Need some help with a potentiometer as a rheostat - UPDATE page 2

[fiatfan]

I'm trying to build a speed control for the trains using a 13.6v transformer and a 10K potentiometer.

 

First up, can it be done with just those two components?

Second, if it can be done, how do I wire it up?

I've already blown a couple of pots trying various wiring combinations.  Do I need a different value for the pot?

 

Any information, guidance, or advice will be greatly appreciated.

 

Tom
 

[cacole]

A potentiometer alone cannot handle the amount of current you're passing through it. You need additional components that allow the pot to control a power transistor's throughput.  A transistor throttle circuit is what you should be looking for.  There was one mentioned on these forums a couple of weeks ago.

[fiatfan]

 Thank you, sir.  I will do some searching.

 

Tom

[richg1998]

Look in this forum for: DC Tethered Walk-around

Maybe two days ago. There are some good links with throttle possibilities.

Rich 

[Seamonster]

A pot can't handle the current a locomotive draws and will burn out. A rheostat is an entirely different animal, built with wire windings instead of a carbon element like a pot. It can handle higher currents. Besides, with the setup you describe you would be applying AC to the track and the motors require DC. They wouldn't like pure AC at all. Buy a good transistorized throttle and you'll be much happier with it.

[fiatfan]

 Thanks for  the additional information, everyone.

Bob, I forgot to mention the rectifier I have between the transformer and the pot.  I have a transistor throttle based on a circuit from www.awrr.com but it eats up too much voltage.  I have 19 volts coming in and less than 10 volts coming out.

The engine is a P2K with the QSI DC sound chip and I'm trying to get more voltage to the track.  Right now the top speed is about 3 mph.

 Thanks again to everyone for your help.

 

Tom

[Seamonster]

fiatfan

 Thanks for  the additional information, everyone.

Bob, I forgot to mention the rectifier I have between the transformer and the pot.  I have a transistor throttle based on a circuit from www.awrr.com but it eats up too much voltage.  I have 19 volts coming in and less than 10 volts coming out.

The engine is a P2K with the QSI DC sound chip and I'm trying to get more voltage to the track.  Right now the top speed is about 3 mph.

 Thanks again to everyone for your help.

 

Tom

Okay, with the rectifier you're getting pulsed DC. You would need a large value capacitor after the rectifier to get pure DC.

Which AWRR circuit is it? As for it eating up too much voltage, I suspect that it has a variable voltage regulator in it, like a 317T. They are designed to put out zero to some maximum voltage no matter what the input voltage is (so long as it's at least 2 volts more than the max. output and less than 35). The reason you get only 10 volts out of your throttle is that 10 volts is a good maximum for your locos. You wouldn't want the full 19 volts on the track.

What's the voltage on the track when the throttle is turned up and the loco is going only 3 mph? If it's 10 volts, I would suspect a fault in the loco. If it's only 3 or 4 volts or so, then something is sucking down the voltage, maybe a loco fault, maybe a throttle fault.

[fwright]

fiatfan

 Thanks for  the additional information, everyone.

Bob, I forgot to mention the rectifier I have between the transformer and the pot.  I have a transistor throttle based on a circuit from www.awrr.com but it eats up too much voltage.  I have 19 volts coming in and less than 10 volts coming out.

The engine is a P2K with the QSI DC sound chip and I'm trying to get more voltage to the track.  Right now the top speed is about 3 mph.

 Thanks again to everyone for your help.

Tom

Tom

You don't say which circuit you are using, so it's tough to analyze what the problem might be.  Some guesses:

• How are you measuring voltage in?  RMS from the transformer?  Peak to peak from the transformer?  DC input voltage to the throttle?  Your full wave rectifier should be dropping close to 1.5 volts from the RMS output of the transformer.  This could be higher with filters and/or a voltage regulator in the rectification circuit.  Also, be aware transformer output voltages can drop significantly under load.  A 19V AC (RMS) output can easily be 14-16V AC under load - especially if you are using an old Lionel transformer.
• Does your throttle circuit use transistors (typically Darlingtons) or voltage regulators or op amps to vary the output voltage of the throttle?  A Darlington package will have a voltage drop of another 1.5 volts - and 0.7 volts more if there is a driver transistor in front.  Notice your total drop is about 3.5 volts from the transformer output in this case.  I'm not familiar with the voltage drops associated with the op amps or voltage regualators, so I can't advise you there. http://www.awrr.com/throtl5.gif is a transistor circuit designed for a DC input.  With a full wave bridge in front it would work off the 16V AC terminals of a transformer.  Hooked to 12 volts fixed DC terminals, it's going to max out at less than 12 volts output.
• Is your pot a quality linear taper pot?  An audio (non-linear) taper is what you normally get unless you ask and pay for the linear taper.  The cheaper audio pots will often have or develop dead spots and other problems, and might not go smoothly to either end of the resistance range.  I would measure the resistance of your pot (disconnected from the circuit), both at the ends, and maybe a couple of places in the middle to see if you are getting what is advertised.
• Finally, a DCC sound decoder operated on DC (your situation) has issues.  The usable range of your throttle is very small because the 1st 7-9 volts is taken for sound, and is not available to power the motor.  Most of these locomotives start moving at 7-9 volts, and may or may not reach their DCC top speed on 12 volts DC.  Those who use these DCC decoders can tell you more than I.

Hope this helps

Fred W   

[rrinker]

 This is the case with all the sound decoders. It's not a design flaw, it's simply the way it has to be - otherwise your loco would be moving at half speed before the sound even came on, hardly realistic. It's a compromise. Sound and DC really don't mix that well, for just this reason. And note that more than 20-22 volts to the decoder will probably fry it.

                             --Randy

[modelmaker51]

Make sure you're using a linear pot, audio pots are non linear, the first 2/3 rotation is only about 1/3 to 1/2 of the output.

[richg1998]

The sound part of the decoders use about 5 volts DC to run the sound chips. The motors require more voltage to start moving the motor armature.

Some motors require more voltage, some less. I have put a 'Scope on the motor decoder wires and saw pulses before the motor armature started turning.

When running on DC or DCC, the motors get pulse power so the motor is dependent on the micro processor to put out enough to rotate the armature. There is mechanical resistance of the gearing and weight of the loco.

Rich

[fwright]

fiatfan

I'm trying to build a speed control for the trains using a 13.6v transformer and a 10K potentiometer.

 Tom

 

Tom

I didn't catch the part about a 13.6V transformer (I assume center tap) in my first reply.  With a normal 3.5 volt combined drop across the diode and transistor junctions, a maximum DC output voltage of 10 is about right.  A transformer output rated at 16-18V is recommended for most HO to O transistor throttles to allow a maximum throttle voltage of 12+.

Your transformer will likely read substantially higher under no load; 19V is possible.  Or your 19V reading is peak to peak, not RMS.

hope this helps

Fred Wright

[richg1998]

How about telling us everything you are using?

Are you using a single rectifier which is half wave?

Are you using two rectifiers which is full wave?

Are you using one bridge rectifier which is full wave. Four rectifiers in one case.

Are you using a filter capacitor and if you are, what is the value?

Show us a digram of what you are trying to do?

How is the transformer wired?

Up tot his point, you seem reluctant to give us complete details of your design.

My DC throttle has 13.6 volts DC out of the full wave bridge rectifier and I can get 12 volts DC at full throttle running an older open frame motor loco. I am using the design I posted near the beginning of this thread.

About 14 volts AC is applied to the bridge rectifier. I am using a 470ufd filter capacitor and all the old DC throttles in our club are the same design we used for some years.

Rich

[fiatfan]

 Wow!  Thanks for all the input everyone.

OK.  Here's what I have.  I'm using the VST w/TWAC and momnetum and an old American Flyer transformer.  The transformer is rated at 16 volts and 40 watts.  I'm using the fixed output taps and nothing else is hooked to the transformer.

Output is 16.5 volts no load and 16.3 volts full load from the transformer when measured using a cheapie digital multimeter.  I must have misread the meter last night.

