Max # of motors on a typical decoder

[Lemosteam]

In the N Scale Pennsy T1 thread a question was asked regarding if an average decoder, say a TCS M1, could power three n scale motors under load?  Those who know me here know that I've not the brain to hande this question.  For example, it would surprise me if O Scale dual-motor diesel folks had to spend for dual decoders in each diesel and speed match them.

And I have two additional questions to this-

• Is there an easy way to step up a decoder's power output?
• Can a resistor be used to slow a motor (what are the pitfalls/dangers/recommendations of doing this if possible)?

Many thanks in advance for any thoughts, suggestions or opinions on these three questions!

[TiVoPrince]

Back
in the day I built a pair of SD9043MAC locomotives (HO) with paired Overland 'green' motors.  I tend to really overweight locos, so too much current draw is a concern.  Overland motors of the era are happy to draw a full amp so I used a Digitrax DH83FX decoder because it was rated 2amp and had dich light options.  Until one of the motors failed it was a great combination.  Overland motors are made of unobtanium so replacement is the natural course of events.  Right now they rest on my workbench ready to receive pairs of Kato HM5 motors and improved LED lighting when the mood to work in HO strikes.  The 3mm LED and fibre optic combination that was so revolutionary and cool when originally built needed updating with discrete 'direct view' components since I was already 'under the hood'.

Generally Decoder amperage ratings should be respected because failures are costly in so many ways.  Repair time, melted shells or burned out motor/deoder are all concerns.  Slowing a motor with a resistor might work but heat and other factors usually mean that replacing the motor is a better option even if it looks expensive up front.  Atlas parts department slow speed motors are moderately priced and plentiful to the point that I'm pretty confident that if I fluff one up replacing it won't break the budget.  I have found that they draw less than 0.4amp when stalled so two on a 1amp Nscale decoder would fill me with confidence not concern...

[peteski]

John, answers to your questions are:
1. no
DCC decoders usually use a form of an H-bridge, consisting of 4 transistors (usually MOSFET), to drive the motor. It wouldn't be trivial to replace those transistors with ones rated for larger current. Besides, the rectifier diodes which convert the DCC signal to DC power for powering the entire decoder would also need to be replaced with ones with higher current rating. Then, depending on the implementation, the BEMF circuitry might be affected by these changes.

2. yes but..
A resistor can be used to dissipate some of the power applied to the motor, but that power would have to be dissipated as heat.  Depending on what you have in mind, the resistor could get quite large and quite warm (under certain conditions).  And plastic locomotive shells do not deal well with heat.  Another problem is that a motor with a series-connected resistor will behave in a non-linear way.  The speed of the motor with a resistor will be affected by the load on its shaft in a much greater way than in a motor without a resistor.

As far as powering multiple motors from a DCC decoder goes, it is possible to do that.  Specs of every decoder clearly state the maximum allowable current it can supply to the motor (and to the function outputs).  As long as the combined motor load does not exceed that current, that setup will work.

For example, if the motors you will be using each consume a maximum of 250mA (0.25A) of current under load (locos wheels slipping) at 12V (the current is greatest at the highest operating voltage) and you have a decoder with its motor output rated at 1A, then you could safely run 3 motors using that decoder.   I said three (not four) because it is wise to have a margin of safety.  Normally, you will not be running your loco with its multiple motors running at their maximum speed with the loco's wheels slipping, so the current load on that decoder will be even lower. But that safety margin I presented does not consider a possibility of stalled motors (which with 12V applied to them will consume more than 250mA of current each).  In this example, if all 3 motors stalled with 12V applied to them, that would exceed the decoder's rating and quite possibly let its magic smoke out.  It is up to you to decide how much of a safety margin you want to have. You can always try to get a decoder rated for higher output current.  Best way to determine the motor's current is to simply measure it under load, running on DC power, using a multimeter (inserted in series with the motor under test).

The other thing to consider is that with multiple motors being driven from a single decoder, its BEMF circuit might be affected. Assuming that the motors are not running in prefect sync, the BEMF will not be able to precisely read each motor's speed.  It will be a more of an average.

[Lemosteam]

TiVoPrince,

Thanks for the real life example! 

Peteski.

Awesome and exactly what I was looking for.

Just some follow ups.  I looked at the TCS site and the M1 has a 1A Contiunous, 2A Peak rating.  Would that mean the the safety factor you mention is kinda built in?  So in your example I would have a 1.25A safety factor with three motors using a TCS M1 decoder?

If I were to use a resistor on two of the motors (they are identical, from two newer Bachmann Northerns), what would I need to learn about the motors electrically, and how would I determine the correct resistor values to use?  Trial and error?  I think I might have adequate room around them to prevent shell destruction.  I may also be able to bury them in mechanism metal as a heat sink.

I am trying to electrically speed match (but slightly faster) the two northern motors with a LifeLike FA2 chassis.

