The jump is because the 'old way' of getting a pulse output was to tap off a leg of the rectifier, which, in those days being either copper oxide or selenium had a FAR greater diode drop than a silicon diode - on the order of volts. So switching the pulse out would cause a significant change in the track voltage.
I'd be curious what PWM would do with the constant lighting. Probably work, since the pulse is always positive going, unlike unfiltered DC which has a negative going half of the sine wave which is what I think is why the constant lighting didn't work. Would give good pulses to the motor though.
Lots of good throttle circuits here on Rob Paisley's site
http://www.circuitous.ca/Throttles.html#1
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
I put together a transistor throttle with parts I had on hand...used the 15 vac from my old A.C. Gilbert transformer to power it. That set up, along with the Hobbytronics directional constant lighting circuit solved the dim lamp problem. I've got bright lights before the engine's motor begins to turn, which is good, but the motor revolutions are jerky at low speed. The result is that I sacrifice the smooth motor operation at low speed that I can get using the dimmer/transformer combination for bright, constant lighting.
I understand the jerky low speed problem can be solved with pulsed DC (?), but I've read that when the pulsing is switched off when normal operational speed is attained the engine can suddenly (and unrealistically) jolt forward (or backward, as the case may be). I've also read that there are circuits that can gradually reduce the pulsing as track speed is reached. I've done some internet searching, but haven't found anything yet. Anyone have a schematic for such a circuit?
-Chuck
If you have a couple of transistors, or an LM317 adjustable voltage regulator, you don;t need any heavy duty potentiometer. Lots of circuits for simple transistor throttles all over the internet.
Think you may have mis-read.....I've got the dimmer ahead of the transformer, i.e. between the household AC and the transformer, as it would be if used to control a house ceiling, or other, light.
Wish I had a scope so I could see the output from the rectifier to check for possible AC coming through. I tried a 8 uf. cap. across the rectifier output (i.e. track power), but there was no change; the lamps are still only getting ~.3 volt with the motor not, or barely, moving. You mentioned that today's motors run at lower voltages than earlier ones. My motor dates to ~'72, so it was manufactured about 46 yrs. ago.
I don't have another power pack to use to compare results. I've got parts I might be able to use to put a DC power supply together. Not sure I have a heavy-enough potentiometer to use for voltage control, so I'd probably use the dimmer for that purpose.
I'm surprised you get control with a household dimmer, since most of them are SCR or Thyristor based and NEED AC to work - the zero crossing of the AC is needed to switch the SCR off.
If all you have in the circuit is a bridge rectifier, you are probably getting a LOT of AC pulsing through to the track, which helps the motor start sooner - not exactly what we want here.
Modern motors draw a lot less current and can run at lower voltages, The resistence of the windings is much different, which is why many new HO locos cannot be controlled by old HO rheostat power packs, the current draw in the motor is so low, little voltage is actually dropped across the rheostat. All but the cheapest train set junk is transistorized now, the voltag output is not dependent on the current drawn.
The circuit I used with the capacitor charged up on pulse power because the pulses were present but too narrow to make the motor turn - the effective DC voltage was under the motor's minimum, but those pulses are at full voltage - 12V or more. It's the same concept as a smoothign capacitor in a DC power supply - if you use say a 6.3V transformer feeding a bridge rectifier, you get a bit under 6.3 volts out measured with a DC meter, but there's significant ripple that gets through. Add a capacitor and the measured DC voltage will be greater than 6.3 volts becauser the capacitor charges with PEAK voltage, what you were previously measurung was the RMS voltage. The 6.3V output of the transformer is the RMS values, the PEAK is 1.44 times that. (square root of 2) (for sine wave AC).
I don;t think the power pack back then used PWM, but DCC decoders do to drive the motor. That uses a square wave, the peak is the actual voltage, but the apparant DC voltage is based ont eh duty cycle. 12V square wave at 50% duty cycle will read as 6 volts DC.
All the lighting and stuff works with DCC because the track is always powered. There is a constant ~ 15V square wave on the rails at all times. So always on lighting is trivial.
Do you have an ordinary model train power pack you can test with, or maybe a variable bench power supply that puts out smooth or at least somewhat smooth DC? The negative going component of the rather pulsed DC coming out of your setup is probably responsible for the lower reading across the diodes.
