TRIAC circuit to control PW

An easy way to do this would be to reverse engineer the rather simple 12938 PowerStation controller to suit your needs.

 

You could try using a MTH Z controller too. 

 

I've tried searching for triad control of AC circuits but all I found was for 120v light dimmers. 

Gentlemen, I am very grateful for your replies.  Kind of surprised more people would not want to do this as industrial surplus transformers can be obtained inexpensively. 

I have seen 400w ac triacs for lighting control... Wondering as well if they might work for 24vac rather than 120vac.

 

i think ADCX Rob has a good idea, and while I am handy, I lack the skills and background  to design my own circuit.  If I were going to reverse engineer a 12938 I would want to add PSX-1circuit protection and maybe beef up the capacity to be in excess of 135w..

 

I like Flash's approach, just wondering  about waveform issues on an inductive load... And might want to fine tune output.. And that is assuming I could use a 24vac input.

 

i downloaded the TRIAC specs for a Q 4040K7... But no dice on a schematic to control it.

Seems to to me this might be possible as the components are not very expensive.

 

 

i am also looking for a 12938 controller, but they are normally part of an expensive set including the power brick.

 

here are some ideas along the lines of what I was thinking

 

 

 

 

 

http://www.instructables.com/f...B9HLA110NK.LARGE.jpg

 

the TRIAC is about$5.00

 

thanks!!!

Buy a set on eBay and sell the brick.  You might come out almost even.

I do know that household dimmer switches will not work for 3-rail 18vac train. I've tried.
That circuit from instructables does not look like it will work for controlling trains. I'm not an electrical engineer so I can't quite explain why. All I can say is that it's doesn't look right.
I've been down your path before and came up empty when searching for DIY circuits. My advice is to buy a MTH Z controller and reverse engineer it or just use it as is. Also you are going to have to figure out a way to limit that 24v transformer to 18v.
Ultimately I just decided to go TMCC all the way and have a 18v transformer brick connected directly to the track.

Not on the secondary side of the transformer, but it will work, for a while anyway, on the primary side.  A ceiling fan speed control will work even better.

Flash.. Thanks for letting me know about the dimmer switch being a non-starter.

 

i am wondering why I have to limit down to 18vac.. The tpc400 accepts inputs from 9vac to 30vac.  I assume the TRIAC circuit accepts the same input?

 

maybe you were referring to the MTH Z?

 

ADCX Rob.. Good idea. Might consider that if the price is right.

It's the output to the track that should be limited to 18v. Some of the train electronics and lightbulbs do not like excessive voltage.

This may help some (I hope)

http://www.bristolwatch.com/ele/triacs.htm

 

 

Thanks flash and Rob.

Flash.. I can see the limitation for modern trains, but I think the postwar stuff is more robust.   If I were running modern stuff I would use a voltage limiter pr a tpc 400 to limit the track voltage.. Your point is well made.

 

Rob .. I am a bit concerned about hysteresis losses in the transformer by limiting the input voltage.. Somewhere in the corner of my grey matter I remember some concern about this... But I like the idea.. Problem may be practicality for a 3kw transformer.  

Garfield

 

that is what I'm looking for!!  Now to build it and see if it works

Garfield

 

that is what I'm looking for!!  Looking where to place a control potentiometer

on this circuit

A variac on the input mat do what you want.

 

Al

A rule of thumb is that 3KVA is equal to about 75% or 2.3KW, "all things being equal".  For 24VAC at 2.25KW, you will have about 93A of current, in other words an arc welder!

An industrial fan controller should be used on the primary side, but coming up with a 240/277V model may be hard and $$$ (check industrial electrical companies for prices and availability). 

 

Best bet, if you want to continue would be to wire it to 120VAC and use an 120V industrial fan controller or heavy variac.  A transformer is designed so that at full designed load it puts out the desired voltage.  As load decreases, the voltage rises up eventually reaching an open circuit value, and vice versa.  A well designed transformer has alot of iron to keep voltage controlled between open and full load. Thus with 120V input and controller you would still be able atain 18VAC for rails, but you best have fuses and a fast solid state circuit breaker before the track.  One derailment at 93Amps would be spectacular without those safety devices!!

