Subject:
|
Re: Smooth power or PWM for Lego train motors?
|
Newsgroups:
|
lugnet.trains
|
Date:
|
Thu, 4 Sep 2003 06:55:49 GMT
|
Viewed:
|
1192 times
|
| |
|
|
In lugnet.trains, Amaury Jacquot wrote:
> Mark Bellis wrote:
> > I've seen the DCC mod some people have made to motors, and read about
> > the overheating problem.
>
> the overheating is not due to the use of PWM, but because the task asked
> from those motors was too great. why do you think long trains use 2 or
> more engines, with sometimes even engines in the middle or at the end ?
> One of the questions I had for those using DCC is, Do you make sure only
> 9 volts is given out by your DCC system ????
Being British I'm not too familiar with American trains (yet), Though I know
about the multiple engines in real life. I would intend to triple head trains
in a similar way when I have made enough American engines!
The motor duty limitation is one reason I monitor the current from my power
supply.
> > Does DCC provide smooth DC to the motor or PWM? If it's
> > PWM, the motor will heat up twice as much as usual.
>
> that depends.
> PWM is the only proper solution but the frequency at which the pwm is
> done is very important. also, lego motors are not really adapted.
> You see, most high-end model train manufacturers (fleishmann comes to
> mind) use Faulhaber 5 poles motors. the lego motor is only 3 pole.
Yes, I see your point. I tested the PWM frequency of the RCX and it was 164Hz,
not much better than half wave rectified power. I think it's OK for the Technic
gearmotors it was designed to drive, but not for train motors (unless it's a DCC
command station).
> > I once tried a 9V train motor with a 12V transformer up to 9V but the
> > buzz was horrible!
>
> the buzz you hear depends on the frequency (you heard the basic 50 or 60
> Hz of your mains supply)
>
> > This used full wave rectified power in a similar way to controllers
> > from non-Lego railways.
>
> well, duh, that was the 50 or 60 Hz you were hearing. proper pwm is done
> at a much higher frequency (10 or 20 KHz).
> by the way, this was probably a VERY CHEAP transformer
> Ever been on a recent public transportation train ? the (sometimes loud)
> high pitched noise is the PWM you are hearing.
Yes, I would think of using a PWM frequency of above 22kHz since that's the
threshols of human hearing (might still disturb a dog though!)
I've not been on a public electric train as they are mostly diesels round our
way in the middle of England.
> > I concluded that 9V motors need smooth power and have used that ever
> > since.
>
> the premisses were flawed, hence the conclusion too
Yes, OK, but I always intended to research further to improve my conclusion!
> > I use a dual 30V 3A smooth DC power supply to power my Lego trains,
> > especially since some have four motors and one happily draws 1 Amp
> > round the curves.
>
> this sounds like a nice power supply. don't go over 9v though
> oh, and by the way, this is where the lego controller fails to deliver...
> This thing is designed with an LM317T voltage regulator. This part is
> not designed to handle more than 1 amp of *resistive* load. An electric
> motor is part resistive, part inductive, and thus, they would have
> needed a much larger voltage regulator (national semiconductor makes
> some that can handle up to 8 amps). Also, the use of a wall adapter
> (just a transformer in a box really) is not proper. True, it makes the
> boxes cheaper, because you can ship just a different transformer for the
> various countries (100v for japan, 120v for the US and 230 for the rest
> of the world). However, this small transformer can only give out about 1
> amp max, thus, there's no way it will be able to properly power more
> than 2 trains (as you say below, the motor requires 300mA minimum)...
Yes, I've had a 9V conroller apart and drawn the diagram. I was thinking of
using an outboard pass transistor (2N3055 or similar) in accordance with the
LM317 data sheet. Horowitz and Hill show how it's done, but I haven't had time
or enough motivation to build it yet. The transistor would need its own
heatsink outside the box, so it might get untidy, needing more effort on the
enclosure. I think the max output of the transformer is 300mA. The limit on
the British wall transformer is 7VA - 0.78A resistive at 9V, minus the losses in
the controller.
> > I reckon the motors are 3 Watt ones, so I allow 300mA per motor when
> > running the trains (in proportion to the applied voltage though).
> > Some of my steam engines have Technic gearmotors (also 3 Watt motors)
> > to turn the driving wheels and I make sure the load on them is small
> > as I build.
>
> nice and all, but there is no way you will be able to have realistic
> speeds with variable voltage. the lego motor doesn't even start turning
> if you feed it less than 2.5v (at least)
>
> quoting from :
> http://stage.itp.nyu.edu/~tigoe/pcomp/motors.shtml
>
> The rated voltage of a motor is the voltage at which it operates
> at peak efficiency. Most DC motors can be operated somewhat
> above or below their range, but it's best to plan to operate
> them at their rated voltage. Dropping below rated voltage
> reduces the motor's power, and operating above the rated voltage
> may burn the motor out. Plan on the motor's top speed being at
> rated voltage, and slowest speed at no more than 50% less than
> the rated voltage.
>
> you can see that you should not go below 50% of 9v for the lego train
> (or technic motor), which is 4.5v
>
> now, acording to:
> http://hot.ee/sonnich/lego/9vcontroller.htm
>
> the lego controller has the first 2 steps under those 4.5 volts
> (respectively 2.9v and 4.3v).
Yes, I understand that realistic speed for many American trains is a crawl, as I
see HO ones at a railway club I go to sometimes. British goods trains often run
at 45, 65 or 75mph and passenger ones at up to 125mph. My small engines can
crawl in the yard on a Lego controller when they don't have trucks behind them.
Funnily enough, putting a gearmotor in parallel with the train motor actually
speeds up the train motor under no load conditions - perhaps the gearmotor is
acting as a parallel choke and storing the spikes from the train motor.
> Now, to put a final nail in the coffin, the RCX uses PWM, and I never
> had a motor overheating....
>
> Conclusion:
>
> Your post shows no knowledge of physics whatsoever, and not even
> research about it's subject on the web (which has tons of educative
> resources at your fingertips...)
Sorry, I don't have the time for extensive web research - I snatch 15 minutes a
day online and I just want to chat about Lego to fellow enthusiasts. I would
rather talk to people with ideas that have worked for them than spend ages in
books (or equivalent web pages). Surely the aim of this place is to encourage
others in the Lego hobby, not put them off by patronising. We're not all
professors of quantum Lego physics! :-)
> PS: google is your friend :D
If I get the web at home (which I hope to soon) I'll spend more time looking,
but for now, thanks for the pointers to other useful pages.
Mark
|
|
Message has 1 Reply:
 | | Re: Smooth power or PWM for Lego train motors?
|
| (...) You could also simply swap the LM317 for a heavier type (LM317K for instance), which could give you around 1000mA uncooled without much issues. Together with Klaas Meijaard we cut up an old home computer power supply, giving 2*11VAC at 22 VA, (...) (21 years ago, 4-Sep-03, to lugnet.trains)
|
Message is in Reply To:
| | Re: Smooth power or PWM for Lego train motors?
|
| Mark Bellis wrote: > I've seen the DCC mod some people have made to motors, and read about > the overheating problem. the overheating is not due to the use of PWM, but because the task asked from those motors was too great. why do you think long (...) (21 years ago, 2-Sep-03, to lugnet.trains)
|
5 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
