SouthPenn riogrande5761 Pulls up lawn chair and popcorn. This should be interesting. And don't forget the fire extinguisher.
riogrande5761 Pulls up lawn chair and popcorn. This should be interesting.
Pulls up lawn chair and popcorn. This should be interesting.
And don't forget the fire extinguisher.
But then we have to bring our own heat source for the popcorn.
LAX1DUDE I just bought an old oscilloscope at a garage sale, and one of my first projects was to look at DCC through it. One thing I noticed is that there is litterally no transition between + and - when the wave moved, which gave me an idea. Considering I'm just getting started with DCC, I only have a Digitrax Zephyr to control my trains with. This works great with N scale, but it might be a problem when I start to use it with some of the S gauge trains I have been converting. 3 amps was nothing when someone first designed those things in the 1950s, so power consumption would definatly be an issue if I want to use more than 1 at a time What I saw through the oscilloscope gave me an idea: What if I bought a large 12 volt DC transformer and rigged it up with transistors, so its like an amplified version of what comes out of the DCS51? I definatly know how diodes and transistors work, and how to set them up, but what I'm not sure about is how big of a possibility there is for it to malfunction and destroy a DCC decoder. All I can say is, the Soundtraxx decoder from my other post wasn't actually the first one I bought, and I don't want to replace it AGAIN. Look, that happened because of a stray wire touching the track, but it still blew up the decoder. I didn't expect it to be so delicate, and I don't want to blow up ANOTHER because of a similar risky project. Tell me what you think and what could be possible issues with this idea. Oh, and if anybody knows exactly where I am supposed to buy a transistor or diode of that size, send me a link... Sorry Digitrax, but if I succeed, you may notice a few differences in the number of customers buying your boosters...
I just bought an old oscilloscope at a garage sale, and one of my first projects was to look at DCC through it. One thing I noticed is that there is litterally no transition between + and - when the wave moved, which gave me an idea.
Considering I'm just getting started with DCC, I only have a Digitrax Zephyr to control my trains with. This works great with N scale, but it might be a problem when I start to use it with some of the S gauge trains I have been converting. 3 amps was nothing when someone first designed those things in the 1950s, so power consumption would definatly be an issue if I want to use more than 1 at a time
What I saw through the oscilloscope gave me an idea: What if I bought a large 12 volt DC transformer and rigged it up with transistors, so its like an amplified version of what comes out of the DCS51? I definatly know how diodes and transistors work, and how to set them up, but what I'm not sure about is how big of a possibility there is for it to malfunction and destroy a DCC decoder. All I can say is, the Soundtraxx decoder from my other post wasn't actually the first one I bought, and I don't want to replace it AGAIN. Look, that happened because of a stray wire touching the track, but it still blew up the decoder. I didn't expect it to be so delicate, and I don't want to blow up ANOTHER because of a similar risky project.
Tell me what you think and what could be possible issues with this idea. Oh, and if anybody knows exactly where I am supposed to buy a transistor or diode of that size, send me a link...
Sorry Digitrax, but if I succeed, you may notice a few differences in the number of customers buying your boosters...
If you are bound and determined, there are plenty of schematics on the internet if you want to build your own booster.
Most locomotives today would probably draw about 500mA worst case. So a huge power supply to supply enough current to weld with isn't needed. Along with proper wiring, a 5A booster should be adequate, and if you need more, dividing the layout into power districts with their own boosters would solve that. You could probably run 10 locomotives on a 5A booster with no issues.
Trying to compensate for poor or inadequate wiring with a big power source is just asking for trouble.
Well, he is using S scale, so probably more than 500ma per loco, but since they are remotored and not the original Flyer universal motors, they shouldn't be 4 amps each, either. Though there was that one Atlas O diesel with dual can motors than drew over 8 amps.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
rrinker The correct solution is to bypass the rail joints as the power carrying part of the circuit, either by sldering or installing jumper wires across each gap
That was already part of the plan. I just don't want to solder everything and then find that my layout is crap. I want to do a mix so I don't end up with a loss like that...
