous) on the driven axle should not affect the symptoms you describe. The lateral motion you drew is rotation, in this (vertical axis) case called yaw. The height of the gearbox is not relevant. There should be no yaw. If you are observing motor yaw relative to the chassis, either the rod centres are not aligned with axle bearing centres, or crankpin throws are inconsistent, or there is excessive play between axles and their bearing bores.

 

In your upper diagram, you do not have 'up and down' movement of the motor. It is rotation (about an axis - the driven axle in this case). This pitching is normally controlled by a torque reaction arm, or equivalent. It sounds as though your "piece of insulated wire soldered to the motor terminal at one end and a chassis spacer at the other" might be doing something toward this objective, but it seems a strange way of doing torque reaction. Your "nickel silver towards the bottom" might also be helping, but putting a torque reaction control rod near to the axle axis will negate its intended function. Are you observing any twitching in the motor pitch rotation when the chassis is running without a body on?

[griffgriff]

Now if you had a cradle mount for your motor you could have secured it with a nice blob of mastic glue (it's also quite good for sound proofing) it's a very unsophisticated fix but often highly effective, 

[Captain Kernow]

 

Ok. What was confusing me was your terminology.

 

On your lower diagram, the lateral 30 thou sideplay ('slop' is ambiguous) on the driven axle should not affect the symptoms you describe. The lateral motion you drew is rotation, in this (vertical axis) case called yaw. The height of the gearbox is not relevant. There should be no yaw. If you are observing motor yaw relative to the chassis, either the rod centres are not aligned with axle bearing centres, or crankpin throws are inconsistent, or there is excessive play between axles and their bearing bores.

 

In your upper diagram, you do not have 'up and down' movement of the motor. It is rotation (about an axis - the driven axle in this case). This pitching is normally controlled by a torque reaction arm, or equivalent. It sounds as though your "piece of insulated wire soldered to the motor terminal at one end and a chassis spacer at the other" might be doing something toward this objective, but it seems a strange way of doing torque reaction. Your "nickel silver towards the bottom" might also be helping, but putting a torque reaction control rod near to the axle axis will negate its intended function. Are you observing any twitching in the motor pitch rotation when the chassis is running without a body on?

Thank you for this, Miss P.

 

Correct engineering terminology never was and probably won't ever me, a strong point, but I understand what you mean here. There is no yaw as far as I can see, and I have now replaced the motor and removed all sideplay, by insertion of brass washers on the outside of the chassis, between bearing and wheel on each side.

 

My proposed brass bracket functions, I would imagine, as a kind of torque reaction device. I have now made and fitted this as described above. There is a 6.5mm diameter hole in the face of it, which fits over the similarly-sized plastic bearing (?) on the rear of the motor. There is now no rotational movement whatsoever.

 

When power was applied, the wheel rotation ('in the air') was reasonably smooth, but not perfect, so I've left it running for a while to see if it beds in.

 

With regards to the nickel silver pieces I've added to the gearbox, this diagram shows what I've done:

 

 

I doubt I'd have done this, had the sides of the gearbox not been so long, but it does impart helpful additional rigidity.

[34theletterbetweenB&D]

...I've also had the motor/gearbox and centre axle running nice and smoothly in the chassis, with the drive (on centre axle) in place. The leading and trailing wheels have also run smoothly, when rods fitted, one minute, but then the next minute, I'm getting some resistance - for no apparent reason, enough to produce a faltering gait to the loco (this is with the motor directly wired to lightweight wander leads from the controller, so it's not the pick ups either).

 

...However, I'm wondering if there is some kind of 'law of physics' going on here, that governs the relationship between the motor/gearbox and the driven axle. ... Anyone got any good ideas out there, please, because I'm fresh out!

 

The more close fitting your assembly, the more likely you are to see this; the description italicised is what some describe as a 'wind up'. What's going on can be very subtle and difficult to observe, involving multiple components. All the machined surfaces have an ultrafine screw thread on them, in rotation these act to shift moving components along the axis of rotation. If they happen to reach a point where the various small inaccuracies produce a bind, then you see the effect described. When power is switched off, the moving components generally restore to 'neutral' position; for the problem to recur when motion is restarted.

