LGB 2060 Tips

This LGB 2060 industrial switcher originally came in a starter set. It doesn't look too much like the original due to the black paint and weathering. It was originally rather bright orange color. LGB fanatics may cringe at the changes, but LGB locos make good razor saw fodder because they typically run well so that the eventual bash runs well too even though it may look trashed.

The purpose of this page is to document the installation of DCC into the loco and then the eventual removal of this equipment from the loco due to decoder related performance issues. Then batteries and a radio control receiver were installed. Sound also went in. Then, many years later and with a much better decoder, DCC went back in in addition to all the other stuff. This time around, it ran fine on DCC.

• Isolating A Three Terminal LGB Motor Block
• DCC Decoder Installation
• Sound Installation With A Soundtraxx DSX Sound Only Decoder
• DCC Removal 11 Jun 10
• Battery and Radio Control Installation in the 2060
• Sound Installation in the 2060 14 Sep 09
• DCC, One More Time 3 Oct 09

Many LGB locos are wired to allow them to accept a DCC decoder. The motor leads are brought out of the power brick separately from the track pickups. These locos are usually designated with a "D" at the end of the model number or on the bottom of the brick or by some sticker that indicates DCC or MTS compatibility. However, this little 2060 that came in an industrial starter set is the older version with only three terminals on the top of the brick. It is not difficult to modify the brick to isolate the motor.

To gain access to the wiring, most of the loco must be disassembled. First remove two screws at the end of each hood. Then remove the two steps under the cab and two more screws hidden way up in the tanks. These screws are accessed through holes in the bottoms of the tanks. With the cab loosened, the hoods will come off easily, then you can actually remove the cab.

There are two strap like brackets on the underside of the loco that hold the brick to the bottom of the frame. Remove the four screws that hold these brackets in place and pull out the brick. Then remove the wire terminals from the brick by pulling on each one. Note the colors of the wires that go to each post. Then remove two screws that hold on the brick cover and remove the cover. The motor can then be just pulled out of the brick.

There are three metal posts sticking out of the top of the brick. They are labeled br (brown), gr (green) and ws (weiss or white, but actually the wires were black). The black leads are the common leads and this is the area that needs modification to accept DCC. This engine doesn't have a motor switch, so the brown and green terminals are jumpered.

The motor terminal tabs rest against the two outer posts. Electrical contact is assured by the spring action of the motor tab edges against the posts. The single inner post goes to the power pickups on one side. The outer post by itself goes to the power pickups on the other side. The other outer post doesn't go anywhere. Power is delivered to the post from the top through a green jumper wire. DCC installations absolutely require that BOTH motor leads be isolated from all other wiring and connected only to the DCC decoder itself. A simple and reversible modification is needed to isolate the motor.

The motor tabs leave the motor at an angle with respect to the posts so that an edge presses against the post. All that is necessary is to use a pair of needle nose pliers to gently bend the motor tabs so that they are parallel to the posts and do not touch them. Wires are then soldered to the ends of the tabs and insulated with heat shrink tubing.

With the engine disassembled to this level, this is a good time to lubricate the engine with gear grease on both axle gears and oil on the axle bushings.

These wires are then routed by the posts and up through an access hole in the brick cover and along with the rest of the wiring up into the shell. I didn't bother to color code my motor wires because I usually figure out the correct polarity by test anyway. The LGB wiring is unmodified at this point and the original wires can be pressed back on the proper posts after the cover is reinstalled.

[ Top ]

The 2060 and most other starter set engines are so simple the that DCC wiring is a snap once the motor is isolated. If directional lighting is not desired, then all that is necessary is to tap into one set of lighting wires for the DCC power input and to wire the two new wires to the DCC decoder motor output.

I chose to wire the headlights to the decoder to allow directional lighting. This is done by cutting the headlight wires about 1.5" away from the bulb sockets and splicing extension wires on them and wiring them to the decoder per the decoder manufacturer's instructions. I chose to extend the lighting wires because the rear headlight wires were too short to make it easy to get the rear hood back on. The remaining long wires that went to the front headlight were insulated and left ready to hook to a future sound system. The wire stubs that went to the rear headlight were used to power the DCC decoder.

The decoder (a Digitrax DG580L) was installed under the front hood. The decoder is mounted crossways on the rear of the weight with some foam tape. A large storage capacitor (47,000 uF, 25 V) is also mounted on the weight. Refer to my DCC Tips page for more information on the capacitor. I didn't cut any of the decoder wires so the extra length is simply looped around to get it out of the way.

