[H0] Hybrid Automated DC and 'normal' DCC system and bypass capacitors...

Robert, I don't fully understand your description but it sounds like a very complicated exercise, I would suggest more normal ways of doing block sections win DC will give better results more easily. Check at Keith

Make friends in the hobby. Visit Garratt photos for the big steam lovers.

KN> On Sun, 10 Apr 2005 17:20:32 +0200, Robert Heller KN> wrote: KN> KN> >I have one section of track, specifically a closed loop trolley track that KN> >is to be purely automated. The trolley cars will NOT have DCC decoders, KN> >they will be straight DC with live overhead wires. They will be KN> >operated by a computer program and not under human control. I am using KN> >the same power source (a DCC booster) for the trolley track, using KN> >computer controlled relays and series rectifiers to drop the voltage -- KN> KN> Robert, KN> I don't fully understand your description but it sounds like a very KN> complicated exercise, I would suggest more normal ways of doing block KN> sections win DC will give better results more easily.

I am actually doing this, just using the the DCC booster as the DC power source. There are to be no manual throttles for the trolley loop. Each of the trolley loop's blocks has a specific *fixed* speed, determined by how many rectifies are shorted out by the relay contacts for that block. I want to use a single base track power bus (a DCC booster for each layout section) and one model of track occupancy detector board (simplifies things in terms of spares). I already will have a computer control system, handling other control aspects (signals, switches, and so on).

KN> Check at KN> Keith KN> KN> KN> Make friends in the hobby. KN> Visit KN> Garratt photos for the big steam lovers. KN>

\/ Robert Heller ||InterNet: snipped-for-privacy@cs.umass.edu

|| snipped-for-privacy@deepsoft.com /\FidoNet: 1:321/153

Up to you of course but its a very expensive DC power source compared to a straightforward transformer/rectifier.

So long as you use a Diode drop type detector it will work equally well with DC or DCC but current transformer detectors won't work on DC.

How many trolley cars will you have running on the loop? At MERG we do a DC throttle specially designed for unattended automatic block working that includes built in detectors and a ramp up and ramp down feature, it is easy to incorporate stops at defined places and could do all you want much better than diode switching. The kits cost about $16 and you would need one more than the maximum number of cars you want to run. Keith

Make friends in the hobby. Visit Garratt photos for the big steam lovers.

KN> On Sun, 10 Apr 2005 22:50:13 +0200, Robert Heller KN> wrote: KN> KN> >

KN> >I am actually doing this, just using the the DCC booster as the DC KN> >power source. KN> KN> Up to you of course but its a very expensive DC power source compared KN> to a straightforward transformer/rectifier.

I will actually be using DCC on other parts of the layout section. Rather that have *two* power supplies, I just want the one power bus.

KN> KN> > There are to be no manual throttles for the trolley loop. KN> >Each of the trolley loop's blocks has a specific *fixed* speed, KN> >determined by how many rectifies are shorted out by the relay contacts KN> >for that block. I want to use a single base track power bus (a DCC KN> >booster for each layout section) and one model of track occupancy KN> >detector board (simplifies things in terms of spares). KN> KN> So long as you use a Diode drop type detector it will work equally KN> well with DC or DCC but current transformer detectors won't work on KN> DC.

I know. I want to one only one type of detector -- this simplifies things like spare parts.

KN> KN> How many trolley cars will you have running on the loop? At MERG we do KN> a DC throttle specially designed for unattended automatic block KN> working that includes built in detectors and a ramp up and ramp down KN> feature, it is easy to incorporate stops at defined places and could KN> do all you want much better than diode switching. The kits cost about KN> $16 and you would need one more than the maximum number of cars you KN> want to run.

Two for now, maybe four later. Maybe I'll look at your kits.

KN> Keith KN> KN> KN> Make friends in the hobby. KN> Visit KN> Garratt photos for the big steam lovers. KN>

\/ Robert Heller ||InterNet: snipped-for-privacy@cs.umass.edu

|| snipped-for-privacy@deepsoft.com /\FidoNet: 1:321/153

Your implication originally was that you were going to dedicate a booster to this purpose. In that regard, it would be an expensive proposition. I'd use 12V relays and use the 12V from that poweruspply for the trolley power. This will insure that if the trolley shorts out somehow, it won't affect the rest of the layout. I'll also note that if you are using diodes and relays (multiple) for each block, you already have a lot of expense in the control circuits. I'd more consider using a computer for this work (PIC processors are cheap) and control each block with a DAC-08 for each block driving a transisto output and a single relay for direction.