I think a plan is starting to form
Thanks for all the help!
I think what I'm going to do is a mix of everyone's sugestions: First, I will set up a 15 amp booster connected to my Zephyr with an H bridge or similar circuitry. Then, I'm going to make sections of around 6 track pieces that are soldered. Each section will have it's own feeder wires. I'll find some large 15 amp 12 volt (180 watt) breakers for safety, but I don't think I need some lightning fast single use fuse because the power is conducted through an entire truck, not a bunch of indivisual wheeles connected with paper-thin wire. The types of shorts everyone is worried about would have to melt the entire truck...wait...scratch that. I'll find really fast fuses because I just noticed this might weld the axel or wheel to the truck, which would not be good, especially considering that some of my locomotives would have to be destroyed if I needed to replace something like that... I'm aware of the amp problem, okay? If someone wants to talk about it, send me a link to a fuse or don't say anything at all, unless you're sure its something I haven't already been told about or know. Finally, I'll connect the sections together, have one last look through the oscilloscope, and pray the locomotive won't pop and smoke when I turn it on.
Thanks everyone for all the help!
If you think I should do anything differently, leave a comment!
PS. Any comments but "Seriously Dude? OMG Don't do it! You are going to ruin everything! Don't tell us we didn't warn you!"... I'm not buying 3 DB200s
PPS. I'm just brainstorming, I'm probably not going to make anything real for the next 6 months. This is an ongoing disscusion
Maybe a fast trip breaker would make things easier.
I didn't see a mention of changing motors, sorry. And in this day and age I suggest you might be able to get a reasonably powerful motor at a reasonable price that draws WAY less than 3A.
Disclaimer: This post may contain humor, sarcasm, and/or flatulence.
Michael Mornard
Bringing the North Woods to South Dakota!
Yes, you want an electronic circuit breaker, NOT single use fuses, unless you plan to buy stock in a fuse company.
And the whole point of doing this is to use breakers rated at LESS than the booster capacity, to divide the layout into sections such that if a loco derails in one area and shorts out, it doesn;t shut off the entire layout.
You will need very heavy wire, #12 at least, for the power bus. Even #12 will exhibit noticeable voltage drop over relatively short lengths at 10 and 15 amp current levels. Again, the whole point of NOT having one mosnter booster is to prevent the voltage drop. There is a low power signal (in the Loconet cable, it's the Railsync lines) that are used by each booster. That signal is what is used to drive the Booster's output to replicate the same DCC waveform at full power for the rails. Even if there is some reduction in amplitude of the Railsync signal over a large distance, as long as it exceeds the threshold of the input device on the booster, the booster will be able to generate a full power DCC signal to match. Thus even the furthest point on the layout gets full power without strining long lengths of expensive heavy gauge wire. At your current levels - maybe 10 feet of #12 in each direction from a boostr, so 20 feet of railroad, per booster with 5 amp boosters. Minimum voltage drop with a maximum of 20 feet for a complete circuit.
This is one reason they don't make 15 or 20 amp boosters. Another is the false economy - DCC can be inexpesinve because many of the components are used in other industries where they make more of them than have ever been used in the model railroad hobby. One such item is the power supply. Keep in mind, you will need MORE than a 12V input if you expect to get 12V tot he rail, and it's more typical to have a rail voltage of 14-15V for HO "12V" motors, O and G scale commonly use 18V on the rails. To get that kind of voltage out, the input must be higher, to compensate for losses in the circuit. It so happens that 15-19V power supplies capable of 5 amps are very common - for laptops. 15V at 8 amps? Let alone 15V at 15 or 20 amps? Not so much. Oh, they make them, but they are generally expensive because they aren't commonly used anywhere.
