Speaking of cooked motors, "pulse power", etc.:

BA:

It looks like the Cooler Crawler could do most of that, actually:

You could probably do the same with any simple transistor momentum throttle, too.

Cordially yours: Gerard Pawlowski President, a plywood world with dime store trees.

Bernhard Agthe skriver:

Try....

But thee is no need to change that....

Klaus

That's what I'm afraid of ;-)

Good idea, see, if you ask you'll get good answers! Thanks. I'll look into it.

Well, going digital is not "boring" - as I said I have a few locos that might prove a challenge ;-) I'm not even sure how to open them, yet.

On the other hand, for a digital system I would likely spend a few hundred bucks and then I'll still be able to drive two concurrent trains only. And then, my control panel is really great in usability, you wouldn't get that with a DCC system: all switches are controlled by a touch-screen!

;-) You see, I seem to love challenges ;-)

Thanks!

Most of that, yes. It still uses the reverse polarity feature, so I'd still have to use two transformers...

There's a rather interesting page with lots of different throttle ideas at which has several variants doing most of what I want, but none doing it all ;-)

But then I still need the "fire extinguisher" button ;-)

Thanks!

BA:

Hmm...if you mean, can you build several throttles, powered by one transformer winding, to control multiple blocks in a common-return setup, you can't...they must be separate windings, or you'll get a short. If you find a transformer with two separate secondaries, this will work.

If you mean you are using a system that runs half the locos on positive voltage and half on negative voltage, with the common ground at 0, this has been done. I think it's called the "Detroit Method" or something, and requires rectifiers on the motors to block the current that isn't used for each. The motors ran on half-wave DC pulses. I could definitely see this being combined with a pair of PWM throttles, but you'd still get half a wave, if split evenly between both motors.

To reverse, you could build something similar to a Lionel E-unit, activated by interrupting the current, which would switch the motor connections internally to the loco, because you can't reverse by switching polarity...if you tried, you'd just switch control to your other loco.

This is neat. I like reading about clever projects like this. Nowadays so much of our electrical work is described as "plug in the box and follow the instructions". That sort of thing just isn't much fun for me.

Cordially yours: Gerard Pawlowski President, a plywood world with dime store trees.

Well, I partly agree - you can power several throttles from one transformer winding (you need a tap in the "middle", though). The middle tap is the common return and all trains are operated from a voltage between the "ends" of the transformer.

If you try to reverse polarity by switching the wires (with a switch or relais), you'll get a short, right ;-)

Actually my layout is small enough and the electrical blocks are set in a way, that two locos will always be in separate blocks, anyway. So I don't need a system to run two locos off the same electrical signal ;-)

Yes, that is mostly it. I will continue to think about this while I finish the wiring of the layout (I still miss some turnout motors, anyway). Once I got something feasible, I will come back here and report my findings ;-)

Thanks to all!

Sorry that I haven't been around to keep up on this thread but my eye has been full of blood lately and it takes a while to get it out of the field of view.

Yes, there is heat generated by any electronics. BUT, there is a big difference in heat generated by a switching MOSFET and a transistor of any kind being used in a linear fashion. Technically, the 30mOhms or os of the on resistance of the MOSFET will generate some heat and the switching time will also generate some heat but that is down in the low milliwatts of power while driving several amps of current. A linear power control will have watts of power dissapation with more dissapation at lower voltages - 3amps times 10V is 30Watts of power, a long, long way from maybe 30mWatts of the switching power supply. For the common wiring layout, the best thing is to provide a power supply for each direction and then use which one you need for the job. This gets a little bit more difficult with electronics (power reostats work best in this case) but what you can do is to provide a bidirectional throttle where there are 3 voltage levels at which the output can be. This is the sort of thing done with audio amps and there is a Class D amp design which is a digital PWM power design. The power is either off or on to one side or the other, depending upon the speed of the loco desired. A SPDT switch can be used to drive either one or the other driver as needed.

-- Bob May

rmay at nethere.com http: slash /nav.to slash bobmay http: slash /bobmay dot astronomy.net

Get well soon!

Yes, that's why switched power supplies work so well ;-) However, I wanted to point out that there is some - though very little - power loss by heat - which needs to cool off ;-) In most cases it will be sufficient to let the heat aggregate during operation and leave it to cool over night or to provide a small ventilation hole (or two).

As mentioned, a modern computer CPU is the extreme example - it dissipates more heat per area than an electric kitchen stove... But all it does is switching low-power currents on and off ;-)

That's basically what it boils down to, though I would eliminate the switch (to switch between + and -) and rather have the driver work the full range in one sweep with the input control set up for switching... Using all the ideas I gathered here and there I would end up with about a hand full of small parts - and still be rather simple compared to the majority of designs I've seen. At the moment I'm finishing my last track connections and working on scenery a bit, but I'll come back to the topic soon and develop a circuit. Perhaps I can post it here and hope for some useful comments?

Thank you all!

Ciao..

Coincidence here too. Back around 10 years ago there was an extensive discussion on rec.models.railroad regarding the Cooler Crawler, and I added some wise words in which I basically said I had no idea how the thing worked, even though I'd built one and tried it out. It seems as though, by some mysterious method, it acts as a back-EMF feedback device, where if the train is near stalling it gets a boost in voltage to get it moving. Yet the circuit seems too simple to have any such feature. I'd definitely recommend it to anyone wanting to build their own analog throttle.

The club I'm in uses a PWM throttle system which (ahem) I designed. But it's only PWM for convenience in generating the signal via a microcontroller: from there on it uses analog components to put the voltage on the rails, and the waveform is shaped so that it's not a square wave. We have back-EMF feedback, but it's done digitally by the processor, with the signal sampled via an A-D converter. Our basic control scheme is "one throttle per block", where each block has its own reversing relay, so the difficulty of generating a positive or negative signal doesn't arise. And reversing loops are so simple we're hardly aware that they exist.

One very handy feature of our design is that the actual power-handling element is an LM317 voltage regulator, which is short-circuit proof and overheat-proof. I'd like to say "bomb and bullet proof", but it probably isn't quite that good. Still, we have over 100 of them in use and we've never lost one yet.

Read all about it here: