All these years, from the time when they were new to present, even with yearly cleaning /lube, etc. of the motors & E-units, I've just accepted the loud buzzing/humming sound from my 2243AB and 2343AA Santa Fe F3's, whether in forward, neutral or reverse. I just thought that the sound was just part of the toy train experience, with the sound coming from the E-unit and/or motor. However, I recently obtained an all-original (no modifications done) 2344 New York Central F3 AA, and it's as quiet as a mouse. It runs perfectly smoothly in all positions, and at all speeds, with NO buzzing/humming whatsoever. Just the sound of the wheels on the tracks. How can I get my other engines to behave similarly? Thanks. EdN
The noise is unlikely to be coming from the motors. The fact that you hear the buzzing in neutral demonstrates that the e-units are at least involved.
What you can do is to swap out e-units until you find a couple of the very rare ones that are quiet. Or, if you're willing to modify them, install rectifiers and maybe capacitors on the coils of the existing e-units.
Bob Nelson
The most common fix is a bridge rectifier and a small electrolytic cap. Running the E-Unit on DC will quiet it right down.
Is there any problem with the E unit coil overheating when it is run on DC? The current will be higher since now the inductance is not a factor.
I have a few locos with quiet E units, but most of them buzz. I have never tried to modify one to eliminate the buzz because they don't come apart too easily.
Bruce
The inductance is significant. However, the only case where I've had to reduce the voltage to limit overheating (with a series lamp) was when running two motors in series, which effectively doubles the track voltage.
There's no need to disassemble the e-unit to attach the rectifier.
Well, I did it to one and I didn't have an overheating issue, but that's certainly a possibility. The coil got warm with extended running, but never very hot, so I deemed it good. Extended was about 20 minutes, don't know if long term it would get hotter.. I'd be tempted to use a constant current source to power it if I were going to run for hours. You need enough current to cycle the E-unit at a fairly low voltage, but not so much as to overheat it at running voltages.
If I ran conventional more, I'd probably do some experiments, but 95% of all my stuff is command/control and that's not an issue.
I have pondered installing the bridge rectifier on a NW2 I have. But what is the purpose of the capacitor? And what size would be recommended?
If a bridge rectifier alone gets it quiet, that's all you should use. Without a capacitor, the solenoid winding sees a full-wave-rectified sine wave, which has less AC content than the full track voltage and therefore might do the job. If it doesn't, you can add capacitance as needed. Because the capacitance will raise the DC voltage on the coil as it reduces the AC, it's best to use as little as possible. You could start with, I don't know, 10 microfarads and go up by factors of 3 (33, 100, 330, ...) until the noise stops. Be sure to match the capacitor polarity to the output of the rectifier.
The previous response nailed it. You may get away without the cap, I used a 33uF as I recall. I'd start with just the rectifier bridge and see if that does the trick, it probably will.
You could start with a plain diode, as long as it pulls in the E-Unit at a low enough voltage, it would have less risk of overheating, since it would only be hitting it with half-wave current.
The catch with a single diode is that the force on the solenoid slug would be fluctuating at 60 hertz instead of 120 hertz, and more likely to shake it. Adding capacitance to overcome that will get you to the same DC coil voltage eventually as if you had used a bridge. So, if a single diode doesn't work, I would go directly to the bridge rectifier before adding capacitance.
Actually, adding capacitance to a single diode output will not bring the voltage up to anything like a bridge, you'd only have half the average current available. The only time the voltage would approach the peak value is if there was no load, and since there's always the coil there, that's not the case here.
I have to disagree, the single diode will have a different effect. I'd suggest you might want to try the experiment and see for yourself.
John, either rectifier circuit will charge the capacitor to the peak of the sinusoidal track-voltage waveform. Assuming that the single diode's cathode is connected to the generator, the differences are that the single diode will charge it every 1/60 second to the positive peak voltage less one diode drop, while the bridge will charge it every 1/120 second to either peak voltage, less two diode drops. The voltage will decay exponentially between charging events. But, with enough capacitance, the average DC voltage will be higher with the single-diode circuit.
Again, that totally depends on the load. Yes, if you had a few farads of capacitance across the coil, they would be equal. However, in practical terms, the single diode will not develop the same amount of voltage across the coil as the diode bridge, simply because the load of the coil will discharge it during the half cycle it's not being charged. I can't explain it any clearer, but if you just actually try it, it'll explain itself.
If you still insist there's no difference, we'll just have to agree to disagree, and I'll go take my EE Diploma off the wall.
Thanks everyone for all your responses. I think I'll try the bridge rectifier and see if that works.
I don't think we need nearly that much capacitance to get the full voltage across the coil.
Let's assume that we use enough capacitance to have 10 percent peak-to-peak ripple. Then the coil current is about 6*vpeak*C, where vpeak is the peak track voltage and the approximate coil voltage. The coil power is 12*vrms^2*C, where vrms is the root-mean-square track voltage. If we assume a track voltage of only 10 volts and we need only (!) one farad of capacitance to get the ripple down to 10 percent, the coil must be consuming 1200 watts!
Like I said, try actually measuring it both ways, don't take my word for it.
Well, I can't add anything to THIS debate, but I would like to make a progress report...
I started by putting a diode in series with the e-unit coil. Given what I now know about diodes, the result was not unexpected. My e-unit went from buzzing to making sounds which are positively RUDE. Kind of like "pbtbtbtbtbbt...!"
Not having any capacitors handy, I removed the diode, and tried a bridge rectifier--quiet as a mouse. If I listen real close, I can just hear a faint rattle.
I don't run the engine for hours on end, so hopefully overheating will not be an issue...
Thanks for the help.
The capacitor would have made all the difference with the single diode.
Our community is FREE to join. To participate you must either login or register for an account.
Get the Classic Toy Trains newsletter delivered to your inbox twice a month