DC fast, DCC slow

[Torper]

I have a kit built E4 2-4-0 fitted with a Zimo decoder that has always seemed very sluggish with poor haulage capacity.  Today I decided I could tolerate this no longer and, as a first step, removed the decoder and tried the loco on DC.  It was something of a revelation - the loco ran well and pulled a rake of coaches that it would have had difficulty moving previously.  That's easy, I thought, faulty chip, so I fitted a different Zimo (which i knew worked), put the loco on the track, and it was back to its old sluggish self which came as a considerable disappointment.  Chip out, back to DC, loco fine again.

 

My layout is DCC so I want this loco to run properly under DCC control!  It's the only one of my locos that behaves like this. Has anyone any ideas why it runs fine under DC, but is slow and weak when tried with two different decoders?  As it's kit built, there are no extra electronic things - connection is straight from the decoder to the Mashima motor.  I'd much appreciate any assistance.

 

DT

Edited February 28, 2018 by Torper

[Phil Bullock]

Hi there

 

It may well be that the default decoder settings are not right for your needs

 

CVs are listed here http://zimo.at/web2010/documents/MX620MX63MX64E.pdf dependant on the decoder you have fitted

 

Hope that helps

 

Phil

[Torper]

I've kept the decoder setting simple.  CV 29 is set to 34 at the moment, ie 128 speed steps and long address.  That's my basic setting for my kit built locos (almost all with Mashima motors) and it normally works fine.  CV 3 is 2, CV 4 is 1.  Anything else is the Zimo default.  The decoders were different ones, one the latest Zimo "cheapie", the other a rather more expensive one.

 

DT

Edited February 28, 2018 by Torper

[Phil Bullock]

Have you tried cranking up cv5 for top speed? Then CV3 and CV4 to control acceleration and deceleration once you are happy with top speed?

 

Phil
 

[smokebox]

What controller are you using? Quite a few reports of sluggishness when using the NCE Powercab, usually due to a lower voltage power supply than ideal.

[Torper]

Phil, CV5 is at 1, which is the default and is apparently the same as 252.

 

Smokebox, as it so happens I am using an NCE PowerCab.  The supplied power supply connected to my layout has an output of just over 13 volts which I've always found sufficient.  I also have a 15 volt 2 amp PowerPax power supply I got from Coastal DCC which I generally use with my rolling road.  Somewhat to my surprise I did find that changing the layout power supply to the 15 volt one did indeed make a difference to the E4, but still it wasn't nearly as good as DC running.  I did some timing with the E4 over a set distance and the results were:

DC:  11.5 seconds

DCC 15 volts:  20 seconds

DCC 13 volts:  26 seconds.

The DC running was not only faster, but was also of considerably better quality than the others.

 

In view of the above results I also tried another of my locos on the 15 volt supply and it ran slightly faster although not markedly so.  Speed, I shoiuld add, is not an imperative on my short end to end layout.  I have always worried a bit about using the PowerPax 15v supply in case in doesn't have same inbuilt protection as the official NCE one does although I'm sure that CoastalDCC wouldn't have sold it to me if it didn't.

 

So I'm still baffled.  I think the next step is to undo and clean up all the loco and tender wiring in case there is some minor short and see if that makes any difference.

 

DT

Edited February 28, 2018 by Torper

[GoingUnderground]

What DC controller are you using? If it's an older one then it may be supplying over the nominal 12 volts.

 

Also try turning off the Back EMF on the decoder as that sometimes causes problems.

[Nigelcliffe]

I'd suggest the following test first:

 

Take a loco which performs properly on DCC, with similar drive arrangements, and remove decoder.   Run that on DC, noting the comparable timings as done for the E4.    That should show whether the problem is actually in the decoder settings, or whether the E4 is just a bit slow in general, which could be the motor, or some other factors.   

 

Next, decoder settings.  CV2,5,6 isn't going to increase the speed *if* its a decoder issue. What needs changing are the motor control parameters.  I'd use the current Zimo manual as a guide:

http://www.zimo.at/web2010/documents/MX-KleineDecoder_E.pdf

Page 16,  CV57 for motor voltage regulation, may help.  Set it towards the lower end of things. 
Page 16, CV9, motor sampling rate.  Try a value of 55, and then try reducing the digits figure and see what happens. 

Page 16/17, CV56, the P and I values for the motor regulation.  Again, starts at 55, but may need different values, which works (between 01 and 99) takes time to find, though table at bottom of page 18 may give some clues. 

