RFS

They do work with non-bogied stock provided that you're using the flexible NEM pockets with "butterfly" attachment. I have a number of close-coupled freight trains using this method.

I recently purchased E5022 from Rails. Everything seemed fine until I tried it on a substantial load when the loco encountered a lot of wheel slip. Close observation revealed that only the wheels of one bogie were slipping, and that the wheels of the other bogie could be spun with my little finger. As this model has both bogies powered there was clearly a problem. The model was replaced by Rails and the new one is fine. So if you have a Hornby 71 make sure you can't turn any of the 4 axles by hand, since if you can there's a problem with the gears and the loco is working off only one powered bogie instead of 2.

Both models have NEM coupling pockets and separately-supplied valances, if you choose to fit to one end in lieu of a coupling. They have different methods of fitting head codes but you can do so on the Hornby without taking it apart. Both are more difficult to fit head codes to than the Heljan 33 where the plastic front window simply unclips. All models these days have excellent slow-running characteristics (so long as you don't fit a cheap decoder!) and the DJ 71 is no different in this respect to the Hornby. My newly-acquired unrebuilt Clan Line, with flywheel-driven motor, also has superb slow-running characteristics. I needed my DJ 71 to take its turn on my Golden Arrow representation, but at 10 bogies (7 lighted Pullmans) this was way beyond its capabilities. Regretfully I had to make the decision to return it to Kernow and replace with the Hornby version E5022. This version has no problem with speed and haulage, and its motor is silent unlike the noisy gears of the DJ 71. DCC fitting, with a Lenz Silver Direct, was no different to fitting a 21-pin to the DJ 71. I appreciate there are pros and cons for both versions and everyone will have their personal choice. My need was for haulage and, although the DJ 71 is a nice model in many respects, it didn't deliver that. EDIT: remove reference to DJ 71 having a flywheel. Only the Hornby one does.

Bit 6 is the EMF control for high-efficiency motors, whereas bit 7 turns EMF on or off. This is the description of bit 6: "The decoder also has what we refer to as a EMF switch which makes it possible to further adjust the decoder to different motor types. Depending on the motor type used, it is possible that a digitally controlled locomotive cannot reach an adequate maximum speed compared to a locomotive in conventional operation. If this is the case, activate your EMF switch by setting Bit 6 in CV 50. The locomotive will then reach a higher maximum speed while the minimum speed is also slightly increased." ​So you can try turning it on, possibly with motor type 1 - so set CV50=33. I found that the loco started at walking pace on speed step 1 of 128, but this improved a little with motor type 1.

Couplings can be rigid​ (Roco, Bachmann "twin-pipe") or flexible (tension locks, Kadees). NEM pockets can be rigid​ (eg moulded as part of bogie) or flexible​ (eg CCM that is completely separate from the bogie). To couple effectively, at least one component must be flexible - eg Roco coupling + CCM, or tension lock/Kadee + fixed NEM pocket. Both components can be flexible too, but both must not be rigid. Modern coaches with flexible CCMs allow any coupling to be used - rigid or flexible - but work best with a rigid coupling. If your coaches do not have this independent CCM then you can either use the Keen or Symoba conversions, or else use a flexible coupling such as Kadee or tension lock.

There are also the close-couplings provided by Symoba which I have successfully used on my Bachmann Bulleid coaches. http://www.dccsupplies.com/search/results/?search=symoba

With a Lenz Silver you can try setting bit 6 in CV50 to increase top speed (CV50=32), described as "back EMF control for high efficiency motors". Also top speed is reached at about speed step 100, and you can actually reduce CV5 to around 190 and it will still do around 60mph. As a result you get a better speed curve.

