So finally I have this issue resolved. Long story short (no pun) I needed to do 2 things. The first was suggested by many here, and that was move the switch to in front of the Yard PSX, then every time I power up the Yard, whether from a cold start or during a session, the PSX sees it as a system cold boot. BUT, it only worked after the second component of the solution. I spoke with Eric at Tony's Trains where I purchased the boards and he instructed my how to make the boards Digitrax friendly; that is, install jumpers on every board on J7-3 to J7-4. Once both procedures were done all works as it should. Much appreciation to all of you who helped with your advise, as always.
http://cs.trains.com/mrr/f/744/p/248254/2769589.aspx#2769589
Alton Junction
[quote user="DigitalGriffin"]
you could use a power up delay circuit which I posted here a couple months back.
That way only 1/2 of your yard tracks would power up at a time. I created this circuit to deal with current inrush on my 18 stall roundhouse + 18 stall diesel shop.
I couldn't open the link, but I would be interested in learning more if moving the switch doesn't solve the issue. Thnx for the suggestion.
The switch is connected to the output of the PSX, so after it. When I use it during sessions the PSX is already powered up and not a cold command station boot so it sees the inrush as a short. So my plan (Randy's original idea) it to move the switch to in front of the PSX, connected to the output of the Super Chief. Then whenever I switch it on I am powering up the PSX (even if the command station is already powered up) and it thinks it is a cold command station boot, and hopefully responds to the inrush accordingly. I'll advise in a few days.
Don - Specializing in layout DC->DCC conversions
Modeling C&O transition era and steel industries There's Nothing Like Big Steam!
Is your switch ahead of the PSX. DCC Specialties reply in NYC post states that the PSX woun't trip if it gets powered up at the same time as the track.
If your switch is after the PSX, it sees the current inrush as a short.
Rich, and all others, I haven't forgotten to reply. I just have been tied up with non-railroading issues but this week I will move the switch to before the PSX and report back.
Tophias So Randy, as usual, I think you are spot on with my situation. After another day of testing/experimentation I think the central issue is the number of sound locos being powered up at one time. The BLI Hudson I have previously mentioned is a current hog, but it's not the primary concern. The primary concern is the TOTAL number locos beping powered up AND when they are powered up. Long story short, I think your suggestion of placing the Yard switch BEFORE the PSX is the ultimate solution. I'll be away for several days but when I return I will re-route the switch wiring and let's see what happens. As always, thnx for your advise.
So Randy, as usual, I think you are spot on with my situation. After another day of testing/experimentation I think the central issue is the number of sound locos being powered up at one time. The BLI Hudson I have previously mentioned is a current hog, but it's not the primary concern. The primary concern is the TOTAL number locos beping powered up AND when they are powered up. Long story short, I think your suggestion of placing the Yard switch BEFORE the PSX is the ultimate solution. I'll be away for several days but when I return I will re-route the switch wiring and let's see what happens. As always, thnx for your advise.
Rich
rrinker Shouldn't be an inrush issue with just 1 loco. About the only defect the switch could have (since it does pass current, so it's not completely broken) would be high resistence in the contacts, which can cause issues, same as insufficient feeders. That's more of a "device doesn't trip when there actually IS a short" sort of thing though, when the wire resistence is such that a dead short doesn't actually draw more amps than the breaker is set for. But a higher than expected resistence onthe contacts could cause the inrush to last longer, since it will take longer for the capacitor to charge up. Do try without the switch, and also make sure the wires under the terminal strip going to the track are all clean and the screws tight. It's strange that it works fine if everything is turned on, which leads me to believe it's NOT the switch - jumpering past the switch should be the same as having the switch turned on, and we already knows it works fine if the switch is on at the beginning. Center off or not on the DPDT doesn't really matter if only the center and one set of end terminals is used, even if it does make before break it's only connecting a set of unused terminals. You're just using it as a DPST to turn off both rails. I'm almost tempted to think this is normal, if it's just a momentary short indication. By turning the switch on, it is causing a sudden change in the output of the PSX, which is may indeed interpret as a short and then when it realizes it's not, goes back to normal operation. The amount of time it indicates a short is sort of the minimum cycle time for short detection, verification, and control from the microcontroller. If you put the switch before the PSX, then it will be like turning on the track power with the existing switch already closed, which seems to be fine. --Randy
Shouldn't be an inrush issue with just 1 loco. About the only defect the switch could have (since it does pass current, so it's not completely broken) would be high resistence in the contacts, which can cause issues, same as insufficient feeders. That's more of a "device doesn't trip when there actually IS a short" sort of thing though, when the wire resistence is such that a dead short doesn't actually draw more amps than the breaker is set for. But a higher than expected resistence onthe contacts could cause the inrush to last longer, since it will take longer for the capacitor to charge up. Do try without the switch, and also make sure the wires under the terminal strip going to the track are all clean and the screws tight. It's strange that it works fine if everything is turned on, which leads me to believe it's NOT the switch - jumpering past the switch should be the same as having the switch turned on, and we already knows it works fine if the switch is on at the beginning. Center off or not on the DPDT doesn't really matter if only the center and one set of end terminals is used, even if it does make before break it's only connecting a set of unused terminals. You're just using it as a DPST to turn off both rails.
I'm almost tempted to think this is normal, if it's just a momentary short indication. By turning the switch on, it is causing a sudden change in the output of the PSX, which is may indeed interpret as a short and then when it realizes it's not, goes back to normal operation. The amount of time it indicates a short is sort of the minimum cycle time for short detection, verification, and control from the microcontroller. If you put the switch before the PSX, then it will be like turning on the track power with the existing switch already closed, which seems to be fine.
--Randy
No Randy, no spoke. So after having dinner and clearing my brain, here's the result of my last test. If I switch OFF the yard district before system power up I can power up with up to 19 sound locos. My example was this - yard switch off, 5 locos (including the old LLProto2K gp's with QSI Quantun decodes) on main layout district ON, powered up command station OK (5 locos). Then turned ON Yard switch and everything OK. That's a total of 19 sound locos (I wasn't aware I had that many ). Of course, thats wth BLI Hudson in a block OFF on main layout. So, I can live with this. I'll just have to finagle how to power up the Hudson, which I don't use too often anyway. I'll review the previous suggestion solutions regarding jumpers on the PSX and the CV changes, though I'm not sure I understand that concept. Be ready for a new thread next week! . As always, I very much appreciate everyone's help and advise.
Do you have the smoke generator turned on in that BLI loco?
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Well I'm back after 45-60 minutes of testing of all kinds of different senarios of first system power up, including with the Yard DPDT switch off and on and also Yard circuit power up with the DPDT switch after the system has powered up. My conclusion is there is nothing wrong with the switch. Second, I found the central issue is Inrush involving one loco, a BLI 4-6-2 Hudson. If the Hudson is in a block (from my original layout block system) that is off I can power up the layout (all 4 circuits) with up to 17 sound locos, most of which are Tsunamis and a couple of LokSound (plus 3 old P2K circa 2001 with QSI Quantum decoders that I thought might be also an inrush problem). If the layout is fully powered up, and I switch off the Yard circuit with the DPDT switch, I can then power up the block with the Hudson along with at least 3 other locos (the 3 P2K's with the QSI Quantums; I didn't try anymore because all other locos were in the yard). I can then togglle the DPDT switch for the Yard and it sucessfully powers up. So it's that blasted BLI Hudson causing the issue, I believe so far. I guess it has one of those high energy capacitors? One last thing, there is a minor issue I can live with and that is after the number of Locos in the total being powered up exceeds around 7 the PSX beeps for about 2 seconds then everything is OK, as my original post started. I can live with that though I can further look at some solutions suggested earlier by you. Again, to you all, thanks for all your ideas and suggestions. Whatever I think I know about DCC I have learned here! DC was much simplier!!!!!
Possible guess:It is possible that the PSX delays a short indication when it detects a power up condition (as in your case) as there tends to be line sagging as everything powers up. With a fully running system, your voltage is immediate when you flip the DPDT.THAT said you don't have to do this. I believe QSI, Soundtraxx, and possibly loksound have a timeout value that will AUTO MUTE the engine after a certain amount of inactivity. (No speed & Function button presses after so many minutes:seconds)
I have eight of the older Power Shield breakers and I have had the same experience with having them not want to reset after a brief short. Sometimes I would "bridge" the rail gap with something metal between power districts and this would allow the tripped one to reset.
Thinking it was high "inrush" from having many sound locomotives in one concentrated area I jumpered the trip current to the max. but this didn't seem to help with the reset function.
I have the "disable" switches and LED indicators wired to mine.
What I did do that made a big difference is to make sure I had plenty of feeders to the rails on each section of the power districts. Currently I only have one district that still gives me trouble when resetting and I am adding additional feeders to alleviate the problem.
My PSX breakers perform flawlessly but the reset problem has been vexing at times. Take a look at your feeders and the continuity of the wiring to your track to be sure it is beefy enough and that may solve your problem.
http://www.dccspecialties.com/products/pdf/man_psx1.pdf
Have you looked at the Weak System Boost jumper?
J7-4 to J7-3 is the Weak System Boost enable jumper. When the jumper is installed, the breaker will use a turn on algorithm that is designed to assist weak system boosters in starting high inrush current loads. It will also help low power boosters such as the Digitrax Zephyr and the NCE Power Cab start difficult loads. When CV53 is zero, the jumper will enable/disable this function. If CV53 is a nonzero value, then the Weak System Boost function is enabled regardless of the presence or absence of a jumper. On Digitrax systems, setting OPSW #18 to “CLOSED” will extend the short circuit shutdown to 0.5 seconds. This will help prevent the booster from shutting down during a short circuit.
Good Luck, Ed
Tophias And if your asking, why not just silence locos with f8 when they're in the yard?
And if your asking, why not just silence locos with f8 when they're in the yard?
So I guess NYC H10b's explanation as quoted from DCC Specialties explains it, except for my testing with just one engine Seems to conflict with it a bit. So I will try to test by by-passing the switch with movable jumpers and see what happens. And if your asking, why not just silence locos with f8 when they're in the yard? I can, but most are consisted and only the lead unit will silence. Regardless, nome of this is a huge problem. I can always do as DCC Spec. says with cv's or I can just use the main command station track power function to toggle off (which will silence my locos which are all primarily Tsunamis and I use the jmri setting of no sound power up until the address is called) and then toggle on. As a lone wolf this won't interfere with other operators -there aren't any others (). I will do the jumper test today and report back. In the meantime, thanks to everyone so far for their suggestions and inputs. Being "on an island" you folks are a wealth of knowledge to me and I really appreciate it.
And, as Randy points out, the placement of the kill switch relative to the PSX may make a difference.
I have several Radio Shack DPDT switches I used to reverse power on turntables and a wye that are NOT center-off. I also use PSX circuit breakers. But I've had nothing like your problem. RS stuff isn't top end, but it is usually workable and reliable when it comes to components like this.
As Rich noted, might try subbing in a SPST to see if that helps.
Mike Lehman
Urbana, IL
Could a DPDT be the problem? Why not just use a SPST?
http://cs.trains.com/mrr/f/744/t/228155.aspx
Thnx Mike, I was starting to think it might be the switch as well. There is nothing connected to the other set of poles. Tomorrow I will test it with the switch removed from the barrier strip and using jumper wires. If it doesn't exhibit what I've described then it's the switch. If it does then it's probably the inrush issue.
I was experiencing this issue with the PSX breakers back in August of 2014 and contacted DCC Specialties directly. Here is their response:
"Don't turn the blocks on and off; leave them on. Here is what is happening. Most of today's decoders have fairly large filter (or hold up) capacitors to ensure reliable operation even with dirty track, unpowered frogs, etc. When you turn off the block, this filter capacitor discharges. When you turn the block on, the discharged capacitor is a dead short across the rails and draws a large current. As the capacitor charges up, this current decays exponentially. The PSX detects this high current spike and trips. Depending on the specific current draw of the engines involved and the size of the input filter capacitor, some combinations may generate enough surge current to trip the breaker and some may just sneak under the wire. You may correctly note that when the breaker turns off after tripping, these capacitors will again discharge, so you might ask why when the breaker turns on again is doesn't simply trip? The answer is that after it trips, the breaker KNOWS that it is turning on and it is looking for this current surge. It has special software built in that allows it to monitor and control this surge so that it can turn on with a large number of decoder loads without tripping. The difference is that you are imposing this start-up load AFTER the breaker is on so it does not know that it is a start-up load and it does its job of turning off to protect the layout. So, the simple answer is to stop turning these blocks on and off. If, for operational reasons, you need to do this, then the other method is to use the PSX to turn the block on and off. If you do it this way, the PSX knows the block is coming on and will not allow the start-up surge current to trip it. By default, all PSX devices can be turned on and off using accessory address 2042. Accessory command 1 turns it on, and 2 turns it off. Following the directions in the PSX manual, you can change this default address to any value between 1 and 2042. One other thing you can try is to set CV49 to 4. This will increase the trip current threshold to around 6 amps. It will still trip with a true short (the NCE can easily output more than 6 amperes for a short time), but by increasing the limit, you are somewhat defeating the reason for the breakers in the first place: to protect the layout from current spikes and overloads."
So as a workaround, I increased the number of PSX breakers from 4 to 7. I installed the additional three breakers where I often throw block switches and do not want interference with other operations. For example, the engine terminal now has its own PSX so that throwing a block switch there won't interfere with a train running on the main line.
It may very well be the DPDT switch. Since there is no center-off, there may be just enough momentary overlap when it switches poles to cause the PSX to sense a short.
A DPDT center-off would be better for what you want to accomplish, because if you're not wanting sound, center-off is where it's dead. I understand you have track power through one set of poles on the DPDT. What's on the other set of poles? Nothing? Or track power with reversed polarity?
Just an update - I cleared all of the locos from the district except one (a BLI Paragon 2 steamer) and it did the exact same thing. So is it still an issue of inrush? With just one loco? Thnx all for your help
No adjustment - if it works when you power the layout with the switch on - it's because the PSX can handle the inrush when power is first applied. But once it's already powered, it has a harder time detecting inrush vs a short. You should put the switch in front of the PSX and it should work, or even connect the switch to the PSX, one of the terminals is for adding a remote control switch to turn the power on and off.
There is also a "low power system boost" which originally was only for low power DCC systems but this will work for others, it delays the reaction to the inrush current of starting up all the sound locos.
Thnx for the replies. But if that is the issue then I need to ask why does it not do it with the switch in ON position and I boot the Digitrax command station and apply power to the track? Would that not have the same effect? Not being critical, just asking. Thnx for your help.
I may be wrong but I think you don't have a short at all.My guess is that your stored sound equipped locos are the cause here,sound decoders need an extra input of current on initial power-up to charge their condensers and that's what your breaker actually sees as a short.Once the condensers are charged,current draw is minimal.