I know this sounds like an insane way to power a kitbash, but you knw that I am an unconventional thinker...

The FA2 will power the tender, while the two northern mechs will be cut ahead of their rear two drivers and spliced together for the boiler chassis.  The northern mechs seem to run faster than the FA2 chassis.

I'm trying to do two things:

• Power all three with a single decoder
• slow down the northern mechs just enough to have some slight wheel slip at times

I think as long as the boiler mech runs slightly faster than the tender mech, I should not have problems, except in reverse where I do not think the boiler will be able to overcome the tender.

[peteski]

John,  I know that you don't want to head it but it would *REALLY* be much easier to do this with separate decoders. Not only it would be easier to do the initial speed matching, it would be so much easier to do any additional fine tuning adjustments later. Remember, it is so much easier to modify software (CV values) than hardware (unsoldering components).  I'm not even talking about custom speed tables - just 3 CVs: Vmin, Vmid and Vmax. Just messing around with those 3 CV, you can achieve pretty good speed matching.  Plus, you wouldn't have to deal with all wiring between those 3 units.

But since you're stubborn  , instead of resistors, use diodes.  Each silicon rectifier diode drops around 0.7V across it when it conducts current.  That voltage drop is almost constant regardless of the current passing through the diodes. Voltage drop across a resistor will vary depending on the current passing through it.  Here is a simple drawing showing how using diodes to drop voltage. Top example will drop around 0.7V and the bottom circuit will drop roughly 1.4V.  You can add more diodes into the circuit for additional voltage drop.



The diodes are rated  for 1A of forward current and 1N4000 has a reverse voltage of 50V (quite safe to use in your 12V circuit).

[peteski]

Just some follow ups.  I looked at the TCS site and the M1 has a 1A Contiunous, 2A Peak rating.  Would that mean the the safety factor you mention is kinda built in?  So in your example I would have a 1.25A safety factor with three motors using a TCS M1 decoder?

If I were to use a resistor on two of the motors (they are identical, from two newer Bachmann Northerns), what would I need to learn about the motors electrically, and how would I determine the correct resistor values to use?  Trial and error?  I think I might have adequate room around them to prevent shell destruction.  I may also be able to bury them in mechanism metal as a heat sink.

I didn't really answer those questions in my last post.
Yes, assuming that a motor uses 250mA with wheels slipping with 12V applied (per my lasat example), then with 3 motors a 2A peak rated decoder would give a good safety of margin.  You will at some point need to measure the actual current of your loco's motors to get the real world values.

If you end up using resistors then once you have some measurements of the motor current at few voltages (lets say at 3, 6 and 12 volts), then we can figure out some rough range of values for the resistor (both resistance and power rating).

[Lemosteam]

Peteski, thanks a lot!  I think I'll meet you half way, but I'm still thinking about the diodes...

So I'll use one to run the two bachamann motors that are identical, and one to run the LL chassis.  Then I can speed match the two chassis.  I probably wont get independent slip on one of the two motors in the boiler, buyt that's OK- I should be able to get good performance.

Could you give me a brief description of how I would measure the load on a motor at slip?  I see Max and others make reference to this but I am not sure how to set the multi-meter and where I should put the probes, etc.

Peak load would be max grip at near motor stall, right?: Or do I just run the thing and capture the highest number?

[peteski]

Sounds like a plan John.  I'm glad that I convinced you to come to the bright, um, DCC side.   
Slip or maximum grip, there is not too much difference. In N scale this is not an exact science.  You just need an approximate idea of what kind of continuous current could be flowing through the motor while the locomotive is still pulling the train.

When measuring current, the ammeter has to be installed in series with one of the leads coming  from the power pack/power supply.  Doesn't matter which of the 2 leads.  If the meter is digital then the polarity doesn't matter either (it will just read negative if it is connected in reverse).

The connections go as follows:
1. Connect one lead from power pack to the track.
2. Connect one lead from the ammeter to the other track.
3. Connect the other lead from the ammeter to the other lead from the power pack.

I don't know what current ranges your meter has so I can't tell you exactly how to set it up.  Most multimeters have low and high current ranges.  If the highest of the low range is less than 1A (or 1000 mA) then you need to use the high range (usually 10 or 20 Amps).  That won't be as accurate but it'll at least give you a usable reading without blowing up the meter or blowing a fuse. 

Place the loco on the track.
Turn up the throttle to its maximum (assuming that the throttle supplies about 12 volts) while holding the loco with your hand (so it is slipping). Note the current reading on the meter.
You can also take another reading with the loco stalled. Just push down on the model until its wheels stop turning, then note the current reading on the meter.  Do this quickly (couple of seconds at the most, so you don't damage the model).  Why? It is good to know the maximum current of the motor at the maximum operating voltage. 

If you would like to attempt another reading, right before the loco starts slipping), be my guest.  It won't be too far off from the slipping current.