I typically don't pay much attention to the polarity of the motor or decoder power wires until I am ready to wrap up the installation. I just connect them and test the engine. If the engine goes the right way when running on DCC, the motor wires are correct. If not, I just reverse the motor wires and do a final clean up on the connections.

I then check the direction of the engine in analog mode by putting an unconverted engine on the same track. If the engines go the same way under regular track power, then the power leads are correct. If not, the decoder power leads are reversed and the connections are cleaned up.

Large scale locos use the opposite polarity convention as compared to HO, so the manufacturer's instructions about the left and right rails must be reversed. I've tried to keep the instructions straight, but I usually get one of the connections wrong anyway so I've adopted the cut and try approach. It works for me.

[ Top ]

This is a little engine, so I chose to install a little sound system in it. I selected a Soundtraxx DSX sound only decoder. This is a DCC decoder that doesn't have any motor controls in it, it reads the DCC packets and creates and controls sounds instead. I used and EMD 2-stroke type sound system, probably not appropriate for this loco, but it sounds like a diesel anyway. The audio output power is low by large scale sound system standards, but since I was going to drive a small speaker, a lot of audio power would not be usable anyway.

The Digitrax motor decoder and the Soundtraxx sound decoder are wired in parallel to the track. Each is programmed to the same address so they both respond to commands directed to the engine. Wiring two decoders in parallel pretty much wipes out the capability to program either one on the programming track as the command station sees too much load. In any event, the storage capacitor also makes programming on the programming track impossible anyway. I set the decoder address independently before they were wired together so that once they are installed, OPS mode programming can be used to program both of them. I'll only need to mess with the wiring to disconnect them if I choose to change their addresses.

The biggest problem in this installation is finding room for a speaker. I fished around for quite awhile before I concluded that a normal speaker just wasn't going to fit. This 1" speaker that came with a Dallee system. The photo is about twice its real size. I determined that it could marginally handle the power of the DSX and sound marginally acceptable IF it was mounted in a proper enclosure. I also determined that the "long" hood of the model would make an adequate enclosure if a few holes were plugged up and the speaker was mounted so that it projected into the cab. The sound could then escape the engine from the cab windows. Since only one window is normally open, I had to remove the rear window piece to allow enough sound to escape.

To properly seal the hood area, the wiring had to be cleaned up so that it was channeled along one side of the frame through a gap near the cab floor. The wire bundle nearly fills the gap and plugs it as good as it is going to get.

Some small tabs of styrene were used to plug the remaining gaps. A long gap between the wall that the speaker is mounted on and the floor is plugged by a strip of foam tape mounted to the floor. The speaker itself is mounted with hot glue over a hole cut in the wall of control stand in the cab. The DSX decoder is wedged in near the speaker and held with a couple of globs of hot glue. The DSX requires an external blocking capacitor in series with the speaker. It is mounted under the DSX and hot glued in a corner. At large scale track voltages, the DSX also needs a 39 ohm resistor in series with the track power leads. It is inserted in one of the power wires. I used DB-25 type connector pins to allow the cab assembly to be completely separated from the frame in the same fashion that I use to interconnect power between engines and cars.

In operation, I found that the DSX has some problems. The low audio output power coupled with a small speaker leaves something to be desired in sound level. Second, and more serious, the DSX requires excellent power continuity to operate properly. The 2060 doesn't have particularly good power pickup and the DSX looses power and resets often resulting in a series of pops and clicks when the track and wheels are less than clean. Other large scale sound systems with batteries can ride over power outages with little difficulty, the DSX cannot.

[ Top ]

Even with all this work this engine's performance with DCC just didn't meet my expectations. After a year or so, I removed the DCC equipment and the engine became somewhat more tolerant to dirty track. I would appear that DCC is more suited to larger locos with better power pickup. In the future, I'll use DCC for the larger current hungry stuff for its precise control, virtually unlimited power and auxiliary function control. The smaller stuff will be left track powered or converted to battery power. A set of high performance batteries will drive a small loco well enough with the relatively light loads that this kind of loco can handle. Further, it will still coexist with the DCC stuff and it could run on track with any kind of power.

Update 11 Jun 10

Note that many years later, DCC was added back in with much better results, see below. The biggest change came from a better DCC decoder. The original DG580L was not a very good decoder and it responded poorly to track power interruptions. The much newer DG583S worked MUCH better.

