Gearboxes that will drive back

[bertiedog]

Adding a little more, Hobbytown of Boston specialised in using 1:1 cross helical drives in their HO Diesel loco transmissions in the 1950's to 60's in the States.

 

These used a column of spur gears from the motor level to connect to a drive shaft with the 1:! gears fitted on the axles.

 

The drive was in theory reversible, but they had problems, which still apply to all reversible drives with helical skew gears.

• The drive shaft has very high end thrust, which wore the washers used, and lead to a lot of play.
• The high thrust pressure wears the gears, brass in this case unless regularly lubricated.
• The spur gears needed very careful setting up, if badly done it could be noisy.

Now they also had a giant flywheel and a centrifugal clutch, so that the motor was disconnected as it stopped, and the whole unit at speed would coast many feet, my own could manage12 feet from high speed, and 2 feet at low speed, down to a couple of inches at shunting speeds with a train in tow, which added momentum..

 

Sounds marvellous, until you realise that stopping on a slope becomes more difficult, and the trains speed up dramatically as they go over the crown of a hill.

 

Hobbytown then dropped the clutch, keeping the motor connected to the flywheel, which cure most issues, but a well run in example was difficult to stop exactly, and could run away backwards with a heavy train when power was cut.

 

But the main problem was wear and tear, the Hobbytown gears did not last long in use, unless properly looked after. The gear were enclosed in grease which helped.

.

The HPC steel gears would last, they have more teeth and could be grease lubricated to ease wear and tear. But it would need very careful attention to side and end thrust movements on the layshaft and the driven axle, any sideplay would cause the whole axle to be driven sideways under power one way and in reverse hard the other way. The driven axle would really have to be set with no side play whatsoever.

 

It really needs ball raced thrust bearings to get the best from crossed helical gears.

 

The ABC designs use far bigger tooth counts and diameter, and do not suffer so much side thrust in operation, but the space problems in 4mm are eased so much in 7mm it makes them a very practical solution. with a good feedback controller then control is good although a heavy train can "push" on cutting power, and may not stop on a slope, but then this is driving skill, not an engineering issue as such.

 

Stephen.

 

,

[bertiedog]

In the ABC design the side thrust of the helical gear is dealt with by having the helical as the first stage of the gearbox, so that side play can be tolerated on the axle gears. This means the first shaft has to be a very precision unit, with races to get the best operation.

 

He also uses 2:1 ad 3:1 crossed helical gears which give a quicker ramp up in ratio, then uses spur gears to multiply the ratio to the required level.

 

If you build a box in this style it may be best to use spur gears first, then use the 1:1 simply to turn the gear drive at right angles, this version would suit 4mm versions better, as it keeps the final gear size down, and the speed of rotation of the gear lower and therefore quieter.

The whole thing can be reversed with the 1:1 skew helical gear at the top , then 2 or three stage spur gears to the axle.

 

Unfortunately the 2:1 and 3:1 gears used in the 7mm ABC box are not made in smaller sizes or would be specials and costly for 4mm, so there is no instant way of using the design, which is good for ABC of course!! and lucky for 7mm users.

 

4mm has been poorly served with precision gearboxes over the years, the Portescap was really the only complete unit, with Exactoscale's boxes the nearest spur types off the shelf.

 

A combination that works well for shunters in 4mm is one of the tiny gearhead motors with say 50:1 gearhead, and simply use the smallest HPC gear to turn the drive at right angles. I have a BR 08 with this design, in theory reversible, but in practice only able to coast a bit, but just enough to overcome bad pickup. It is a bit "noisy" at highest speed range, but silent at crawl.

 

Stephen

[Horsetan]

.... most 4mm modellers will not like spending 150 quid plus on a transmission and motor.

 

They're already well on the way to spending that on eBay every time an RG4 pops up...

[bertiedog]

There's a bit of a difference between MOST, and the very FEW who use a Portescap, at a guess, 1,000 to 1....the average user is a Mashima, gear frame and gears about £30/£35 or so, not the £150 plus that making a 4mm ABC would cost.

 

Stephen.