-- Why isn't there an Ozone Hole at the NORTH Pole?

"M> Your implication originally was that you were going to dedicate a booster to "M> this purpose. In that regard, it would be an expensive proposition.

No, one booster per layout section, which in this case happens to include the trolley loop, along with a small dock side switching RR (using a DCC loco).

"M> I'd use 12V relays and use the 12V from that poweruspply for the trolley "M> power. This will insure that if the trolley shorts out somehow, it won't "M> affect the rest of the layout. "M> I'll also note that if you are using diodes and relays (multiple) for each "M> block, you already have a lot of expense in the control circuits. I'd more

1 relay per block: each block only has two speeds: 0 (stopped) and some fixed speed (one particular speed for that block). The rectifier diodes are pretty cheap.

"M> consider using a computer for this work (PIC processors are cheap) and "M> control each block with a DAC-08 for each block driving a transisto output "M> and a single relay for direction.

The trolley loop is 'one way', all trolley will travel counter clockwise.

I'm already going to use a process on a Linux computer for the trolley loop. Other processes will manage the small dock side switching RR's turnouts, handle the Rail Driver cab control and speak to the DCC unit to control the small dock side switcher loco.

"M> "M> -- "M> Why isn't there an Ozone Hole at the NORTH Pole? "M> "M> "M>

\/ Robert Heller ||InterNet: snipped-for-privacy@cs.umass.edu

|| snipped-for-privacy@deepsoft.com /\FidoNet: 1:321/153

I'm also confused...

If you use a DCC booster to power a DC motor without a decoder, it will not work without the DCC signal being recified first. Sounds like you are doing that part (even though your description sounds very confusing).

But I'm not sure how you are planning on using capacitors to help in train detection.

DCC signal is not a sinewave AC signal. It is more a square wave. Capacitor wouldn't work the way you expect. But if the trolley is running, it will draw current (from the DCC booster via the rectifiers), so it should be detected by the DCC detectors (again, I'm sort of confused here).

Honestly, for about 5 dollars, you can pick up a cheap wall transformer to power your trolley - much simpler!

And instead of using relays and rectifiers to control the speed, use a simple LM317 volt. regulator with several resistors (controlled by your computer) on the regulator's voltage adjustement side. That will get you a nice variable step voltage for your trolley.

Peteski

"W> I'm also confused... "W> "W> If you use a DCC booster to power a DC motor without a decoder, it will "W> not work without the DCC signal being recified first. Sounds like you "W> are doing that part (even though your description sounds very "W> confusing).

It is my understanding that the DCC 'signal' is pulses over a base DC power. And the DCC output is a polarized pulse-wave power source, not unlike the PWM type DC power packs, except the pulses carry a digital message serially. This should drive a DC motor directly without a problem. Just that the DCC pulses are a fairly high frequency relative to the frequency range used by PWM type power packs.

Sigh. I guess did not really explain things well...

"W> "W> But I'm not sure how you are planning on using capacitors to help in "W> train detection.

I want to make sure that the *high frequency* DCC pulses themselves survive well enough for the detector to pick them up properly.

"W> "W> DCC signal is not a sinewave AC signal. It is more a square wave. "W> Capacitor wouldn't work the way you expect. But if the trolley is

I know. The square waves just have a high frequency component. I effectively need a high-pass filter of some sort, but I am not sure.

"W> running, it will draw current "W> (from the DCC booster via the rectifiers), so it should be detected by "W> the DCC detectors (again, I'm sort of confused here).

The Occupancy detectors need the DCC pulses because it uses an inductive sensor (current transformer). I know, I can use a batch of DC only detectors for the trolley run, but then I would need two batches of detectors, which complicates the spare parts issue (ultimately I'll have

*lots* of detectors).

"W> "W> Honestly, for about 5 dollars, you can pick up a cheap wall transformer "W> to power your trolley - much simpler!

I'll already have the DCC booster -- needed for the other parts of the layout. I only want one track power bus per layout module.

