I use a Pi SPROG 3 DCC system controlled via JMRI on an N scale layout. I do not have Loconet. I plan to add a booster (or upgrade to a Pi SPROG 3 Plus) so stationary decoders are able to throw switches to clear shorts on the track bus. I am looking for stationary DCC decoders that will throw Kato Double Crossovers and other Kato switches, and the discontinued DS64 looked like a good, cost-effective option, until it was discontinued. I'm not willing to start out with a discontinued product with no drop-in replacement. But the sparse documentation for the new DS74 only mentions DCC control with a subset of functionality, and does not identify what functions are/not in that subset. Anybody have some hints or directions to more/better information on using a DS74 w/o loconet? If this new DS74 is intended to be a (not drop-in) replacement for the DS64, it would be behoove them to provide documentation for use with DCC exclusively (sans Loconet). Alternatives: I could add Loconet to my JMRI setup to control the DS74, but would rather avoid that hassle and cost for this purpose. It is also not clear to me whether JMRI would let me control a DS74 on Loconet from my Engine Driver throttle. I'm also looking at a ZIMO MX820V dual output, Kato-compatible stationary decoder, but it would be quite a bit more expensive, since its individual outputs do not have enough current to drive the dbl xover w/o separating the switches out (and using 4 outputs to drive them). Two such Zimo's are needed to drive one dbl xover, but one DS74 would do it, without modification.
While there is no separate sans Loconet section in the manual, it does mention in several places 'it will do this without Loconet' or 'it will do this with Loconet'. I haven't done a thorough read of the manual, but what I gather is the only features that don't work without Loconet are features that cannot be implemented without it (which I believe is the same for the DS64). Since the DCC track signal is a one way communication, that would be anything that requires communication out from the DS74 to either the command station or another device, including other DS74's. For example, if you use the DS74 inputs it would need Loconet for it to report the input states back to the command station (the inputs will still work to throw the turnouts). Also, input triggered routes won't work across DS74's without Loconet, because there is no way to get one DS74 to send commands to another one (you could still have command station or throttle triggered routes work across multiple DS74's by having routes in separate ones triggered by the same turnout command).
Robert, Thanks for the encouraging reply! What you say is kinda what I expected, but I still cannot find HOW to do it with only DCC. And AFAIK, there is no DS74 decoder definition for JMRI yet. The document defines all kinds of OPSW entries as either T or C (assume T=thrown=1, C=cleared=0?), but in DCC are those five consecutive 8-bit registers, or what? Is this manual you are mentioning and quoting available online? The online DS74 "Instruction Sheet" (only the link to it identifies it as that, the doc title appears to be "DS74 Quad LocoNet Stationary/ Accessory decoder for turnout control.") does not say "without Loconet" at all. The only other pdf online I can find is a one-sheet description with very little documentation. Slightly enlightened, but still confused,
Yes, the manual I'm referring to is the "Instruction Sheet" on the DS74 page. I guess I should have stated that I was paraphrasing when I mentioned that it talked about with and without Loconet. For example, under "3.0 Setup DS74 Ez-Routes:" it states "If this DS74 does not have a LocoNet connection but operates from DCC Track power on the TRKA/B terminals, then it will ignore Route Entries that do not fall within the DS74's Base address and next 3 SW addresses." When doing OPSW programming, you don't deal with bits and registers, you send turnout commands to the address that corresponds to the OPSW you want to change. For example, if you want to enable the capacitive discharge in pulse mode (it might be needed for that double crossover) you need to set OPSW 16 to C. To do that, enter OPSW programming mode (Section 2.0 step 2). Send a 'C' or 'Closed' command to switch address 16. Exit OPSW programming mode (Section 2.0 step 5).
I have been in the same conundrum in deciding which decoder to use. The DS64 seemed to be the better choice compared to the NCE SwitchKats if for no other reason than price as would need to buy 4, total almost 2x the DS64. Stumbled upon the Digitrax intro video in its first week of release, then the online documentation. Left with the same thoughts; the DS74 is marketed with locoNet focus as the docs mention practically nothing in its use in a non-loconet environment. Don't recall if it was CSXRobert or RBrodinsky who provided some insight in how one can use the DS74 with MOM toggles since the DS64 supported, quite clearly in the docs thru OPSW 12, and the DS74 without a respective OPSW. Subsequent email to Digitrax and reply indicated push buttons only. As I'm yet to finalize decision on MO pushbuttons vs toggles because the layout hidden panel for whichever one is selected needs to be built differently. Kinda waiting to hear more feedback from DS74 users. Having a double crossover, limits options as you know but will probably when ready to proceed, take a closer look at the Switchkats since they seem to be designed for connection to DCC track, and supports either type of switches-Push or Toggle.
I remember this discussion, but I can't seem to find it now. I was thinking I had explained this but maybe not: the DS74 can be used for momentary toggle control of the turnouts, but since it doesn't seem to support the OPSW 12 option, it's a little more complicated. You have to configure the inputs for route control and then program each route to throw the wanted turnout in the desired direction. For example, route 1 throws turnout 1, route 2 closes turnout 1, route 3 throws turnout 2, route 4 closes turnout 2, etc.
