rrinker The 'trick' for the waveform is those two extra diodes. If you look carefully at the schematic you'll see thea the output of the full wave bridge goes directly to the final output stage, whereas the two individual diodes feed the low power first stage transistors. I suppose a 'cheater' version would use 4 diodes instead of a bridge and tap off the half wave feeds, but that might cause an unwanted feedback.
The 'trick' for the waveform is those two extra diodes. If you look carefully at the schematic you'll see thea the output of the full wave bridge goes directly to the final output stage, whereas the two individual diodes feed the low power first stage transistors. I suppose a 'cheater' version would use 4 diodes instead of a bridge and tap off the half wave feeds, but that might cause an unwanted feedback.
rr:
I thought there was some sort of half-wave adding effect too, but if you look carefully, both sets of diodes act as half a bridge rectifier, and Q4 gets full-wave power like the rest of the circuit...however, your comment about feedback has made a light go on...
Yesterday, I dug up a thread from 1997 on RMR on this subject, which turned out to be very illuminating! There seems to have been a consensus that the CC throttle worked very well, but nobody, including Rich Weyand, its designer, was absolutely sure /how/ it worked. (That makes me feel a little better. )
In this 1997 thread, however, through batting ideas back and forth, a very plausible theory had been found:
http://groups.google.com/group/rec.models.railroad/msg/0c4b4c10eb5067f4
which was then demonstrated by Ken Willmott in a couple more circuit designs here:
http://web.archive.org/web/19990506041638/http://wits.on.ca/rail/electronics/throttle/throttle.htm
(Page archived with text writeup, but the circuits are, sadly, lost.)
I vaguely remember reading about this rheostat-capacitor method in an old book. According to Willmott, the Cooler Crawler is an improved, solid-state version of it. He described the rheostat-capacitor throttle this way:
Ken Willmott 1997 At the lowest setting, when power is applied the cap charges through the load until it reaches almost the peak value of the applied half wave rectified power supply. Subsequently, the cap spends most of the cycle slowly discharging through the rheostat, and a small portion charging rapidly through the load. It is the latter portion that propels the loco.
At the lowest setting, when power is applied the cap charges through the load until it reaches almost the peak value of the applied half wave rectified power supply. Subsequently, the cap spends most of the cycle slowly discharging through the rheostat, and a small portion charging rapidly through the load. It is the latter portion that propels the loco.
Here, capacitor C2 acts in the same way with the power transistor Q4, and that's where your comment about feedback comes in! The diodes are there to keep C2 from discharging through the control and amplifier circuit!
Now here is where the really diabolically ingenious part comes in. The cap is charging through the loco motor. Should the loco momentarily lose contact, it will charge less, but the instant contact returns, the cap will start to charge again through the loco, providing a jolt to get it moving again. Is that not cool?
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Their reason for using two separate transformers seems to be related to the availability and current rating of transformers at the time the circuit was designed, more than how the circuit functions.
Their explanation of the circuit pretty well covers it all. The output is regulated by the voltage across R1 and C1, which subsequently controls the bias on Darlington pair Q1-Q2. The Zener diode is only to prevent overvoltage or a runaway that could destroy Q4, which is actually an audio output transistor designed for use in power amplifiers.
I built 10 of these circuits for use on our HO scale club layout in 1995, and they performed okay for the first 10 years or so, and then they slowly began to fail. The weak point of this circuit is the 2N3904 Transistors which, over time, burn out. Using a different transistor with a higher voltage/amperage rating may prolong the useful life of the circuit.
I modified the circuit when I built the units for our club layout and completely did away with DB1 and the two transformers. Instead, we power them from a 35 Amp central power supply that delivers a flat, filtered DC at 18.5 Volts.
We hardly ever use the Cooler Crawler now since everyone prefers DCC operation using the NCE PowerPro Wireless system. The layout is wired so operators can use either system because we sometimes have visitors or new members with older DC locomotives who want to try them out. Most members soon ask to have decoders installed once they see the difference.
Folks:
I like to play with transistors, and I was thinking of building myself a Cooler Crawler:
Reference for diagram, output graphs, and writeup:
http://users.rcn.com/weyand/tractronics/articles/ccartcl/ccartcl.htm
Before I got started, I wanted to understand its function, because that's how I work. Unfortunately, I am an electronics n00b, and while I understand the basic idea, I have a few questions:
1. Why on Earth are there two separate, parallel, positive, and equal voltages supplied by DB1 and diodes D1 and D2? The only thing I can come up with is that it separates the large and small traces on the PC board, but then why have the separate diodes?
2. How does the circuit come up with the waveform shown in the article, with spaced pulses at low output voltage, and full wave dc at higher voltages? It just looks like 4 (well, 5) transistors chained together and wired as a common-collector voltage follower. For a moment I thought the zener diode D4 did something, but since it's just shunting the small base current from Q3, I don't see how it could affect the output.
What am I staring at and missing here?