Lighting Trains

[Sheffie]

I think your first idea was correct, Jeff. leave the capacitor wired across the output of the rectifier, so it gets the full voltage. If it's wired after the resistor, it will be charging more slowly and discharging more quickly. If it helps.... think of it as part of the power supply (rails and rectifier), rather than the lighting (resistor and LEDs).

Edited February 26, 2019 by Sheffie

[cteno4]

Ok reading a little more on caps good the machete to a few of the cobwebs! The cap will still charge to the output voltage of rectifier either before or after the dimmer rectifier. It will just charge slower on the inside of the dimmer rectifier (which you don’t want actually, you want it to recharge fast to cover the the next flicker). 

 

On discharge the cap will be at the same voltage in either location so my logic was correct that if the cap is on the inside of the dimmer resistor (next to the leds) it will result in a larger current to the leds and thus brighten, but if the cap is located outside the the dimmer resistor (between the dimmer resistor and the rectifier) it should result in the leds getting the same current off the cap discharge as from the track power thru the rectifier and thus not change brightness.

 

my reading did bring up one little forgotten bit in that to be safe with caps it’s usually best to slow the inrush current on charge a tiny bit with a very low value resistor so you don’t get all your caps sucking up a lot of current at once. With these little caps and only firing randomly now and then it’s probably not an issue though. Still a little foggy on how much of it’s needed in this instant.

 

thanks for validating this sheffie, your post just came in as I was writing this!

 

jeff

[gavino200]

30 minutes ago, cteno4 said:

It might also mean you could use 16v caps. Caps are strange beasts at times in the logic.

 

If that's the case it would be very helpful. It would be easier to get physically smaller caps at a lower voltage.

 

30 minutes ago, cteno4 said:

 

My main electronics education was age 5-20, that’s a long time ago for me and the cobwebs get in the way at times.

 

but putting the cap between the rectifier the dimmer resistor I know it should perform correctly, just not sure the alternative will work exactly the same.

 

jeff

 

I could put this to the test with a few simple experiments.

[gavino200]

23 minutes ago, cteno4 said:

Ok reading a little more on caps good the machete to a few of the cobwebs! The cap will still charge to the output voltage of rectifier either before or after the dimmer rectifier. It will just charge slower on the inside of the dimmer rectifier (which you don’t want actually, you want it to recharge fast to cover the the next flicker). 

 

On discharge the cap will be at the same voltage in either location so my logic was correct that if the cap is on the inside of the dimmer resistor (next to the leds) it will result in a larger current to the leds and thus brighten, but if the cap is located outside the the dimmer resistor (between the dimmer resistor and the rectifier) it should result in the leds getting the same current off the cap discharge as from the track power thru the rectifier and thus not change brightness.

 

I follow the theory. However, on the train I've already done, the cap placed after the reistor. There's no increase in brightness. 

 

Placing the cap after the resistor would be preferable as that would make a design possible were two caps are simply soldered to the middle and far end solder pads. the rectifier would be soldered to the near set of solder pads. This would be much more compact and a LOT faster to construct. 

 

I can run a test of all of these designs to see if there's any difference. 

 

Most important would be to work out if a lower Voltage capacitor would be safe if the cap was placed after the resistor, as it would allow for use of much smaller caps.

 

Quote

 

my reading did bring up one little forgotten bit in that to be safe with caps it’s usually best to slow the inrush current on charge a tiny bit with a very low value resistor so you don’t get all your caps sucking up a lot of current at once. With these little caps and only firing randomly now and then it’s probably not an issue though. Still a little foggy on how much of it’s needed in this instant.

 

I'm sure this is correct, but there's no way I'm doing this.

Edited February 26, 2019 by gavino200

[cteno4]

Sorry the caps need to be 25v.

 

i doubt you would notice the brightness increase, but I’m pretty sure it will be discharging your cap faster and thus not as good flicker control.

 

cant you just put the cap across the rectifier output and then dimmer resistor off one side of the cap that goes to one led lead and other led to other side? You could even fold the cap under the rectifier to make a shorter block wire leads. Also formthe leads from the chassis connection to the rectifier you can use very fine wire 30g or smaller is fine as you are only using about 3ma total for your strip. Also another lead that works well is 1/8w resistor leads. They are fine and flexible and solder easily. You can buy spools of uninsulated buss wire like this as well but it can be expensive. I have a load of 30awg buss wire if you want some.

