DCC loco on DC problems

Post all your DCC only problems, solutions and discoverys here.
User avatar
fourtytwo
Posts: 153
Joined: Mon Oct 31, 2016 7:41 pm
Location: Norfolk (Bodecia's country) UK

DCC loco on DC problems

Postby fourtytwo » Fri Oct 22, 2021 4:25 pm

I recently bought a s/h Bachmann Ivatt 2-6-2T fitted with DCC, I have a DC layout with 100hZ sine control.

I found the loco had a considerable windup delay, that is the controller had to be advanced quite high for a period (perhaps half a second) before the loco would start and then had to be immediately backed off. It also had an unpleasant inertia as if it had a huge flywheel, it also refused to crawl.

On taking it apart it had a Zen black? (colour of sleeve) decoder that I successfully removed, and then the loco behaved normally (on DC).

This is just to remind others that DCC fitted loco's may not behave very well on DC layouts and if as in my case it will only ever run on a DC layout it may be worth removing the decoder to improve running.
A fresh start in OO, DC Steam

User avatar
Mountain
Posts: 5388
Joined: Mon Oct 24, 2016 3:43 pm
Location: Somewhere in Wales, UK.

Re: DCC loco on DC problems

Postby Mountain » Fri Oct 22, 2021 5:06 pm

DCC decoders are said to work fine on DC as long as one does not have a feedback controller or use electronic track cleaners.
I personally use a simple old Hornby third amp trainset controller from the 1980's to test locos on and they behave fine. I set my DCC decoders so that they would work on DC or DCC.

Now what happens if a DC controller is not suitable is that the DCC decoder will think it is picking up a DCC signal and act accordingly, and some controllers which use a pulsed type of DC control will have the decoders assuming they are running on DCC. It is these sine waves that the decoder picks up which confuse it.
If one uses a more conventionan DC controller that works by slowly increasing the voltage itself to increase the speed, they will work fine with a DCC loco.
The feedback type DC controllers which do not work with DCC decoders work in a different way as they will send the same max voltage (Usually 12 volts) but send it in pulses, so less pulses gives the loco less speed etc. Very clever stuff but with DCC decoders they don't like it as they think they are working with DCC but they then can't find a suitable DCC command signal (Or they may take a pulse as being a DCC signal and act accordingly!)

I maybe wrong in what I write as what I write is my understanding of how things work, so those more technically minded in electronics can correct me or explain more.

Bigmet
Posts: 9119
Joined: Tue Aug 14, 2007 2:19 pm

Re: DCC loco on DC problems

Postby Bigmet » Fri Oct 22, 2021 6:50 pm

fourtytwo wrote:...I found the loco had a considerable windup delay, that is the controller had to be advanced quite high for a period (perhaps half a second) before the loco would start and then had to be immediately backed off. It also had an unpleasant inertia as if it had a huge flywheel, it also refused to crawl...

The problem you experienced largely relates to the set up on the decoder, which of course isn't adjustable unless you have a DCC system. The inability to crawl at dead slow is often observed: the decoder doesn't supply any motor current below a threshhold voltage value, and if that threshold is higher than the motor needs for dead slow, then dead slow isn't available.

It's a little surprising that DCC manufacturers haven't now moved to abandon DC compatibility as a poor compromise. It's won the fight to become globally established, and really doesn't need to offer this any longer.

User avatar
Bufferstop
Posts: 13489
Joined: Thu Mar 11, 2010 12:06 pm
Location: Bottom end of N. Warks line

Re: DCC loco on DC problems

Postby Bufferstop » Fri Oct 22, 2021 7:54 pm

Bigmet, your highlighting of the DCC on DC being down to the historic need to the fledgling DCC in the DC world goes hand in hand with the other big compromise DC on DCC. It works by shifting the baseline of the DCC waveform so that it has an average voltage of plus or minus a few volts, which makes DC motors turn, whilst doing nothing to protect them from the alternating peaks whilst they are standing still. Any one hoping to run mixed fleets of DCC and DC locos should really consider having separate isolated sections one for each system. It's not such a big problem, they are likely to have section switches on their layout. Grouping some to each system isn't going to be that difficult.
Growing old, can't avoid it. Growing up, forget it!
My Layout, My Workbench Blog and My Opinions

