Binding Gearbox

[dal-t]

It has always surprised me that fold-up frames came so late and have been taken up by so few designers.

This still isn't helping Gareth much, but I've always assumed that was because it's commercially crippling - don't fold-up frames have to be Gauge-specific, so if you go for any one of the 4mm Gauges you're foregoing sales to people working in either of the other two?

[Tim V]

The history of chassis construction is mildly interesting, and I could certainly contribute, but it is off topic for this thread which is supposed to be about binding gearboxes. Perhaps the extras could be split off?

[Will L]

The history of chassis construction is mildly interesting, and I could certainly contribute, but it is off topic for this thread which is supposed to be about binding gearboxes. Perhaps the extras could be split off?

I agree and given Bill has for once chosen to honour us with more than his usual pithy one lines, I feal he deserves a full answer. But this isn't the place. I am in the process of starting a new thread else where, but as it will take a little while can anybody who would like to add there two penny please await the new thread. Which is here

[garethashenden]

I picked this back up a couple of days ago. With the motor removed I worked on the chassis and found a couple of spots that were binding. The connecting rod bush on the rear axle needs to be a long bush rather than a standard one as the rod is quite thick. I also slightly enlarged the holes in the front of the rear rods. After this the chassis rolled quite freely on both the workbench and layout. I replaced the motor today and was full of hope, but it is pretty much the same as it was before. It runs fine at full throttle, but badly at low speeds. I took a few videos of it running up against the bufferstop.

The first one is speed step 20, then increasing to 30 and then 40.
[youtube]https://youtu.be/kmeehEzrg3I[/youtube]

The second one is speed step 10. You can see that there are two spots it binds. It can get through the first, but not the second.
[youtube]https://youtu.be/zWtt5I_QDVk[/youtube]

For the final one I removed the rods from the rear axle so that it can run on its own. Again, starting at 20 and going up to 40, then down to 10 and finally 1. It stops at the end, but it still has power.
[youtube]https://youtu.be/xh6RUHGQcmM[/youtube]

It's really looking like I need to take the gearbox apart and examine each section in turn. I'm hesitating because of the Gibson wheels, but it seems to be the next step to take.

[garethashenden]

I picked this back up a couple of days ago. With the motor removed I worked on the chassis and found a couple of spots that were binding. The connecting rod bush on the rear axle needs to be a long bush rather than a standard one as the rod is quite thick. I also slightly enlarged the holes in the front of the rear rods. After this the chassis rolled quite freely on both the workbench and layout. I replaced the motor today and was full of hope, but it is pretty much the same as it was before. It runs fine at full throttle, but badly at low speeds. I took a few videos of it running up against the bufferstop.

The first one is speed step 20, then increasing to 30 and then 40.


The second one is speed step 10. You can see that there are two spots it binds. It can get through the first, but not the second.


For the final one I removed the rods from the rear axle so that it can run on its own. Again, starting at 20 and going up to 40, then down to 10 and finally 1. It stops at the end, but it still has power.


It's really looking like I need to take the gearbox apart and examine each section in turn. I'm hesitating because of the Gibson wheels, but it seems to be the next step to take.

[John Palmer]

I get the strong impression that there is significant fore and aft movement of the driven wheel nearest the camera. This is particularly noticeable in Video 1, in which I estimate the range of movement to be of the order of .015" to .020". It's not so bad in Video 3, but still apparent - perhaps in the .005" to .010" range. There may also be some lesser fore and aft movement on the middle wheelset.

Assuming there's no slop either of axle in axlebox or axlebox in horn guides, it would be worth establishing whether the wheel is eccentrically mounted on the axle. If it is, a new wheel with concentric bore would be in order. But if there's slop in axle or axlebox (or both) then that needs to be cured. In the worst case there 's a combination of eccentric wheel(s) and excessive slop. If the visible fore and aft motion isn't eliminated I fear this chassis will be a source of continuing frustration.