Another important change was that in the interveaning years, I had developed a more effective and much easier track cleaning process using drywall screens dragged under a couple of track cleaning cars.

[ Top ]

Even running on straight track power and even with the sliders, the 2060 still doesn't have good enough power pickup to run on the same dirty track that the larger engines will run on without difficulty. There just aren't enough power pickup points to provide consistent power when each pickup point has low reliability by itself. There were two courses to follow, either clean the track or convert the loco to battery power with radio control. The track power option is cheap, but it doesn't allow command control. The radio control option involves some capital investment, but will allow the loco to run consistently and have command control. I choose to convert this loco to battery power.

I had two choices to make, what kind of batteries and what kind of R/C gear to install. The second choice was easy, I had recently converted an FA that had an Aristo ART-5490 mini-receiver in it to DCC so that the 5490 was available. I also had all the transmitters that I would need so that the receiver was a zero cash option, an easy pick. The 5490 isn't real small and the loco is so that there wouldn't be a lot of room for batteries. After testing the top speed of the loco with track power, I determined that I would need at least 16 volts to reach the speed that I desired. The 5490 has about 2 volts of drop so that implied about 18 volts worth of batteries. After playing with small battery packs made of dead alkaline AA cells and duct tape, I determined that I could get 15 AA sized cells in there with some room to spare. The cells would be arranged in a "flat" pack of 5 cells stacked 3 flats high. With the battery flats nested into each other, 15 AA cells just fit in the long hood with the front weight removed. The weight would be supplied by the batteries. 15 cells of NiCad or NiMH batteries gives a no load voltage of 18.75 volts.

The receiver would fit right behind the batteries, about 1/3 in the hood and 2/3 in the cab. The engineer had to go and part of the interior cab walls had to be hacked away, but it would just fit. The short hood was left alone with the weight left in place. A sound system might eventually fit in there.

The 2060 was stripped down to the frame to make room for the new stuff. Since the motor and power contacts had already been wired up separately, the original harness was left in place. The lead weights are held in by one screw each accessible after the motor block is removed. Vertical clearance for the RX was going to be tight, so the engineer was removed and his mounting pad was ground off. The engineer is held in with a medium strength adhesive. I found that by bending him to one side, one corner of his mounting "puddle" lifted and I could pry it up to break the glue joint.

In most battery installations some sort of power switch is desirable to prevent the battery from being flattened while the loco is not in service. This can be done with disconnecting type power jack, but a switch is even handier. I used a sub-miniature toggle switch with a 3/16" mount. This switch is tucked into the very front corner of the long hood going down through the floor. The switch handle is accessible through a hole behind the front step. Since the lower and upper tier of batteries is offset half a cell width to the rear, there is room for the switch body and also the headlight wire coming down from the front of the hood.

A disconnecting type power jack is installed in the front step well on the other side from the power switch. The jack mounting tabs were too large to fit so they were trimmed and the jack was epoxied into place. I prefer to mount switches and jacks in the floor of the locomotive so that there is minimal wiring between the frame and the removable parts of the loco.

Since the batteries will be bottled up inside the loco, external access is desirable for connecting to the battery charger. The jack as mounted can be accessed through the round hole behind the other front step.

I choose to use a high capacity NiMH battery consisting of 15 of these tabbed cells. The tabs make construction of a battery pack much easier. I have had poor luck with the reliability of NiCad batteries in the past in many other applications. The manufacturers claim that NiMH technology is more reliable and more tolerant to partial charges and discharges than NiCad technology. The energy density is higher, but so is the cost. This cell has a rated capacity of 1650 mAh. By my tests, this kind of cell is good for at least 1500 mAh, at least when new.

After the cells were individually charged in a regular AA NiMH charger they were soldered together into a pack. The rubber band was just to hold them together while they were being handled and before they were made into a complete pack. The pack was tested to make sure that it had the right open circuit voltage to make sure that no cells had been wired backwards. For freshly charged NiMH cells, you will expect about 1.4 volts per cell. The pack made 21.1 volts. Under load, it will drop quickly to 1.25 volts per cell and stay there for most of their discharge.

A piece of 3" shrink tubing was used to wrap the pack permanently. This stuff shrugged off the puny hair dryer that I use to shrink small tubing. I used a gas match on this stuff to get it to shrink around the cells. After shrinking, this pack is solid.

The pack was test fit into the hood. Initially I attached it in the hood, but later I determined that it was just as effective to attach it to the frame so that the wiring would not be stressed and the engine could be more easily tested without the hood.