Also kee in mind that for this to work, the input power has to be well filtered. You end up feeding the H bridge with DC, and the driver circuit generates the square wave DCC output. But any ripple or noise on that DC input will appear in the output and distort the DCC signal. Clean, smooth power is important, take a look at any of the schematics for DIY boosters, you will see plenty of filter caps adn bypass caps to ensure power stability. This is NOT a trivial circuit.
I'm not trying to discourage you from trying to build your own booster (I'm building a lot of DIY things for my layout), but rather than try to design something from scratch, consult one of the many published schematics. You'll notice they are all very similar - there's a reason for this. Some of your responses lead me to believe you don't fully understand the electronic principles involved and doing this wrong could end up generating an arc welder circuit rather than a DCC booster. After going through the effort to remotor those Flyer trains, I don't think you want to blow them up or melt them down.
even though you may use a single booster doesn't mean each section can draw the maximum power from the booster.
you can build your layout to support some total number of operating locomotives, the total of which require P watts. But you can divide your layout into sections within which you only expect to operate a few locomotive. Each section would require something much less than P watts.
Each section would only need a circuit breaker for the max power in that section, perhaps only 5 amps. probably cheaper and safer
greg - Philadelphia & Reading / Reading
Safer, most definitely. There was a limit placed on the power output of "toys" in terms of total wattage - the old familair Lionel ZW put out 250 watts - the new version of it is not as high because 250 total watts, even at low voltage, was considered to be unsafe. Pretty sure that's why CVP discontinued their "Big Boost" 10 amp booster, although NCE still sells theirs. But as I said, the apparent economy of scale is not really - in this eexample, using MSRP, the 10 amp NCE booster is $110 more than the 5 amp booster, significantly less than double the price - so doubt the power for a 68% cost. However, the 5 amp power supply is $60, the 10 amp supply is $190. The total cost for 10 amps is $460, the total cost for 5 amps is $220. So you can buy 2x 5 amp for less than 1x 10 amp!
If you can source parts, you MIGHT be able to build 2x 15 amp booster with power supplies for less than 6x 5 amp boosters and power supplies - MIGHT. But start going much higher and the cost of the MOSFETS for teh H bridge start to skyrocket. And the greater the differential between the load being switched and the driver side fo a MOSFET, the more complex the driving circuit has to be - too great a differential and MOSFETs tend to not fully turn on, or turn on slowly. The former results in a lot of power wasted as heat in the MOSFET, the latter results in poor DCC waveforms. This is fixed with intermediate driver stages with regular transistors such that the one actually conencted to the MOSFET are handling power levels closer to that of the MOSFET, allowing a quick and complete turn on.
This is not all that different from an audio amplifier, except it deals with square waves and not sine waves. In fact there was one DIY booster years ago that used one of the integrated audio amp chips as the driver - I still have one of the chips, I was going to build it but never did. But even a MOSFET audio amplifier does much the same thing, there is an intermediate stage or two before driving the MOSFETs which actually switch the signal going to the speakers. The stereo I had in college blew a first or second stage driver transistor at least twice - on low volumes, all the power was dissipated as heat and the thing could fry an egg. Cranked all the way to full power, it would barely get warm, even after hours of play.
This is also why I recommend looking at schematics of existing designs - a DCC booster looks simple in block diagrams but there is pretty much to it. Even a 5 amp one, at 15 volts is 75 watts, that's not an insignificant power level.
Good point, 15 or 20 amps through a MOSFET is a LOT!!!
rrinker You will need very heavy wire, #12 at least, for the power bus. Even #12 will exhibit noticeable voltage drop over relatively short lengths at 10 and 15 amp current levels. Again, the whole point of NOT having one mosnter booster is to prevent the voltage drop. There is a low power signal (in the Loconet cable, it's the Railsync lines) that are used by each booster. That signal is what is used to drive the Booster's output to replicate the same DCC waveform at full power for the rails. Even if there is some reduction in amplitude of the Railsync signal over a large distance, as long as it exceeds the threshold of the input device on the booster, the booster will be able to generate a full power DCC signal to match. Thus even the furthest point on the layout gets full power without strining long lengths of expensive heavy gauge wire. At your current levels - maybe 10 feet of #12 in each direction from a boostr, so 20 feet of railroad, per booster with 5 amp boosters. Minimum voltage drop with a maximum of 20 feet for a complete circuit.
