Q: Track conductivity at extremely low voltages

[VonRyan]

I would agree that your orriginal assumption (that the problem is electrical contact) is correct.  If you crank the DCC throttle to the max and the loco is not moving then I suspect that the power is not being delivered to the loco rather then a stalled motor. Unless there is some serious bind, a 12V motor with 12V applied to it is generating maximum torque it is capable. Remember, permanent-magnet based DC motors generate maximum torque when stalled (at full supply voltage).

Since you have a DCC locomotive, assuming it has a working headlight, you can easily see whether the problem is electrical or if the motor is stalled.   When your loco doesn't move while the throttle is all the way up, see if you can turn the headlight onn and off. If not, then the decoder is not getting any power from the track.

From all the explanations so fay I thin that jagged ben's makes most sense.  At slow speeds there is very little inertia in the gear train to allow the model to coast through the dirty spots in the track.

I chopped the wire for head and tail lights off of the decoder to save space. And also because the headlight wouldn't have been on during the day.

The dead pole would appear accurate, but when I originally ran it on DC there was only startup issues on dirty track. That made me think that either the motor doesn't like DCC or the very slight tilt in the gear tower is effecting startup. I'm hoping that it is just the motor, as I can live with nudging the engine if need be.
I can't relocate the decoder because the wires have been cut as short as the length needed and putting it in the tender would mean modifying it as well as the fact that I wouldn't be able to put it in its box.


-Cody F.

[mmagliaro]

...
I know the engine is making contact as the engine makes an almost un-noticeable slight-movement when the throttle is turned up to a certain point.
The engine doesn't always require a nudge, but I would say that at least 50% of the time it does.

-Cody F.

Ah wait, this is different.   I thought it did this often when you would go run it for the first time
in a while.  Now I am suspecting a bad winding, armature slots loaded up with carbon, or a bum
brush spring.

When I got the H10 I have, it had a burned out motor (and it is a straight-wound 5-pole, by the way).

[VonRyan]

Aren't these tested at the factory?
When I got mine, it had hardly any run-time probably just enough to seat the gears reasonably well.
I have clocked on a good bit of run time since I wired it with DCC, but hardly enough for any major wear issues to arise.
The only thing I can really surmise is that the slight tilt in the motor (which results in a slight tilt in the entire mechanism) has created a significant need for a good deal of torque (or momentum) to be generated on startup, in which case I can live with nudging it since I can't move the decoder, and taking the shell off and then putting it back on is quite the laborious process.


-Cody F.

[mmagliaro]

Cody, the short answer to your question, in my experience, is "No", they are not tested at the factory.
Maybe somewhere, somebody tests the motors, or  1 out of every 100 or 1000 motors, but as a general rule,
no.  Nobody is testing every engine that comes out.

I am fuzzy on just what you did to cause this misalignment.   If there is something misaligned enough to keep
the motor from starting up, no matter how high you turn up the speed, there is something really wrong in there.

Do you have an ammeter and are you able to measure the current draw when this engine is running?
It might appear to run ok once it gets going, but something could be binding it that is really working
the motor hard.  No matter how you slice it, a loco that will not start up no matter how high you turn the throttle
has something really wrong in there that should be addressed.

[victor miranda]

hi Max,

At slow speed, almost anything will cause a stall.
There is little spare inertia in the mechanism to drive the loco over the electrical stall.

The real trick for good slow speeds is weight on the wheels doing electrical pickup.
If you can get about .5 oz on each, any stalling is truly dirty track or mechanism conduction related

victor




 

[mmagliaro]

Well hello, Victor... how the (*#)(*@  are you???   

Yes, stalling is more of a problem for all the reasons you and others have cited.
What I'm really after here is this.   Forget about motion or lost contact for a moment.
Assume that this is a stationary object... a light bulb on two brass blocks on the rails...

At lower voltages, does the current have a harder time breaking through the oxidation layer,
and more importantly, does that magic break-over point occur at a voltage that "matters", i.e. between 1 and say,
4 volts.   In other words, if the engine were set up with dropping resistors or diodes such that it
would not move until there were 4 volts on the rails instead of 1v, would that alleviate the problem
of electrical stalls?   