Mark, the DS64/74 cost advantage is even better than that! They will control FOUR Double Crossovers each (one dxover on each of the four DS64/74 outputs.) And no need to modify the Double Crossover's internal wiring to separate the 4 coils and bring out 3 more pairs of wires. Thanks Robert, you've been very helpful! If I understand you correctly, if I have multiple DS74s on the same DCC bus, I just put one in OPS mode (via pushbutton), and then send commands to the switch address corresponding to the option. If there happens to be a real switch decoder at that address, it will harmlessly throw or close that switch too (I probably shouldn't be running trains on my layout while trying to configure a DS74!) Not being a Digitrax system user, some of their nomenclature was not familiar to me. I don't blame Digitrax (or any other DCC system manufacturer) for emphasizing the benefits of their ecosystem, but I'm very happy with my Pi SPROG 3 system, run with JMRI, on my small layout (albeit complex, with 11 switches and crossovers, or 19 if each individual switch in every crossover had to be controlled independently). I'm leaning towards an on-screen JMRI control panel, rather than a physical switch panel. Or, I'll just use throttle commands to throw the switches/crossovers.
Hi Robert. Oh, you explained it perfectly. Thank you! Your approach was great, I was really commenting on the lack of non-loconet discussion in Digitrax's docs. If anything, I don't know, maybe its me, would have thought they would have promoted non-loconet more than 64..
Since the MSRP for SwitchKats is $33 and would need 4 is $132 compared to the DS74 is $79 is how the estimate was arrived at. I have 1 D X-over, 2 RT #4s and 1 #2 wye so the DS64 or 74 would be perfect. And has been on top of my list. While I don't blame Digitrax for promoting their own ecosystem, much like Apple, the standard, Apple still provides important info for interfacing or using with non MAC-OS or IOS devices. Oh well, one thing for sure, I haven't seen or read much in terms of success or failure of a DS64 in a non-loconet environment so holding off for now. Currently using Pi4-JMRI and DCC++EX Mega and plan to stay with that hardware. The SwitchKats from what I have been told can handle 1 d x-over. Never saw any sense or would even consider opening D x-over to separate solenoids. I'm most concerned about consistent operational functionality and reliability. Whatever product gets me there at the most fair price is where I want to be. Hope it works outs for you and will be eager to read about the outcome. Thank you!
I have not heard that. But it would not surprise me. How many times it can do it, in a given time, at a given ambient temperature, may be another matter. A lot of electronic components can exceed their specified capacity, for a while, especially at room temperature. That's one of the benefits of capacitive discharge. It can be self limiting on duty cycle (it takes time to recharge the capacitor between uses), which protects other components.
Thank you for the feedback.... Bumped into one of the DCC guys at the store some time ago who was using Switchkats for Kato D x-over and seemed satisfied with its performance but I have no direct experience with the NCE product or have discovered much other info on the stationary decoder best for the layout and Kato D x-over. Hard to judge without having them physically, and without schematics or detailed specs though I've found, generally with electronics and at any given price range, component quality about equal. Remember when buying hifi stuff with multiple pages of specs, today lucky to get a half of page. Guess its just a sign of the times. No question that the design and proper selection of components plays important role. Most devices rated at, say, eg 1 W at ambient temp of x is generally spec'ed at 100% duty cycle so that same device operating at 50% duty cycle can dissipate 2W. As you know, the charge and discharge cycle is what is defined as the RC time constant and is based on the value of capacitance times the resistance in the charged-discharge circuit. eg, It will take 50 secs for a charged capacitor rated at 1ufd to essentially fully discharge thru a 10M Ohm resistor. Proper capacitance discharge is used for human safety concerns too! Ensuring proper discharge of filter caps in power supply crucial for safety whenever troubleshooting inside stuff. At any rate.... Your initial post sort of hit a sore spot with me as newbie because documentation and product info seem to be lacking throughout the hobby. Very frustrating at times for a new person. (3rd MRR but last 1 was in late 80s so consider myself new) Actually, consider Digitrax among the best along with Woodland Scenics since researching began last September. Sorry for the little venting. Very interested in DCC control of switches and the 74 so looking forward to future postings.. promise, no more venting on the topic.
"50% duty cycle" also includes 24 hours on and 24 hours off... Therefore duty cycle alone is not enough to tell whether a device will overheat. Thermal mass, and any heat sinking thermal conductivity (and capacity of "cold source") also come into play. Optimizing for non-continuous utilization is it's own art/science. My former (I'm recently retired) employer's customers had safety regulations for maximum time for unsafe voltages/charges to dissipate to safe levels after removal of power. It stinks when you have to design (and consume cost, volume, and weight for) additional circuitry just to turn the thing off!
