Nigelcliffe

Yes, good point. And the "PC" could be just a Rasberry-PI, rather than a big computer tower or an expensive laptop. A WiFi access point can be a tiny little box (smaller than the LocoBuffer), or with the right configuration, a PI can be the WiFi access point and run JMRI.

Uhlenbrock Daisy-2 Throttles work fine on Digitrax systems. Throttle costs about the same as a DT402 (give or take). Only significant limitation is that I don't think you can do any programming of a loco on a Digitrax system from a Daisy-2 handset, and that would include setting CV19 to form consists. If CV19 were set on another handset, then the Daisy can then control the consisted loco. I doubt Daisy can access the command station consists within the Digitrax system either. Daisy is nicer to use than a PowerCab, so the only risk is that you also sell your PowerCab and get the Daisy-2 system. The Fred mentioned above is OK, but I think now discontinued. Fred doesn't do accessory devices. The Fred is originally a DIY design, and there are kit forms of it available. - Nigel (Edited to mention programming limitation)

Either as Nick suggests, alter the pulse angle sent, or modify the sketch to drop the servo output signal after about 2 or 3 seconds. Dropping the pulse output is what a number of commercial designs do, as does the MERG kit servo controllers. - Nigel

There are no replacement wheels (unless you find someone with a very old set of "Neil's wheels" which are spare from 20+ years ago). Re-profiling is an option, but I don't think it will work well on this loco. There is a replacement chassis etch available from Shop3, which builds up to a complete chassis with new motor, new wheels, new gears, etc.. Part 3-640, but do note the list of parts needed to go with the etch. Or, there may still be parts available from someone's spares drawer for the chassis upgrade which wrapped new frames (from Printed Circuit Board) around the Farish chassis block - a process which does work, but can be very fraught in that the original Farish block needs filing down to very thin sections in some areas, so can easily get (disastrously!) bent in the process.

An example of a book not distributed through the usual wholesalers. Consequently Amazon, Wordery, etc.. can't easily add copies to their stock lists, so Amazon's stupidity algorithm (and chancer resellers working through Amazon) rates it as a very rare book. It is rare at Amazon warehouses ! - Nigel

All of them (0-29 , the manual talks of accessing thousands of functions, which is way more than current decoders support). Support for 0-24 is better than the others, as for 0-24 unique function icons and latching can be programmed for each loco in the roster. Being able to set the latching / not-latching for functions is very useful for sound decoders. - Nigel

Small branchline layout, Coldfair Green, uses an Uhlenbrock Daisy-II. Fits hand incredibly well. I've not met anyone yet who hasn't been impressed with the handset compared to what they used from other makers.

That would be typical of the failure of a small coreless motor when subjected to a crude feedback controller, such as the HH. Proving the failure was down to the HH is nearly impossible. - Nigel

KKPMO are the source of many of the gears in the 2mm Scale Association shop. - Nigel

Hi Norman, I think you are wrong. Both do the same thing in this area. If you can show an example where DecoderPro doesn't behave correctly, please send me the details and I'll get it fixed. This stuff is just CV reading/writing, but its an awfully large number of CV's involved. LokProgrammer/DecoderPro instructions: In an ESU decoder, functions (sound, lights, loco behaviours) are controlled by a large table with 40 mapping lines. Any line can control anything, and function keys, outputs, sounds and behaviours can appear on multiple lines. This is really important, it is very different to the way functions are setup in other maker's decoders. So, if you have output AUX3, then the first thing to check is whether any of the existing lines are controlling AUX3. Any existing mapping lines will remain active regardless of what you do; so if you have left an old mapping line where FnKey7 turns on AUX3, then that line has no direction limitations and will apply even if another line has direction limitations (and the light on AUX3 will appear to not have direction control). This is checked by reading in all the existing CV's for the 40 mapping lines. The first image is a cut-up screen shot from DecoderPro function mapping. In this, I've taken two rows and used them such that F7 in Forward operates AUX3, and F7 in Reverse operates AUX4 The second image is a crop from a screen shot of the LokProgrammer function mapping, set to the same thing. Note that the default display in the LokProgrammer software implies each Line is F1 to F20(something), but the user is free to change any Line to do anything they like. If you want things to only operate when stationary (eg. cab lights) then that is within the option marked "motion" (LokProgrammer) or "driving" (DecoderPro). It can be combined with direction of travel. Delay time in the light coming on (or going off) is set elsewhere, as a CV associated with the output (Aux3, etc..). As I said, if there is a function allocation issue where DecoderPro doesn't do something which the LokProgrammer software offers, then please provide details and I'll get it to the attention of the DecoderPro developer. This function stuff is only complicated CV changes. - Nigel