Page 17, CV58, Back EMF intensity.  Possibly a lower value (say 170-230) may help.  But turning it off (values near zero) seems to defeat the point of having a decoder which does a good job of BEMF regulation.   Alternative (or additional) might be to use page 18, and CV10+CV113 to define a speed (in CV10) where the BEMF is reduced to the value defined in CV113. 

[RAF96]

Running a loco on DC with a decoder fitted is usually the slow performer in that it takes a lot of throttle to get it going. This is due to the decoder needing a specific voltage to get it to crank up before it recognises it is seeing DC and switching control logic.

Some decoders are worse than other at this.

Rob

[MiDAS]

What DC controller are you using? If it's an older one then it may be supplying over the nominal 12 volts.

 

Also try turning off the Back EMF on the decoder as that sometimes causes problems.

Agree with above with regard to DC controllers some nominal 12v supplies gave out in the region of 15 volts at 1.5 to 2 amps

[dhjgreen]

I was experiencing deja vu, and then found this phew!

http://www.rmweb.co.uk/community/index.php?/topic/92841-dcc-20-slower-than-dc/

[Torper]

Interesting.  However, in my case, as can I think be seen from the fiigures I mentioned, the loco is not only running slower, it is running abnormally slower and much slower than my other DCC locos, most of which have similar configuratiions - almost all my locos are kit built with High Level gearboxes and Mashima motors, and the decoders (all Zimo) are hard wired.  Its not therefore a simple case of unplugging the decoder to run in DC - I have to remove it completely so there is no conflict there. The E4 should in fact be slightly faster than most of my other locos as it has bigger driving wheels, but running in DCC it is not only much slower, but it is also weak with little pulling capacity and it is laboured in its running.  As mentioned earlier I have tried it with two decoders and it exhibits the same symptoms with both.

 

My DC controller is Bachmann one - it doesn't seem to have a model number but it is a small grey square shape with a big blue control knob.  Its stated output is 12 volts.

 

I'll go and refit the decoder and see if I can do anything with Nigel's suggestions.  My own gut feeling, however, is that the fault does not lie with the decoder or its settings, but we'll see.

 

DT

[Nigelcliffe]

Interesting.  However, in my case, as can I think be seen from the fiigures I mentioned, the loco is not only running slower, it is running abnormally slower and much slower than my other DCC locos, most of which have similar configuratiions - almost all my locos are kit built with High Level gearboxes and Mashima motors, and the decoders (all Zimo) are hard wired.  Its not therefore a simple case of unplugging the decoder to run in DC - I have to remove it completely so there is no conflict there. The E4 should in fact be slightly faster than most of my other locos as it has bigger driving wheels, but running in DCC it is not only much slower, but it is also weak with little pulling capacity and it is laboured in its running.  As mentioned earlier I have tried it with two decoders and it exhibits the same symptoms with both.

 

My DC controller is Bachmann one - it doesn't seem to have a model number but it is a small grey square shape with a big blue control knob.  Its stated output is 12 volts.

 

I'll go and refit the decoder and see if I can do anything with Nigel's suggestions.  My own gut feeling, however, is that the fault does not lie with the decoder or its settings, but we'll see.

 

DT

 

I also build kits up, and scratchbuild, and there's a Gibson E4 kit to P4 standards on my to-do pile.   I appreciate its a pain in the neck to remove decoders and refit when they have to be soldered in and out to carry out tests.

 

Other suggestions:

 

Measure the current consumed by the E4.  

 

Ideally I'd measure on DC,  but I think its possible to get a reading for DCC current consumed from the PowerCab display. 

 

If you can get it over to DC without a decoder present:   Put a multimeter in series with the track feed, set it to an appropriate DC current setting (start at the highest available current setting on your meter, and work down, if it offers "10A" through a different lead connection use that !).   I'd expect to see below 250mA, but a bad motor, or mechanical issues, could send it to a much higher reading.   If you have an analogue meter, the needle can flick up/down indicating tight spots in the mechanism, or problems with the motor.    **** Note that used to measure current can blow an internal (often glass) fuse if the current goes off the scale in quite a lot of multi-meter designs, so know how to change the fuse !  ****

I always test the locos I build on DC with an ammeter to watch the current, I use the flicking needle as part of diagnosing mechanical issues.  

 

 

If you have a spare motor, try a motor swap.  But, if the worm is glued on, that can be another pain in neck job !