Bigmet
Posts: 9119
Joined: Tue Aug 14, 2007 2:19 pm

Re: DCC loco on DC problems

Postby Bigmet » Fri Oct 22, 2021 10:32 pm

I learned the problem with the way DCC provides a DC voltage on 'address 0' the hard way about two days after getting my DCC system. I had used all my initial purchases of decoders to fit selected locos, and tested them, all good and up and running. Right, let's try an unfitted loco on address 0. Used a large motored loco (Mashima 1830) with a very free running drive, and got an immediate cloud of smoke without it even moving. Superb steam loco effect, most of the smoke expelled from the 9F chimney, but rather a high price for the few seconds it lasted. All the other locos running on DCC didn't miss a beat, nor did the system trip. Never tried address 0 again.

User avatar
Mountain
Posts: 5388
Joined: Mon Oct 24, 2016 3:43 pm
Location: Somewhere in Wales, UK.

Re: DCC loco on DC problems

Postby Mountain » Fri Oct 22, 2021 10:45 pm

Strange. I have in the past tested quite a few locos on address 0 with no problem, but I do not to my knowledge have any Mashima motors in them (Unless a RTR manufacturer put them in).
Lima and older Hornby models have been tested for a thousand hours on address 0 with McKay models who used to be the importer of Lenz DCC, but they do have nice heavy windings compared to some more delicate motors. The main "Loco" I used to run on address 0 was the Bachmann Gandy Dancer. It worked fine. I could not think how to add a decoder to the thing without it being too visible!

As a general rule though, I would only really use address 0 on older RTR models with tougher motors. The Hornby Ringfields and Lima pancake motors can cope with a lot.

I do understand why it is better not to use this feature. My greatest concern was with using HST dummy powercars as their lights were being fed with more then 12 volts. Thwy seemed ok though. I was preparing myself for some blown bulbs!

User avatar
Bufferstop
Posts: 13489
Joined: Thu Mar 11, 2010 12:06 pm
Location: Bottom end of N. Warks line

Re: DCC loco on DC problems

Postby Bufferstop » Fri Oct 22, 2021 11:58 pm

The problem of DC on DCC is that when the baseline voltage drops back to zero and the loco stops there is still an 18V waveform on the tracks. A good old clunker like an X04, one of the old Romfords or an Original H-D is going to have a solid iron core in the motor and the AC will make it hum a little, fail to make it move and get it a little warm. Do the same with a more modern motor, it's armature will vibrate like crazy and it will get very warm, to the point of emitting smoke, do it with a coreless motor and the armature will rip itself to bits if you leave it standing in a powered siding.
The only time I've had DCC on DC was a couple of Bachmann Gregs. The were basically a faceless Percy fitted with a decoder and formed part of their digital starter set. Running on DC was "interesting" I think they would have been ok run off batteries, but the heavy100Hz ripple on a conventional controller got the decoder confused and they jumped and stuttered along the track, other people have commented on poor DC performance of various models. I think it depends on the controller you are using and the worst misbehaviour comes if you try to run them from a PWM controller which takes AC and chops it up into short positive or negative going pulses guaranteed to confuse a decoder looking for a DCC waveform.
Growing old, can't avoid it. Growing up, forget it!
My Layout, My Workbench Blog and My Opinions