Between the transformer and the transistor throttle I'm using a full wave bridge rectifier rated at 4 amps 50 volts (Radio Shack part number 276-1146).  For the pot I am using a linear taper 10K ohm (R/S part #271-1715).  All parts in the circuit are as specified by the designer.

The sound chip turns on at about 5 volts and the engine starts moving at around 8 volts.  Max voltage to the track is about 9.5 volts.  The engine is moving at about 1 scale mph (42 seconds to travel one foot).

On the transistor throttle, I have cut off the momentum (switch S3 on the circuit).  With the momentum on, max voltage drops to about 9.2.

I sincerely apologize if I seemed vague in my first post.  That was certainly not my intent. I was simply looking for a way to get more voltage to the engine.

Thanks again to all who have posted.  I really appreciate it.

 

Tom

[CSX Robert]

fiatfan

...For the pot I am using a linear taper 10K ohm...

The circuit you linked to calls for a 5k pot, using the 10k could be the reason you're not getting enough voltage out.

[richg1998]

Show us a diagram of your circuit. What you are telling us does not show the actual circuit. You should not have that large a voltage drop. I suspect the pot in your circuit is handling too much current is why I would like to see the actual diagram.

I have a very small voltage drop in my DC throttle using a 10k pot and two transistors.. The 10k pot controls a NPN Darlington low power transistor that drives a high power NPN transistor. I lose about one volt.

I have a circuit that uses a 10k pot to drive a single NPN Darlington power transistor. This circuit is even simpler than my present circuit.

Rich

[CSX Robert]

richg1998

Show us a diagram of your circuit...

He linked to a diagram of the circuit(http://www.awrr.com/throtl4.gif) in his last post.

[fiatfan]

 Thank you, all, for keeping me on my toes.

I just went down and opened up the control box and it is indeed a 5k pot.  The 10K pot is for another project and I got the two mixed up.  

My mistake.

 Rich, I would love to see that diagram of the single Darlington and 10K pot.  It sounds like something I'm looking for.

All, I also ran a P2K SW9/1200 (straight DC) and the meter showed as much as 10.5 volts so apparently the DCC chip is eating up a lot of voltage.  That's why I'm trying to find something that will get ~12 volts to the track.

Tom

[richg1998]

 Very simple throttle. The 2n6040 Darlington is on ebay for $2.25, Buy it Now. I find and buy a lot of new stuff like this on ebay. Just check the sellers record. Clamp it to a piece of aluminum for a heat sink. Use a Mica insulator between the transistor and heat sink.

Rich

[fwright]

Tom

I agree with Rich on the simple Darlington transistor throttle.  I built one myself in the '70s based on a circuit in Model Railroader, and I use an old AHM trainset power pack with broken rheostat and rectifier as the power supply.  I am still able to get a full 12 volts out - the Darlington drops perhaps 1.2 volts across both junctions in a normal emmiter ground configuration.

You can probably boost your input voltage somewhat with a large capacitor filter across the rectifier bridge output.  This should pump up the input to close to the peak-to-peak value if the input impedance of your circuit is reasonably high.  Use a minimum 2200ufd or similar capacitor rated at 25 volts or better.  Observe polarity.  The disadvantage of the filter capacitor is the loss of any pulses in your output - output will now be filtered DC.

Fred W

[fiatfan]

 Good morning!

Rich, thanks for the circuit.  It's identical to the one I was trying to build (with no success).  More on that in a minute.

Fred, the circuit you referred to is, I believe, Peter Thorne's throttle in the August 1976 MR entitled "The Simplest Transistor Throttle?".  

The circuit calls for a 2N6040 which after some checking, cross referenced to a TIP120.  However, the 2N6040 is a PNP type and the TIP120 is an NPN type transistor.  Does this cause a problem?  I used the TIP120 in the circuit and I got a constant ~11 volts output no matter where the pot was set.

Fred, I have a 4700 uf 35 volt cap leftover from another project.  Is that to large to use as a filter?