The couplings are different: look at the picture at the top of the post and there's a solid black square of plastic in the middle on the Hornby version that is absent from the shorter Roco one. 40270 simply refers to a pack of 4 couplings, and 40271 to a pack of 50. Both are the shorter Roco-version. Hornby's R8220 is a pack of their longer version. These couplings are designed to be fitted to an NEM pocket that is part of a close-coupling mechanism, or to one that swivels. You cannot glue them rigidly in position or you will get the problems you are experiencing. I use the true Roco ones in Hornby Maunsell and Pullman coaches, and the supplied Hornby-Roco ones in Bachmann MK1 coaches where the NEM pocket is further back. The coaches are close-coupled in sets with a Kadee on the outer ends to couple to locos. I also use both types on wagons which have the "butterfly" type of swivelling NEM pocket. Which I use depends on how far back the NEM pocket is, and hence allows vehicles to traverse the curves on my layout.

It doesn't do this when it updates my PCs.

There's no problem fitting a decoder to the Hornby 71, so long as you use a direct-fit decoder such as the Lenz Silver Direct which is what I have done. You will have problems though if you try to fit a decoder with a bulky wiring harness.

Unfortunately it doesn't seem to make the loco go any faster. In fact top speed was reached on mine on speed step 100, and increasing to speed step 126 didn't cause the loco to go any faster.

If you're using a polarity switch for the frog power then the frog must be fully isolated - ie IRJs on the vee rails, and the links back to the blade rail cut underneath. So long as this is done there's no risk of momentary shorts occurring when you switch the points. You can use one GM500 to power both frogs on a crossover, something you might find more reliable than the Seep's own built-in switches. Certainly better than a pair of PM13s and no more expensive. With a diamond crossing you can connect the GM500 to the motor for the turnout that sets the route over it. The frog furthest away from the turnout needs the same polarity as the turnout's frog: easy on DCC (they can be wired together) but on DC you may need an additional switch depending on how your sections are wired. So the GM500 could control both frogs of the diamond crossing with the Seep's own switch controlling the turnout's frog. You'll just need to ensure that when a train takes the other route over the diamond (ie not from the turnout) that the turnout is not switched to the diamond route. Otherwise trains will short on the diamond.

If you only intend to run the loco in one direction you can insert a white panel in the rear slot that has a bit of black masking tape over it. There is very little light bleed, epecially if you use the decoder function to turn down the brightness.

What you need is a Gaugemaster GM500 - http://www.gaugemaster.com/item_details.asp?code=GM500&style=main&strType=&Mcode=Gaugemaster+GM500

To AUX3. See the DJModels page here - http://djmodels.co.uk/?page_id=253 - scroll down to the list of decoder functions.

The head code lights are indeed wired together.

Kernow's original listing of E5004 was one of it drawing out of Victoria with the Golden Arrow. Now replaced by the actual model of course. Here's a later picture of E5004 on the train, albeit now with small yellow panels. http://railphotoprints.uk/p287659169/h78a57f58#h78a57f58

The speed you are recording is in line with what others have seen. This is similar to what I recorded with a Zimo MX600R, although bear in mind that this decoder does not have the functions to operate the head code lights. A Lenz Silver 21 recorded only 63 mph for me even when optimized for coreless motors. My DCC system is Lenz with default track voltage, but the use of occupancy detectors costs 1.5v reduction. I did try increasing track voltage by 1.5v to compensate, but the top speed of the loco did not improve. The speeds I recorded were light engine. Once you start adding vehicles it starts to drop further. I have a representation of the Golden Arrow - Hornby bogie luggage van, 2 Bachmann second opens, and 7 Hornby lighted Pullmans. My 71 E5004 found this a real struggle and recorded only a maximum of 43mph on straight and level track. Plus there was a lot of wheel spin getting the train started. Clearly the this load was well beyond its capabilities. I would have expected Back EMF load compensation to maintain a constant speed but obviously the motor does not have the reserve of power to support this. This train is normally the province of my Hornby Britannia "William Shakespeare" which has no problem with this load. Nor does my new unrebuilt "Clan Line". I'm fortunate to have a layout whose scenic section is 32fft long, which allows me to realistically run scale-length trains. In my opinion the loco is seriously underpowered. Although 60mph would be acceptable for me, it must be able to do that under load. The prototype, at 2550hp, was designed for heavy haulage for the likes of the Golden Arrow and Night Ferry, but this model is unable to replicate these trains. I'm very disappointed now with this loco and am seriously considering returning it as unfit for purpose.

