Date   
Re: Hot Snubber

redking56@...
 

LOL

DONE !!!


---In WiringForDCC@..., <dvollrath@...> wrote :

If you have substituted a different R & C network in that same location and it does not get 'hot', the original capacitor is either the wrong value or faulty. Please throw it out and let's get off this subject.


DonV

Re: Hot Snubber

dvollrath@...
 

The definition of 'hot' is subjective. If you pinch a resistor between your fingers for 4 or 5 seconds and it is dissipating even 1/4 watt it will feel 'hot'. 1/4 watt for 4 seconds (1 watt-second = 1 Joule) and is enough heat energy to raise the temperature of one pound of water one degree F. Compare the mass of your fingers with a pound of water and you can guess how long it will take to raise the temperature of your finger to a sensitive or even uncomfortable level. If the capacitor is shorted, the resistor will see full track voltage across it. 14V x 14V / 100 ohms = nearly 2 watts.


If you have substituted a different R & C network in that same location and it does not get 'hot', the original capacitor is either the wrong value or faulty. Please throw it out and let's get off this subject.


DonV

Re: Hot Snubber

redking56@...
 

OK, here is what I just did.

I disconnected the bus wires just to the right of the #4 snubber.  The resistor was very warm to the touch.

So, I reconnected the #4 snubber and disconnected the bus wires just to the right of the #3 snubber,  Waited several seconds, then touched the resistor on the #4 snubber.  It was cool to the touch.

Then, I reconnected the bus wires next to the #3 snubber, waited several seconds, and then touched the resistor on the #4 snubber.  It was warm to the touch once again.

Thoughts?

Rich

Re: Hot Snubber

redking56@...
 

I tend to think that it is the capacitor because yesterday I did that test with some new 1W resistors in place of the 1/2 W resistor and the problem persisted.

Rich

Re: Hot Snubber

Flash Gordon
 

I believe you have some bad resistors or capacitors. I would get rid of any that get that hot.

The Ramp meter gives average voltage, it does not read/ give the spikes. Only an oscilloscope will do that, but they are there.

Ed S

At 11:32 AM 3/15/2014, you wrote:


LOL

Ed, you may stay up. I should have mentioned it at the outset but, Yes, I am HO scale.

I do have a RRampMeter on my layout, and it shows a constant 14.7 to 15.4 volts. It is wired in-line, though, so I do not use it portably to check for voltage spikes or drops.

Incidentally, I did replace the hot snubber yesterday, and the replacement runs a lot cooler, or maybe I should say, less hot.

Just now, I went down and inserted the hot snubber on my bus wires over by snubber #6 where that snubber is warm at best. Very quickly, the resistor on that hot snubber got extremely hot, for whatever that is worth.

Rich

Re: Hot Snubber

redking56@...
 

LOL

Ed, you may stay up.  I should have mentioned it at the outset but, Yes, I am HO scale.

I do have a RRampMeter on my layout, and it shows a constant 14.7 to 15.4 volts.  It is wired in-line, though, so I do not use it portably to check for voltage spikes or drops.

Incidentally, I did replace the hot snubber yesterday, and the replacement runs a lot cooler, or maybe I should say, less hot.

Just now, I went down and inserted the hot snubber on my bus wires over by snubber #6 where that snubber is warm at best.  Very quickly, the resistor on that hot snubber got extremely hot, for whatever that is worth.

Rich


---In WiringForDCC@..., <eschwerkolt@...> wrote :

Rich,

That should work, now cut the bus loop. Replace the hot R/C
Filter. It should help.

Layman's explanation of spike. I think we are all assuming you are
in HO scale. If you are in Lionel O gauge 3 rail, then I am going back to bed.

On a perfect layout the constant voltage on the rail is around 14
volts. When a spike occurs the voltage can jump to 24 to 38 volts As
per Gartner.

This would fry decoders. The R/C Filters (snubbers) soak up the
excess. What can cause the spikes. Your 200 feet of wire holds a lot
of electricity, it is moving back and forth.

Now a big guess.... the wave of spikes are being absorbed by the #4
R/C filter.