[bertiedog]

I am not querying any of the ABC design principles, but there is a curiousity of the arrangement in that it uses a Maxon Motor and I asume it is a corelsss motor, so how can the helical skew gears be arranged on the shaft of the motor,as the Maxon has NO thrust bearings,in fact most coreless motors can take no end thrust at all, one of the reasons for the bevel gear in the Portescap design.

 

Does the ABC motor gearbox have an auxiliary thrust bearing in between the motor and the helical gear? The thrust must be very high, due to the nature of the gears, anfd far too high for a coreless type.......so is it coreless or a special coreless with thrust bearings or a conventional motor with thrust bearings?

 

By having the spur gears first, then the helical gears, the problem is eliminated at the expense of having the driven axle with no sideplay.

However you could have a spur to helical middle gear, and then a final spur to the axle ,allowing full sideplay.

This could be made with stock Ultrascale helically cut spur gears for the first pair, and then the HPC 1:1 skew helical, and finally Ultrascale straight cut gears for the final drive. For really low speed a double reduction on the final pair would be best as these will be turning at lower tooth speed.

 

The smaller HPC gear pairs fit easily within a normal 4mm frame space, even 16.5 (00).

 

Stephen

[hollywoodfoundry]

From the photographs, the motor is a 352923, which according to the Maxon web site is a 19mm diameter coreless motor in the A-Max range, of 2.5 watt power with precious metal brushes. The no-load speed is quoted at 9750, which is a good speed for a model railway motor. The site does not say what bearings are fitted to the motor, but being a single ended motor, it is possible that there is a thrust bearing fitted to the blind end of the case. The drawing of the motor shows a fairly substantial part at the shaft end that could be a decent bearing. Maxon do recommend this unit for use as a fan motor, which would require thrust bearings, so perhaps this motor is fitted with suitable bearings.

 

However,putting on my sceptic’s hat for a moment, I am doubtful that such an arrangement would provide any real drive-back in a 4mm scale model, particularly a steam locomotive model. Steam locos have a lot of friction in the axles and rods, plus valve gear if fitted, and with a 26:1 ratio, I doubt that a 4mm scale model would have sufficient adhesion for drive-back to occur.You would need optimum adhesion, using steel wheels on steel track and a lot of weight in the locomotive to make it work. I can imagine it working in a larger scale, say from 7mm upwards, but my feeling is that in a 4mm scale model, the weight of the train would simply cause the locomotive to skid for a distance with the wheels locked.

[Buckjumper]

Cool but think about it, the real thing wouldn't stop that quick if it was having an accident! With a simple dcc decoder you can coast for scale miles if you want to but with a panic button if you suddenly realise you have got it wrong. The other advantage is when the gimmick wears off and you realise that it's a royal PITA on an exhibition layout you can turn it all back off!

 

 

I know this is a 4mm thread, but just to clear this little one up; the reason I use ABC with a helical first stage isn't for the drive back facility (that's just an added bonus), but because they are so efficient. ABC claim 75% efficiency, but I understand independent test show they're actually between 80 - 90% efficient - compare with standard worm/wheel gears which are 30% efficient at best.

 

The benefits of this are enormous, both in a DC set up, but especially so in a DCC environment.

 

This example was cited elsewhere: Brian Clapperton used to refer to a model of a King which would haul 16 coaches and yet was driven by a Mashima 1833 through an ABC box and experienced about 0.3 A continuously. By contrast a Sunset Scotsman with a Pittman used over 1A to do a similar job.

[ozzyo]

Porterscaps do (did?) and they have been arround for decades. They sound pretty horrible though. How quite are these ones? Do they have the same horrible whine?

Cheers

Jim

 

 

Hello Jim / all,

 

as far as I know the Portscaps (RG4s etc.) are no longer produced.

The gearboxes are pretty quite, you get some noise from the spur gears, this will quieten down with use and a bit more lubrication. If we could get skew cut spur gears in these sizes this would cut the noise down even more, but at what cost?

 

OzzyO.

[bertiedog]

With Reference to #31 from Hollywood foundry, yes, the Maxon is rated for thrust, I found several references on the net, and being a high quality unit should stand the high end thrusts from the skew helical gears.