Here is an out-of-date description of my overall layout:

This is slightly out of date -- I need to revise this sometime. The trolley run is on the Counterweight City module. Also on the Counterweight City module will be a dock side short line connected via a pair a exchange tracks to the main line.

"W> "W> And instead of using relays and rectifiers to control the speed, use a "W> simple LM317 volt. regulator with several resistors (controlled by your "W> computer) on the regulator's voltage adjustement side. That will get "W> you a nice variable step voltage for your trolley.

Don't need a variable speed. Each block has a *SINGLE* speed (not all blocks have the same speed). There is only ONE relay per block, not multiple relays per block.

"W> "W> Peteski "W> "W>

\/ Robert Heller ||InterNet: snipped-for-privacy@cs.umass.edu

|| snipped-for-privacy@deepsoft.com /\FidoNet: 1:321/153

No, DCC is AC at the tracks. A good on-line primer is at:

-- Bill McC.

Robert,

I'll try to be nice here.

There are some things you CAN do, but aren't worth the effort. This, I think, is one of them.

Why build a Rube Goldberg contraption when you don't have to? DC trolley operation is painfully easy. The technology is simple. Why try to do something involving DCC that it isn't intended to do?

You're going to spend more time and money than it's worth for such a simple thing. If it's just the challenge to see if you can do it, OK, but that doesn't seem to be the issue.

Mike Tennent "IronPenguin"

Robert, it seems to appear that you have your mind stuck on making your hobby more difficult than it has to be. Oh, well...

If you place a straight DC locomotive on a DCC powered track, it will not move an inch - it will just buzz and the motor will get warm. That is because the averaged DC component from the DCC is zero volts.

If you did end up somehow rectifying the DCC voltage going to the trolley then it would see the full voltage across the motor (but in a single direction only). No matter how you design your circuit - if there is any current draw from DCC signal (through the trolley motor), the detectors will function just fine (no Caps needed to bypass anything).

Peteski

"W> Robert, "W> it seems to appear that you have your mind stuck on making your hobby "W> more difficult than it has to be. Oh, well...

How is it more complicated to have a *second* power supply and an extra set of wires and separate batches of spares? The computer, etc. will already be in place for other reasons. I know I AM a computer geek -- this is just an additional function for the computer. Older Intel-flavored systems are cheap and Linux is *free*, so the computer cost is fairly low, compared to everything else (I picked up the computer I will be using for $300, used, and it is seriously overpowered

-- has lots of spare compute cycles). The computer programming part is the 'easy' part (for me, since computer programming is what I do all of the time).

"W> "W> If you place a straight DC locomotive on a DCC powered track, it will "W> not move an inch - it will just buzz and the motor will get warm. That "W> is because the averaged DC component from the DCC is zero volts. "W> "W> If you did end up somehow rectifying the DCC voltage going to the "W> trolley then it would see the full voltage across the motor (but in a "W> single direction only). No matter how you design your circuit - if "W> there is any current draw from DCC signal (through the trolley motor), "W> the detectors will function just fine (no Caps needed to bypass "W> anything).

I will be rectifying and using a number of *cheap* rectifier diodes in series to drop the voltage (lower voltage == lower speed).

Thanks for a *definitive* answer.

"W> "W> Peteski "W> "W>

\/ Robert Heller ||InterNet: snipped-for-privacy@cs.umass.edu

|| snipped-for-privacy@deepsoft.com /\FidoNet: 1:321/153

Robert,

If you use a system that doesn't use block detectors, it's not only simpler, but cheaper.

All you need is a board like I make, which uses no detectors. It uses diode stopping sections, not detectors, and a timer/relay to reverse polarity to the track. The effect will be EXACTLY like your computer run system, since both will slow the trolley in steps by using additional diodes.

Total cost? $29.95 for the board, pennies for a few more diodes, whatever an old power pack costs.

Results - exactly the same. No detectors to malfunction and replace. No wiring other than the block feeders. No programming.

If you enjoy the challenge and the extra work, fine. But I have to agree with Pete - you're making it much harder than necessary. If you were designing a system that actually slows the engine smoothly instead of using diodes, it'd make more sense. But to do all that, just to have diodes do the speed control, well...

Rob Paisely also has some circuits on his site that will smoothly control the speed, if you want to get a bit more complicated.

Mike Tennent "IronPenguin"