[Torper]

I have two multimeters.  One is an ancient analogue one which I tend to use for ensuring there are no shorts, and the other is a digital one that I have never got to grips with.  Half an nour ago, however, I tried the latter with the Bachmann DC controller and it indicated a full speed output of 23.6 volts!  My first thought was that I must have connected something wrongly, so I got out an extremely old Hammant and Morgan that I'm actually quite frightened to connect to the mains, and turned it up to full speed. 13.4 volts, said the multimeter, which would seem OK.  Just the same connections as with the Bachmann one.

 

I still haven't reinstalled the decoder in the E4, so the next step was obviously to try it on DC with the Hammant and Morgan and time it down the relevant section of track. 24 seconds its laboured run took, ie virtually the same as with the 13 volt DCC NCE PowerCab.

 

So 23.6 volts from a Bachmann controller that states it has an output of 12 volts?  No wonder my poor little E4 shot off in the way it did.  But what it does indicate is that the fault is in the loco, and not the decoder or the wiring.  I think an examination of the motor, followed probably by a dissemblement of the chassis, are likely to be the next steps.  Sorry if I've wasted everyone's time with what seems to have been a wild goose chase, at least as far as DCC is concerned.

 

DT

Edited March 1, 2018 by Torper

[smokebox]

If you were measuring the voltage off load then the high readings aren't that unusual. If you were getting the readings while 1 or more locos were on the track and running, meaning that the controller had a load on it, then that would be unusual.

[Nigelcliffe]

OK, good to have some clear numbers from the controllers.  As smokebox has said, an off-load reading on a basic DC controller isn't quite the same as one with the load from a loco.   But, stupidly high DC outputs from basic transformer based DC controllers is "normal".    

 

Current consumption of the loco might also still be useful, because if its way out, it indicates a problem.

 

What's the gear ratio used in the E4 ?   Is it similar to those used in other locos, or is it a lot higher ?    If the gearing is the same, then the top speed should be wheel-diameter related, and slightly higher on the E4 to, say, a J15.    If the gear ratio is a lot higher, then the E4 will run a lot slower. 

 

And, if the loco does pass all reasonable tests, then is the actual DCC top speed acceptable in use on your layout ?    If so, then the characteristics of the loco can be adjusted with speed curve changes, so that it gets towards its top speed at an earlier point on the DCC throttle. 

[Torper]

This is quite a learning curve!  Smokebox, you are of course quite correct.

 

Readings are (all at "full speed"):

 

Bachmann:  Mulitimeter connected directly to controller - 23.5v  (H&M 13.4)

                    Connected to layout, no load or loco:  19.5  (H&M 13.2)

                    Under load with loco running:  16.3  (H&M 12.3)

 

It's amazing the difference that extra voltage made - the loco ran nicely with the Bachmann, poorly with the H&M.

 

Nigel, I think the gear ratio is probably similar to the J15 (I've got one of them - Gibson - and it runs well!) but I'll have to check that.  And no, the top speed isn't acceptable, but even if it was the performance (except with the Bachmann controller) certainly isn't; the loco is laboured in its running and its haulage capacity is pathetic.  I'll be checking the free-running nature of the chassis itself though I don't anticipate that that's going to be problematical as it ran so nicely with the Bachmann.

 

DT (looking out at a blizzard)

[Nigelcliffe]

It sounds like a mechanical or motor issue, so, as you say, diagnosing that first is key to any progress.  
I'm happy to take a look at it - possible Perth in April at the 2mm Scale Association open day  (send a message nearer the date if planning to do this to check I'm definitely at Perth ).    http://www.2mm.org.uk/events.html

[10000]

I have a Powercab on 15 volts and had the same problem with a Zimo that was fitted in a secondhand Kato H0. Took the the Zimo out and fitted a Lenz (until sound was fitted) and it ran with no problems after that.

[Torper]

Problem resolved.  It was nothing to do with the motor, the decoder, or the chassis which was and is free running.  Instead, it seems that a miniscule foreign body had lodged itself in an insulating gap in a section of circuit board that I had installed to anchor wires running between the loco and the decoder in the tender.  That was enough to cause a sufficient short to adversely affect but not stop running.  The circuit board has been replaced (with larger gaps) and I took the opportunity to simplify the the wiring.  As a result, the loco now runs properly.

 

Many thanks to all who contributed.  Although I was initially barking up the wrong tree, I have nevertheless learned a lot from the contributions made.

 

DT