Garden Railway Forum

As trains have become increasingly heavy on a steeply graded line a very 1:1 scale problem has become apparent, the need for a working train brake. I'm wondering if anyone has dealt with this problem before, and what solutions did you come up with? The control system I use is micron radios and the motor cars have all axles driven by independent traction motors like the prototype. I was thinking some sort of rheostatic brake using the spare channel on the radio and a relay might be the easiest to implement, though not very controllable. A friction brake using a servo is also an option, though probably more complicated.

I can't offer any help, but look forward to hearing from 'someone who knows'.
Fantastic photo, by the way. It almost looks like a winking face with that offset hooded window.

How about combining the two?

Having a servo that operates a pot. So you have rheostat braking you can apply sloftly or hard.
Maybe you could even fit a pushbutton to be hit by the servo arm at the end of its travel for the emergency short circuit brake.

I have this emergency shortcut brake on my 5" gauge models. Works fine.

A friend also had this fitted to his 5" gauge loso. This loco has 24V geared motors. Even at twice walking pace, if you hit the button, the loco stops right where it is.

For a parking brake:
You could really do with something with an over-centre spring.
This would hold itself, without continuous power, which could be a problem.

By rheostatic braking, do you mean applying a little power to the traction motors? - These will be at stall, so the currents could be quite high. This will heat both controller and motors.

Phil.P

Scrat wrote: ↑Thu Aug 24, 2023 9:46 am How about combining the two?

Having a servo that operates a pot. So you have rheostat braking you can apply sloftly or hard.
Maybe you could even fit a pushbutton to be hit by the servo arm at the end of its travel for the emergency short circuit brake.

I have this emergency shortcut brake on my 5" gauge models. Works fine.

A friend also had this fitted to his 5" gauge loso. This loco has 24V geared motors. Even at twice walking pace, if you hit the button, the loco stops right where it is.

I was kind of hoping that braking might have been a programmable feature on the motor controller, but after reading through all the documentation it doesn't look like it. I'm leaning towards a servo operated friction brake at this stage, it occurs to me that the problem with using the motors for braking is that you really need to have it interlocked to cut out motor power first to remove the possibility of the braking resistors being a dead short across the motor controller. You could do it with a relay or servo operated switch but it starts to get complicated.

Phil.P wrote: ↑Thu Aug 24, 2023 11:35 am For a parking brake:
You could really do with something with an over-centre spring.
This would hold itself, without continuous power, which could be a problem.

By rheostatic braking, do you mean applying a little power to the traction motors? - These will be at stall, so the currents could be quite high. This will heat both controller and motors.

Phil.P

Yeah the ability to hold the train stationary on a grade would be useful, something that can't be done at the moment! By rehostatic braking I mean connecting the motors across a variable resistor, so increasing resistance (and braking force) could be applied. You can do a crude version of this already just by putting it in reverse, but it locks the wheels up straight away as soon as you apply a tiny bit of power and the big current spike wouldn't do the motor controller any good.

I'll do some experiments, there's another motor car in production at the moment so I might try fitting a friction brake on at least two of the axles.

Lonsdaler wrote: ↑Wed Aug 23, 2023 4:55 pm I can't offer any help, but look forward to hearing from 'someone who knows'.
Fantastic photo, by the way. It almost looks like a winking face with that offset hooded window.

Thanks, this is a Melbourne suburban swingdoor set, former loco hauled cars that were converted to electric sets in 1920 and ran in service right up until the 1970's. The reason I'm having this problem at all is the line was originally built as a tramway with tramway grades but recent advances in technology have allowed more and more heavy rail rollingstock to be built.

I think I would try a band-brake, tensioned by a servo.

Simple technology, and modern servos do not need a lot of current to hold their position.
There might even be something adaptable, from the RC truck world?

Phil.P

That's a nice looking model of a 2 car Doggie set.

To be prototypical, Doggies and Taits only had Westinghouse automatic air brakes. The Butterboxes, being locos, had the addition of straight air brakes for the loco only. The first use of regen brakes on the VR electric fleet was the L class I think..... Dynamic brakes appeared in the fleet with the B class and I assume that is what you describe as rheostatic brakes.

A simple parking handbrake would be to fit a manually operated DPDT switch that selects either the ESC or a dead short across the traction motor brushes. The motor cars won't roll far with the traction motors shorted out and it can't blow the ESC, as it won't be connected to anything when the parking brake is selected.

I've not seen any form of electrical braking on commercial ESC units used in garden scale, possibly because most electrically driven loco models in the garden scales use worm gears and have the free rolling characteristics of a brick when not under power. I've used DCC decoders in HO that had simulated braking, but it relied on the worm gear drive to provide the physical effect.

There are controllers for battery locos in the ride in/on scales that are sophisticated enough to have regen braking that can recharge the battery on long downhill sections. They'd be physically too big to fit in garden scale models, but show it's possible.