Absolutely 110% agree! That's so funny.. What is really off? Years ago, one thru a manual switch, a true disconnection from the AC input, and there was no question that power was its status Today, I think about all the surge suppressors and outlet strips bought just to actually shut things off. Can't imagine how many billions in wasted power and money with all the devices that are somewhat "OFF" Maybe I'm too old?? The duty cycle relationship is based assuming a given ambient temperature usually obtained from the device's data sheet. The device's ambient temperature can be controlled via heat sinks and appropriate thermal compound along with proper design. One of the reasons 2 of my 3 purchased L298 motor shields use the H packaging, though now slightly moot as a TAM Valley DAB002 booster has been added. Wanted the larger heat sink because it allows the L298 to run cooler. Used to teach electronics troubleshooting so always mentioned looking where high power is being dissipated with corresponding increase in heat, as a good place to start, eg visual inspection of components. Design or repair to prevent active device thermal runaway or passive component failure. Or as in the ham days. prevent maximum smoke and minimal signal. Been away from electronics for a couple of decades.. 1 of reasons got back into MRR and its great to talk electronics again...
Jake, Mark etal... Did you ever solve this problem implementing the DS74 into a Kato non-loconet environment? I currently have an NCE PowerCab but am strongly considering the ESU ECoS command center and using their layout map interface to control turnouts. Like you both, I am on the fence (although I do have the products in my shopping carts at 2 vendors) until I have more info. What turned me on to this solution was the nscalewonderland video of a few years ago, but using DS64s at that time. Ron
Hi Ron, Forget your reason for going with either a DS64 or DS74. Driving the 4 solenoid Kato double crossover was my primary reason. No, decided in January to buy the DS64. While there was some evidence here and on other boards that the DS74 could work out of the box when only one is used, it appears that a DT602 with July 2021 firmware or later is the only way to program additional DS74s. Like to have 3 independent references or sources when researching for purchase. The ESU looks really nice. The non-Loconet DS52, and its ability to throw a 4 solenoid double crossover, is what I'd select if needed another decoder for turnouts. Did not see it till mid February or might have bought that instead of the DS64. It can easily be found for under $25. If you haven't looked at it, see attached link, page 3 about 1/2 way down, references the double crossover. I have one in the cart in preparation for a small switching layout with a Double Crossover. Also, want to have options in the event I do another full layout or the DS64 malfunctions. DH123D (digitrax.com) Interested in what you decide to do...
I don't have any DS74s yet so I can't speak from experience, but according to the manual, that's not the case and you can, in fact, use multiple DS74s without a DT602.
I started a thread (https://www.trainboard.com/highball...stationary-decoder-on-dcc-w-o-loconet.140735/ ) asking about the DS74 use on non-digitrax systems without loconet. Long story short, it is possible to configure DS74s on the layout, one at a time, via DCC. I have not progressed to the point on my layout where I have tried the DS74s yet, but plan to in the future. I do not know if the DS74's have the same issue: losing their configurations when commanded to throw a switch to clear a track short, when the decoder is controlled via the same DCC bus that is shorted. Note that even if they did not corrupt their configuration, they would still not necessarily be capable of executing the command and throwing the offending switch, even if the command is correctly received via the shorted DCC bus. Thus it is important to operate all such stationary DCC decoders controlling track switches (turnouts), that could cause/clear DCC bus shorts, from a DCC bus that is not compromised in the event of a track short. That often means powering the stationary decoders off the main DCC output, upstream of boosters (or possibly short circuit protectors) driving the track DCC bus. This is another reason I am hesitating deploying any stationary DCC decoders for track switch control just yet.
As stated earlier in the thread, not only is Loconet specifically part of the product name, but one essentially has to parse the DSD74 instruction sheet to find reference to programming w/o Loconet. The DS64 and DS52 instructions are quite clear on how to program, step by step, w/o loconet and also highlight, can work out of the box assuming default addresses 1-4 does not conflict, to connect and presto, you are in business. Can't find anything similar procedurally in the DS74 instructions. Actually, I did not say it would not work, only said that based on the DS74 heavily focused LocoNet emphasis and the lack of concrete user experiences, leaves some question for programming w/o loconet as compared to the fairly straight forward DS64 and 52. Track isolation via a booster is good practice regardless of the track connected stationary decoder used whether DS64, DS74, DS52 or other unless a separate bidirectional bus eq;LocoNet is employed. As a consequence of using Tam Valley DAB002 booster, the DS64 has been operating magnificently, even upon a few "forgot to throw the switch" shorts. No hiccups, no resets. Lastly. can't imagine why 2 new DS64s sold on ebay yesterday sold for around $100 each when the newer DS74s can be bought from online dealers less than $70. Thinking in a non-loconet environment, the "non-Loconet user friendly" dual decoder DS52, MSRP @$28 a piece is less expensive than the DS74 @79 MSRP. Most of the DS64s sell for more than the DS74 street price. In the end... Maybe the LocoNet in name, lack of clear track programming instructions, is by design---Digitrax wanted to produce separate higher powered decoder for non-loconet users???