[David Knight]

Gareth,

From watching the videos it strikes me that the gearbox is just fine and the bind is elsewhere in the mechanism. There seems to be a bit of a 'squirm' showing up in video 3 which suggests an eccentricity in the rear driver set. If this is true of any of the others that may well be your problem. Have you tried viewing from directly above? I had a loco which misbehaved in a similar fashion, it took replacing wheel sets, side rods and correcting the position of one of the hornblocks to get smooth running but it did finally work.

Cheers,

David

[Philip Hall]

Clearly it isn't the gearbox, it runs just fine and the ever so slight hesitation is down to the fact that the wheels are sliding on the rails rather than revolving in the air. The wheels on this axle are also concentric. However, the chassis does wobble quite a bit in the first video; all chassis will tend to do this to some extent when run 'against the blocks' or on a rolling road, but not this much, which points to eccentric wheel(s).

I would rather see the flywheel a little closer to the motor bearing - should there be any slight eccentricity here it will be exacerbated by the flywheel being where it is. Also, if the coupling rods are so thick it would be preferable to slim down the boss a little so a standard bush has sufficient clearance. The bush you have used there is the usual one for a centre axle with a connecting rod, and possibly this may actually be too long, so the rod is moving out and in as the engine moves.

I rather think that the problem lies with the rods not matching the axle centres and I would check this to start with. However, I would first check that the crankpin bushes actually have their holes exactly in the centre, sometimes you get one that doesn't. Also check that the crankpins are at right angles to the wheel face. I wouldn't worry about taking the wheels off just the once; it's doing it several times that loosens the grip on the axle.

Philip

[Philip Hall]

Just another thought. Although I'm not a DCC user myself, the chips being added by the new owners when they get their engines, I have always understood from others that it is best to test an engine on plain DC first. I believe tuning is possible with DCC, but it's better to start with a decently smooth running chassis first.

Philip

[Jol Wilkinson]

Gareth,

pushing a chassis won't always show up a problem. Power isn't being transmitted by the rods, the wheels are moving them.

Vid 3 shows a small hesitation at one point so I don't think the gearbox is entirely fault free.

As the big "lurch" happens with the rods fitted , but not with the rear coupling rods disconnected, then I would suggest that it's a wqheel quartering or rods/axle centre mismatch. Try fitting only the rear rods only and try. If no good, that's where the problem is. If okay fit the front rods. If the problem returns, that's where to look. Of course it could be both sets of rods, so sort the rear set and then the fronts.

I hope this helps.

Jol

[Will L]

I'm pretty sure the problem relates to the rods not the gearbox. The kick so evident from the motor in vidio 1 happens when the motion goes stiff and the motor try to revolve around the axle instead. If the gearbox was the problem that wouldn't happen.

When rods bind because the axle centres and the rod centres are slightly different they do so when the rod tries to go past the wheel centre line. Vidio 2 suggests it doing this on both side, although I'd quite like to see the same thing from the other side, and running the other way to be sure.

Like Jol I would suggest you try running the chassis with only half the rods on (same half both sides) as it is quite possible the error only applies to one half part of the rods and it helps to tie down which end it is.

[Andy W]

Also you could try removing the crank pin bushes. By doing this axle by axle it should show if there's a problem with one set of rods.

[Allan Goodwillie]

Andy is right in that it can be a problem with the bushes and crank pins, the throw can be marginally out on one wheel or more which will only show at one point during the turn of a wheel. It only takes one crank pin to be out. Another fault can be that the bush is slightly short for the thickness of the rod boss and if the nut is tightened too hard there will be binding. Just slackening the nuts one by one and turning it over will show that this may be the problem, eventually the nuts will fall off, but by then you will have eliminated the culprit if this is the problem - Will suggesting separating the two coupling rods is useful unless it is the middle axle's bush that is binding, as you may simply think that it must be OK just because it turned OK when fitted to the drive with one coupling rod missing as it would only show as a problem when both couplings are fitted.

There are ways of making sure that these causes can be eliminated during the construction stage. More difficult to eliminate once built, however deeper bushes can be used, but it is always easier to reduce slightly the thickness of the coupling rod bosses until the bush stands slightly proud leaving a working clearance when the nut is tightened.

