1. How intuitive the operation was; and,
2. How easy it was to turn the knob to "Stop."
Obviously, as long as you can see the
train, you'll know when it stops; however,
once you've adjusted the knob so that the
12 o'clock position produces zero volts out,
any setting between 11 and 1 o'clock will
stop the train -- easy.
Then my wife (a non-railroad person) came to inspect my work; she picked up the throttle and began to run a train (she'd never done such a thing before!) She pronounced it "really easy to operate....and kinda' fun!." The best testimonial I've ever had for any of my creations. I've found it to be quite nice to use and particularly handy on my SWITCHING LAYOUT.
If your trains can disappear from view, there's an optional "IDLE" indicator circuit that will tell you when you've halted. There's also an "Emergency Stop" switch, and a "MAIN/YARD" switch which allows you to scale down the voltage range for yard switching operations, yielding better throttle resolution at slow speeds (very handy!). And....there are two different power supplies shown for this throttle -- one of them using that spare MRC power pack that you have laying around. This project can be built as a true memory walkaround (that's the way I did it), or integrated for fixed use.
1. Transistors Q1 & Q2 are "Darlington" power devices; it's very important to have good heat sinks on them, as they do get quite toasty!
2. Capacitors C1 & C2 are position-critical with respect to the op amp (U1); they should be located as close to the device pins as you can get them. Otherwise, construction is fairly forgiving.
3. If you going to use this throttle on a switching layout, or if you never run your trains faster than sixty or seventy scale MPH, you can make a resistor change to further increase the "resolution" of the throttle (while limiting the maximum speed): in place of R19 & R20, try substituting a 10 Kohm resistor for each. Train goes too fast in the MAIN position? Try changing R1 & R2 to 10K ohms. Not fast enough? Try something around 4K ohms for R1 & R2 (yes, I know; 4K isn't a standard value -- put a 100 or 220 ohm unit in series with a 3.9K; just don't go below 3.9K) Any value greater than 3.9 Kohms is OK -- feel free to experiment.
4. Speaking of R1 & R2, they should be fairly-well matched; 5% resistors are technically OK, but you should try to do better. Measure a half-dozen on a digital ohmmeter, and use the two that are closest in value. No ohmmeter? Then you could (a) parallel two 10 Kohm resistors in place of a single 4.7 Kohm unit (averages out the values), (b) use 1% resistors (or 2% arrays), or (c) you could just not worry about it (but the train might creep when the "Emergency Stop" switch is thrown). Also, resistors R19 & R20 should be closely-matched (same options as above; lacking an ohmmeter, try paralleling two 15 Kohm units in place of each). And speaking of matching resistors, R15, R16, R17 & R18 should all be as close in value as you can manage.
5. Like to follow your trains? Then you'll want to implement this project with a tethered MEMORY walk-around. The hand-held unit can be constructed in a "project box" of a size that just fits in your hand, and contains R1, R2, R19, R20, the pot (R6), toggle switches S2 & S3 and the LED. All this requires only 4 wires from the base unit. I used good 'ol modular telephone jacks and a long, color-coordinated coil cord. You can unplug the handheld unit, move down the layout, and plug it into another jack; meanwhile, the train just keeps going at whatever speed you'd set when you unplugged (like all good things, however, this won't last forever -- if you take several minutes to plug in again, the train will start to slow....and eventually come to a stop; it just physics, so don't tarry). Resistors R15 thru R18 ensure that if the handheld is left semi-permanently unplugged, the trains will stop....and stay stopped; they should be as closely matched in value as you can manage. NOTE: IF YOU DON'T EVER PLAN TO PLUG AND UNPLUG THE HANDHELD, you can omit resistors R15 thru R18 and capacitor C9 -- assuming you don't want momentum (see #9 and #10, below).