As a long term ogauger, EE and designer, my best advice is to forget about using a triac. I have designed and built several loco RC systems. One of my earliest used a triac. The circuit worked, but performance was not great and I switched to other control approaches. The triac circuit will deliver a variable amount of AC power, but  electronic E units appear to be overly sensitive to the triac output waveform.

I helped design a power controller for one of the major O gauge manufacturers. It uses a Q6008L5 thyristor. I used a PIC to control the output waveform for throttle and could shift the negative and positive alternations to give the whistle and bell functions.

Forman.. Yup ... Gobs of juice... Welding is not something I wish to do... At least not with the trains.  Would use a psx1 electronic circuit protector for each track

 

thanks for the input !

Last response was for rrman. Darn auto correct!

Bob walker.   But the lionel tpc 300 and tpc 400 use triacs...the 400 uses a Q40407k.. How do they get around e unit issues??

thank you for your response

Gary e.  I think this is what I am asking... would this resolve the TRIAC issue bob walker raises?  Can u fill me in more on how I could build something like that?

I assume a thrysistor and a TRIAC are the same thing... sorry for my lack of knowledge .

so wondering why bob walker was experiencing electronic  e unit issues

 

thanks for the input... Now I just want to try to bread board it

Gunrunnerjohn... Thank you for your reply.  I am grateful to you and all those here sharing your knowledge as I try to learn.  

 

I understand many of the concepts but lack the foundation in ee beyond the basics of resistors capacitors Diodes etc.I get frequency.. Wave forms ... You know... The basics tought to a mechanical engineer...   I enjoy learning.   

 

I have to assume that if lionel did this with a thrysistor or TRIAC in the tpc 400 there must be a way to do it reliably and inexpensively through a potentiometer controlled TRIAC circuit or similar control circuit. Besides ... I have this immense 60 lb industrial 10:1 transformer and can reliably make all the 12 or 24 vac power I could ever need.

 

 

 

 

Just note that any triac circuit that cuts down the voltage from the transformer is going to result in a sawtooth waveform.  That isn't a problem for a lot of model train stuff, but it is for some. 

 

As a young EE I'm a bit confused on why you are trying to do this. Is 240VAC all you have available? Why not use a standard train transformer? A 3KVA transformer may be a bit of a fire hazard when not properly protected... Have you considered starting with a smaller transformer like a transformer out of an older car battery charger? What about running your old trains on DC? You could make a simple AC to DC supply with a variable output...

If part of the exercise is to experiment with and learn about triacs, then by all means stay the course.  But one "technology" that was not economical or feasible when triac-based controllers came out is Class-D switching amplifiers. 

 

http://en.wikipedia.org/wiki/Class-D_amplifier

 

So rather than chopping the nice 60 Hz sinewave once (or twice) per cycle which leads to "ugly" sawtooth-like waveforms, a Class-D amplifier chops the input hundreds of thousands of times per second.  With modest conditioning (filtering) of the output to remove any remnants of the high frequency chopping you get a clean signal...so clean that this is how most high-power consumer audio amplifiers for home theaters and cars now operate.

 

 

To be clear I have NOT tried this myself but as a concept for discussion, consider this module (several on eBay for about $20 shipped).   2 x 100 Watt outputs.  Note the modest-sized heatsink for a 100 W output device.  You supply low-voltage (vs. line voltage) AC as the power supply.  So the concept is to feed a 60 Hz sinewave as the audio input.  Obviously you have a source of a clean 60 Hz input signal from the low-voltage transformer itself!  The volume control knob becomes your voltage control.  So you have a 100 Watt hum generator since 60 Hz is in the audio range.   Yes, there are other considerations with current limiting, short-circuit protection, etc.. but these are factors you also need to consider with a roll-your-own triac controller.

 

If someone has already done this as a way to roll your own pure-sine controller for our trains I'd like to read about it.

I thought of that Stan, and I never considered looking for a packaged solution, so I didn't mention it because of the complexity of rolling your own.  That's a great idea!

 

Looking at them, I see supplying power to them might be a trick...