Checking the crank throw is the same on each wheel is more problematic, a small square can show up if the crankpin is off and needs to be set at 90 degrees. Carefully tweaking with a small pair of pliers can help. If the throw is out it can be checked using dividers. Difficult to eliminate at this stage - most modellers will simply open up the holes in the coupling rod, although this is like introducing wear into your system to get it running, although it will allow the engine to run OK.

(To eliminate such problems occurring in the first place, I cover this in my starters guide to building chassis. Much of it is to have a system of checks made during construction using simple jigs that can be made at home. This thread has shown just how difficult it can be to track down where a problem is once a chassis has been built and having a system of checks that work as you go through the stages of construction I think is essential. The idea of putting it in the starters section is to reduce problems like this and help those starting to get a successful outcome, but I have discovered that there are far more modellers who have been working for years reading it and making use of it in their projects.)

I and the others too I am sure could make a list of where faults occur and their symptoms as well as solutions. My thread on chassis construction does cover this so I might add just such a list if you think that would be useful? Maybe the information is available in a compact form elsewhere?

[Le Corbusier]

Andy is right in that it can be a problem with the bushes and crank pins, the throw can be marginally out on one wheel or more which will only show at one point during the turn of a wheel. It only takes one crank pin to be out.

Allan,

Do you cover in one of your threads a methodology to ensure that the positioning of the crank pins are identical on each wheel. On my barney the AG wheels I got had no pilot for the cranks, and I did the best I could with square and dividers. However I have to admit to being guilty of the

- most modellers will simply open up the holes in the coupling rod, although this is like introducing wear into your system to get it running

gambit on two of the rods as a result.

Tim

[garethashenden]

Thank you all for you help. I checked all the crankpin bushes and they appear to be concentric, at least to the eye. Four of the crankpins were crooked, one in two directions. These have now been straightened. I stuck twp paint tins on the chassis with some blutack to give it a little more weight and it is much better. There are two things that I have noticed that are a cause of concern.
First, both wheels on the middle axle aren't square to either the axle or to the chassis. They both had crooked crankpins and appear to be roughly parallel to each other, so that may have been a problem with the way they were pressed on. Hopefully that can be fixed.
Second, the locomotive goes about half the speed backwards as it does forwards. If all I change is the direction, it seems noticeably worse when running in reverse.
The motor kick is much less than it was, although it will still do it when up against a bufferstop. I filmed it running back and forth, once it has uploaded to YouTube I will post it here.

[garethashenden]

And video!

[dal-t]

T
Second, the locomotive goes about half the speed backwards as it does forwards.

You realise this requires a re-write of the fundamental laws governing the universe? Until now, every chassis ever built has run faster in reverse than in forward gear!

[Tim V]

I would agree that it looks like a mismatch between coupling rod and wheelbase centres.

Fear not!

Suggest you look at Snooze 142 - article on John Brighton's method of getting it right - I've used it with great success. No major dismantling needed.

Edited to add - you might have cooked the motor/chip running it in the way you show in those videos - could explain the mismatch in speed.

[Allan Goodwillie]

Hi Gareth and Le Corbusier,

You will find a simple adaption of one of my jigs in the "Livingston starters Group" section 3 which allows you to use it as a test rig for testing your chassis running upside down and the other way up allows you to see how best to load your chassis and trial it on track. The original construction of the jig which is also used to make your coupling rods accurately you will find in the first section of the "West Scotland Group Build a loco" section.

viewtopic.php?f=39&t=666

Le Corbusier what you are asking about can also be found on this thread - in fact if you follow the threads you will be able to build accurate models straight off after a small investment both in time and cash. The Livingston lads have all made the jigs and are using them. The cost per person has been about £12 and a couple of evenings work each to make the jigs. It is worth looking at the Livingston Starters Group thread as it covers the use of basic tools and some basic information on how to use them.

It also has a look at various common types of motor and gearbox combination and what they are best for as well as how to put them together accurately, also what advantages each has as well as some disadvantages. This is not stuff to do with the two localities, it is more to do with using a system of construction which has been worked out logically. There is a lot to take in, but if you start at the beginning it leads on to an organised way of building, in this case from scratch - all the way through to building the bodies and painting - everything in fact. There are three "West Group" threads altogether which can be found in the section for beginners - but more widely used as can be seen by the number of hits for each section - clearly there are a lot of people using the information.

