I have a nce pro ab with older powerhouse and a homebrew power supply. The power supply trans former power switch and circuit breaker are labeled Tony’s train exchange upon a call to them they said they used to sell those parts for making a power supply. It is a simple 123 volts ac measured at the wall into transformer 18.5 volts ac out can anyone shed light on this?
designing the device (e.g. NCE Powerhouse) for AC power allows a less expensive power supply to be used
most electronic devices that accept AC power need to include the “guts” of a DC power supply which include rectifiers, filter capacitors and a voltage regulator circuit. even when DC power is used, a voltage regulator may be used,especially if a processor is involved.
an AC power supply is nothing more than a transformer. separating the power supply from device allows it to be powered from various sources. some larger layouts have their own power bus. other countries have different line voltage levels.
Loy Spurlock retired and closed up shop in 2007. I still have three-ring binders with the handy DCC charts and things he made up for customers. A DCC pioneer for sure.
I understand on how to build power supplies I have built a few 5000 volt 3 amp ccs supplies. I was curoius if this was a big thing at one time and what advantages and disadvantages it provides. I should have been more clear. I called tonys train xchange and they said they no longer sell this anymore.
it works fine but I am rewireing it with better wire not speaker wire and it has a breaker. I apreciate all answers and thank you all i am still new to dcc.
My club has three Loy’s Power Supplies for our 8 amp boosters. We have two spares in the toolbox if they should ever be needed. Working fine since 2004. I replaced the main power rocker switches in two of them a couple of years ago with parts from Digikey. I was able to match the switches up using the specs on Digikey’s website.
DCC involves modulating voltage at over 410kHz frequency, variable pulse duration, with minimal distortion of square waveform.
If you’re discussing input of 14VDC to a separate unit that does the switching, it may need to have both +14V and -14V connection to a common center rail. Although the load ‘sees’ only 14V at any given moment, I think separate power-supply rails (i.e. with nominal rail-to-rail swing of 28V) need to be present for the NMRA modulation scheme to work and not transfer charge across the connection over time. This may be done electrically within the controller.
(As previously noted this was a concern with high-speed modem design back in the old V32.bis days when ‘constellation’ didn’t mean heavenly star patterns… many POTS lines were not set up to transfer DC electricity…)
If you just have something like a wall-wart supply that outputs reasonably-smoothed 14VDC, then yes, a properly-filtered power supply will do the job, probably better than the dedicated built-to-a-price device.
The maximium DCC freqeuncy is about 10khz, and that’s the command stations and boosters that have to be able to do that, not the power supplies.
The power supply is just a regular AC or DC (depending on system) power supply of the appropriate volatge for the scale you are using, I don’t know of any DCC system that requires a bipolar (+V and -V) supply.
I was going to edit the post, but will just add some here:
I made a fundamental mistake in not realizing the ‘power supply’ in question puts out low-voltage AC, not DC. Since all the subsequent rectification, smoothing, and modulation is relative to that input, the answer to the question collapses to the same ‘yes’ as the short answer immediately after mine for that case.
The same would be true for any system that has a ‘single’ DC supply input.
The only point about the high frequency is that the system needs relatively ‘quick’ power to full voltage in order to produce a ‘clean enough’ square wave without ringing effects for the decoders to “read”. That is orobably done with energy storage in the system, and not more inductance/capacitance in the power supply, but it might be a concern calling for a higher output rating at the supply than the nominal DCC voltage might imply.
Don’t know where I pulled that 410KHz from. It is surprisingly difficult to find an actual peak ‘bandwidth’ in the references, perhaps because the modulation is not a sine wave, not symmetrical with cycles, and has variable pulse timing. But this shot clearly puts the “frequency” in perspective for the (PicoScope 6, Hornby) system being tested, which I believe is NMRA 9.1-9.2 compliant.
Even the ‘best’ PWM frequency to model railroad motors barely breaks a tenth of that figure I used – about 43KHz the last time I looked at it.
According to NMRA Standards and Recommended Practices S-9.1 a “1” half bit duration can be from 55 to 61 µs (microseconds) and a “0” half bit can be from 95 to 9900 µS, so the shortest total possiblle total bit time is 110 µs. The reciprocal of 110 µs is 9090.9 HZ
I think Overmod is talking about the PWM motor drive not the DCC signal.
My RMC Prodigy DCC PWM is 15.5 volts at about 8K, very clean square wave. I have mostly MRC decoders and their motor drive PWM is about 16K. I have one TCS that I measured at about 40K. I don’t get into the PWM rate only the square wave on my scope.
The RMC decoders look pretty good the TCS has a slight trailing delay, all I’ve looked at are cleaner than I expected.
Mel
Modeling the early to mid 1950s SP in HO scale since 1951