 

If anybody tries a 4mm version however, then a thrust bearing really must be added, as smaller coreless motors do not have thrust bearings, and even with a conventional Mashima the high thrust pressure may wear the main shaft spacers faster than usual.

 

To summarise, a worm gear induces end thrust in the worm shaft, but it is not as high as a 45^o skew helical, which is about 60% or more end thrust pressure.

The gears are more efficient in transmitting power, but some of ABC's excellent efficiency also comes from the precision assembly of their units.

 

The coasting effects from Hobbytown wheel were due to two things, the 1;1 helical gears and the massive flywheel, and would not be possible in steam outline sized mechanisms.

 

It is not so much the reversed running or coasting, but an easing of the strict control of speed that the ordinary gearbox imposes, the better ABC gearboxes respond in a different style, and locos slow under load, and slow on up gradients,,and speed up on down slopes in a more realistic manner, and with a controller with inertia, or DCC, will pull away and stop in a more realistic manner all round.

 

Stephen.

[bertiedog]

Hello Jim / all,

 

as far as I know the Portscaps (RG4s etc.) are no longer produced.

The gearboxes are pretty quite, you get some noise from the spur gears, this will quieten down with use and a bit more lubrication. If we could get skew cut spur gears in these sizes this would cut the noise down even more, but at what cost?

 

OzzyO.

 

It is possible to get skew cut spurs in similar sizes from Ultrascale, but they are not direct replacements. An "own design" can be made from the stock gears they do, with the UIltrascale crown wheel replacing the bevel.

 

The problem with the Portescap is not the spur gears, it is the bevel that makes the noise, along with dry spur gears!, The pinion was often set too tight to the bevel face, which wore it, or it was set slack, and also wore! The lifetime lubrication claims made the whole issue a problem, as the gearbox was open to dust and dirt geting into the whole thing and binding, or causing a lot of wear. Once the spur gears ran dry they wore and the noise just got worse.

 

I had several early units and these were well assembled, but as time went on it was obvious that standards slipped, due to wear in the jigs used for manufacture, and the standard of drilling the sideframes and bearings went a bit slack.

 

The plastic bevel gear finish also got worst as time went on, I was told the cost of a new mould was prohibitive and they used what were at the bottom end of the tolerances on later production.

 

Adding together the slack in the bearings, bevel wear or bad surfaces, plus the open construction allowing dirt in, and lack of any requirement for oiling, you have a noisy and progressively deteriorating motor and gearbox.

 

The cure was easy when new, strip, and rebuild with a cover over the lot, and fill with Labelle grease. The exposed ends of the shafts were covered with shim sideplates, supported by the existing bolts.

 

I have still got four Portescap originals altered in this way, which are still silent and powerful, but other ones from that period have long worn out that have gone through the workshop.

 

The Portescap was a good design, but poorly implemented in practice, no thought for lubrication, and no thought about dirt getting at the gears. In good condition, excellent, but in average hands they could be very poor indeed due to wear and tear rapidly catching up on them..

 

Stephen..

[bertiedog]

................and I would add that the ABC combos could do with a decent overall casing to keep dirt out and oil in!..purely a.personal opinion, not an attack on the design!! It could of course be added by the owner, and in the case of ABC would be worth considering to maintain the condition at 100%

 

Stephen.

[ozzyo]

Hello Jim / all,

as far as I know the Portscaps (RG4s etc.) are no longer produced.

The gearboxes are pretty quite, you get some noise from the spur gears, this will quieten down with use and a bit more lubrication. If we could get skew cut spur gears in these sizes this would cut the noise down even more, but at what cost?

OzzyO.

 

 

Hello Birtiedog all,

 

the reply was to two questions, the first about RG4s

The second about the ABCs.

 

Hope this clears up any confusion

 

OzzyO.

[bertiedog]

No confusion, ........but the ABC do not make much gear noise, I have used several in locos for other people, and with grease,Labelle silicon with PTFE,, they are dead quiet. I don't think putting all skew gears in the ABC would make much practical difference, they are so good already.