A "hard stall", where the circuit isn't complete no matter how high you turn up the throttle, isn't going to be helped
by this, obviously.  I'm after the cases where the engine is not mechanically stalled or bound, and it won't run
at 1v, but when you hit 4 volts, suddenly, the current can jump across the oxidation layer and complete the circuit,
making the engine lurch forward a little.

I don't think anybody knows the answer to that question, but I was wondering if there was anybody who understood
the dielectric properties of the metals and oxidation that we typically have on our tracks, who just might be able
to shed some light on this.

Recall that there were electrical track cleaning devices that relied on this idea.  They would put very high voltages
(at severly limited current, of course) onto the rails so as to arc over the oxidation, complete the circuit and burn the
oxidation off.

[GaryHinshaw]

Max, I think the only way you'll get a reliable answer to this question is to try it and see. 

There is a sizeable literature on the ESD breakdown of highly-controlled silicon oxide layers because they are used as insulating layers in most micro-chips.  In that application, of course, they want to avoid breakdown.  Here is a recent paper that gives some test results:

http://www.scu.edu/engineering/centers/tent/upload/publication42.pdf

They typically find breakdown voltages in the 8-10 V range, depending on film thickness (in the 2-3 nm range!), area, and duration of applied voltage, but they have no compelling model to explain these results.

Since the insulating layers on our tracks vary widely in composition, thickness, etc., it would be difficult to even predict a rule of thumb.  So give it a shot and report back. 

-gfh

[victor miranda]

Hi Max
I am fine.
you asked a question that I have tried to answer for myself.
and if darned near anyone else had asked I would not have answered.

It is a little odd the answer is reliable slow speed is driven by weight on wheels.

voltage at the rails varies somewhat depending on the electrical draw in the loco.
if the loco electrically stalls the voltage at the rails will rise.

a b-mann with a decoder starts at about 4 volts...
my experience is they stall at slow speeds.

ok you are onto a question that I have found also.
what is the reason a loco is getting power,
is in good repair (no burnt winding or such),
and will not start until a lot of power is supplied.
I called it a powered stall.

here is my answer... eliminate possible causes
and what is left will be the answer.

the real eye opener was that spotlessly clean track had no real effect
on a powered stall.  a loco that did it, did it on CLEAN track...

I could not find a way to test static vs dynamic friction in the journal bearings.
also it is hard to tell the above from commutator bridging.

curing the issue... in the order I found most often effective
reshape the brushes of a three pole motor, fix loose bearings,
swap out the motor and replace gears.

if you can see the gear faces are flat, you can be looking at the cause of the problem.

victor

[peteski]

Max, I think that you are over-thinking this entire "very thin crud on the rails" problem.  I also think that the problem is more complex than it appears.

The other thing is that DCC (or whatever other remote control system will be developed in the next decade, or later) will most likely use full voltage on the track, or contact-less inductive energy transfer, so there will be no need for improving connectivity in the low voltage range.  I suspect that the conventionally-powered (DC) models will fade away, except maybe for a small fringe group of followers.

But then again, you might be onto something. And you seem to be enjoying it.  I agree with Gary: give it a shot and report back. 

[mmagliaro]

I guess since I started all this, I owe it to you to try it.
The thing that made me notice it was this.
My 4-6-2 project engine starts and runs at 0.9v... pretty darn low, but then, it's got that
coreless gearmotor in it.
But I noticed after test running it for a while that it wasn't quite as nice at startup.
For a few hours, I checked the things you'd expect... dirty track, something in gears, a bind I hadn't noticed,
etc.

Finding nothing, I then started noticing that when this happened, I could turn up the throttle and at around
2-3 volts,  "bam!"
the engine would take off, so ... I thought... aha... it's a mechanical bind.
But THEN I noticed that every time this happened, the ammeter read zero... DEAD ZERO until
all of a sudden it would jump when the engine took off, so...
loss of contact.

EXCEPT that it was loss of contact that could be overcome with higher voltage.

And that's how I got to where we are in this thread.

I took out the tender wheels, cleaned the axle tips and cups and the wheels, and the
good performance came right back.  SO.... it was electrical and it was electrical that
could be overcome by higher voltage.  I'm not sure I can reproduce it to even know
if it gets better by installing a resistor as I proposed.

I'll report back.

Oh, and Peteski... I am one of those fringe-element DC users.  There are a few of us left.   