Designing something like that is well above my pay grade. Do you have any mates that are electronics engineers? About 50 years ago (in the days when a 2N3055 transistor was hi-tech) an EE that I worked with at the time designed a speed control circuit for us that we used for controlling 5" gauge battery locos.

I've seen pics of your powered bogies but I'm intrigued to know the grades you have on your track. My battery powered locos have bevel final drives and won't start to roll until the grade is 1 in 12......

Regards,
Graeme

There are small electromagnetic clutches available, which could be set up to work as a brake. If it was driven with an ESC you could likely get variable braking force out of it. It would probably have to be fitted in an unpowered bogie under a trailer, or driving trailer, as it's about the size of a traction motor.....

This is just an example, there'll be other smaller variants if you search around. From the look of the output gear, this one looks like it might be for use in a printer of some sort.

https://www.ebay.com.au/itm/26621895636 ... gIfg_D_BwE

The issue with trying to drive a brake system with servos is that servos are a variable position device, not a variable force device. Using one to drive a mechanical brake gives you a brake that is either on or off. OK for a parking brake, less so for controlling downhill speed.

Graeme

GTB wrote: ↑Fri Aug 25, 2023 12:11 pm That's a nice looking model of a 2 car Doggie set.

To be prototypical, Doggies and Taits only had Westinghouse automatic air brakes. The Butterboxes, being locos, had the addition of straight air brakes for the loco only. The first use of regen brakes on the VR electric fleet was the L class I think..... Dynamic brakes appeared in the fleet with the B class and I assume that is what you describe as rheostatic brakes.

It's essentially the same thing, in that resistors are connected across the traction motors. For some reason usually called a rheostatic brake on the sparks that have it like the comeng.

GTB wrote: ↑Fri Aug 25, 2023 12:11 pm A simple parking handbrake would be to fit a manually operated DPDT switch that selects either the ESC or a dead short across the traction motor brushes. The motor cars won't roll far with the traction motors shorted out and it can't blow the ESC, as it won't be connected to anything when the parking brake is selected.

That'd definitely be simple to implement, the main problem is I really need it for controlling speed when running especially for heavier trains.

GTB wrote: ↑Fri Aug 25, 2023 12:11 pm I've not seen any form of electrical braking on commercial ESC units used in garden scale, possibly because most electrically driven loco models in the garden scales use worm gears and have the free rolling characteristics of a brick when not under power. I've used DCC decoders in HO that had simulated braking, but it relied on the worm gear drive to provide the physical effect.

There are controllers for battery locos in the ride in/on scales that are sophisticated enough to have regen braking that can recharge the battery on long downhill sections. They'd be physically too big to fit in garden scale models, but show it's possible.

Designing something like that is well above my pay grade. Do you have any mates that are electronics engineers? About 50 years ago (in the days when a 2N3055 transistor was hi-tech) an EE that I worked with at the time designed a speed control circuit for us that we used for controlling 5" gauge battery locos.

I've seen pics of your powered bogies but I'm intrigued to know the grades you have on your track. My battery powered locos have bevel final drives and won't start to roll until the grade is 1 in 12......

Regards,
Graeme

The grade averages about 1:20, with some short stretches slightly steeper. With power shut off the two car set rolls freely on the straight sections but the friction on the tight curves stops it reaching a high enough speed to derail. Adding any more unpowered cars would be a different story though! I haven't had a chance to test what the maximum load for one motor car in the uphill direction is yet but it is more than two unpowered cars and I'd want to run the same train back down again safely. I'm not much of an electronics person either so it seems like a mechanical friction brake is the easiest option to investigate. I've got a prototype under construction, I'll see how it goes. It also has the fun nerdy aspect of being a miniaturized feature of the real thing, which of course is why we're all here.

GTB wrote: ↑Fri Aug 25, 2023 1:02 pm There are small electromagnetic clutches available, which could be set up to work as a brake. If it was driven with an ESC you could likely get variable braking force out of it. It would probably have to be fitted in an unpowered bogie under a trailer, or driving trailer, as it's about the size of a traction motor.....

This is just an example, there'll be other smaller variants if you search around. From the look of the output gear, this one looks like it might be for use in a printer of some sort.

https://www.ebay.com.au/itm/26621895636 ... gIfg_D_BwE

The issue with trying to drive a brake system with servos is that servos are a variable position device, not a variable force device. Using one to drive a mechanical brake gives you a brake that is either on or off. OK for a parking brake, less so for controlling downhill speed.

Graeme

Great find! If my servo operated friction brake is unsuccessful this seems like it would be the next thing worth a try. Good point on the variable position not variable force, I'm hoping a spring between the servo and brake linkage will work around that problem.

As it turns out this problem had a rather mundane resolution. I was in the process of designing a servo operated friction brake, but before I had a chance to test it in service I ended up cooking the motor controller on 8M, the first completed motor car.