I have very limited knowledge of DCC, but from what I understand the many functions available are possible because the associated electronic circuits (e.g. a loco motor, constant lighting, a steam exhaust sound effect) are each activated by a discrete signal sent from a main control. Another hobby I have is amateur radio so I'm familiar with the basic theory of radio waves and the fact that they are AC and use a carrier wave (modulated various ways) to send information. I first thought that a given DCC signal was sent as a carrier wave over the basic track circuit to its destination (e.g. a loco whistle sound circuit), but then remembered that the track circuit is DC (isn't it?), which couldn't support an AC carrier.
My track power is from a 100 watt A.C. Gilbert model 8B transformer manufactured about '48/'49 that I got with a Lionel toy train set when I was 4 (!). It puts out 15 volts AC which I convert to DC with a full-wave diode rectifier. I control the DC level to the track by using a standard 300 watt-rated house wall light dimmer ahead of the transformer. It allows me to vary engine speed smoothly from barely a crawl (about 1.8 track voltage) to max. speed (about 12 track volts).
In Mark Rollins' lighting circuit the 6 diodes are wired in series with each other so as to form a complete circuit themselves (anode of one to cathode of the next) and the connections to their circuit are 3 diodes apart, one from one side of the track and the other from one side of the motor. I connected my lamps just as Mark shows, one side to the motor and the other to a diode each side of the track connection. I measure the .3 volt (minimum needed to bring the lamp on) across each lamp's connection to the circuit (...boy, this would sure be easier to describe with a schematic diagram!....).
Yeah, my meter has a diode test function. The diodes I'm using measure between .55 and .60 forward voltage. According to Heath, the maker of my VOM, those readings fall within the range for silicon diodes, .5 to .9 volt.
So, was the reason one of your club locos began moving after the beacon came on the capacitor pulses, or a motor that wouldn't turn until the track voltage was up to something above what the beacon needed to light? I've read that some motors won't move until the voltage is up to 3 or 4.
I always get full brightness witht he locos standing still because I use DCC.
What sort of power pack do you use? At the club I used to belong to, I helped one of the other guys wire up a constant voltage circuit to power a rooftop beacon on one of his locos. We had 2 big walkaround DC power systems on our layout, one main ran off an MRC Controlmaster 20, the othe line ran off a CMI Hogger. His loco worked great on oen fo them, you could crack the throttle, beacon would light, but loco wouldn't move. On the other line, same sort of thing as yours. The becaon would work, but the loco was already moving. That constantly lighting circuit used a capacitor as well as the diodes, whichever of those two power supplies had stronger pulses was the one that worked, as the cap would charge up to the track voltage because the pulses were full voltage but too short a duration to spin the motor. That shouldn;t matter here since there is no capacitor.
AO you have this: rail---1->|->|-2->|--(motor)----rail
and the voltage you are measuring between points 1 and 2 is only .3 volt? Hmmm, this is quite strange. Does your meter have a diode test function, a real diode test range will show the forward voltage of the diode. It should be around .6-.7V
Hello Randy-
I added 2 more diodes for 6 total and wired them per Mark Rollins' circuit. The directionality function is fine, but I'm still not getting enough voltage for the lamps. In fact, it's only ~.3 volt when the track voltage is ~1.8 and the loco's motor is just barely moving. Even at full track voltage of ~12 the max. lamp voltage is .95. Some factor is limiting delivery of a full 1.5 volts to the lamps.
At what track voltage do your locos start to move? You evidently are getting full lamp brilliance (at 1.5 volts) before your engines begin moving. Re the current rating of lamps, mine measure 73 ma. which, at .3 volt is real dim (.02 watt). If I were to substitute a 30 ma. lamp, for example, the wattage at .3 volt would be .009 watt; I probably wouldn't be able to see any light at that wattage.
You need to wire it up more like this:
http://www.mrollins.com/constant.html
The extra diode NOT in the lamp circuit allows the lights to come on before the motor, and make it directional to boot.
