Capacitors. blah blah blah

[peteski]

Chris,
John's explanation is pretty good.

You mention that you are running the Plymouth (with 6V motor) at around 1.5V. What is the current draw?  What is the voltage at which the model will stall?

Then you have the boxcabs with 2 12V motors (is that 2 motors per boxcab or one in each?) which coast much further using the same supercap.  At what voltage are they running while you testing them? What is the current draw?  What is the voltage at which the model will stall?

I suspect that the current draw and the stall voltage will be in play here.

A battery (regular cell or a rechargeable battery) has a fairly flat discharge curve where it supplies a steady voltage for most of its rated capacity. Only when you get close to fully draining it, the voltage will roll off fairly steeply.

A capacitor OTOH has a discharge curve which is a slope down right from the start.  So if you are running your loco using 1.5V, the supercap is also only charged to 1.5V. As soon as the track power is cut, the motor starts draining the supercap and the voltage starts on the downward slope.  If the motor will stall at let's say 0.5V then there isn't much time it will take for the cap to discharge down to 0.5V.  But from that point on, the perfectly remaining energy stored in the cap will simply drain through the stalled motor - it will be wasted.  So you only have a not very big window of the engine running from 1.5V - 0.5V - good part of the energy in the cap is wasted.

But if you uses a stay-alive supercap in DCC it works much better. Why? because the window of the cap supplying usable power is much larger.  The supercap in the keep-alive is charged to whatever the DCC track voltage is (more or less). So that is around 12V.  But the decoder's and motor's operating voltage range is much wider than in your model.   The decoder and motor will work from the full voltage (12V) down to about 3V.  As soon as the track power is cut, the supercap sill start supplying the power. But in this case the operation window is 12V-3V=9V. That is much longer than in your DC model. The cap is able to provide much more useful energy to keep the mode running before it discharges to under 3V.

In your case, the model will start slowing down as soon as the track power is cut, where in the DCC keep-alive, the model will stay running longer without any appreciable slowdown.

[Chris333]

The boxcabs. There are 2 boxcabs. One 12V motor from a Kato 11-105 type chassis is in each boxcab. They are wired together with one cap:
https://picasaweb.google.com/ErieChris333/Nn323TBoxcabs#5392659408631748178
https://picasaweb.google.com/ErieChris333/Nn323TBoxcabs#5392659468580007986
https://picasaweb.google.com/ErieChris333/Nn323TBoxcabs#5392659315139367122
(wiring was later tidied up  )

In this (boring) video they are running off a 1.5V D cell battery:

/>
Yesterday I was running it at 1 volt. I have a video of it, but all it really shows is the train running, doesn't prove anything. But either locos speed seems to be most pleasurable around 1-1.5 volts.

The boxcabs have a pretty low gear ratio, like around 20:1. So the slow speed is all motor.

The Plymouth has a Searails Power Max chassis under it:
http://www.searails.com/powermaxproducts/powermaxmotorchassis.html
With a 6 volt very small motor and a 60:1 gear ratio.

I'm sure you can guess that anything this small is light. They have tungsten wedged in where ever possible. Of course the boxcabs have twice the wheels for pick-ups as well.


As for current draw and stall. Well I've tried many times to figure that out with various locomotives. Victor explained it, Max explained it, you have explained it, I've looked it up. And I end up with a goofy number that does not compute. Either my meter sucks or I'm just dumb. I'm guessing there is a reason you put that part in bold type and I am no help.

[peteski]

Well, without the parameters I asked for there is not good way to even semi-scientifically answer your question as to why the supercap provides longer running in the boxcab engines.  That is why I used bold text.

When I say stall voltage that is not stall current (which is a maximum current the loco will consume at the maximum operating voltage - which is often mentioned in model reviews).  I just wanted to make that clear. I am interested in what is the voltage at which the running motor stops spinning and the loco stops on the track.

I'm also curious as to what throttle are you using. Can it throttle down the voltage smoothly down to zero volts?   What type of voltmeter are you using? Is it built into the throttle or a regular (digital or analog) multimeter?

You did mention that you do have a (multi)meter. Can you post a picture of it (the faceplate)?  Measuring voltage and current shouldn't be very difficult (unless your throttle outputs unfiltered pulsing DC).

[Chris333]

This is me trying to measure the current draw. Both locos are running nice and slow.



When the Plymouth was running I pushed it down till it stopped running and the meter read 0.12.  Is that what you mean?

This is my controller RC02 running with an AC cord:
http://www.rokuhan.com/english/products/controller/

[Chris333]

I have the controller plugged into a 12V wall adapter. With no trains running I get 15-17volts at the rail when going 0 to full speed 

An MRC pack on my big layout reads 0-14volts at the rail going 0 to full speed.

[peteski]

So, when you took these measurements the boxcabs and the Plymouth were running at the speed you like (nice and slow).

At that speed the boxcabs are consuming 20mA (0.02A as the meter indicates).  So that is most likely 10mA per each motor, which is a very low current (not that there is anything wrong with that). 

The Plymouth is consuming 50mA (0.05A as the meter indicates).  That is twice the current that the boxcabs consume!

This alone is a pretty good indication that the same super-cap used in these models will discharge much quicker in the Plymouth.  I suspect that this is due to the 6V motor's windings having lower resistance than the 12V motor windings.