 

Jeff

[chadbag]

5 minutes ago, cteno4 said:

Sorry the caps need to be 25v.

 

 

Why is this?  As long as the voltage feeding the cap does not go above the cap voltage spec you should be ok, right?  (Having some extra headroom of course is great -- but 16V gives you that over 12V ?

 

I have some 16V caps and will be putting a zener diode in  to make sure we don't over push the voltage 

[gavino200]

47 minutes ago, cteno4 said:

Sorry the caps need to be 25v.

 

Even if the resistor is before the cap?

 

Quote

 

i doubt you would notice the brightness increase

 

 Nope. Consider this proven. 

 

 

Quote

 

, but I’m pretty sure it will be discharging your cap faster and thus not as good flicker control.

 

I agree. This is logical. I could put it to a practical test to see if it's important. At the moment, with a 2.2k Ohm resistor placed before a 220uF cap there's zero flicker.

 

Quote

 

cant you just put the cap across the rectifier output and then dimmer resistor off one side of the cap that goes to one led lead and other led to other side? You could even fold the cap under the rectifier to make a shorter block wire leads.

 

Actually I'd need to buy larger rectifiers. Mine are tiny and don't line up with the capacitor solder pads. Need a very short wire on one side. Not a big deal, but an extra step that takes time.

 

But the main problem with the capacitors is their HEIGHT. There isn't enough room for them between the carriage roof and the raised interior. Walls, dividers etc. It's actually quite difficult to place them so that they fit AND aren't visible through the windows.

 

Also, if the caps can be placed after the resistor then a pair of smaller caps can be placed on the remaining free solder pads, while the rectifier can be placed at the pick up end. That would be an extremely quick setup to make. Like about 10 minutes.

 

Quote

 

Also formthe leads from the chassis connection to the rectifier you can use very fine wire 30g or smaller is fine as you are only using about 3ma total for your strip. Also another lead that works well is 1/8w resistor leads. They are fine and flexible and solder easily. You can buy spools of uninsulated buss wire like this as well but it can be expensive. I have a load of 30awg buss wire if you want some.

 

Yes. I have tons of 30g wire. That's what I've been using. I use resistor legs if I want more rigidity. But for the most part flexibility is more important. 

 

 

Edited February 26, 2019 by gavino200

[cteno4]

Nope the rule of thumb is to double the input voltage for the cap voltage. When you charge a cap close to the rated value it’s when it can no pop!

 

tantalium caps with their high charge density make them more sensitive to potential failure with higher voltages so they usually reccomeded to devalue even more, down to a third even. So really 35v would be playing it really safe.

 

jeff

[gavino200]

12 minutes ago, cteno4 said:

Nope the rule of thumb is to double the input voltage for the cap voltage. When you charge a cap close to the rated value it’s when it can no pop!

 

Agreed. However space is a premium here. How close is close?

 

 

Quote

 

tantalium caps with their high charge density make them more sensitive to potential failure with higher voltages so they usually reccomeded to devalue even more, down to a third even. So really 35v would be playing it really safe.

 

Quote

 

 

I'd have to check again, but I believe at with a lower voltage like 16V we can use ceramic caps, no problem.

 

BTW, after a 2.2k Ohm resistor the voltage must be pretty low, right?

Edited February 26, 2019 by gavino200

[gavino200]

16 minutes ago, chadbag said:

 

 

(Having some extra headroom of course is great

 

Amen to that. Wait till you try this. You'll see just how great that would be.

[gavino200]

A six-pack of the beer of your choice to the person who can reliably theorize with subsequent experimental proof, what the lowest capacitance that can be used in this set up is, either before or after a 2.2k Ohm resistor. (provided that the answer is lower than the rule of thumb value).

 

I would but would rather not use use zener diodes. I'm trying to make a simple quick and cheapish design. 

 

What size are zener diodes anyway? And what do they cost?

Edited February 26, 2019 by gavino200

[gavino200]

Another question. My knowledge of rectifiers is extremely limited. These are the rectifiers I bought. I just copied what Dani listed. However mine are definitely smaller than his. This is an advantage or disadvantage depending on the specific install. 

 

A prize - but this time just one beer - to anyone who can explain the important parameters at play here when choosing rectifiers. What do all the numbers mean? What do I need? What's the size code? How can I shop knowledgeably for these things.