Current DecoderPro versions have the full function mapping which is available in a LokProgrammer. Laid out in near enough the same way. So, if you want to make settings, you can make them with DecoderPro. In a given row, you define which outputs are used, and which buttons (and direction / movement) are required. Yes, it can be done with individual CV settings. It is incredibly tedious and complicated, but possible. - Nigel

If using a car battery, please use some sort of fuse or circuit breaker to limit the current on the battery output ! Car batteries can deliver hundreds of amps, which is good for welding, cutting through rails and burning out all the layout wiring. - Nigel

Err, no. The above quoted material is wrong. CV29=14 means 28/128 step running, DC running enabled, RailCom enabled. A sensible arrangement for a default ESU decoder. CV29=16 would mean 14 speed step running and the 28 step speed curve, a setting unlikely to be sensible. Note that changes to the speed curve settings (CV2, CV5, CV6, or the 28 step curve) will alter how the sounds play out. This can mean changes might be to your advantage or disadvantage, so go carefully. If the 28 step speed curve is required, then bit 4 (value 16) needs to be added, so CV29=30. If you do decide to use the 28 step curve, then I think (from the JMRI forums) that there is an error in the ESU manual in that the end-values of the 28 point speed curve are still set by CV2 and CV5. - Nigel ( writer of the CV29 explanation and calculator at http://www.2mm.org.uk/articles/cv29%20calculator.htm )

My view, for fourteen turnouts, the MiniPanel is the better value option. The Mini Panel will just handle your requirements, it has 30 channels, and for 14 turnouts arranged in the simplest manner, you need 28 channels (28 push buttons). Regardless of decision between Alpha and MiniPanel, you will also need decoders for the Seep motors to operate. The panel is only one side of the requirements. So, this isn't a cheap arrangement no matter which approach is chosen. Basically yes, either will be another "cab" device in NCE terms. Not needed to run the locos. Whether you need more power for the turnout motors depends on the accessory decoders chosen. I'd be looking for accessory decoders which have an external power supply, so they place minimal load on the DCC output - if nothing else, power supplies for the decoders are far cheaper than DCC Boosters ! - Nigel

The circuit with "3 diodes and a schottky" is a bit more efficient in having a lower voltage drop than the 5 diodes above. The 1N4001S might be a under-powered for large scale locos pulling 2.5A. The "3 diodes" just need to be fast recovery rectifier diodes, 1N5001 or similar is fine. The Schottky just needs to have enough current capacity. The Schottky goes in the diode string directly between the track feeds, the 3 diodes form the alternative longer route string. (Read the post carefully, it does make sense !) - Nigel

Or, switch the frogs with conventional switches attached to the routes through the turnout and the problem goes away. The extra bit of powered rail is then not needed. In the situation of an installed juicer, replacing one output with a conventional switch which changes with the route set through the turnout should solve the issue. - Nigel

Not sure how well the Z21 has implemented LocoNet, when I looked at it (some time ago) a lot of LocoNet devices didn't play well. But possibly some upgrades to the Z21 have improved things. I'd be fearful for your BDL168's as they can be sensitive to things not being "right" and can be damaged if things are wrong. It *is* possible to do RailCom with Digitrax equipment, but needs third party hardware. I have a DCC4PC RailCom Cutout device (around £60) which will add the RailCom cutout to the output from a Digitrax command station. This produces the "short" RailCom cutout because the pre-amble on Digitrax equipment isn't long enough for a "long" cutout. This is sufficient for single loco identification within a block, but not other RailCom data. You may want to check the current handling with DCC4PC, as my hardware (a Zephyr) is much lower current than you have with the DCS200. For RailCom communication with software, I have built two Hans De Loof designed RailCom reading devices which read the loco address from a block, and transmit the address over LocoNet as a (Digitrax) Transponding message. Thus, a loco with a RailCom decoder (eg. Zimo, ESU, Lenz, etc) appears as if it has Transponding, and software can subsequently deal with it. This works seamlessly into JMRI software, but I don't know if TrainController can handle Transponding messages or not as I don't use TrainController. - Nigel

