I have begun setting up separate power districts on a layout controlled by a Digitrax Super Chief system. A PM42 power manager has been installed and a couple of districts established. The problem I'm having seems to be isolated to Athearn locomotives equipped with Soundtraxx Tsunami decoders. If one of these units is operating on either district, a short circuit on the opposite district also affects the Tsunami. The result is that this locomotive now stops every three seconds, restarts and moves, then stops again. Once the short is removed, the Tsunami unit sometimes "runs away" at higher speed. I've tried this test with MTH and Broadway limited units, and they do not experience this fault.Things tried so far:
1. Use another Tsunami loco- tried a dozen, all act the same, even in consists.
2. Disconnect the two districts and test track and wiring for short circuits- all checked fine, separate circuits.
3. Swapped PM42 with another- same problem.
4. Adjusted current settings on PM42- no difference whether it's 3 amps or 9.
It would seem the Tsunami decoders are overly sensitive since other brands work OK. I looked through the Digitrax and Soundtrax manuals but find no reference to this. Anyone have an idea?
Do you have the PM42 connected to the loconet? If so disconnect it and try the loco's again.
If the DCS100 is shutting down, then the sensitivity of the PM42 power district needs to be adjusted. You can set the DCS100 to a long delay before shutting down, and the PM42 to fastest shutdown. We had a similar issue at the club(16 PM42's/64 power districts).
Set the DCS100 OpSw 18 from 'T' to 'C' - this will increase the delay from 1/8 second to 1/2 second. This is described in the chart in you Super Chief manual. Now set the PM42 power district to 'Fastest' using the chart(table 4) on page 11 of the PM42 manual.
At home, I have a 5 amp DC100 and a 5 amp booster. The PM42 power districts are set to 4.5 amps for the 'trip' current, and Fastest for the 'sensitivity'. I have had no problem with Soundtraxx decoders.
Jim
Modeling BNSF and Milwaukee Road in SW Wisconsin
Also try disabling analog on the decoder (CV 29), and on the DCS 100 (one of the Op switches, not sure which one).
When the system comes back up after a fault, sometimes there is enough of a analog signal on the track that fools a decoder, and makes it run goofy.
Mike WSOR engineer | HO scale since 1988 | Visit our club www.WCGandyDancers.com
Thanks to all of you who took time to respond. The PM42 is not connected to the Loconet. The settings adjustments recommended will be tried this weekend at our next session, after which I'll let you know. Sounds like a potential solution. I've also been working with Soundtraxx on some suggestions about CV settings. It must be a decoder sensitivity issue, as we've got 14 "Tsunami" equipped units (some self-installed, others installed by a professional) on the layout on various power districts and they all react to shorts anywhere. The MTH, Proto, older Athearn, or Broadway Limited locomotives also on the layout are unaffected.
Jim,
I tried the settings you recommended. With PM42 set as fastest and DCS200 set slowest as suggested, all four PM42 circuit breakers began cycling upon restoring track power. They didn't stop, so I went down one level on PM42 to "Faster", and things got better. The Tsunami equipped units now only stop briefly when a short on another district occurs, then resume operation. Definitely an improvement. Any thoughts on why the PM42 doesn't work on fastest response?
Victor Jim, I tried the settings you recommended. With PM42 set as fastest and DCS200 set slowest as suggested, all four PM42 circuit breakers began cycling upon restoring track power. They didn't stop, so I went down one level on PM42 to "Faster", and things got better. The Tsunami equipped units now only stop briefly when a short on another district occurs, then resume operation. Definitely an improvement. Any thoughts on why the PM42 doesn't work on fastest response?
Probably the inrush current when the power is reapplied to the sound decoders. Many of them basically have the (poor!) circuit design of having a capacitor directly across the power supply which acts to keep the sound alive over small pickup glitches. With no other components, this capacitor initially appears like a short until it aquires a charge.
A slight modification using 1 resistor and 1 diode prevents this from happening. This was known long before the Tsunami saw production so why SOundtraxx didn't incorporate this I have no idea. The Digitrax sound decoders have this,a nd have no inrush problems - and they program on a regualr program track without a special booster. This inrush issue is also why so many sound decders have problems programmign on program tracks. As far as I can tell with a few hours with a new Quantum Revolution decoder it seems they also don't have the problem. The OEM QSI dcoders in BLI, Atlas, and P2K do seem to have it. I also did not have issues with a Loksound 3.5 in a PCM loco, so maybe Loksound incorporate the proper circuit as well.
And a pox on those who bail out with the excuse that "there's no NMRA standard for it". There is common sense engineering however.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Randy,
That explanation makes sense. Can you give me the information about the resistor and diode please?
Bill
Long ago Mark Gurries posted the schematic on one or more of the Yahoo groups related to DCC. I'm not sure exactly where anymore, but if you belong to any of those (as a Digitrax user you should at least be in the Digitrax one - lots of helpful people there) you should be able to find it. Basic theory is this (remember the keepalive capacitor is on the DC side of the decoder - after the rectifier bridge): the diode and resistor are arranged so that when the capacitor is being charged, the current flows through the resistor, reducing the inrush load. When the capacitor is being discharged, such as when the loco encounters a dirty spot on the track, the diode allows the current to flow out of the capacitor unrestricted by the resistor.
BTW a keep-alive capacitor can be added to just about any decoder if you can find the right terminals to attach it to, and also aren't concerned about the warranty on said decoder. It works for motor-only decoders as well as sound. Oh, here you go, Marcus has updated his we site and it shows the inrush limiting circuit as well as how to add keep-alives to other decoders: http://www.members.optusnet.com.au/mainnorth/alive.htm
Once again thanks for all your time. I'll try this next.