[peteski]

Finding nothing, I then started noticing that when this happened, I could turn up the throttle and at around
2-3 volts,  "bam!"
the engine would take off, so ... I thought... aha... it's a mechanical bind.
But THEN I noticed that every time this happened, the ammeter read zero... DEAD ZERO until
all of a sudden it would jump when the engine took off, so...
loss of contact.

Oh, and Peteski... I am one of those fringe-element DC users.  There are a few of us left.   

Max, at this point the DC users are still quite numerous. But in about a decade I can see that number dwindling.   

As far as the symptoms you described go, is it possible that the open circuit (until you got to 3-4 volts) was not between the track and the wheel, or between the wheel and the pickup contact, but in the motor, between the commutator and the brush?

BTW, I keep meaning to ask you: from what ebay seller did you get those cool blue-backlit LCD digital meters you installed in your power supply?

[mmagliaro]

Max, at this point the DC users are still quite numerous. But in about a decade I can see that number dwindling.   

As far as the symptoms you described go, is it possible that the open circuit (until you got to 3-4 volts) was not between the track and the wheel, or between the wheel and the pickup contact, but in the motor, between the commutator and the brush?

BTW, I keep meaning to ask you: from what ebay seller did you get those cool blue-backlit LCD digital meters you installed in your power supply?

This may set a record for thread drift...  it is drifting in 3 directions simultaneously.

First, my own privately held suspicion about DCC is that in 10 years' time, something ELSE will replace DCC
as the control mechanism of choice.  Everyone who has migrated to DCC will find themselves migrating yet again.
(Maybe when those rechargeable Heli batteries get even better, we will carry a radio-controlled, battery-powered controller
on the engine, and at last be free from the bane of dirty track!)

Motor/ commutator /brush problem....   Very unlikely.  If it were an old motor, or a conventional motor, maybe.   But this
is a Faulhaber.  I have yet to see one of those be anything but perfect.   More scientifically, cleaning the wheel cups made the problem go away, so it was not the motor, it was the pathway between the track and the motor.

Those cool blue meters...  I am having trouble getting eBay to show me my prior purchases right now (they are
having a problem).   But a quick search shows that many sellers have that exact same meter.
Some ship from China, and some ship from a drop location in California.
Item 161045251073     is one of the voltmeters I used, and you can buy a similar ammeter.

The only thing to be careful of... these meters are powered by a 9-18v ISOLATED source.  You cannot
power it from the same transformer or power supply that is feeding the inputs of the meter!
Also, you cannot power two meters from the same power source (i.e. you can't hook them both to the same
9v battery!)   Yep.  I tried it, and blew up one of the meters.   For now, I just have them
each powered by their own separate 9v battery, but you could make a power supply with
separate ISOLATED outputs on it to provide 2 outputs.   

Other than that, they have been very
handy.  Nice bright displays, and they measure pretty close to what my DMM says (i.e the current
meter will read within 2-3 mA of what the DMM reads over the 0-100mA range
 and the voltmeter is within 0.1v of the DMM over the 0-10v range).    That is probably nothing more than
calibration.  I did notice some trimmer pots on these, but I did not mess with them.


Ahoy there, thread!   Where did you come from? 

[peteski]

Thanks Max!

To continue setting the record-setting thread drift:
1.  As you noticed, I did not paint myself into a corner with only DCC as an alternative to DC.  No way would I do that.    I specifically mentioned that in the next decade or later, most models will utilize some sort of control system which will use either full voltage at at the track or even wireless (inductive) power delivery. I guess batteries are also a possibility, but I still think that it doesn't make sense to solely use batteries on a model which is in constant contact with the solid surface it travels on.  So there! I'm not a DCC-snob. 

2. I should have mentioned this in my last post (but I can't edit it now):    Like I mentioned to Cody, if your loco in question has a headlight (which does illuminate around 3V), did you check whether the headlight was on when the motor wasn't yet running?  That would have been an important clue to show whether the open circuit was at the track or at the motor.

3. Instead of going through the ebay purchase history, I usually just look through my feedbacks to find my old transactions.  To me that works much quicker (and it is glitch-free so far).  In either case, thanks for the pointer to that listing.

As far as digital meters go,  there are ones which are capable of measuring their own power supply voltage. Not sure about current. But many cannot. Thanks for the warning.