Alan, I admit to being confused by the manual. However, Digikeijs say connect the LocoNet-T cable to a Z21, so assume it works if you follow the manual ! LocoNet connection gives you access to all the control features Digikeijs describe - the ability to configure the device through LocoNetCV values (LNCV), the ability to turn the booster on/off from commands issued centrally, act as a reverser based on commands on the layout (eg. reverse when a turnout is thrown, much better than a short-circuit detecting reversers), and so on. - Nigel

Alan, Chances are that the boosters will need "LocoNet-B", which is a LocoNet signal cable with one pair of wires carrying "RailSync", this being a copy of the DCC track signal. The alternative is "LocoNet-T" which is for throttles and has a DC power feed replacing RailSync. A booster would struggle to work correctly on LocoNet-T, whereas its simple on LocoNet-B. On a Digitrax system, the sockets on the rear of a command station are effectively LocoNet-B with a little bit of power overlaid allowing throttles to connect, and those from a powered facia such as a Digitrax UP5 are effectively LocoNet-T. On an Uhlenbrock system, the LocoNet-B and LocoNet-T are clearly labelled. My guess (I said guess!) is the Z21 is more likely to have a LocoNet-T presentation, and my guess is that won't be suitable. But, such questions are probably best answered by the makers of the hardware. - Nigel

Piano wire is a tough springy steel wire usually used in radio-controlled aircraft. Comes in different diameters, suggest 0.7mm as a starting place for turnout mechanisms. Probably related to the stuff used to string pianos at some time in the past. You could take a few from the home piano, depends if anyone would notice :-). I don't think the linear servos offers much advantage over the rotary type. The issue is still making a linkage which can allow full range of servo movement to prevent damage to the tiebar and turnout. Omega loops can help with over-driving the distance, as can just winding the wire around the servo axis creating a spring which can tighten or slacken. - Nigel

It is the evil empire Apple's doings. They put orientation tags into images, which is sort of fine on a device which rotates around. But those are not necessarily handled well by other software. It can be sorted by nerds who edit the EXiF data in the image file. - Nigel

A minor comment on that picture is the wire may be a little fat and thus stiff. There are two factors which can be used in such mechanisms, the position of the fulcrum to alter the throw, and the springiness of the wire. If the servo was mounted vertically, rather than horizontally, the rowing action at the end of the rod will be reduced, this might be advantageous. - Nigel

Its my view that directly pushing a tie bar in 2mm with a servo arm is a risky proposition. If the servo twitches, the tie bar is over-strained. Therefore I'd use some sort of mechanical drive from the servo which allowed fuller movement at the servo motor arm, and a smaller movement at the tiebar. Examples would include the servo mount kits from MERG and numerous other similar arrangements which use a combination of flexible wire and a shift in the pivot fulcrum to control the throw. Then, a microswitch can be mounted at the servo arm (like Ed's image), with the arm moving through a decent range, and the microswitch will be reliable in its operation. - Nigel (another 2mm modeller).

Next areas to look at: - is there an issue to the fly-cranks. When resetting the wheels outwards, the fly-cranks are not moved outwards. So something could be catching in that area, or on the wheel closest to the fly crank. On my 04 I've removed the plastic fly crank blocks, and just have the over long rods moving behind the steps. Even though its been the subject of close-up movies to illustrate my couplings, I've never had anyone comment on the absence of the crank parts. - check the pickups, are any catching on the backs of the wheels ? - Nigel

Probably the rods/crankpins catching on the steps. That's a common issue if setting the original wheels to a wider back-to-back after thinning the flanges. - Nigel