Hi all,
I hesitate to post because we are getting away from the original topic. However, I think I should be clear about what makes MTH PS2/3 and Lionel Odyssey 2 speed control superior to what is offered in HO.
To date, all HO decoders with electronic speed control, except for MTH PS3, have used BEMF. For those who don't know, BEMF works thanks to the natural properties of a DC electric motor. When you apply voltage and the motor spins. However, if the motor is spinning and you remove the input current, the motor becomes a generator and gives back an electro magnetic force (aka a voltage). The faster the motor is spinning the higher the voltage. Modern motor control electronics on both O and HO use pulse width modulation to govern motor RPMs, which provides an opportunity to measure BEMF from the motor many times per second. You will often see HO decoders marketed as having a high BEMF frequency, suggesting superior motor control. BEMF works great for basic motor control, but it has some real limitations. You will see many HO engines with good BEMF decoders that will run VERY slow. However, good speed control requires more than slow speed. It requires REPEATABILITY. This where BEMF runs into trouble.
The back voltage from a spinning motor is dependent not only of the motor's speed, but also on the resistance in the motor windings, air gap between stator and armature, and a few other lesser factors. Seemingly identical motors come off the production line with subtle variations in all of these factors. As a result, no two motors are exactly alike. What's more, as the temperature of the motor changes these same electrical characteristics change. Occasionally, you will see well versed HO decoder programmers refer to "the warm motor effect." As a motor warms up the back voltage usually decreased. Unless you account for these variations in production and operation conditions, BEMF has a wide margin of error. Speed step 25 might be 15 scale miles per hour at one time and 16.5 smph at another on the same engine. Two engines that are well matched when cold may not be well matched when one or both are warm. In industry, you will sometimes see BEMF used for motor control where low to moderate accuracy of RPM control is needed. In the best cases, the motors are carefully tested to document their electrical characteristics across a range of operating conditions. The electronics that govern the motors are then programmed for specifically for each motor to maintain as tight a tolerance as possible. But even with all these efforts, the best BEMF systems are only accurate to within about 0.5%. Not surprisingly, model manufacturers don't both with such costly and time consuming designs. They take the motors as they come and call it good. Most HO decoders are only accurate to within 10% on RPM control. That doesn't mean they can't run 1 smph, it just means they can't reliably run the same speed every time.
High frequency BEMF makes things worse at slow speed rather than better. Most HO engines these days use 5 pole skew wound motors. With only five poles, as the motor speed decreases there is a subtle fluctuation in the back voltage. Low frequency motor controls can ignore this fluctuation because the motor spins more than one pole while current is shut off. High frequency picks up the fluctuation and confuses the motor controller. So why wouldn't manufacturers use a lower frequency speed control? Why would they in fact brag about having high frequency controls if they are all less accurate at slow speeds? The answer is WHINE. Remember the the noise from Train America EOB boards? That was low frequency BEMF at work. It did a great job controlling the motor, but you can hear it working. Decoder manufacturers push their motor control to frequencies we can't hear in spite of the diminished accuracy.
So, is there a better way that is more accurate? Yep! Optical and magnetic encoding. Both MTH and Lionel use optical encoders (tach readers) to measure actual motor RPMs, rather than infer them. It doesn't matter if each motor is slightly different or if it is hot or cold. The electronics simply adjust the pulse width to produce the commanded speed. Since 2000 when MTH released PS2 and 2007 when Lionel released Odyssey 2, both companies have used this highly accurate motor measurement to adopt a consistent speed map for all their locomotives. They account for the drive wheel diameter and gear ratio on each engine. That allows any two PS2/3 engines to consist with each other regardless of prototype. The same is true for Legacy engines with Odyssey 2.
So what stops other HO manufacturers from adopting their own optical or magnetic encoding systems? Cost, disparate manufacturing, and market demand. Encoding systems are more expensive and they require additional time to install on the production line. BEMF saves time and money. In HO, most model manufacturers don't make their own electronics. Athearn, Atlas, and Bachmann make models while Soundtraxx, QSI, and MRC make the electronics that run them. That makes it more difficult to coordinate an effort to make the electronics and their programming fit any given production run. In contrast, MTH and Lionel keep both functions in house, so coordination is easier. To date, the only HO manufacturer to offer products with a consistent speed map is MTH. They have been gaining popularity in HO and some people are starting to see the advantages of such a system. But, for the most part, people in HO still don't know what they're missing.