 

 https://www.ebay.com/itm/100pcs-New-MB6S-B6S-0-5A-600V-Rectifier-Bridge-Rectifier-SMD/152781676348?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

[cteno4]

21 minutes ago, gavino200 said:

Even if the resistor is before the cap?

 

Yes. These were the neurons I dusted off the resistor / cap interaction only effect charging time of the cap, voltage is not effected, so you will be charging the cap to 12v If placed on either side of the dimmer resistor. With the dimmer resistor between the cap and the leds will make it slow the cap discharge. This may mean you could use a lower value cap.

 

ceramic caps actually need to be derated more than electrolytic caps. Some say down to a third the rated value. 16v ceramic caps with 12v fee is not reccomeded. The denser the charge the bigger the boom when they go pop! Tantaliums also blow shorted usually not open like electrolytics so it will short your dcc if they do blow.

 

Jeff

 

[Sheffie]

I can answer a little of that.

MB6S and B6S are almost certainly manufacturer specific codes. They might tell you something about the materials or package size or other spec.

0.5A is the rated current for the component. If the items downstream of it draw more than this, it can overheat. I'm not sure how permanent that would be.

600V is the breakdown voltage. I'm pretty sure that going over this is a one time occurrence.

The terminals are nicely labeled, at least. + and - are clear enough, and the other two are for the AC input.

[gavino200]

2 minutes ago, cteno4 said:

 

Yes. These were the neurons I dusted off the resistor / cap interaction only effect charging time of the cap, voltage is not effected, so you will be charging the cap to 12v If placed on either side of the dimmer resistor. With the dimmer resistor between the cap and the leds will make it slow the cap discharge. This may mean you could use a lower value cap.

 

 

Ok. I'll buy that.

 

Related question. Can I place my caps at the two empty solder pad locations. If so What resistors will I need to wire in series at these points? The small tant 100uF "DIP" caps could work in this position. They're on legs. I could cut one leg shorter and solder in a resistor. Not much extra trouble but with a lot of benefit i terms of getting all the components "onto the strip" so to speak.

[cteno4]

4 minutes ago, gavino200 said:

Another question. My knowledge of rectifiers is extremely limited. These are the rectifiers I bought. I just copied what Dani listed. However mine are definitely smaller than his. This is an advantage or disadvantage depending on the specific install. 

 

A prize - but this time just one beer - to anyone who can explain the important parameters at play here when choosing rectifiers. What do all the numbers mean? What do I need? What's the size code? How can I shop knowledgeably for these things.

 

 https://www.ebay.com/itm/100pcs-New-MB6S-B6S-0-5A-600V-Rectifier-Bridge-Rectifier-SMD/152781676348?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

 

Gavin, 

 

this stuff is easily designed out in basic circuit design. Usually you build it to be robust as possible and as efficient as possible, then try to get elegant to reduce the components and pathways. 

 

Each of your 3 led strip modules are running at about 1.4ma so like 1/12th the rated max in your circuit. So your leads can be tiny as this is a very small amount of current. 

 

on the rectifier it’s rated to 500ma at 600v, way past your 3ma at 12v!

 

jeff

[gavino200]

6 minutes ago, Sheffie said:

I can answer a little of that.

MB6S and B6S are almost certainly manufacturer specific codes.

 

Or size codes maybe? I've noticed that all these smd components have standard metric size codes? What makes you almost certain?

 

6 minutes ago, Sheffie said:

 

They might tell you something about the materials or package size or other spec.

0.5A is the rated current for the component. If the items downstream of it draw more than this, it can overheat. I'm not sure how permanent that would be.

600V is the breakdown voltage. I'm pretty sure that going over this is a one time occurrence.

 

Are these fairly standard values? is this sort of a one size fits all small electronics? Do i need this much slack? 600 V???

 

6 minutes ago, Sheffie said:

The terminals are nicely labeled, at least. + and - are clear enough, and the other two are for the AC input.

 

I had to work this out the hard way.

[gavino200]

9 minutes ago, cteno4 said:

 

Gavin, 

 

this stuff is easily designed out in basic circuit design. Usually you build it to be robust as possible and as efficient as possible, then try to get elegant to reduce the components and pathways. 

 

Each of your 3 led strip modules are running at about 1.4ma so like 1/12th the rated max in your circuit. So your leads can be tiny as this is a very small amount of current. 

 

on the rectifier it’s rated to 500ma at 600v, way past your 3ma at 12v!

 

 jeff

 

 

Excellent. 