Just knowing the above is enough to tell us why the super-cap is not as effective in the Plymouth.  The stall voltage I was asking about would be measured across the track (DC volts).  You would run the loco at the desired speed, then gradually slow it down until it ws no longer running (stalled). The voltage at that instance would be the stall voltage I was asking for.

What you did by pushing down on the loco until it stalled was to measure its stall current.  It read 0.12A which is 120mA. It seemed reasonable, but the stall current (as given in model reviews) is usually measured with full throttle voltage applied to the motor (which is 12V for standard N scale models). However, for what we are dealing with here we don't really care about the stall current.

[peteski]

I have the controller plugged into a 12V wall adapter. With no trains running I get 15-17volts at the rail when going 0 to full speed 

An MRC pack on my big layout reads 0-14volts at the rail going 0 to full speed.

Hmmm... so with no load (a locomotive) on the track the Rokuhan controller seems to send 15-17V to the track regardless of the position of the speed knob?  But if you measure the track voltage under load (with a loco on the track), it goes down to zero volts when adjusted for minimum speed? If so, that most likely means it uses a simple rheostat instead of a transistorized controller.

The Rokuhan web page you linked to does not specify the throttle's internal design.

The 17V is not surprising if you use a cheap wall-wart type of power adapter. Those are usually unregulated and even if they use a 12V transformer, the filter cap inside them will charge up to the peak value of the rectified AC voltage (which would explain the 17V).  Under heavier load, the voltage would drop down close to 12V.

EDIT: I reviewed the Rokuhan web page and I don't know how I missed that "DC voltage controller with digital voltage control which generates less heat."!  So this throttle most likely uses PWM (Pulse Width Modulation) to vary the average voltage delivered to the motor (similar to how DCC decoders work).  That makes measuring the true DC voltage more difficult.

Maybe you should get a true DC variable voltage supply (like http://www.ebay.com/itm/271559759526 ). That would make things easier (as far as taking measurements).

EDIT #2:  After gettign some sleep, I got thinking:   If the Rokuhan is indeed a PWM throttle that means it sends full voltage pulses to the motor at all times (even on slow speed setting). It is just that those pulses have very low duty cycle (so the motor "sees" them as low voltage). But the super-cap would still "see" those pulses at their full voltage.. I'm surprised that the super-cap is able to withstand such an overvoltage without getting damaged.  But all of this is my speculation based on the description of the Rokuhan throttle.

This could all easily be checked with an oscilloscope (which I have but you don't).    Try putting your multimeter on AC voltage range and then check how much AC voltage it sees at the track (with and without the loco on the track).

[jdcolombo]

John,

I'll order those caps and try it. (just ordered 10) My space is 12mm wide, 10mm high, and 7mm long.

Since the caps are 6.8mm in diameter (you might have to cut the mounting tabs back some - they appear to stick out quite a bit), you have plenty of room for these.  With 12mm width and 10mm height, in theory you could use 3 of these wired in parallel, making a "sandwich" that is 9mm thick, and 6.8mm in diameter.  The diameter should just fit the 7mm length, so you orient your sandwich L-R.  This would give you nearly a Farad of capacitance.  A whole Farad is a lot.

John C.

[Chris333]

I'm not sure either loco will run much slower than I'm running them, but I will try later on to get min current.

So with the caps John listed do I still need a diode?

[peteski]

I'm not sure either loco will run much slower than I'm running them, but I will try later on to get min current.

So with the caps John listed do I still need a diode?

It is still is a polarized super-cap. So, to do it right - yes. But you have had success with running your super-cap equipped models without a diode and even in reverse polarity applied to the super caps, so I guess it is up to you to decide. 

[Chris333]

OK they charged me $11 for shipping, but they are here already. Just a mock-up, but looks like I can get 4 in there easy. I will cut down the mounts and solder them up.

[jdcolombo]

OK they charged me $11 for shipping, but they are here already. Just a mock-up, but looks like I can get 4 in there easy. I will cut down the mounts and solder them up.

Did they work?  Holding my breath, waiting to hear . . .

John C.

[Chris333]

Well they did work, but I can't really notice any difference. Like noted the faster I'm running the better they work, but now I need to limit to 3.3volts.

Also I'm not sure what the tabs are made of. They were a pain to get solder to stick. I sanded each surface first too. They cut easy so I doubt they are stainless. Oh well.

Think what I need is a double worm reduction drive with a 12volt Kato motor vertical in the cab. This way the voltage will be higher.

[jdcolombo]

Yeah, the voltage - I was afraid that might be an issue.  The amount of energy stored in a cap is a function of both the capacitance and charge voltage.  Peteski's right that those of us using DCC don't have to worry about the charge voltage, because the way DCC works, the caps will get a constant voltage from the track (after rectification by the DCC decoder board) of 10v or more (depending on the track voltage).  The caps work better when you're running faster because you are supplying more voltage to the rails, which charges the caps at a higher voltage (though, as you know, don't go past 3.3v).

I wonder if you could use a couple of diodes in series with the motor leads.  The diodes drop voltage, usually something like .6v per diode.  If you get best running performance with the motor set at 1v, if you added 3 diodes (dropping 1.8v), you could now supply 2.8v to the track, charging the caps at a higher voltage.  I don't know if this would work electrically - maybe Peteski can chime in on whether this idea makes any sense.

John C.

[Chris333]

They even sell 2 diodes to use with the chassis to limit the motor to 6volts. I didn't install them because I just keep the speed (way) down.