I lamented my bad luck and replaced the dead Deltang RX65 with the only spare I had, a micron MR603B. As it turns out this motor controller seems to have braking as a built in feature, the speed remains very steady in varying load conditions including while running down hill. Some sort of friction brake might still be useful as a parking brake, but the importance for normal running is much reduced.

Melbournesparks wrote: ↑Fri Sep 29, 2023 10:25 am I lamented my bad luck and replaced the dead Deltang RX65 with the only spare I had, a micron MR603B. As it turns out this motor controller seems to have braking as a built in feature, the speed remains very steady in varying load conditions including while running down hill. Some sort of friction brake might still be useful as a parking brake, but the importance for normal running is much reduced.

Interesting. Andy at Micron is probably the only person that could confirm if the MR603 is programmed to use some form of dynamic braking.

The MR603c seems to have feedback control at least, as the one in my Motor Mule has very good low speed control and more or less ignores both up and down grades.....

For what it's worth. The motor driver ic used was originally designed for automotive use controlling the throttle in drive by wire road vehicles. According to the ic manufacturer's docs it can do dynamic breaking (sic) as well as closed loop control, among other tricks.

Graeme

GTB wrote: ↑Sat Sep 30, 2023 2:37 pm

Interesting. Andy at Micron is probably the only person that could confirm if the MR603 is programmed to use some form of dynamic braking.

The MR603c seems to have feedback control at least, as the one in my Motor Mule has very good low speed control and more or less ignores both up and down grades.....

Phil P. of RC Trains may also know. He'll no doubt be along fairly soon.
He did say to me a couple weeks ago that there were a few bits to sort out with the 603C, so I wonder if this is one of them?

Melbournesparks wrote: ↑Fri Sep 29, 2023 10:25 am As it turns out this problem had a rather mundane resolution. I was in the process of designing a servo operated friction brake, but before I had a chance to test it in service I ended up cooking the motor controller on 8M, the first completed motor car.


I lamented my bad luck and replaced the dead Deltang RX65 with the only spare I had, a micron MR603B. As it turns out this motor controller seems to have braking as a built in feature, the speed remains very steady in varying load conditions including while running down hill. Some sort of friction brake might still be useful as a parking brake, but the importance for normal running is much reduced.

Good and bad news .......
Glad you found a solution, a pity you had to fry an Rx65 to get to it though ....

Rik

A generational change in technology..
The motor driver on the Rx65, did not even have a way to monitor back-emf, there is a lot more 'intelligence' in the new receiver/controller.

There is plenty of memory left, and more features will be added over time.

Phil.P

ge_rik wrote: ↑Sat Sep 30, 2023 4:54 pm
Good and bad news .......
Glad you found a solution, a pity you had to fry an Rx65 to get to it though ....

Rik

Yeah that was a pity, especially since I'm not sure exactly what caused it.



The only visible damage was this component that got hot enough to burn through the heatshrink, the black colouring there is paint to make the leds less bright. It was just operating normally up until it suddenly went dead. The application here has always been a bit off label, with power supplied by overhead wires instead of a battery, so despite the capacitors and voltage regulators it is more prone to interruptions. This was also the first run in a while where there can be a lot of arcing off the pantographs that might cause some stray voltages somewhere.

Phil.P wrote: ↑Sat Sep 30, 2023 10:00 pm A generational change in technology..
The motor driver on the Rx65, did not even have a way to monitor back-emf, there is a lot more 'intelligence' in the new receiver/controller.

There is plenty of memory left, and more features will be added over time.

Phil.P

Interesting, good to know!

A little more on this:

The motor controller is programmed to hold, rather than be 'off', when you return the speed control to zero. - This can be altered in programming.

One thing I have noticed, compared to the Rx65, is that the speed-change is a lot more responsive (instant) this is something I have discussed with Andy, and I think I will have a little inertia, programmed into the RC Trains configurations? - Gentler on gears and drive-trains.
This is also something that can be altered in programming.

Phil.P

Phil.P wrote: ↑Tue Nov 07, 2023 10:13 pm A little more on this:

The motor controller is programmed to hold, rather than be 'off', when you return the speed control to zero. - This can be altered in programming.

One thing I have noticed, compared to the Rx65, is that the speed-change is a lot more responsive (instant) this is something I have discussed with Andy, and I think I will have a little inertia, programmed into the RC Trains configurations? - Gentler on gears and drive-trains.
This is also something that can be altered in programming.

Phil.P

Sorry for the late reply, good to know. Another interesting property I noticed compared to the Rx65 is the braking effect still works when a motor car is being hauled dead, with no power to the motor controller. It rolls freely at first, but then I assume the voltage from the motors powers the controller and causes the braking to cut in? The wheels briefly lock up, but then of course the generated voltage disappears and it rolls again. It's not a major issue, I'll just rewire the main power switch for the receiver to break the traction circuit as well.

There isn't a lot of information out there on these kind of edge cases, like braking, hauling powered units dead, how to configure multiple unit trains with motor cars facing in opposite directions etc so it has been an interesting voyage of discovery.