 

So now mainly I'm concerned about the physical size of the rectifier. I'd like to have a couple of widths. My tiny ones are great for tucking in next to the pickup strips. 

 

But what would also be usefull would be a rectifier that I could solder directly to the capacitor. Mine is too small to do this. I need to link it with wire. An extra  step.

 

I AM being nit picky. But if I get this to work well, I'll probably end up repeating it about a thousand times. So extra steps are important.

Edited February 26, 2019 by gavino200

[chadbag]

1 minute ago, gavino200 said:

 

Or size codes maybe? I've noticed that all these smd components have standard metric size codes? What makes you almost certain?

 

The company who makes the MB6S makes other MB6x caps (I did some research).  I bought some MB6F which are the same basic rating but not as tall (I want to create "blue" wires from decoders and wanted a not as tall unit).  I believe, form what I saw, that they are manufacturer specific.  Though there may be manufacturers who make compatible parts that call them similar or the same.  Especially from China.

 

1 minute ago, gavino200 said:

 

Are these fairly standard values? is this sort of a one size fits all small electronics? Do i need this much slack? 600 V???

 

A few days ago I looked up rectifiers from a large electronics house online catalog.  There are probably all sorts of ones we could use with lower values, etc.  I think Dani chose the ones we got (I also got my MB6S today) as they are very inexpensive and common and easy to buy.

 

1 minute ago, gavino200 said:

 

 

I had to work this out the hard way.

 

My MB6S have the + - marked.

[Sheffie]

As for the code, I'm sorry, I wasn't really certain. That was some sloppy language on my part.

 

Rectifier values, I'm not sure about the output. Obviously it's not a lot of current. It's enough to run a couple of LEDs and presumably to charge the capacitor too. (Putting the capacitor "after" the load resistor is going to help keep the total current draw down compared to putting it right across the rectifier output.)

 

Voltage. My one real hands-on experience with a rectifier was building a bass guitar amplifier, and we spent a little more to get something with an 800V rating. This was only providing about 9V, but my friend who was seriously into guitars and electronics recommended a high safety margin given the potential heavy loading involved (and the risk to the loudspeakers, which cost a bit more than LEDs and even N gauge coaches). I would say that in this context 600V qualifies as "nice and safe". I wouldn't worry about a higher rating, and I wouldn't try too hard to find a lower one unless you desperately need something smaller.

Edited February 26, 2019 by Sheffie

[gavino200]

3 minutes ago, chadbag said:

think Dani chose the ones we got (I also got my MB6S today) as they are very inexpensive and common and easy to buy.

 

That's what I thought. However his are a good bit wider than mine. How wide are your (in relation to the width of the LED strip? 

 

3 minutes ago, chadbag said:

 

My MB6S have the + - marked.

 

Mine too. But not which sides are input and which are output.

[chadbag]

Just now, gavino200 said:

 

That's what I thought. However his are a good bit wider than mine. How wide are your (in relation to the width of the LED strip? 

 

I'll take some pics next to a ruler as I may have different LED strips than you

 

Just now, gavino200 said:

 

Mine too. But not which sides are input and which are output.

 

The + - is always the output as the input is AC and does not have a + or -

[gavino200]

1 minute ago, chadbag said:

 

The + - is always the output as the input is AC and does not have a + or -

 

Duh!

 

(gavino200 face-palms at his own idiocy)

[gavino200]

So how about this gentlemen:

 

Imagine an LED strip. It's really two joined units torn from a roll. (see pictures on previous page). There are 6 LEDs. This strip has three solder points - one in the middle and one at each end.

 

How about I wire it like this. 

 

rectifier soldered to one end with a dimming 2.2k resistor after the rectifier.

 

A 100 uF cap soldered at each of the remaining solder points (middle and far end) 

 

Each cap has a resistor. it's soldered to one of the "legs" of a DIP tantulum cap (looks like a blob on two legs and is very small)

 

Will this work? An what resistance would each of the cap-leg resistors need to have?

 

 

I believe this would be a very simple to make and versatile design.

[Kiha66]

Little late to the party, but that sounds pretty good.  The resistor on the cap isn't totally necessary as the resistor to the lightboard will already limit inrush current.  Imo adding these extra resistors wouldn't really add much for the time, money (and internal space!) they would cost.  Adding a 100uF cap to each point should work well, and will give you a total of 200uF (caps in parallel are additive!).