What Tim has mentioned in his latest post gets eliminated as well in the process.

You will realise when you look at the thread, that I have put many hours into the posts as I realised that when I was doing the courses it was possible to photograph each stage and write it up as we went along. It allowed the others who were building the locos to look up what we covered at home, so that they could come along to the next meeting a stage further on.

I am now running the second course and using the same thread for the Livingston starters group so they too may have a good loco working within a reasonable time. This group are not scratch building locos, but are building kits and finding that the equipment and the system works with a wide range of locomotives. Some of their questions also appear as I have been trying to get them into doing this - there is no such thing as a stupid question from my point of view, for a beginner it may all seem that building an accurate model is like some sort of alchemy, whereas it is simply logic, some common sense and years of experience building many models that has led to a system that gets good results and eliminates problems as you go along.

Although there is some theory - everything is by example and can be seen just how it works.

We have had a short discussion about building a working chassis in a day elsewhere on the forum and it is possible using the simple jigs which I use.

[garethashenden]

Gareth,

Vid 3 shows a small hesitation at one point so I don't think the gearbox is entirely fault free.

Jol

I agree that the gearbox is not innocent. Having come to that conclusion I just took it apart and found a problem. I'm sure there's a better way to describe this, but lacking the correct term, this gearbox has a straight section and a dangly section. This was held in place by superglue, a technique I have used successfully in the past. Unfortunately, the joint had failed this time allowing the gearbox to flex in the middle. I have corrected a couple of minor problems with rods/crankpins so hopefully a rigid gearbox will result in a smoother chassis. I'll clean it up and solder the two parts together.

[Enigma]

To obviate the 'nodding donkey' phenomenon with the motor and gearbox torque reaction I do the following.

Bagnall Running Chassis.1.JPG

Bagnall Running Chassis.2.JPG

A small half round recess is filed into the chassis sides and the gearbox and some fine brass tube soldered across both . The tube is cut away between the gearbox sides and between the frame and the gearbox side and a length of close fitting wire inserted. This holds the motor/gearbox assembly in its 'correct' position. In the photo a small drill is in temporary use while testing. I have only used this on compensated chassis where the driven axle is fixed. A sprung axle might need some other idea.

I don't know if this idea will help with the problem in hand but it might be of use somewhere!

[garethashenden]

Thanks, I've done something similar but less well executed on a compensated chassis before and it works well. Unfortunately, it doesn't work on a sprung chassis as it effectively locks the driven axle in place. The brass wire torque strap I have fitted here works well under normal circumstances, but this situation was too much for it.

[garethashenden]

I'm pleased to report that the flexible gearbox was the source of the problem. With the gearbox carefully reassembled, the axle reassembled, and everything reinstalled performance was much improved. There was still a bit of cogging at very slow speeds however. It then occurred to me that I have heard the 1220 described as the worst motor Mashima make and this is a 1220. I have another North London locomotive project currently scattered around a box. It is fitted with a High Level RoadRunner+, also 54:1 but with a drive stretcher and, of interest here, a Mashima 1432 motor. Since the drive is the same on both gearboxes I was able to easily swap the motors and this produced a very smooth mechanism that I am quite happy with. The only downside is that the 1432 is too long to fit inside the tanks. I think the best solution is to fit a 1428 which will fit in the available space.

Thank you all for you help, I have learned quite a bit along the way. Further progress shall be documented in the Bow Works thread. viewtopic.php?f=128&t=5075

[Will L]

Which all goes to show is that what works at this game is perseverance.

[Will L]

Thanks, I've done something similar but less well executed on a compensated chassis before and it works well. Unfortunately, it doesn't work on a sprung chassis as it effectively locks the driven axle in place. The brass wire torque strap I have fitted here works well under normal circumstances, but this situation was too much for it.

And it won't work om a compensated chassis which dosen't have a rigid driving axle either. What is required is a "Torque reaction link" which we have discussed before most effectively in this thread Torque Reaction Links - Suspending The Driven Axel