User avatar
SRman
Posts: 1148
Joined: Fri Oct 03, 2008 2:26 am

Re: DCC loco on DC problems

Postby SRman » Sat Oct 23, 2021 1:20 am

Bufferstop wrote:The problem of DC on DCC is that when the baseline voltage drops back to zero and the loco stops there is still an 18V waveform on the tracks. A good old clunker like an X04, one of the old Romfords or an Original H-D is going to have a solid iron core in the motor and the AC will make it hum a little, fail to make it move and get it a little warm. Do the same with a more modern motor, it's armature will vibrate like crazy and it will get very warm, to the point of emitting smoke, do it with a coreless motor and the armature will rip itself to bits if you leave it standing in a powered siding.
The only time I've had DCC on DC was a couple of Bachmann Gregs. The were basically a faceless Percy fitted with a decoder and formed part of their digital starter set. Running on DC was "interesting" I think they would have been ok run off batteries, but the heavy100Hz ripple on a conventional controller got the decoder confused and they jumped and stuttered along the track, other people have commented on poor DC performance of various models. I think it depends on the controller you are using and the worst misbehaviour comes if you try to run them from a PWM controller which takes AC and chops it up into short positive or negative going pulses guaranteed to confuse a decoder looking for a DCC waveform.


I found this on my old layout, where I could swap from DC to DCC and back again. Running the DCC-fitted locos on DC was fine with an old H&M Duette controller, which put out a fairly pure DC form - that controller is still in use to this day. However, when I acquired a much more modern H&M 2000 controller (A Hornby product), that put out some odd wave forms that confused various DCC decoders, so when increasing speed a bit they moved off, increasing a bit more, yes, they loco went faster, increase more again and the loco slowed down. :roll:

The answer for running DCC locos on DC is to make sure the controller puts out pure DC, not the chopped waves as described by Bufferstop. That is, of course, if removing the decoders is not practical or desirable.

User avatar
fourtytwo
Posts: 153
Joined: Mon Oct 31, 2016 7:41 pm
Location: Norfolk (Bodecia's country) UK

Re: DCC loco on DC problems

Postby fourtytwo » Sat Oct 23, 2021 10:42 am

Thanks for every bodies contributions, as a non-dcc user it helps me to understand the behaviour. For me I shall just carry on removing decoders if encountered as I wouldn't compromise my controllers that enable me to creep shunt where pure DC would not. Obviously people would make different choices depending on there circumstances but I am a committed non-dcc modeller who has no desire to run there loco's on anybody else's dcc layout.

Thinking some more I guess the modern can motors are a lot more sensitive than the old X04 style so normally start moving at a very low voltage threshold but the dcc decoder requires a reasonable track voltage to power up and switch the track voltage through to the motor hence the observed startup delay and initial high setting required, also the inability to creep on "DC". Once again many thanks :D
A fresh start in OO, DC Steam

User avatar
Bufferstop
Posts: 13489
Joined: Thu Mar 11, 2010 12:06 pm
Location: Bottom end of N. Warks line

Re: DCC loco on DC problems

Postby Bufferstop » Sat Oct 23, 2021 11:29 am

Yes I think everyone has some experience of this problem, the only thing I would disagree on, and then ii's only a matter of degree, is an H&M Duette putting out a fairly pure DC, if we look at the waveform put out full power is an AC sine wave with every other hunp turned right way up, half power looks like Nessie's Humps on a mill pond. If you are competent to open an old controller fitting a large capacitor after the bridge rectifier and before the reversing switch will improve the smoothness of the output, maybe to the point of being acceptable as DC to most decoders. Of course I must issue the standard warning, don't open any transformer/controller unit unless you are sure of your ability to work on a mains powered device. The rivets you have to drill out are there for a reason. It will render the unit illegal to sell on! But then if you are competent you know that :wink:
Growing old, can't avoid it. Growing up, forget it!
My Layout, My Workbench Blog and My Opinions

User avatar
RAF96
Posts: 785
Joined: Tue Feb 19, 2008 1:39 pm
Location: Dereham, Norfolk, UK
Contact:

Re: DCC loco on DC problems

Postby RAF96 » Sat Oct 23, 2021 12:19 pm

What I don't think anyone has mentioned yet in answer to the OP question is a decoder is like a computer and has to boot up when it is first powered, hence when you apply DC power to it, it needs a certain voltage level to even crank it up, then it has to decide it is not being fed a DCC signal but an analogue voltage, so it puts that duck in the row as well and off it goes. Next time you stop the loco the decoder switches off. Repeat.

