more DCC questions/ramblings....

Your problem isn't DCC, it's the mechanism of the locomotive. The clue is that it happens in the same place(s) in each wheel revolution. No decoder is going to fix this.

Check out your drive train... probably in the last gear or two, or the axle.

Bachmann WD A small decoder will fit very nicely on top of the motor, wired directly to the motor. TCS M1 Digitrax DZ 143 size decoder is no problem.

No - feedback did cure this in DC. The bachmann J39 chassis is hardly precise. Even in OO there is significant fore-aft wheel movement which slight affects the quatering effect. This is down to the axle slots which are slightly over-wide. Of course in EM, this effect increases slightly. However, as suggested in my first query, the Pentroller does seem to cure this in DC for slower running. Thanks for the reply though

Feedback may have _treated_ the problem, but it didn't _cure_ it. It's mechanical, not electronic. A decoder with BEMF might _help_, and treat the problem just as your analog system did, but to _cure_ it you're going to have to do some mechanical work.

Joe Ellis wrote:-

As another DCC beginner I couldn't help noticing that John (Clements) is using an N-gauge chip with a maximum rating of 0.5 amp. Is it not possible that when maximum back-EMF is applied the current load could exceed the maximum rating for the chip?

(kim)

I doubt it. Most _good_ N scale locos draw no more than .25 amp. (I wouldn't necessarily put Bachmann in that category without actually measuring the draw on the specific loco - Kato and Atlas, yes...)(Yes, we Yanks are spoiled when it come to running quality.)

I've never seen an N scale loco _stall_... wheel slip happens well before then, even with traction tires.

If that concerns you, both Digitrax and Lenz make N scale decoders with a

Joe Ellis wrote:-

John Clements is using the N-gauge decoder with an EM-gauge loco, presumably to save space. EN is a finer scale version of 00-gauge.

(kim)

Back EMF is measured, not applied. The measurement of back EMF is used to adjust the drive to the motor.

The fact remains that there are numerous N scale decoders out there rated higher than .5 amp. It's quite common to put them in HO scale locomotives, so if his loco has a good motor it should be fine.

However, as I stated before, NO decoder can compensate for a bad mechanism. THIS is where his binding (and possibly overloading the decoder) comes in. Fix that problem, and it'll run better on DCC OR analog. Throwing more power at it doesn't _fix_ it, it _compensates_ for it, and may cause problems down the road.

ICK! Poo! Not _I_! Got no use for that Horribly Oversized stuff - I'm a Normal modeler. ;)

What about shimming the slots to get a better fit?

The sad fact is, that locomotive just isn't going to run _well_ until its mechanical deficiencies are corrected.

Try playing with the VMin, VMid, and VMax CVs, and the "Kick Start" CV. (Check your decoder documentation - I don't recall exactly what the are at the moment...)

"John Clements"

We are indeed. [Gloat]

-- Cheers Roger T.

Home of the Great Eastern Railway

Joe is correct. Not having seen your particular locomotive the advice I give may or may not be of any help. Sometimes getting rid of excess sideplay in the locomotive driving axles, can cure a poor running side rod driven locomotive as described. An easy way to do this is to use KD fibre washers, cut a quadrant out of the washer and slop over the driving axles. If this easy solution does not work driving axle bearings can be shimmed, always place the shim on the same side of each bearing. If this fails, re quartering of your driving wheels may be necessary, but check the problem in not the gear train or pick up before attempting this last resort before replacing the mech.

Thanks for replying Terry. The real crux of the matter that I was addressing was really the comparison between DC using Stewart Hine's Pentroller (as used by the Pendon museum and up until recently by me) and using DCC with various decoders. The J39 is one of two J39s I have converted to EM gauge and they both look the better for it. With hindsight, you are perfectly correct in suggesting that the problem with the chassis needs looking into. I think that I padded out the chassis slightly too much and I have dis-assembled the chassis several times to make sure there is no binding on the inside of the wheels. Some of the problem has been subsequently sorted out by doing this. But the other J39 was not DCC converted and did run beautifully so, in contradiction to what I said earlier, it can't be inherent in the propietary production of the chassis. (although their technique for this loco does leave little room for error when converting - particularly in quartering). After considerable testing with Lenz 1000, Lenz 0511 and Lenz 0521 chips, I have come to the conclusion that for perfect slow running, and the emulation of a steam engine slowly accelerating, you can't beat the Pentroller - and that is after testing several locomotives both in OO and EM. I found that the best decoder was the LE0511 which did produce great running at speeds over a scale 15mph for the J39. The Bachmann WD which ran superbly when controlled by the Pentroller in DC, was noisy and kicked somewhat when accelerating after it had been converted to DCC using a LE0521 decoder. So my locos are now all disconnected from DCC although I'll be keeping the equipment for the time being. (I still really like the idea of DCC and if slow running can be improved with new decoders then I may still look at converting in the future) I'll continue to mess with the J39 that sticks a little at very slow speed until I've sorted it out, although it won't be a massively high priority as I'm planning a new railway at the moment and want to begin building soon. But to repeat myself thanks again for going to the trouble of trying to help. Everything both Joe and yourself said was entirely correct, but DCC would be a retrograde step and is not for me, for the moment anyway.