 

 

Right.  This is by no means a slam-dunk.  What caught my eye when glancing at these modules is they take an AC power source and you see two bulk caps suggesting they generate + and - DC supplies.  Thus, there may not be a problem with tying common "grounds" in train applications that you'd have with a unipolar DC supply and a BTL (bridge-tied-load) output stage generating "AC".  The hum generator is also a unique situation since the amplified signal is the same shape as the power supply.  This would suggest that less bulk capacitance is needed since the output is zero when the input is zero, and there is maximum available power supply magnitude when the output is to be at the maximum magnitude.

I think we've ventured into the realm of "why would I go to all this trouble?".

Especially with 3KVA and 240VAC.  If you don't know how to build a device for those specs, from your EE background, I don't think I would be looking to the internet for help on this one.  Could be a Darwin Award winner.  G

Stan. What a cool idea!  Thank you for the thoughtful reply

 

gunrunnerjohn ... Yes this may end up as an academic exercise.. But learning new things is part of why I'm looking into this... That and I am pretty sure I could just get a tpc 400 and a cab 1 and use the 24vac input.  I guess I'm back to what ADCXrob was saying.. To use a 12938 lionel power controller as another fall back.  It just seems that the circuit would not be overly complex.

 

Matt A.  I have 120 and 240 vac .. But I would presumably be varying the 24vac transformer output.   This is basically what I think lionel power bricks do.

i just got a good deal on a big transformer .

 

GGG. Yeah I could see this all going badly... While I lack the ee background to design the control circuit I have wired in a 12.5 kva generator.. Had my electrician friend guide me. I have lots of respect for making sure to not put myself between the hot and the neutral or ground. while the transformer is 3kva.. The output is 24v, and I plan on using fuses, circuit breakers .. And some even faster electronic protection.  I'll try not to accidentally weld something.

 

To all :   i saw a demo video of a psx1 circuit protector that looked like it used a large fixed transformer .. I'll find it and post it .

 

thanks to everyone for the input

Hi Bob I might have one of these 12938 controllers without the brick that Im not using 30thstreet

Interesting how much conversation the original posting generated. Bottom line though seems to be what I suggested, forget about a triac based circuit. I based this on my experience and observation that over the past several years, there have been a number of train magazine readers who have expressed problems with triac based AC power packs.

30th street - thanks, Ill try to email you

 

Bob Walker.  sounds like you have tried to go down this road unsuccessfully, but isn't this how most modern TMCC stuff works?

 

Is it the home grown stuff that has been a problem or have the manufactured systems been unacceptable?

 

Other than electronic e unit issues, what other difficulties did you have?

 

Thanks

Most modern stuff is moving to FET drivers and away from triacs.

 

In the reverse engineering category, I opened up a Z-controller and took these of the circuit board. Obviously this has bells and whistles (literally!) that do not apply to your PW triac controller application but perhaps you can glean something from it. 

 

 

To GRJ's point, IF a "new" design were to come out from a manufacturer, I'd think it might be a FET-based switcher with a pure-sine output.  But with the growth in fixed-track-voltage command-control, I wonder how many manufacturers have AC conventional voltage controllers on their new product radar.

 

Actually, there is an artifact of triac-based controllers that apparently helps with some modern electronic smoke units.  For the same RMS AC track voltage, a chopped-sine controller has a higher peak voltage than a pure-sine controller.  There have been several detailed threads about the how-and-why but the net effect is that for a given speed, a conventional engine running on a triac-based controller can produce more smoke.

 

I can put a lower value resistor in and get more smoke too.

 

I work for Lutron Electronics.  Can't disclose too much, but the inside of a CW-80 looks an awful lot like stuff I work with every day.

 

The stuff going to FETs is the LED and CFL dimmers.  Incandescent and Halogen still use triacs for the most part. But magnetic transformers (like in the CW-80) can't handle DC offset so you need a dimmer designed to keep perfect symmetry on the dimming signal...

 

We (and everyone else) make dimmers for Magnetic Low Voltage light circuits (MLV).  One of these days, I'm going to take one of these dimmers, hook it to 120VAC and hook the "dimmed hot" output up to a magnetic 120/24VAC transformer ($9 at radio shack).  Nicer, digital MLV dimmers have the ability to limit high end to keep the output below 18VAC.