 

It would pay with the Portescap though, the gears do whine when dry or just oiled lightly. I think it is the play in the bearings being proportionally a bit looser that means the gears vibrate as they turn, which does not happen on the ABC box.

 

Once a Portescap is worn and dry ,due to not adding lube,, then they are very noisy., the cure is to drill out the steel bearings and fit reamed bronze, silence then returns.

 

Both the ABC and the Portescap should be silent at modest shunting speeds, a whine only sets in at high speeds, but far more on the Portescap, with very little on the ABC.

 

what is wanted is for ABC to make 4mm versions, it must be possible, the market exists, look at the number of Portescaps in use

 

Stephen.

[Guest jim s-w]

 

The benefits of this are enormous, both in a DC set up, but especially so in a DCC environment.

 

 

I fail to see how though? I mean we are not short of power for what we do, Yes it's obviously better but is the advantage a theoretical one rather than one that had any real term benefit?

 

Don't get me wrong, it's an interesting topic in it's own right

 

Cheers

 

Jim

[bertiedog]

I think the reason that ABC do use straight cut on the secondary and final drive is sideplay is allowed. Although sideplay can occur with skew cut, there is the thrust to consider, which as power is applied not only turns the axle, but forces the axle sideways quite hard, reversal will then send the axle hard in the other direction.

None of this matters if the axle can be rigid, but often sideplay is vital to get an all flanged loco round smaller curves.

 

A slight sideways lurch can be induced by skew gears on the axle, as the loco starts up the wheelset with the skew gear is pressed hard one way or the other. This can move the loco sideways in a slight twitch.

 

The reason I know is the 08 diesel I have with HPC 1:1 skew gear drive, the front axle is driven, and at first it twitched sideways as it it reversed, only cured when the play was removed, but this did not affect the 08 diesel as wheelbase is so short anyway.

 

But it would be deeply undesirable on a longer wheel base loco, and the side thrust would show as it could move the valve gear quite visibly on an outside cylindered loco, with the gear on he middle axle.

 

Even the Hobbytown diesel bogie twitched sideways as it reversed, due to the skew gears thrust reaction.

 

None of this applies to the ABC design, it is immune from any side thrust from the skew gears affecting the driven axle, which can have sideplay allowed to very generous limits, another plus point to an excellent design.

 

Stephen.

[bertiedog]

I fail to see how though? I mean we are not short of power for what we do, Yes it's obviously better but is the advantage a theoretical one rather than one that had any real term benefit?

 

Don't get me wrong, it's an interesting topic in it's own right

 

Cheers

 

Jim

The general improvements with a "free rolling" gearbox and adding a decent flywheel is akin to having a shock absorber in the system, sudden changes are evened out, momentum is maintained, or lost etc., at smoother rates,. As gradients are encountered there is a smoother lose of speed, and less sudden burst as the top is driven over. It very definitely works better in DC, and really the same applies to DCC, with extras like momentum adding to the better rolling and response.

 

DCC acts as an electronic shock absorber, but there is still a world of difference between a mediocre mechism being improved by DCC, and an already good mechanism being on DCC, it is simply even better.

 

Stephen..

[Guest jim s-w]

Hi Stephen

 

With a back emf decoder (assuming there is enough power and mechanical grip to get over a hill) there is no change in speed at all. I agree that its simply better is is the difference actually perceptable? Without a perceptable difference and with an increased cost the market is severely limited.

 

Cheers

 

Jim

[bertiedog]

Even back EMF control is only trying to keep the voltage and current stable, and as the load on the motor changes the voltage may remain constant ,and the current is forced to stay the same, but the simple force of gravity will speed up and slow the loco all the same, albeit under far better control than without the feedback,

It is like a car with ordinary suspension damping and a car with active damping, which with active suspension should perfect, but in the real world still bumps over objects.

 

A loco should be as perfect a runner as possible on plain DC before using feedback or DCC, they can't cure bad design, build, or poor engineering, you need a decent gearbox , free efficient running, and good resistance to wear first, the control is secondary

 

A loco with a sound gearbox is a joy to run, it will never be improved by DCC or any electronic control.