Many decoders now come with DC running set to off as default as a result and this also prevents DC runaway, another pest where if a decoder loses the DCC signal and thinks it is on DC it uses full DCC track voltage to crash your most expensive loco.

The other thing is we started talking about DCC on DC but then went off on the other track of DC on a DCC track, which can lead to smoke if a loco is just left standing on track and is one of the reasons why many controllers either do not support DC running or have it set to off as default.
RAF Halton Brat - 96th Entry
http://www.halton96th.org.uk/robs_rails.html
β-tester

User avatar
Bufferstop
Posts: 13489
Joined: Thu Mar 11, 2010 12:06 pm
Location: Bottom end of N. Warks line

Re: DCC loco on DC problems

Postby Bufferstop » Sat Oct 23, 2021 10:31 pm

I think it's fairly evident from people's experience that if a DCC loco is to be used long term on a DC system the best course of action would be to remove or at least disconnect the decoder. The two facilities to run DC on DCC and DCC on DC were introduced to try to sell DCC to the serious established modeller, at a time when the average motor could withstand operating on the "wrong" supply.
The established modeller with a large number of DC locos has to choose one of the optional paths in making the switch, or not! For the newcomer DCC is the obvious choice, there's a fairly hefty purchase to make at the begining in the form of the best controller that an be afforded but thereafter some locos may need a small additional expense for a decoder.
Endless attempts to make these two facilities work faultlessly are just a waste of effort.
Growing old, can't avoid it. Growing up, forget it!
My Layout, My Workbench Blog and My Opinions

gppsoftware
Posts: 306
Joined: Thu Mar 09, 2006 5:34 am

Re: DCC loco on DC problems

Postby gppsoftware » Tue Oct 26, 2021 11:33 pm

Bigmet wrote:I learned the problem with the way DCC provides a DC voltage on 'address 0' the hard way about two days after getting my DCC system. I had used all my initial purchases of decoders to fit selected locos, and tested them, all good and up and running. Right, let's try an unfitted loco on address 0. Used a large motored loco (Mashima 1830) with a very free running drive, and got an immediate cloud of smoke without it even moving. Superb steam loco effect, most of the smoke expelled from the 9F chimney, but rather a high price for the few seconds it lasted. All the other locos running on DCC didn't miss a beat, nor did the system trip. Never tried address 0 again.


Be aware that not all DCC systems support address zero for a DC loco.

I know that my Lenz 100 system does, but my NCE system does not. Haven't checked my Hornby Elite.

Without check instructions, I would say that it isn't safe to assume that address zero will run a DC loco.

I am in agreement with others that this is such a long outdated backwards compatibility thing designed to encourage DC users into warm DCC waters, but it is no longer necessary and address 0 locos never ran particularly well anyway, not withstanding the constant buzz!

gppsoftware
Posts: 306
Joined: Thu Mar 09, 2006 5:34 am

Re: DCC loco on DC problems

Postby gppsoftware » Tue Oct 26, 2021 11:34 pm

Bufferstop wrote:I think it's fairly evident from people's experience that if a DCC loco is to be used long term on a DC system the best course of action would be to remove or at least disconnect the decoder. The two facilities to run DC on DCC and DCC on DC were introduced to try to sell DCC to the serious established modeller, at a time when the average motor could withstand operating on the "wrong" supply.
The established modeller with a large number of DC locos has to choose one of the optional paths in making the switch, or not! For the newcomer DCC is the obvious choice, there's a fairly hefty purchase to make at the begining in the form of the best controller that an be afforded but thereafter some locos may need a small additional expense for a decoder.
Endless attempts to make these two facilities work faultlessly are just a waste of effort.


Seconded on all of this.


Return to “DCC Forum”

Who is online

Users browsing this forum: No registered users and 9 guests