Best Regards

John C. "Terry Flynn" wrote in message news:crcu4i$rnv$ snipped-for-privacy@tomahawk.unsw.edu.au...

Hi Andy,

Thanks for your interest. I wouldn't quite go as far as saying I'm throwing in the towel. I have conducted many experiments over the last month and in particular over the Christmas period. I have tried re-programming the decoders in a controlled way (yes I've tried 128 steps as well as 14 and

I must emphasise that I still have an open mind on the subject but I can't deny the overall results of my 'trials', even though from the start I wanted DCC to be better.

In spite of the extra cost (>£20 per decoder for the better ones) I really warm to the idea of DCC and set out with the idea that I would be converting to DCC. But the most important aspect of running is how closely you can emulate the distinctive movement of steam power. As you know steam engines do not accelerate 0-30 in 10 seconds and in spite of trying different combinations of settings of decoders, the same degree of very slow control attained by the Pentroller could not be matched. The one area where the decoders (particularly the LE0511 - the N-gauge silent decoder) scored slightly above the Pentroller was when emulating a loco running at a steady, say 25-30mph. Whilst I have designed my layout to be double track ex-GN in Notts and will have through freights, it will include a station and so realistic stops and starts and shunting in the yard will take place.

However I would be interested if you find a good decoder which responds to give very slow speed control and convincing very gradual acceleration. Particularly in situations such as where a locomotive is very slowly backing a freight into sidings or a tank engine is shunting. My next experiment is to try a non feedback controller (I've never tried Gaugemaster) to see if it's possible to equal the LE0511 performance with DC when running at a higher speed.

I suppose it's all good fun - wish it wasn't quite so expensive!

By the way I must visit the CVR again when it gets a bit warmer.

Best Regards and Happy New Year!

John C.

You have convinced me to find out more about the Pencontroller. Does any one have the circuit diagram of this controller on the web? One reason why I continue to use DC is because I did not like the control of the earlier decoders without back EMF.

"John Clements" wrote

Oh, Where are you modelling? This is my stomping ground (Ashfield area) and where some of the family worked (Nottingham Low level-London road) I would be interested to hear more! :-)

I think you may need to invest just a tad more... Have a look at Decoder Pro - its a free download software that when connected to either a lenz unit approx £50 or a Sprog II (£40 ish) you can set up the speed curves on your computer a lot easier than you can on a hand set programmer CV by CV ! This will improve your starts and yo can adjust the stopping distances to (Which you can do anyway) but just a suggestion...

E-mail me of list I yo want to talk more!

Its "Pentroller" a short form of "Pendon Controller". No, you won't find a circuit diagram on the web. They are made in a very small volumes by their designer, Stewart Hine (who isn't an internet-connected person to the best of my knowledge);

Some stuff here, (including links to Stewart's address)

- Nigel (disclaimer; Stewart Hine is a friend of mine)

Hi Andy, I could write reams about this but I'll try to be reasonably brief. I was brought up in the 1950s on the West side of Nottingham and my grand parents lived in Kimberley on the Nottm-Derby/Pinxton line. Trying to model an actual prototype has proved very difficult (space-wise) although I did go someway to producing a layout of Awsworth Jct in P4. On moving house and re-evaluating I decided to go EM (to speed things up) and have gone for a fictious station based on Kimberley/West Hallam. Work has prevented me from making huge strides but I do hope to make serious progress this year. The layout templates are all printed out in 'Templot' (assuming I don't change my mind again!!) and I have two-thirds of the visible baseboards built. I would have loved to have known the Leen Valley better but you were blessed with the same Colwick locomotives as I was. I still find the prospect of running J6's, K3's, K2's, O4's, WD's, A5's, L1's, J39's, B1's etc. very motivating. I'm interested in most British Railways and have collected one or two of the recent excellent RTR models from Bachmann and Hornby that have nothing to do with my main project. What do you run on the CV Model Railway??

Best Regards