This is the diagram of the wiring. Its about as simple as it gets. I initially choose to run the lights from the motor output but later I modified the circuit to allow the lights to be directional. If they were wired to the battery, they would burn absolutely constantly and be non directional They would also drain the battery while the engine was on but standing. With the addition of the diodes and capacitors, they are directional, nearly constant intensity (as long as the engine is moving) and only one is on at a time so that the battery drain is minimized. In the photo below, the capacitors can be seen at the right of the photo before they were insulated with shrink tubing.

The charge jack that I used is a disconnecting type. When the charger is plugged in, the loco is disconnected automatically. This is not really necessary because the power switch can be used to disconnect the loco during charging. I have also provided current limiting for battery charging from a 24 VDC source.

The 5490 is attached to the floor with a little Lexel adhesive. The code set switch is attached across the heat sink of the 5490 with hot glue, but if it breaks off, I'll reattach it with Lexel. It is just accessible through the center rear cab window which the engineer has conveniently left open.

The antenna wire was pinched as a result of some past problem and it broke just about in half. The half that is left is not quite adequate as the control range is only about 10'. Some more work was clearly required on the antenna.

When the wire was repaired, and extended to it's original length, no amount of fiddling with it would produce a control range exceeding 15'.

It finally stopped raining so I set up the loco for test. I loaded it up just below wheel slip (just two Aristo box cars with metal wheels) and let it run at full speed. It ran at nearly constant speed for 3-1/2 hours and then abruptly slowed to a crawl. The receiver was not perceptibly warm at the end of the test and the motor was just warm. While this was a little less than I expected, it is good enough.

As an experiment, I cut the antenna to about 6" and soldered the end to a now unused power pickup. The result was rather amazing. The range increased to at least 50' (the full width of my yard) and it appeared that it would work well beyond that. The antenna circuit is DC blocked so that if a track powered train is used at the same time, the DC voltage will not damage the 5490. Even with a DCC signal (23 volts p-p) on the track, the engine ran fine. It also ran fine with a regular loco running on PWC on the same track with all its motor noise injected back on the track. The 2060 also ran fine with another regular loco running as the analog loco with DCC on the track. I can only conclude that the experiment was a success.

[ Top ]

The LGB 2060 was the last loco at the GIRR that ran in stealth mode. I elected to finish the job with an installation of a Dallee Railbus sound system (#632) in the 2060. This sound system doesn't sound as good as other railbus sounds, it has an acceptable engine beat that increases tempo with motor speed but there are no gear shifts or diesel engine transitions. The horn is a loud single chime air horn and the bell is fair. However it only cost $94 shipping included, it makes good sound volume and the Dallee boards integrate well into battery powered environments.

This is the schematic of the installation. I did not disturb the previous wiring at all, this was a complete add on. This 5490 receiver has the pigtail connector for a 5495 accessory board and I had one so I elected to use it to allow the bell and horn to be triggered. I also used one output to act as a sound power switch.

I had originally wired the power to the Dallee board through the "E" reed relay on the ART-5495 board. This was done so that I could remotely turn the sound system off and remove it's idle power consumption as well. I could have elected to just mute the Dallee board, but then it would draw idle current.

Aristo doesn't rate the current capability of the contacts of the reed relays. I tried it and it seemed to work. However, after several power cycles, the Dallee board stuck on. Tapping on the "E" relay would cause it to release and the sound would go off, only to stick on again later. The reed relay on the 5495 clearly wasn't up to the task.

I checked the relay with an ohmmeter and with almost no load, it still seemed to work so I added another small 5 volt relay with better contacts to actually switch the Dallee board. This one has a 10 mA coil, so I added a 1K resistor in series with the coil to drop enough voltage so that the coil wasn't overloaded. Then it worked correctly.

The 5495 accessory board is just the same size as the side of the 5490 receiver. I attached it to the side with some foam mounting tape. I had to hack out some plastic from the cab liner on both sides to clear this board and the Dallee board on the other side.

The 5495 accessory board has five reed relays that are activated by the A through E buttons on the old 27 MHz TE transmitter. Buttons A and B are intermittent, the reed switches are closed for only as long as the button is held down. Buttons C, D and E latch but their state is not remembered after a power cycle.

On the other side, I mounted the Dallee #632 board. It is also about the same size as the side of the 5490 receiver. The wiring on both the 5495 and the Dallee board just fit inside the end of the front hood.