6. Planning to implement the IDLE indicator circuit? You'll need to match the actual values of R11 and R13 to within 1% or so; to do this, you'll need to either (a) measure the candidates using 5% resistors, (b)parallel three 27K resistors for each, averaging-out the variations, or (c) use 1% resistors (usually metal-film types). You could also use resistor arrays, which are generally in the 1-2% range. If you find the "stop window" too narrow for your tastes, you can increase the value to R12 to 560, 680, even 1 Kohms, as you see fit -- it OK to experiment. Two pieces of U4 are unused (it's a quad), and it's important to tie the unused pins to the most-negative voltage, as shown; this prevents spurious oscillation. Finally, capacitors C7 & C8 should be located close to the pins of U4 (similar to what you did on C1 & C2).
7. Including the "Emergency Stop" switch (or not) is strictly up to you. It really depends on how comfortable you are finding the STOP position when disaster looms. It's a sort of security blanket. Being a daring sort, I didn't include it....so far. If you go for it, just remember to turn the throttle knob to the stop/idle position (12:00 o'clock) BEFORE you turn the Emergency Stop switch to OFF!
8. MOMENTUM -- A bit of momentum is designed into the throttle. If you want more, try increasing C9a/b to 220uf/35V. The downside of more momentum is that it's harder to find the IDLE position on the pot by looking at the LED (there's a "lag" -- but that's where the detent will come in quite handy!); the upside is longer memory retention. If you want even less momentum, reduce C9 to 47uf/35V; again, be forewarned -- memory retention will be cut by about half.
9. CAPACITORS C9 (a & b) -- They're multi-purpose beasts (as you've no doubt noted). Here's a summary of how they affect the throttle:
No C9s No momentum, no memory, no unplugging
C9s<100uf Less momentum, shorter memory retention
C9s=100uf Some momentum, adequate memory retention
C9s>100uf More momentum, longer memory retention,
more lag on IDLE indicator
The connection of C9a and C9b is critical (as with all electrolytic/polarized caps) -- by connecting the negative (minus) terminals together, we've formed a "non-polarized" cap from two polarized caps. The equivalent value of two equal value caps in series is one-half the value of each -- in this case, 50uf.1. TRANSFORMERS -- Any transformer with a Center-Tapped secondary whose voltage rating between 25.2 (often specified as 12.6-0-12.6) and 40 (20-0-20) volts will be dandy. Radio Shack has a nice 25.2V unit; Jameco has 25.2s and others. Get one with at least a one amp (1A) rating; 1.5A or 2A is better.
2. The resistor values for R6 thru R10 yield outputs of approx. +/- 14 volts DC. This lets you get close to 12 volts maximum on the rails. If you don't need 12 volts, you could substitue fixed regulators for those shown (and eliminate the resistors). For the positive side, use an LM340T-12 in place of the LM317T; for the negative side, use an LM320T-12 in place of the LM337T; don't forget to eliminate resistors R6-R10, and tie the regulators' "ground" pins directly to ground (!). If you want more than 12 volts on the rails (G-scale?), try using an LM340T-15 and LM320T-15 instead of those mentioned above.
3. It's very important to have heat sinks on U2 & U3; I'd suggest using the same type you used on Q2 & Q4, but a slightly less elegant sink would probably suffice.

If you happen to have an unused commercial power pack (MRC, etc.) lounging about the premises, you can put it to work here. Instead of buying a transformer, simply use the "AC Output" terminals on the power pack (AC outputs are sometimes labelled "Accessories"); just make sure yours has at least a 16VAC output. The schematic below shows how to build the necessary dual supply using that extra power pack. "Alright," you're asking, "What's the catch here? What do I lose by not buying a transformer?" The answer is SPEED. You'll only get about 7.5-8 volts on the rails; plenty for me, and more than enough for switching....but some folks feel the need for speed. If you know that you routinely run trains at speeds which cause more than 8 volts to be measured across the rails, then go for the supply with the new transformer.


If you build it, let me know how you like it. If you have any problems with construction or use, or you have questions, just drop me an Email....I'm Fred Horne in beautiful Santa Fe, NM. Our Basset Hounds "assisted" with the construction of the prototype I'm using, but they DID NOT participate in the circuit design.
A special "Thanks" to Phil Hendry, whose needs inspired me to undertake this project.
may have missed much of what the Armadillo & Western RR Web Site has to offer. | ||
to visit the Home Page. |
|
available & how to get to it. |
To This Page.Updated 1-27-02