 

Stephen.

[Guest jim s-w]

Thats not what i asked Stephen, The question was, is the difference perceptible?

 

Cheers

 

Jim

[Guest jim s-w]

You need a decent gearbox , free efficient running, and good resistance to wear first,

 

 

No i don't. Look at all the rtr stuff out there, hundreds of thousands of models, thrashed to death day in day out and they don't NEED any of that. It's a bit like saying we all need to drive a rolls Royce when clearly we don't. Some people just want something that work within their budget.

 

If you WANT those things that's great and I am chuffed people do but please don't say we need it

 

Cheers

 

Jim

[ozzyo]

If you WANT those thongs that's great

Cheers

Jim

 

 

 

Hello Jim,

 

I don't fancy wearing one of those "thongs" a bit of string up your bum, it don't seem right . .

 

OzzyO.

[Guest jim s-w]

Hello Jim,

 

I don't fancy wearing one of those "thongs" a bit of string up your bum, it don't seem right . .

 

OzzyO.

 

iPad predictive text fail! Sorry

 

Jim

[bertiedog]

Just because some are satisfied, it does not mean that we should all give up seeking better, or should we ask why bother with P4, Trix twin ran well enough? or do we go back to Tri-ang as it ran quite well......

 

The better gearbox will always run better, there's no other position possible. Free running high efficiency boxes or spur gear drives respond better to control and track conditions.

 

I did compare a DJH kit with a delrin chain drive to the ABC, at steady speed, little or no difference, but actual running, pull away, and coming to a stop was smoother, and quieter as well, in this particular case.

 

The running had a "steadiness" the plainer drive did not have, and a slight natural lag as the controller was altered.

 

The controller is a full feedback type with switchable rates and direct control. Even on full feedback there was a perceptible lag as the motor took up the set pace, and although, as it hit a slope, the speed was basically maintained as you would expect with feedback, there was a slight drop and extra effort by the engine was needed to maintain the pace, After the crown of the slope the loco speeded up slightly, not as much as without Feedback on, but more than the DJH chain drive, which stayed the same on the down slope. It was far more like having the control on a real regulator..

 

As the ABC has a Maxon coreless and the DJH a Buhler it is hardly fair to compare the current, but the load was the same, and the ABC was drawing a lot less current than expected under load and slack running. Both managed very slow running on full feedback with PWM., but the ABC was noticably smoother on increase of power from the slow speed, just less effort to chance speed, and with a tiny lag just like the real thing.

 

The difference with the high efficiency box, free running style was that it behaved like the real thing, even with full feedback, which should normally remove the variability, the ABC style box puts back a bit of real locomotive presence into the model loco, not a large amount and very difficult to quantify, but it is just a touch of quality running, and I am sure it is due to the free rolling nature of the design.

 

Stephen.

[Guest jim s-w]

Just because some are satisfied, it does not mean that we should all give up seeking better, or should we ask why bother with P4, Trix twin ran well enough? or do we go back to Tri-ang as it ran quite well......

 

 

Indeed so but then you aren't telling people they need to model p4 are you? It's probably a case of one solution won't fit all. I took my wife to the Severn valley a while ago and the loco that hauled us was far from steady, with a definate pulsing motion when under power. Looking out of the window of the first coach smooth would be the last word I would use to describe the motion of the loco

 

Cheers

 

Jim

[ozzyo]

iPad predictive text fail! Sorry

Jim

 

Hello Jim,

 

I knew what you meant, but it just had to be done .

 

A few posts back you asked was it worth it, well I think that it is. You get a big steam loco with limited space for weight so you want the best out of your motor that you can get.

 

As your modelling A/C electrics, you can power all 4 axles and use all the loco weight for traction, when you think about a steam loco that is say a 4-6-0 you have to try to get the weight on the three driving axles but you need some weight on the two carrying axles as well. So you want to get the best out of the motor that you can, so to me that equals the best gearbox you can get.

The two stage box is going into a Castle that I'm building for a customer so we can test it head to head with another Castle and see how they compare.

 

OzzyO.