This 2" long throw speaker is one that I got with a bunch of others at a swap meet for about $1 each. It comes with a thick foam gasket.

One of the standard hacks for mounting a speaker in a loco that otherwise has no room is to mount it on the inside of the cab roof. With this speaker, it was easiest to mount the front of the speaker to the roof forming a small enclosure with the "front" being on the inside of the enclosure. The sound then radiates from the back of the speaker. This doesn't produce the optimal mount for a speaker, but it is easy. I just dabbed a little hot glue on two spots on the foam gasket and pressed it into the nearly flat cab roof. This compressed the foam at the other sides a little and made a seal all around. It actually sounds pretty good that way.

This is the installation when it was completed. With the exception of the added relay, I only soldered two wires, the ones at the speaker tabs. Everything else was a plug in to the Dallee board or connected to a screw terminal on the 5490 or 5495.

Samples of Dallee sound can be heard at Dallee Tips.

[ Top ]

After the sound was in and working on the bench, I put it on the track for a last test and it didn't work. No sound. Also the loco was running slowly so I assumed that the battery was near flat and there wasn't enough voltage to pull in the sound power control relay. A quick charge proved that point, everything worked again.

I charged it overnight and ran it the next day to see what the battery life was since I had charged it fully less than a week before and not run it a lot since then. It ran pushing one heavy car for 90 minutes, half it's runtime when the batteries were first installed eight long years ago. It is suffering from a common problem with NiMH batteries. After they age, they don't hold a charge very well and their overall capacity also decreases. I'm going to need new batteries sometime soon.

The rear hood of the 2060 is nearly empty and it happens to be big enough to accept a Digitrax DG583S DCC decoder. Further, I just happened to have one in my DCC box. Since I've done two Battery/RC with DCC included conversions recently that worked out quite well, a third one seemed like one path I could take to get out from underneath dead batteries. The newer DCC decoders, like the DG583S, work much better with flakey power pickup than the old one that I put in there 10 years ago and if it is a real problem, I can always either clean the track or switch it back to battery operation. This hasn't been necessary with the other dual mode locos that I have done.

There was one complication with this conversion. The relay that I used to control the power to the Dallee board to reduce the idle current without sound was problematic. The 5490 doesn't have a blue and ground wire brought out and I didn't feel like digging into it to bring them out. Without them, I could not figure out a way to get the relay to be controlled properly from both the DCC decoder and 5490/5495. Instead, I elected to eat the idle current of the Dallee board when muted, which is pretty low anyway, and just use the "E" contact on the 5495 in parallel with F6 on the DCC decoder to mute the sound. As it turns out, the Dallee board only draws 10 mA when muted and only 33 mA at with the idle sound running. With the relay, I was trading lower current with the sound off (0 mA) vs an extra 10 mA to run the relay when the sound is on. The trade is obvious, jettison the relay.

This is the schematic of the conversion. It is pretty straightforward. One DPDT switch determines where the 5490 gets it's power, either the battery or the track. The existing SPST power switch becomes unnecessary, but I'll just turn it on and leave it in place. There is room for these two new switches behind the rear steps. The DCC decoder is permanently connected to the track. The DG583S analog converts gracefully so that the decoder can run from DCC, DC or PWC on the track. The 5490 will run from DCC on the track as well. It won't work well with low track voltage but if I should want to use radio control when on DCC powered track, it will work that way.

The second DPDT switch determines what drives the motor, either the 5490 or the DCC decoder. The Dallee board always runs from the power source selected for the 5490 and it senses the motor directly no matter which motor controller is used. The bell, horn and mute functions are controlled jointly by the 5490/5495 and the DCC decoder. When running on battery power with no voltage on the track, the DCC decoder does nothing. If there is DCC on the track, the 5490/5495 can control the sound too but if the motor switch is set to the DCC side, it won't control the motor.

The loco will run as a full up battery/RC powered loco with sound, a full up DCC powered loco with sound, a DC track powered loco but with flakey sound (the sound won't come on until the track voltage is high enough) or a radio controlled but track powered loco with either full voltage DC or DCC on the track. The response to track power is good enough. The sound system is off when the loco is standing and comes on, with a little buzz or crackle sound, at just about the same track voltage as it takes to get the DCC decoder to wake up and analog convert so the sound comes on just as the loco starts to move. For DC track power, a 9 volt battery could be added in parallel with the 680µF capacitor to allow the sound system to run at idle. However, it would not be possible to turn the sound system off so yet another SPST switch would be needed in series with the battery to be able to disconnect it when desired. The Dallee draws a low enough idle current such that a 9V battery will last quite a while when all it has to do is run the engine idle sounds.