 

Thus washers limiting sideways movement can be successful by preventing the components moving into the bind position, but what if you need the sideplay for curves? Brutal solution. Attach your high speed rotary tool chuck to the rear motor shaft and give the assembly some running with a motor that will not falter. Sometimes you can see from the polishing action just where the trouble lay.

[Captain Kernow]

The more close fitting your assembly, the more likely you are to see this; the description italicised is what some describe as a 'wind up'. What's going on can be very subtle and difficult to observe, involving multiple components. All the machined surfaces have an ultrafine screw thread on them, in rotation these act to shift moving components along the axis of rotation. If they happen to reach a point where the various small inaccuracies produce a bind, then you see the effect described. When power is switched off, the moving components generally restore to 'neutral' position; for the problem to recur when motion is restarted.

 

Thus washers limiting sideways movement can be successful by preventing the components moving into the bind position, but what if you need the sideplay for curves? Brutal solution. Attach your high speed rotary tool chuck to the rear motor shaft and give the assembly some running with a motor that will not falter. Sometimes you can see from the polishing action just where the trouble lay.

Very interesting, thanks for that.

 

Unfortunately the 'brutal solution' you suggest isn't possible in this case, as the rear shaft had to be removed to clear the backhead...

 

Anyway, I've removed the rear brass bracket that I'd fabricated earlier, as the running (just the centre axle, 'in the air') wasn't improving. I've now fitted a torque reaction arm, and re-fitted the motor, and the rotation is now back to the original smoothness.

 

When I feel sufficiently motivated, I'll start putting the rest of the chassis back together again and see how it goes. Thanks everyone who has contributed so far.

 

I'm now going to make the tea, which I suspect is going to be a more rewarding experience...

[ullypug]

Are the gears in the drive train concentric? I've heard of very occasional issues with final gears being slightly out of true.

Are both coupling rods symmetrical in regard to wheel centres?

Are the crankpin throws identical on all wheels?

Are the coupling rods fouling the wheel bosses or anything else? 

 

On High Level gear boxes I solder 2 lengths of 0.45mm brass wire on the inside of the main gear box assembly and then solder the other end of the wire 'tails' to chassis spacers. Acts as a torque reaction rod but allows a little flex for sprung/compensated chassis.

[uax6]

Could it be that the when you removed the rear drive of the motor that you somehow twisted that shaft, so it has a slight bow in it? Is the 'bind' only noticable at low speeds?

 

The other issue could be that the rods aren't quite matching in length, that can produce a slight limpy bind.

 

Andy

[Tim V]

Is this the same motor that you were having problems with before? If it is, well the answer is to ditch it... - or at least try another motor.

[Captain Kernow]

Could it be that the when you removed the rear drive of the motor that you somehow twisted that shaft, so it has a slight bow in it? Is the 'bind' only noticable at low speeds?

I hadn't considered that, but I was careful. Interesting notion, though, thanks.

 

One thing that doesn't add up, now that I've re-assembled the basic chassis last night, is that it now runs pretty well forwards, but the slight 'hesitation' now happens in reverse. The motor and gearbox configuration haven't been changed, either.

 

I do think that the torque reaction arm is helping, but I'm going to try washering between the outside faces of the gearbox and the inner faces of the frames next.

 

 

Is this the same motor that you were having problems with before? If it is, well the answer is to ditch it... - or at least try another motor.

 

Yes, it's the self-same one, Tim, but that problem cleared itself up with further running in.

 

Another motor (a 1220 this time) is on order, with a spare worm to fit the 54:1 gearbox, should the above be unsuccessful. The final solution, short of getting the lump hammer from the shed (it's a good job it's been raining so much!!) is to fit the 80:1 High Level LoadHauler+ gearbox and 1220 motor that I've ordered as a back-up. If the original motor and gearbox come good, then there's another project waiting that will use the 80:1 box. The advantage with the new gearbox is that I should be able to mount the motor a bit further forward, leaving just enough room for a small flywheel behind...