The only Peco Electrofrog points which work as-is out of the box are the plain LH and RH ones: polarity switches are mandatory for slips and 3-way points.

I think you mean Cobalt motors are 80% the size of Tortoises, not 80% smaller ... If you do not modify the points to use a polarity switch you are still reliant on good contact between blade and stock rail. But if you ballast and liberally spread diluted PVA then the glue can seep up into the blade pivot and prevent current passing through that to the frog. So the recommendation to use a polarity switch with Electrofrogs is especially important if you are ballasting and painting.

Some more pictures of 71s here - http://railphotoprints.uk/p287659169 Interesting ones of E5000 at Broadstairs on a passenger service in 1961, and E5017 at about the same time on the Newhaven boat train with Bulleid stock in tow. Increases the "authentic" operations for us modellers!

Just to update on this. The DJ Models Class 71 is configured such that it expects the AUX3 function on the decoder to be a logic output. The Lenz AUX3 is indeed a logic output and it works fine. The MX634D, however, has all 6 of its outputs as normal function outputs which is why it doesn't work as is. However a feature of this decoder is that you can switch its AUX3 and AUX4 outputs to logic (ie turning it into an MX634C) by issuing a CV8=3 change. Despite fully resetting the decoder, and then issuing CV8=3, the lights still do not work. I have been communicating with Zimo and they don't know why this is not working. I have 3 MX634Ds and I tried all of them with the same result. I can only assume the logic output is behaving differently and is not compatible with the DJ Class 71. So my 2-EPB will keep its MX634D and I'll run the DL 71 from the Lenz Silver decoder for now.

Do mean to reduce the brightness levels? In which case it's CVs 116 and 117.

Glad you got it sorted, even though you had to get out the soldering iron! I have been trying to sort this out with the Zimo MX634D decoder on the assumption that AUX3/FO3 required to be normal output, not logic output. The MX634D has this as a normal output so I didn't understand why the lights didn't work. But the loco's PCB expects a logic output and the Lenz Silver has AUX3 as logic level which is why the lights work. But the FO3/FO4 output of the MX634D are supposed to be switchable between the two states, with CV8=3 making them logic level. However I've tried that and the lights still do not work. I can only assume the logic outputs on the decoder are somehow incompatible with the loco's PCB. Zimo have confirmed that CV8=3 should convert the outputs to logic so I've asked them why it still does not work. So I've reverted to the Lenz Silver decoder. There are a couple of parameters with the Lenz for coreless motors but they're not well documented. Setting bit 6 in CV50 increases the maximum speed, and increasing the BEMF sampling in CV9 improves the BEMF accuracy too. CV9 has a range of 0-63 with a default of 15. So I've tried the loco with CV9=63 and CV50=32 and performance is much better. It reaches 62mph but on speed step 110 after which it will not go any faster. Here's the output from Traincontroller's profiling (vertical axis is speed, horizontal is speed step, left is backward, right is forward): With an 8-coach load it barely drops speed and I'm happy with that. As far as the head code lights are concerned they're far too bright. CV 118 on the Lenz reduces the brightness: its default is 255 but I've reduced to only 25 to get it to look realistic. As I intend to run the loco only in one direction (so that all the extras can be plumbed into the front) I've put an insert into the rear blind with a piece of black masking tape over it. There's now hardly any light showing. So now I'm at last happy with the loco!

On the MX634 you can choose whether AUX3 are 4 are logic output or normal function output. Setting CV8=3 makes them logic outputs (ie MX634C) and CV8=4 makes them function outputs (ie MX634D). I've tried both variants and they still don't work and have emailed Zimo for advice. With a Lenz Silver, which has 5 normal outputs (no mention of logic level) the lights work fine.