The DCC decoder response is better with PWC from a 10 amp trackside TE receiver than with pure DC or a regular power pack but the difference is not huge. The DG583S has never liked the flavor of PWC made by a 5490 used as a trackside receiver, it'll run in one direction, but not the other.

Since the 5490 has no directional headlight control capability, I elected to leave the headlights like they were. They work pretty well this way as they come on full brightness as soon as the loco starts to move. They will behave the same when driven from the 5490 or the DCC decoder.

There is little room to add two more DPDT switches unless they are the "sub-miniature" variety. The Radio Shack 275-626 is a little expensive, but it fits. However, there were two bosses on the rear deck that had to be removed. I have no clue what they were for so off they came.

Once I had a clear flat spot to mount the DPDT switches, they went in at the rear corners. I notched the lead weight to clear them and then added a couple of ounces of lead fishing weights to compensate for the missing corners. The loco is still a little front heavy due to the batteries in the front hood.

The DG583S was mounted above the rear weight. It actually floats sort of free in the rear hood, jammed up to the top at an angle. It's not going anywhere and there are enough wires spanning the rear weight so that it cannot actually electrically touch the weight anyway.

I elected to reinstall the sliders in the brick. I managed to find a pair sitting on my bench and they fit. The springs were there too. These MIGHT have been the ones that I took out 8 years ago.

Overall, the installation was a little time consuming, but there were no major difficulties and it all worked. The radio range doesn't seem to be as good as it was when I initially did the installation, but it wasn't great before I started the DCC installation either. The DCC part works perfectly.

I put the engine on a loop of track that had not been cleaned at all in several months. The track was heavily oxidized. As expected, the loco ran poorly. It would go several feet and stall. However, running on the main which had been lightly cleaned in the last week, it ran without a hitch.

I did notice a pronounced loss of radio control range after this conversion. The antenna no longer tolerates being connected to the track when there is DCC signal on the track. Since it tolerated this before the decoder went it, I assume that the decoder itself has conducted emission of RF noise that is jamming the radio link. When the antenna is disconnected from the track, the range doesn't change if the track is active with DCC or not, but it is not very good, about 10'. This indicates that the decoder isn't radiating much noise but there is noise voltage on the track wiring. This was the condition when I first installed the RX and had the antenna disconnected from the track. Either I have to find a better antenna configuration or I will have to put up with reduced range.

With DCC, the loco now runs better than it ever did in any mode. What has changed that caused me to remove DCC and install battery/RC many years ago? I believe that there are two factors. The track cleaning method that I use now is much better and modern DCC decoders have improved performance (or perhaps that it was the less than stellar performance of the old DG580L decoder that I used 10 years ago).

During the first time period that I was experimenting with DCC in this loco, I wasn't using the drywall sander for track cleaning and I didn't have a track cleaning car adapted to dragging along a drywall sander pad. The use of the drywall sander has resulted in consistently cleaner track with far less time and effort expended compared to before.

Other 4 wheel locos like the Bachmann Davenport and the USAT Speeder have poorer pickup than the 2060 (with it's additional sliders) and they now run well so that the track condition is very likely much better than it was before. 10 years ago, I would have never been able to run the Speeder at all.

The other factor is that the DG583S DCC decoder is a FAR better decoder than the DG580L decoder that I was using in those days. The DG580L proved to be a dog in the long run. It behaved badly with intermittent track power and it tended to lose it's mind and just stop working after particular bouts with dirty track. I've blown up one of them and scrapped the other two simply because they gave me so much trouble every time that I tried to use them. The DG583S is vastly more graceful in it's response to dirty track. It has both + and - terminals brought out so that adding the storage capacitor would be easy, but that doesn't seem to be necessary. Higher end Lenz decoders with the external supercapacitor module are even more graceful when dealing with dirty track but the DG583S seems to be good enough and it is far less expensive.

These two factors together make DCC operation quite reliable and negate the need for battery power altogether. The 2060 has the batteries in there and they'll stay in there but it is not likely that I'll actually use them on the GIRR. The batteries always seemed to be dead when I've tried to use them and I haven't learned to anticipate a day in advance of when I'd want to use the loco in a battery powered mode to charge them.