Ed S


At 10:45 AM 3/15/2014, you wrote:
>
>
>OK, I probably have already effectively accomplished that because
>between #3 and #4 snubbers, the rails of my double mainline track
>are gapped for occupancy detection. Wouldn't that be sufficient?
>
>Rich

Re: Hot Snubber

Flash Gordon
 

At 11:08 AM 3/15/2014, you wrote:



This question may have gotten lost in all of these exchanges.

Does it matter how many feeders are between two snubbers? ******* no as long as there are enough, more is better

Does it matter what type of device is connected to a feeder besides rail, such as an NCE Switch-It or Mini-Panel or Circuitron Flasher, or Digitrax AR-1?******** I do not know.

Rich
********** Ed S

_

Re: Hot Snubber

Flash Gordon
 

Rich,

That should work, now cut the bus loop. Replace the hot R/C Filter. It should help.

Layman's explanation of spike. I think we are all assuming you are in HO scale. If you are in Lionel O gauge 3 rail, then I am going back to bed.

On a perfect layout the constant voltage on the rail is around 14 volts. When a spike occurs the voltage can jump to 24 to 38 volts As per Gartner.

This would fry decoders. The R/C Filters (snubbers) soak up the excess. What can cause the spikes. Your 200 feet of wire holds a lot of electricity, it is moving back and forth.

Now a big guess.... the wave of spikes are being absorbed by the #4 R/C filter.

Ed S

At 10:45 AM 3/15/2014, you wrote:


OK, I probably have already effectively accomplished that because between #3 and #4 snubbers, the rails of my double mainline track are gapped for occupancy detection. Wouldn't that be sufficient?

Rich

Re: Hot Snubber

redking56@...
 


This question may have gotten lost in all of these exchanges.

Does it matter how many feeders are between two snubbers?

Does it matter what type of device is connected to a feeder besides rail, such as an NCE Switch-It or Mini-Panel or Circuitron Flasher, or Digitrax AR-1?

Rich



Re: Hot Snubber

jazzmanlj
 

Rich,

I can't explain any better than I've tried! Perhaps someone else can.

Leave the snubbers where they are and if the resistor runs hotter than you'd like then go to a higher wattage! Simple as that! There is no harm in one resistor dissipating more power than another under normal operating conditions.

I've said about as much as I can on the subject for now.

 

Regards,

Len Jaskiewicz

 


 

Re: Hot Snubber

redking56@...
 

OK, I probably have already effectively accomplished that because between #3 and #4 snubbers, the rails of my double mainline track are gapped for occupancy detection.  Wouldn't that be sufficient?

Rich

Re: Hot Snubber

redking56@...
 

To my non-electrical engineer's mind, these comments are way over my head.

Are you saying that my failure to twist my bus wires are the cause of a hot snubber and, if so, why only one and why this one?

What on the layout, or what it is about my layout, other than non-twisted bus wires, could cause this to happen?

Rich


---In WiringForDCC@..., <len.jask@...> wrote :

Len, if I have a "trouble spot", can you speculate for me on what might produce a trouble spot?

Rich,
I'm trying my best to explain this to you and everyone else here! The DCC track power is a switching square wave with sharp rising and and falling edges. Parasite components such as inductance and capacitance due to wiring is inevitable! These cause excessive spiking of voltage at the rising and falling edges!
Twisting of wires or much wider separation, etc. is common practice to reduce the parasite components. It is very common practice to place snubbers(transient supressors) at a trouble spot to reduce the spiking.
I'd like all to at least do the Google search on 'snubber waveforms' as well as a Wiki search on 'snubber' versus 'R/C filter' so we all get a better understanding!
Learning and problem solving go hand in hand! Mere speculation leads to misconceptions.
Regards to all,
Len Jaskiewicz

Re: Hot Snubber

jazzmanlj
 

Len, if I have a "trouble spot", can you speculate for me on what might produce a trouble spot?

Rich,
I'm trying my best to explain this to you and everyone else here! The DCC track power is a switching square wave with sharp rising and and falling edges. Parasite components such as inductance and capacitance due to wiring is inevitable! These cause excessive spiking of voltage at the rising and falling edges!
Twisting of wires or much wider separation, etc. is common practice to reduce the parasite components. It is very common practice to place snubbers(transient supressors) at a trouble spot to reduce the spiking.
I'd like all to at least do the Google search on 'snubber waveforms' as well as a Wiki search on 'snubber' versus 'R/C filter' so we all get a better understanding!
Learning and problem solving go hand in hand! Mere speculation leads to misconceptions.
Regards to all,
Len Jaskiewicz

Re: Hot Snubber

Flash Gordon
 

Rich,

You really should read these two pages.... Part 1 and part 2

http://www.wiringfordcc.com/track.htm

Skip the sections about tail light bulbs.... your circuit breakers will replace them.

