UPS for decoder?

MasonJar

It's not rocket surgery
Was out at a club meet this past weekend, and one guy had a fabulous brass railbus that he had added a Uninterruptible Power Supply to...! It was in the form of a capacitor, rather than a battery. Anyone know anything about this technology, and whether it can be applied to sound decoders?

By the way, this railbus was a beauty. Full interior (!) with a Zscale decoder & UPS in the back baggage compartment, and a tiny underframe motor. Pickup on the driving wheels (2) and the front truck (4 wheels). John intends to add functioning markerlights - but how he will do it I have no idea. The marker lamps are less than 1mm around...

Anyway, thoughts on UPS for sound decoders?

Thanks.

Andrwe
 

pgandw

Active Member
A capacitor can be added as a form of battery, however the battery "capacity" is very small unless the capacitor is very large. A capacitor across a circuit acts as a short to AC, with a bigger capacitor being more of a "short" at lower frequencies. So putting a capacitor across the track will mess with the DCC signals. Your friend has put his capacitor across the motor circuit on the motor side of the decoder. There, it can act as a momentary battery (UPS to cross dirt), and smooth the chopped DC going from the decoder to the motor. The latter reduces motor heating at the price of very slow speed running, and is often recommended for vey small motors by DCC and motor manufacturers.

Sound decoders draw/pass nearly as much current as motors due to inefficiency of our tiny speakers. A capacitor in parallel (needed for UPS action) with the sound decoder input will short out the DCC signal on the track. A button battery (with charging circuit) would be a better solution, but the required decoder DC input voltage is the key to this solution's practicality.

A small capacitor across the sound decoder output will attenuate mid to high frequency sound, and the speakers do not use DC anyway.

I have seen this idea discussed on other forums before, with the non-believers insisting it can be done.
 

60103

Pooh Bah
The October Railway Modeller reviews the Lenz Power 1, which has stored power and a circuit to check for signals. It has to be soldered to their Gold decoder.
 

pgandw

Active Member
Andrew

Disclaimer: Please keep in mind that although I work in electrical engineering, I have never used DCC myself. I have read many of the forums to try to get a feel for it, but being restricted to small HO and HOn3 steam locomotives, very small layouts, and single operator operations, I haven't taken the plunge yet.

After wading through the Soundtraxx web site and finding precious little data on the DSX, studying the various diagrams, here is what I have concluded. Some of the diagrams show 4 wires connecting the DSX, the Tech Bulletin you cited shows 5. In any case, 4 are identified - 2 for input power and signal, 2 to power the speaker. The 5th, if it exists, would logically be a synch signal from the motor decoder. However, I haven't seen any such output on a motor decoder.

Again, going back to the capacitor installation tech bulletin, note that the capacitor is NOT installed across the power leads. As in my earlier posts, a capacitor of that size directly across the track would likely result in DCC signal degradation. And using an electrolytic capacitor - again cited in the diagram and parts list - means that polarity reversal cannot be allowed. This cannot be guaranteed, even in the best DCC wiring. For other capacitors (speaker and motor), Sountraxx requires use of bipolar capacitors to guard against DC polarity reversal. Therefore, the use of an electrolytic capacitor in this specific case further supports my conclusion that the Tech Bulletin procedure is NOT installing the capacitor across the power leads. The Tech Bulletin mandates the capacitor be installed on 2 external pads with polarity markings, and that if these pads/markings are not present, the modification must be done at the factory.

Based on all the above, I assume there is diode polarity protection inside the decoder(and quite likely other circuits) between the power leads and the circuit insertion point of the capacitor. These internal ciruits would limit the capacitance seen by the track, yet allow the large capacitor to function as a temporary battery, similar to the circuits in VCRs and digital alarm clocks that keep the time while the device is momentarily unplugged or drops power.

Bottom line: You want to avoid excessive capacitance across your track. The signal is quite robust, and at a reasonable frequency, so that the little capacitance of the parallel rails and bus wiring can be shrugged off. But I don't think (my engineering guess) your DCC system will function reliably with 1 or more 220ufd capacitors across the rails.

An analog engineer occasionally lost in a digital world...
Digital is all fantasy anyway - the real world is analog!
 

60103

Pooh Bah
Anent capacitors: In Europe & Britain there is a requirement for a radio interference suppressor , usually a capacitor, across the motor leads. There is a debate about whether these interfere with the DCC decoders. A number of people have reported improvment in performance when these are removed.
 

b28_82

Member
Think of DCC as a type of AC.
I am in the electronics career field that deals with digital and analog type of circuits and what I have come up with is that DCC is a form of DC that is pulsed to the effect that it acts very much like AC. For instance when you are checking voltage between the rails on DCC with a multimeter you have to put the dial to AC voltage not DC voltage.
 
Be very careful about adding a capacitor to a DCC system if it is not required. Those who have bought QSI sound-equipped locos may be having problems when using them on DCC. It turns out that the capacitor is very large. This is because the locos are designed to also run on DC. When on DC, the capacitor keep the sound going when there is a decrease or an interrruption in the voltage.

On DCC, the capacitor sucks up a lot of juice when put on the rails. The command station detects this surge as a short and can shut down. When operating, the decoder sends out erroneous signals because of the function of the capacitor. This can result in losing control of all trains.

You may not experience this on your home layout because the command station can manage the initial surge. However, if you are operating on a club layout where there are lots of locos running simultaneously, your loco can create lots of problems.

Bob M.
 

MasonJar

It's not rocket surgery
Interesting info guys, thanks!

Good to have the HOTrak implications noted as well Bob... :)

Andrew
 
Top