There is a posting about a runaway engine which is controlled by a DCS system.  MTH uses a tachometer to monitor the rpm of the motor. If this tachometer fails the engine will attempt to accelerate to the DCS controller "speed" setting. As a result the engine rpm will ramp up to the maximum speed as determined by the track voltage. This could be a problem for any control system that uses a tach tape and tach reader (sensor) to set the motor rpm.

 

One way to ensure that the engine can never run at an excessive speed is to install diodes in series with each motor to limit the upper voltage that can be applied to the motor.  I routinely add up to 8 diodes in series with the motor wiring for each "can" motor.  I actually add two sets of diodes (one for each polarity) for each motor.

 

The result is that the maximum voltage which can be applied to each motor is reduced by about 5 volts. However the voltage applied to the control boards in the engine is not affected. With 18 volts applied to the "track" the maximum voltage which can be applied to each motor is limited to about 13 volts.

 

Note: This maximum voltage setting is the same whether the engine is operated under DCS command control or under conventional control.

 

I routinely add these diodes to my Williams engines since I want to have the headlights and sound system operating at all times. I tap off the "diode string" to provide constant intensity lighting with directional control of the headlights as well.

 

 

 

 

 

 

Original Post

I would not recommend this for a PS-2/3 engine.  If a engine has a failure and takes off, this can occur on Lionel engines also under certain circumstances, there is the emergency stop button.  For conventional operator just roll the throttle back.  Your probably not at full voltage anyway.

 

As far as placing them on a PS-2 engine, the command speed will still not be attained and the system will apply more power via pulses to the motor.  This may effect the integrated sounds and smoke output also.  Your also going to start over working the motor FET.

 

Use of diodes for conventional engines with basic reverse unit is fine.  Placing them on DCS or Legacy systems would be another matter.  You are also going to mess with the scale speed and matched speed accuracy that Legacy and DCS are known for.  G

Why did you feel the need to post this "tip" twice?  I'm not sure it is all that good an idea anyway, but it sure doesn't need to be presented twice.  As GGG says, since the tach is determining the voltage, the logic will just keep cranking up the power until it peaks.  I don't know what effect this has, other than limiting the full speed and power of the locomotive, but it's certainly not a good idea.

The way the DCS system adjusts the speed is to provide whatever voltage is needed to the motors to maintain RPM as detected by the tachometer system.  As more cars are added to the train or as the engine approaches a steep incline the motor would naturally slow down (RPM would go down).  When the DCS system detects a reduction in RPM it compensates by increasing the voltage until the RPM is reached.

 

For any given engine the final track speed is dependent on the engine gear ratios as well as the motor RPM. MTH considers all these factors when calibrating the speed for a particular engine.

I'm fully aware of how speed control works, the discussion is about the value of adding diodes to reduce the motor voltage.  IMO, it's not a worthwhile addition.  It's a solution to a problem that really doesn't exist.

Another albeit rare occurrence is a stuck-on motor FET causing a runaway.  So if the operator is not near the kill switch or is distracted - perhaps from the cheering at a show when the kids see a potential Casey Jones event. A related "problem" which occasionally comes up is to limit operational speed where the anecdotal scenario is handing the controls to a grandson.

 

A possible option is to limit the track voltage with a variable source which eliminates internal engine modification (diodes).  I occasionally run the TIU/DCS at, say, 13V in command mode and have not had any problems.

 

A related application of binary speed limiting using internal diodes is to automatically stop an engine in DCS command mode at a red signal or a bridge-up.  So track voltage at the controlled block is switched to 10V (or whatever) which keeps the PS electronics (sounds, lights) alive but the diodes starve the motor and the engine stops.

 

I have occasionally pondered the speed limiting/stopping "problem" and believe that another viable alternative (for the DIY-er) is to add some kind of circuit powered by the voltage going to the DC motor itself (to simplify hookup) that will electronically clamp, cut-off, taper, whatever the voltage to the motor.  So rather than "dropping" the voltage with diodes that burns power (10 Watts using the OP's scenario of 5V drop at 2 Amps), it would use FETs in some fashion that would eliminate the drop.  Of course rather than only $1-2 of diodes it might be $5-10 per engine.  So in the spirit of a forum to exchange ideas, if there is sufficient interest in this I'll throw my hat in the ring...

 

 

Without intelligence to drive this "cutoff" circuit, you'd simply be limiting your pulling power for the 99.99% of the time there is no problem with the electronics.  I guess I'm not willing to do that.  If you limit the voltage so a bare engine would not derail, having half a dozen passenger cars would result in a very sluggish consist.

 

If I really was concerned about this issue, I'd simply use a remote control button on a lanyard around my neck that could simply remove track power.  This solves it for all locomotives and doesn't require I take apart and modify 30-40 locomotives.

 

I've seen tremendous push-back about a suggesting of adding a TVS across the pickups of the locomotives because it requires you open them up and do a simple mod.  There's a mod that would probably be of greater benefit in protecting the engine electronics.  I'm sure there'd be more push-back for a suggestion that actually compromises the performance of the locomotive to solve some very remote failure possibility and requires even more modification work.

 

Obviously, I'm not interested in doing this, but I have no problem with people coming up with a "solution", I just don't know there's a problem to solve.

 

 

I bought a drag racing parachute for my car in case the throttle stuck, the brakes failed, I couldn't shift into neutral, and the ignition switch failed:-)

 

How about software the knows what speed is ordered, and when voltage is ramped up to full to compensate it realizes with still no speed change that something else is wrong and commands the engine FET to a preprogrammed slow speed voltage setting.  MTH has to do this.