Ed

Re: Hot Snubber

Flash Gordon
 

Rich,

New drawing posted.

Feeders....  rule of thumb.  Every 6 to 10 feet.

Ed S


At 10:07 AM 3/15/2014, you wrote:
 

Ed, if I open the loop at that point, you say to gap both rails at that point.

Help me out with that a little.  Where exactly would I gap the rails?

I very well understand the concept of gapping rails, several reversing sections on my layout and some sections gapped for occupancy detection.  But why and where would I gap the rails when I open the bus wires?

Rich


---In WiringForDCC@..., wrote :

Rich,

I posted a solution in your photo folder. You can remove it if you do
not agree.

As others have said, it shows the loop as open. I can see no
advantage in having a closed loop so this solution will not change
how the layout operates. The track at that point also has to be
gapped, both rails. Just cut with Dremel cut off wheel. Later fill
the gap with plastic so they do not creep together.


Re: Hot Snubber

Flash Gordon
 

Wait I will change the drawing.

Ed S


At 10:07 AM 3/15/2014, you wrote:
 

Ed, if I open the loop at that point, you say to gap both rails at that point.

Help me out with that a little.  Where exactly would I gap the rails?

I very well understand the concept of gapping rails, several reversing sections on my layout and some sections gapped for occupancy detection.  But why and where would I gap the rails when I open the bus wires?

Rich


---In WiringForDCC@..., wrote :

Rich,

I posted a solution in your photo folder. You can remove it if you do
not agree.

As others have said, it shows the loop as open. I can see no
advantage in having a closed loop so this solution will not change
how the layout operates. The track at that point also has to be
gapped, both rails. Just cut with Dremel cut off wheel. Later fill
the gap with plastic so they do not creep together.


Re: Hot Snubber

redking56@...
 

Ed, if I open the loop at that point, you say to gap both rails at that point.

Help me out with that a little.  Where exactly would I gap the rails?

I very well understand the concept of gapping rails, several reversing sections on my layout and some sections gapped for occupancy detection.  But why and where would I gap the rails when I open the bus wires?

Rich


---In WiringForDCC@..., <eschwerkolt@...> wrote :

Rich,

I posted a solution in your photo folder. You can remove it if you do
not agree.

As others have said, it shows the loop as open. I can see no
advantage in having a closed loop so this solution will not change
how the layout operates. The track at that point also has to be
gapped, both rails. Just cut with Dremel cut off wheel. Later fill
the gap with plastic so they do not creep together.


Re: Hot Snubber

jazzmanlj
 

>So effectively, that resistor is dissipating 0.31 watts normally and so you need at least a 1/2 watt resistor?
That is correct and thanks to Don for such a good explanation.
Might I add that it is common practice that components are generally de-rated in use so as not to run them at or near the max. If a 50% de-rating is used then a 1W resistor would be desired.

 

Len Jaskiewicz


Re: Hot Snubber

redking56@...
 

Len, if I have a "trouble spot", can you speculate for me on what might produce a trouble spot?

Rich

Re: Hot Snubber

Flash Gordon
 

Rich,

I posted a solution in your photo folder. You can remove it if you do not agree.

As others have said, it shows the loop as open. I can see no advantage in having a closed loop so this solution will not change how the layout operates. The track at that point also has to be gapped, both rails. Just cut with Dremel cut off wheel. Later fill the gap with plastic so they do not creep together.

Some more back ground.

If you read any of the instruction for DCC you will see that bus longer than 30 feet can cause problems. The problem would be spikes in the signal along the bus. These spikes can harm decoders.

If the wires are twisted, the problem will be reduced, since yours are not I am thinking that is the problem.

In layman's terms, the snubbers (R/C Filters) act as sponges and soak up the spikes.

Also if possible, tie each of the two wires in a bus together with wire ties or tape. The closer the pairs are the less likely to cause the spike problem.

As to the hot R/C filter, it may be damaged and I would replace it to eliminate that variable.

Ed S