Feedback or not

[Biskit]

Fairly straightforward query really: I'm looking at options for a new DC controller for a shunting layout. The critical requirement is for smooth slow speed running, resistance to stalling, and most reliable moving-off from stationary to slow speed. I know that achievement of these things is largely down to locos in good condition, clean track etc, however I want to give things the best chance on the controller side too. Prolonged/continuous fast running will not be required.

Looking at the Gaugemaster panel mounted range, there's the UDF range (with feedback) and the UD range (without feedback). I believe the non-feedback Gaugemaster controllers are PWM (can someone confirm?) which itself I guess will be a significant benefit over a smooth-DC type controller for slow speed and reliable starting. I've heard Morley Controllers recommended too, and had assumed that they were also PWM devices, but having looked at their website just now it clearly states "no Feedback or PWM on any of our controllers..." so I'm a bit confused.

I've also read conflicting advice about PWM and/or feedback controllers 'burning out' certain types of motors, but consensus seems to be that this only becomes a problem in extreme situations eg. prolonged fast running, where other controller types would arguably be better.

Any advice/suggestions?

[Walkingthedog]

All I can say is I have a Gaugemaster non feedback and it is perfect for slow smooth operations. I now use it on my new O gauge layout and it is the same.

[Brian]

You don't normally need feedback for shunting/slow speed movements. Its really used for maintaining a constant speed along the flat, up and down hill. Much like Cruise control does on a car. Feedback tries, up to its maximum ability, to maintain the pre set speed. However, there is no reason why such a controller can't be used and some offer switchable Feedback.
A lot of DC controllers use PWM which of course is not feedback. Feedback reads the Bemf produced by the motor.
Older Coreless motors tended to suffer with getting hot on PWM, but mostly now all locos motors are fine with it. PWM is how a DCC decoder operates a locos motor! PWM gives ultra slow control too as its pulsing the motor with full DC voltage - Less pulses for slow speed and more pulses for faster speeds, unlike a standard controller which applies a low DC voltage for slow speeds and a higher voltage for faster speeds often resulting in jerky movement at low volts due to 'stiction' in gearing and bearings etc.

[Mountain Goat]

I used to have two Guagemaster handheld controllers. One had feedback, the other did not. The one with feedback had much nicer control to it and it was actually cheaper to buy for some reason. It was there HH controller.
While I would say that slow speed control was slightly improved and locos were slightly less likely to stall, I would say that it was on the gradients that the HH came into its own. Out of the two plug in controllers (I had plugs so I could plug them in wherever I needed to on the layout) I hardly used the non feedbacked version!

And my personal advice on DC controllers is to avoid cluttered up controllers that have too much to fiddle with. A simple knob for speed and direction or a single knob for speed and a switch for direction.
In the past I have had all sorts of controllers from the simple to the complex. Take a look at the old H&M 3000 I used to have. All that for a DC controller that controlled a single track! Nice to play with but one is glad just to have a single control knob and a switch. (And I MUCH prefer toggle switches to slide switches as they feel soo much nicer to switch. Sadly Guagemaster have only ever done slide switches).
Whenever I have had controllers with similator and brake I have rarely ever used this feature and to be honest they get in the way if one does not use them.

But as far as feedback goes, as long as one does not want to run decoder fitted locos on the DC setting on occasions, and as long as it does not cost too much more, if I had a choice I would go for feedback every time.

[darkscot]

I have a Gaugemaster handheld controller with feedback option. With my old Hornby locos the feedback makes no difference at slow speeds. However, I got a new Bachman loco from Santa and feedback definitely improves slow speed on that, especially slow starting.

[Biskit]

Really interesting replies so far thank you! It's interesting that of you guys with experience in this area, some advocate the feedback route while some don't. I do have a bit of underpinning knowledge in terms of how the devices work (a degree in electrical engineering) so my query was really just how the different options perform in practice. My thoughts on the feedback option (making some guesses about exactly how it's implemented in this particular application) are that from a starting point of view, they may automatically 'ramp up' the duty cycle, t0 100% if necessary, in order to overcome the 'stiction' which Brian mentions, before immediately reducing it the instant the loco moves (and hence starts to create a back EMF), even with the controller set to a very low speed. Non-feedback PWM should also have the advantage of overcoming stiction by using 'high' voltage pulses, though if attempting to go from stationary to, say, 10% duty cycle, the pulses may not be of sufficient length to achieve this so a momentary 'flick' of the throttle up to a higher setting may be necessary to get it to budge, and backing it off immediately as the model moves, though it's likely that such an action could result in a surge of acceleration in a way I hoped the feedback device could avoid, by doing it automatically and instantaneously.

It's interesting how people's experience varies so much!

[RSR Engineer]

When I was designing my old layout in the late 1980s, with its gradients of 1 in 30-odd, I decided to use feedback controllers to even out the speed up and down hill. The Gaugemaster HH1 was ideal, as it also allowed me to move away from the control panel and watch the trains from any vantage point (e.g. to concentrate on shunting), and worked very well. However some locos, notably Trix models dating from before the merger with Märklin, did not run at all well. This is why I had to include the Codar in the circuitry, to substitute via a changeover switch when needed. This is all old hat now, of course, just mentioning it in case anyone else has any of these old Trix locos and is interested. All my other locos run perfectly on HH1s.

Cheers,
Artur