 

Or a device the records tach output and if it goes to zero opens one of the motor leads (or inserts enough resistance to effectively induce a slow motor speed since no tach signal means full voltage to the motor).  This seems simplest since any tach failure, wire break would kill tach output.  Once sensed protection is enabled.  Stan or John could built this circuit.  I just need to patent the idea in case it is a hit! :-)  G

Originally Posted by GGG:

Stan or John could built this circuit.  I just need to patent the idea in case it is a hit! :-)  G

Why do I not like this idea?

 

Truthfully, the tach sensor would be pretty easy.  Monitor the motor voltage and if it goes past around 4-5 volts and the tach isn't running, shut things down.

Originally Posted by GGG:

I bought a drag racing parachute for my car in case the throttle stuck, the brakes failed, I couldn't shift into neutral, and the ignition switch failed:-)

 

How about software the knows what speed is ordered, and when voltage is ramped up to full to compensate it realizes with still no speed change that something else is wrong and commands the engine FET to a preprogrammed slow speed voltage setting.  MTH has to do this.

 

Or a device the records tach output and if it goes to zero opens one of the motor leads (or inserts enough resistance to effectively induce a slow motor speed since no tach signal means full voltage to the motor).  This seems simplest since any tach failure, wire break would kill tach output.  Once sensed protection is enabled.  Stan or John could built this circuit.  I just need to patent the idea in case it is a hit! :-)  G

Be nice if the system was designed so that if communication was lost with the tach reader,the unit would simply shut down. That would be an intelligent design.

 

Dale H

 

Dale,  I remember a few previous discussions on the forum about whether the train should stop or continue.  Some would not want an engine stopping in a tunnel or inaccessible part of their layout.  Others might.

 

On a serious note, adding safety feature that mean more code, or hardware can just drive up the price and complexity.  Both Lionel and MTH provide the kill switch, or a simple direction change brings this casualty to a stop.  For conventional operators rolling the throttle back is all it takes.  G

Originally Posted by GGG:

Dale,  I remember a few previous discussions on the forum about whether the train should stop or continue.  Some would not want an engine stopping in a tunnel or inaccessible part of their layout.  Others might.

 

On a serious note, adding safety feature that mean more code, or hardware can just drive up the price and complexity.  Both Lionel and MTH provide the kill switch, or a simple direction change brings this casualty to a stop.  For conventional operators rolling the throttle back is all it takes.  G

G

 

I guess it is just my background when I serviced meters and electrical instruments. A system would be 3 to 15 or 5 to 25 ma for example for feedback. In calibration for a 3 to 15 system, 15 ma may mean a valve,damper,or flow measure etc etc is closed or off, 8 ma may mean it is half open and 3 ma may mean it is wide open. However a signal of zero ma would be a system error and could be programmed to shut the system down until it is fixed. This could have been put in the system I think at no extra cost, perhaps even programmed in with software. Really a sloppy design with lack of foresight. In an industrial setting such a system could be a disaster. I would hate to see a steam turbine run away with no safety shutoff for example.

 

Dale H

We have to remember these are not life threatening items, so the remote risks of a runaway probably don't justify the development expense of adding a limit feature.  I agree that it should be possible to do this in software, though it's not quite as simple as one might imagine.  The amount of current to get the locomotive moving at the commanded speed can very widely, depending on the load and the grade.

Originally Posted by gunrunnerjohn:

We have to remember these are not life threatening items, so the remote risks of a runaway probably don't justify the development expense of adding a limit feature.  I agree that it should be possible to do this in software, though it's not quite as simple as one might imagine.  The amount of current to get the locomotive moving at the commanded speed can very widely, depending on the load and the grade.

Not that complicated. If the tach reader is not being read the system shuts down with an error message.

 

Dale H

It's not that simple either Dale.  The tach reader is not active when the engine is stopped.  And, the only way to know the engine is not stopped is by looking at the tach reader output!  The motors may draw considerable current before the engine starts moving, depending on the load and grade.  Remember, this has to work in the most extreme situation, say 35 cars going up a 3.5% grade, etc.

Originally Posted by gunrunnerjohn:

It's not that simple either Dale.  The tach reader is not active when the engine is stopped.  And, the only way to know the engine is not stopped is by looking at the tach reader output!  The motors may draw considerable current before the engine starts moving, depending on the load and grade.  Remember, this has to work in the most extreme situation, say 35 cars going up a 3.5% grade, etc.

It should be,that is the problem,poor design. there should always be positive communication to verify the system is working.

 

Dale H

If you're saying they should think of every eventuality, however remote, and design for it, I'll have to disagree.  I'd rather not pay $5,000 for an $800 locomotive!

Runaways haven't been a real hazard on my layout in the 10 years+ that I've had DCS.  It's easy to teach grandchildren the value and purpose of the red button, and they've never been reluctant to use it.

When running several trains together with both remotes, my grandson loves to punch the red button and stop everything.  I can tell he's getting bored when he keeps doing it time after time, that signals the end of train time.

Post
The DCS Forum is sponsored by
OGR Publishing, Inc., 1310 Eastside Centre Ct, Suite 6, Mountain Home, AR 72653
330-757-3020

www.ogaugerr.com
×
×
×
×
Link copied to your clipboard.
×
×