Good point. I'm starting to see some of the advantages to the traditional system. I still want to try something like my original thought to see if it ends up cheaper, but many of the posts say that it is the wrong problem to solve. I could still try making multiple small boosters that are 8 amps each, but I would have to do some research. I still haven't found a solid price for the MOSFETs, and I don't want to end up with a system that's more expensive than buying some DB200s because it would defeat the purpose. Someone said the system would pull about 60 amps a few milliseconds before the breaker trips. Would the MOSFETS have to be rated at 60 amps or 15 (or 8) amps? How big would the filtering capacitors need to be? How expensive will the thick wire be? I would probably use landscape lighting wire because it comes in groups of two wires already, but the companies who make that stuff always get away with making it expensive because it's 'special'...
If anyone can give me some pointers to what types of capacitors to use or where to buy 60 amp MOSFETs, leave a comment. I'm really starting to wonder if there's a reason people don't make boosters like this. I don't mean "I wonder why nobody makes 30 amp boosters", I mean "I've only been messing with DCC for 2 months, and I'm already having ideas like this. I'm not even old enough to have been around in a time where you needed knowledge of this field of electronics if you wanted to build things! I can't be the first to try something like this! There has to be a reason building your own boosters this way isn't 'the way' if you want to make a large layout".
Is it just me, or am I having 'the great awakening' everyone is anticipating?
LAX1DUDEIf anyone can give me some pointers to what types of capacitors to use or where to buy 60 amp MOSFETs, leave a comment.
I would start with Mouser or Digikey.
My suggestion: Finish the re-motoring project. See what your actual current draw of locomotives ends up being, then move to making or buying a booster. That way you dont end up spending a bunch of money building or buying extra stuff you dont need. Northwest short line is a good place to find just about any type of motor you might imagine. They even have truck mounted motors that are DCC ready. Im guessing you already have your motors.
LAX1DUDE I mean "I've only been messing with DCC for 2 months, and I'm already having ideas like this. I'm not even old enough to have been around in a time where you needed knowledge of this field of electronics if you wanted to build things! I can't be the first to try something like this!
You are not the first. And you will likely not be the last.
LAX1DUDEThere has to be a reason building your own boosters this way isn't 'the way' if you want to make a large layout".
I'll put it this way. When I first got into DCC 3 years ago, I had a very similar thought process (bigger is better). I went to my Local Hobby Shop to buy a NCE 10amp system. The owner would not sell me one because I was going to use it for HO scale. He said I might one day need an extra 5 amp booster, but a single 5 amps would do just fine and keep me from cooking locomotives.
I suspect that it might also be a term called liability. If you buy an off the shelf system and use it in accordance with the manufacturers instructions, you are typically not liable if damage occurs to your property or others property. If you build something yourself, or a company makes a product that is inherently dangerous, you or that company could be held liable for the results.
What are those results? Maybe nothing, you could have years of trouble free operation. Worst case scenario? You could start a fire. Notice I did not say will.
LAX1DUDEIs it just me, or am I having 'the great awakening' everyone is anticipating?
I dont personally think so. There are several DCC manufacturers that make DCC systems with maximum voltages and amperages set for your safety.
MTH makes a (slightly more than) 400 Watt transformer for O scale. Its on board circuit breakers are rated @ 12A. It has two of them. 0-22VAC. This is a UL listed product.
