Pulse width control

[vulcan]

Hope someone less electronically challenged can offer an opinion on this.
I have been given a hand built controller that operates on variable pulse width, it has been tested and is electrically safe but I seem to recall some controversy over the adverse effects this form of control has on small &/or coreless motors. Or am I thinking of feedback control?

Thanx

Ken

[pilotfriend]

I operated pulse controllers in the 1970s. They were wonderful and performance was a good as DCC. My own experience was that I never had problems with my motors which are running well to this day. I am sure some new motors may not like them but other will know more about this.

John

[david788]

I have been given a hand built controller that operates on variable pulse width, it has been tested and is electrically safe but I seem to recall some controversy over the adverse effects this form of control has on small &/or coreless motors. Or am I thinking of feedback control?

From my limited understanding (waiting to be shot down by the electronics experts), feedback control is often achieved using a variable pulse width system, but a variable pulse width controller does not have to include feedback.

I have used both since my EM gauge modelling days back to the 1970s and both can work very well and give improved performance but there are some possible 'health warnings' especially when used with very small motors and 009 locos,

1. With variable pulse width control, the motors can run much hotter than with simple variable voltage control, especially at very low speeds.  This was not a problem with most of the larger motors we used in EM gauge but did create difficulties when I used the same controllers on some 009 locos (N gauge mechanisms) on longer continuous runs. Some motors became dangerously hot within a very short time whereas others seemed OK.

2. We found that some motors, especially Portescap type cans, did not work well on the feedback control from the early ECM Rambler that we used. I understand that this was because of the low inertia of the motor and later versions had switchable feedback to overcome the problem. I guess the same might apply to some of the very small cans used nowadays although no doubt modern feedback control may be much more sophisticated.

Both of these difficulties can be overcome, an electronics engineer where I used to work about 20 years ago built a controller which gave the best control I have ever seen without the problems. I've no idea how it worked but to this day I regret that I was put off trying to build my own copy by the high component cost and lack of confidence that I could build it.
I've no idea but I assume that the feedback control available on DCC systems gets round the problems similarly.

So getting back to your original question. I'd certainly try it. Depending on the design it may work very well, BUT be VERY watchful for any sign of the motors running hot especially on any length of run at low speeds.  I'd be especially wary on items such as the Kato tram chassis which run very fast on 12 v so we well tend to run them on very low voltages.

[Hugh Norwood]

It's all down to frequency... The smaller the motor, the higher the frequency of the PWM controller in order to make it work properly. In't good old days, locos had big motors and so the old PWM controllers worked for them but not for modern motors - they get hot.

I have built a PWM controller - just to see if it worked - and it does - it uses a frequency around 18kHz, I think.

DCC chips drive their motors using PWM, don't forget.

The big advantage of DCC chips is that there is very little cable between the controller and the motor, whereas for a conventional (PWM) controller, you have the signal going all the way from the controller through the track to the motor, which could present problems.

Just a few thoughts...

[gallopingalice]

Hi All,

Pulses are not the problem,
Whereas the peak voltage IS.

You would not run a 3volt motor at 12volts,
and be amazed when it got hot, would you?

A variable mark-space ratio, or pulsed, drive
is used to vary the fraction of the maximum voltage
presented to the motor.

The trick is to select your maximum voltage to suit your motor
and the top speed that you require of it, then chop up that voltage
to control the speed below the top speed.

Or so it would seem to me!

regards gallopingalice

[tedp]

Feedback control is achieved by the controller 'listening' to the back voltage produced by a running motor in the short periods the power is off. You need a pulsed voltage to be able to achieve this. When a motor turns and the power is switched off momentarily during the pulse cycle, it will work as a tiny generator producing a reverse voltage. This is used by the controller to control the circuit which produces the power pulses.

I have KPC 'variable feedback' controllers, with two feedback settings. These are unavailable now, but I'm still very satisfied with their performance. Im not sure how and if the feedback is modified in these controllers, but alternatively it may be the peak voltage and/or the frequency.

On the full feedback setting they are suitable to drive mass produced motors used in many older N gauge chassis, like the Minitrix dock tank and the Fleischmann 0-4-0. None of these will run hot, except perhaps the motors in the old Roco H0e chassis, but these run hot on any controller. However, if you run a Mashima motor on this setting the motor will run a bit rough. Mashimas are better on the reduced feedback setting, although they won't run hot on the full setting. The Kato 4w tram chassis is OK on the full feedback setting of the KPC controller.

Coreless motors as far as I know shouldn't be run on a feedback (or pulse width) controller and I have never done so. Some very small miniature motors like the one in the Caramel chassis run very rough on the KPC with full feedback, and only moderately well on reduced feedback. But the Caramel motor won't run hot provided you run it at normal speed.

If you use a very high frequency in a pulse width controller, as hughn says, you will encounter problems 'piping' the power a long way on a big layout, as the HF voltage will easily be lost in the wiring and the track.

[Hugh Norwood]

I have made a feedback controller that is not PWM - it is an adaptation of the Lindlsey-Hood circuit for speed control of tape-recorder motors. It monitors the back-emf from the motor.

Yes, I have also made my own DC controller. In fact I use this one most as it fits nicely in the hand. It has three settings on it which limit the maximum voltage to (if I recall correctly) about 5 volts, 9 volts and 12 volts. A lot of my locos work fine on the lower setting. It would be quite easy to alter the controller to limit the maximum voltage to 3V or whatever you wanted.