I have a power supply (not as fancy as yours) which I built many years ago. I also couldn't find a meter which could measure its supply voltage's value.  I took care of the problem by buying a DC/DC (Yes, DC  ) converter. I found them, for not much money, at one of the surplus places I frequent (IIRC, it was Electronic Goldmine).  The input and output are fully isolated, and it accepts a wide range of input voltages while providing a regulated 9V output.  Its been working well for me but some day I would like to use the cool blue ones you found. 

[mmagliaro]

Thanks Max!

To continue setting the record-setting thread drift:
1.  As you noticed, I did not paint myself into a corner with only DCC as an alternative to DC.  No way would I do that.    I specifically mentioned that in the next decade or later, most models will utilize some sort of control system which will use either full voltage at at the track or even wireless (inductive) power delivery. I guess batteries are also a possibility, but I still think that it doesn't make sense to solely use batteries on a model which is in constant contact with the solid surface it travels on.  So there! I'm not a DCC-snob. 

2. I should have mentioned this in my last post (but I can't edit it now):    Like I mentioned to Cody, if your loco in question has a headlight (which does illuminate around 3V), did you check whether the headlight was on when the motor wasn't yet running?  That would have been an important clue to show whether the open circuit was at the track or at the motor.

3. Instead of going through the ebay purchase history, I usually just look through my feedbacks to find my old transactions.  To me that works much quicker (and it is glitch-free so far).  In either case, thanks for the pointer to that listing.

As far as digital meters go,  there are ones which are capable of measuring their own power supply voltage. Not sure about current. But many cannot. Thanks for the warning.

I have a power supply (not as fancy as yours) which I built many years ago. I also couldn't find a meter which could measure its supply voltage's value.  I took care of the problem by buying a DC/DC (Yes, DC  ) converter. I found them, for not much money, at one of the surplus places I frequent (IIRC, it was Electronic Goldmine).  The input and output are fully isolated, and it accepts a wide range of input voltages while providing a regulated 9V output.  Its been working well for me but some day I would like to use the cool blue ones you found. 

Important point, and NO, my headlight was not on until the engine jumped and started moving.  --- still
all consistently indicates pickup problems.

The place that sold me my meters actually apologized for not making the isolated supply requirement more
clear, and they sent me a free isolation "module"  (yes, as you point out, just a DC to DC converter ).
My power supply is anytning but fancy.  It's an LM317 circuit with fancy blue meters on it.
But it's plenty good enough for model trains.

(For the 15 odd years before that, I used a 2-transistor throttle circuit built inside a plastic deli
vegetable food container, so I figured I owed myself an upgrade!)

[VonRyan]

Cody, the short answer to your question, in my experience, is "No", they are not tested at the factory.
Maybe somewhere, somebody tests the motors, or  1 out of every 100 or 1000 motors, but as a general rule,
no.  Nobody is testing every engine that comes out.

I am fuzzy on just what you did to cause this misalignment.   If there is something misaligned enough to keep
the motor from starting up, no matter how high you turn up the speed, there is something really wrong in there.

Do you have an ammeter and are you able to measure the current draw when this engine is running?
It might appear to run ok once it gets going, but something could be binding it that is really working
the motor hard.  No matter how you slice it, a loco that will not start up no matter how high you turn the throttle
has something really wrong in there that should be addressed.

I recall an article in MR many years ago when they visited Samhongsa and it said that thy test ever chassis (and even showed a picture).

When I installed the decoder, I installed it in the boiler as that is the only place that had room without modifying the engine, plus retain the ability to separate the engine from the tender. The decoder sits right beside the gear tower.
When I put the shell back on, the decoder had just enough room, but for some reason (I think it may be the wires) the whole motor and gear assembly is slightly tilted maybe 10 degrees to the left (decoder is on the right side).
No real noticeable effect on the running quality or the gear mesh as far as I can tell.

Now, the engine doesn't always require a nudge, but I'd say that maybe 50% of the time it does.
And it only needs a nudge when it is first starting up. (Just out of the box, or after a break of about an hour or more)

If it stops mid-running (which is only ever because of a large dirt spot or dead section of track) it may need a hand over the trouble spot, which is normal for any engine.

I don't have any equipment to test anything with. My engines only run at shows as I have nothing else to run them on, let alone anything to make them run.

That's about as much as I can say. I can only speak based on what I understand and what I've experienced while running the engine.


-Cody F.