Reducing the cost of distributing DCC power by spreading boosters and power supplies around is the reason we developed a new electrical standard for NTRAK wiring. Our figures at the time (2005) put the cost of a commercially available 5 amp booster, circuit breakers, and power supply at $250 per power district. Simply increasing the size of the bus from 16 ga to 12 ga and adding feeders doubled the length of a power district. That dropped cost for boosters significantly. We certainly don't need tons of amps to run N scale trains but found that the voltage drop caused by less than robust wiring and track work led to loss of voltage and signal.
A part of our standard that still tends to be missed by many layout builders is the track feeder spec. We included a requirement for feeders ever two feet of track OR feeders on every section of rail that is not soldered to a section of rail that has a feeder. Every two feet is overkill but helps to alleviate differences in skill level between mutiple builders. I see that second part as most important. It could be added to not rely on rail joiners, period. I have crossover sections that have up to six pairs of feeders in only a foot or so of track work. That because I don't solder my switches to the adjacent rails.
Don't skimp on the wiring or track work. Make it as reliable as possible from get go. Much easier then fixing it later. The extra effort on this portiion will save you some expense down the line.
BTW, I purchase our club's 12 ga Lanscape lighting cable at Home Depot in 500 foot spools. Cut price is much higher- nearly double. There are also premeasured packages in their stores of 100 and 200 foot hanks. Still much cheaper than cut. Have not found it cheaper at any other source including several electric supply houses that I have commercial accounts with.
Martin Myers
Indeed, the feeders are every bit as important as the bus. I tend to alternate soldered and unsoldered joints, so I'll have 2 pieces of flex soldered together, then another 2 pieces soldered together, adn the joint between the two sections is not soldered. However, I put feeders to every piece. So if I had, say a 12 foot straight line layout, 4 pieces of HO flex, there would be 3 joints. 2 soldered. One not soldered. and FOUR sets of feeders. Overkill? Maybe, but I also carry this over to turnouts (I was using Atlas) where every turnout had 3 sets of feeders. And I NEVR had stalling problems even though the frogs were not powered.
Likewise the cost of the bus wire. 2 layouts ago, I had an 8x12 donut. The small spools of wire were like 20 feet, and I certainly needed more than 20 feet of each color. The 100 foot spools were less than the cost of 2x 20 foot spools, so even though I didn't initially need 100 feet, I bought the 100 foot rolls. I did have plans to expand that layout that never happened, but the 100 feet would have all be used and then some.
It's definitely easaier to get the wiring in place and done correctly prior to covering anything with scenery and THEN finding you need to rip something out to fix a wiring issue. There is a downside - once the wiring is solid, the layout is operational and at least in my case, I keep putting off the scenery.
BMMECNYC I'll put it this way. When I first got into DCC 3 years ago, I had a very similar thought process (bigger is better). I went to my Local Hobby Shop to buy a NCE 10amp system. The owner would not sell me one because I was going to use it for HO scale. He said I might one day need an extra 5 amp booster, but a single 5 amps would do just fine and keep me from cooking locomotives. I suspect that it might also be a term called liability. If you buy an off the shelf system and use it in accordance with the manufacturers instructions, you are typically not liable if damage occurs to your property or others property. If you build something yourself, or a company makes a product that is inherently dangerous, you or that company could be held liable for the results. What are those results? Maybe nothing, you could have years of trouble free operation. Worst case scenario? You could start a fire. Notice I did not say will.
I meant to comment on this - you have a good hooby shop. There was a guy came in to my LHS once, all confused about DCC. He had gone to a different shop and they sold him an 8 amp radio system with all the bells and whistles - to control a 4x8 layout from one of the Atlas plan books. Nothing like taking advantage of inexperience to drive a bigger sale. Your guy, the ethical one, is the sort of story we usually don't hear about. It's also why I never went in to sales even though I could theoretically make much more money AND get to go on all sorts of company sponsored outings - I just can't look someone in the face and tell them they need this $10k server when the $5k model will more than meet their present and future needs.