The drop across a diode is ABOUT .7V - it varies depending on the current draw, it's not completely linear. The last digit in the 1N400x series indicates the peak reverse voltage it can handle, even the 1N4001 is 50 volts which is plenty for model trains, the 1N4007 is 1000 volts! The motor, while it may draw 1/2 ampo stalled, probably draws only a few mA when running free, if you look at the chart of forward voltage for a 1N4000 series, it can be as low as .6V. Also, what is the rating of the 1.5V bulbs you are using? Some are 15mA, others are 30 or even 45mA. The lower current ones should work best - plus not get hot and melt the plastic.
Thanks Randy and Mel for your replies....
I rigged a bridge rectifier using four 1N4007 diodes I had on hand. The 5-pole motor in my engine draws 1/2 amp. at full speed so the 1 amp. rating of these diodes should be adequate (?). I wired the bridge in series with the motor and connected a 1.5 volt incandescent lamp across the bridge (one lead to the motor connection with the bridge and the other to the connection of the bridge with the "hot" side of the track voltage supply). At this point I'm only checking for constant lamp brightness with one lamp. Later, I'll add 2 more diodes and another lamp to get headlamp/tenderlamp directionality.
When I fired it up the lamp was barely illuminated at just below the 1.1 track voltage when the motor began to move. At that track voltage the voltage across the lamp is 1/2 a volt. Then, no matter how much I increased the track voltage the maximum voltage across the lamp only went to 8/10ths of a volt. So...I'm only getting about half the lamp brilliance I should be getting if the voltage across the lamp was 1.4.
I noticed that one bridge circuit I saw used 1N4001 diodes. Those also are rated at 1 amp., but have lower maximum voltage ratings than the 1N4007 diode. That shouldn't make a difference since those voltage ratings far exceed the 12 volt supply to the motor/track.
Any ideas why I'm not getting 1.4 volts to the lamp?
It probably did work as described with locos contemporary to the device being sold - most modern models have much more efficient motors that start on a far lower voltage. Stick that unit in a classic Athearn Blue Box loco and it will probably work fine.
All it is is a pack of diodes (or a bridge, to get 4 diodes in smaller space) wired across the motor so that the voltage to the motor is reduced (about .7V per diode), so say if there are 2 diodes in each direction (have to have one set in each direction or the loco wouldn;t be able to reverse), the light bulbs would see 1.4V before the motor saw anything more than 0V. Many Proto2000 locos used the same scheme for directional constant lights, on their circuit board they had an array of diodes. Another benefit is that if the lights are wired across the diodes, they see a constant voltage relative to how many diodes there are - no matter how high you turn the throttle. Even at a full 12 volts, full throttle, across 2 diodes it will always be around 1.4V. SO you use low voltage light bulbs and they come on as soon as the input voltage exceeds 1.4V, and never get any brighter as the throttle is increased more.
The schematic for such things is freely available all over the web. They're all effectively the same - companies like Hobbytronics were charging maybe $9.95 or even $19.95 for $2 worth of diodes for the people who couldn;t figure it out from a schematic.
I have a Hobbytronics HO-300 system (yeah, it's old....don't think company's still in business) I'm planning to install in a steam loco. I've temporarily wired it to check its operation before putting the engine headlight and tender back-up light in their respective places and the "power module" inside the tender body. The lights come on (headlight when going forward, tender light when backing), but the engine starts to move simultaneously when I switch on the track voltage (I'm using DC; in fact, don't know if this particular system would work with DCC).
According to the Hobbytronics description, however, the lights should come on before the engine begins to move so there isn't the unrealistic increase in brightness as engine speed increases. The circuit in the power module is supposed to bring the lights up as track voltage is increased from zero to about 4 1/2 volts, at which point the engine should begin to move. At that voltage, and higher, the luminosity should be nearly constant as engine speed goes from zero to whatever speed is desired. Without the power module circuit in place the engine would begin to move when the track voltage is somewhere between 1 and 2 volts (which my engine, in fact, does).
So, by the time the lights are at about their constant maximum brightness the engine is already moving at a pretty good clip with 4.5 track volts applied. Anybody have this system and have same experience? Guess could be a faulty component inside the power module. It's a very small (~1/2 in. long) sealed brass case, though, and the Hobbytronics description doesn't include a schematic diagram.