Again, I really appreciate your help on this project.

 

Tom

[fwright]

fiatfan

Fred, the circuit you referred to is, I believe, Peter Thorne's throttle in the August 1976 MR entitled "The Simplest Transistor Throttle?".

I lost most of my back issues of Model Railroader in the last move so I no longer have the reference or circuit.  The time frame sounds about right - I built the throttle sometime between Jan '76 and June '77 (probably towards the end of the period).  There was also a 2 part series in the '70s on throttle design that may have had the particular circuit I used.  I know I have a small driver (signal) transistor to drive the Darlington in my throttle, and Rich's earlier description mentioned one, too.  I don't think the Darlingtons have enough gain by themselves to keep current levels small enough on the input side for common components without the driver transistor.

I moved the driver, the Darlington and reverse relay to the base unit so full current didn't have to go through my phone handset tether.  I have other pics of the inside at home (but I'm ashamed of my soldering workmanship that showed in the pics).  

The circuit calls for a 2N6040 which after some checking, cross referenced to a TIP120.  However, the 2N6040 is a PNP type and the TIP120 is an NPN type transistor.  Does this cause a problem? 

Shouldn't make a difference, but polarities have to be reversed - and be consistent throughout the circuit.  If a driver is used, and you reverse the Darlington from PNP to NPN, you have to reverse the driver, too.  Same with circuit polarities.

Fred, I have a 4700 uf 35 volt cap leftover from another project.  Is that to large to use as a filter?

Good choice.  You want the capacitor big enough to hold most of the peak voltage against the input impedance until the next peak comes along.  Bigger is more insurance.  Again, make sure polarity is correct - a reversed electolytic can explode with a bang.

Hope it goes well.

Fred W

[fiatfan]

 Fred, I sent you an e-mail with the scan of the circuit and some questions.

 

Tom

[richg1998]

fiatfan

 Good morning!

Rich, thanks for the circuit.  It's identical to the one I was trying to build (with no success).  More on that in a minute.

Fred, the circuit you referred to is, I believe, Peter Thorne's throttle in the August 1976 MR entitled "The Simplest Transistor Throttle?".  

The circuit calls for a 2N6040 which after some checking, cross referenced to a TIP120.  However, the 2N6040 is a PNP type and the TIP120 is an NPN type transistor.  Does this cause a problem?  I used the TIP120 in the circuit and I got a constant ~11 volts output no matter where the pot was set.

Fred, I have a 4700 uf 35 volt cap leftover from another project.  Is that to large to use as a filter?

Again, I really appreciate your help on this project.

 

Tom

 

You are making this quite complicated. You cannot just substitute a NPN for a PNP without wiring changes. The PNP was chosen for a reason.

If all else fails, follow instructions in circuit designs ideas until you understand basic electronics. Then you can experiment.

I told you what voltages I see in a previous message when using a 470ufd cap. Our club used DC throttles with the 470ufd Cap for some years. Yes, there is some AC ripple in the DC track voltage but it never caused any issues. And we had locos using can motors. You might consider it cheap pulse power. I am giving you experience, not opinions.

Rich

[fiatfan]

 Rich, thank you for all your time and information.

At this point, I think I'm going to start all over and see if I can find a simple circuit to build.  

I am going to try the capacitor across the rectifier.  The chip in the engine also has a setting for engine start voltage.  I'm going to try to lower that by a volt or two also.

Again, thank you for all your help on this.

 

Tom

[cordon]

I think you need more than 13.6 volts out of the transformer.  I think the LM317 needs about 2.5-3 volts "head room," and other posters have mentioned above the need for a volt or two to cover other losses downstream.  Therefore I think the voltage under load out of the transformer should be at least 17 volts.

So, to answer your original question, I don't think you can get what you need with a 13.6-volt transformer. 

Recommendation 1:

      Change to an 18-volt transformer.

       Instead of getting all wound up in the arithmetic of the LM317 formula (1.25 * (1 + R2/R1)) and worrying about tolerances of the resistors, I would substitute a 5K pot for R2 and adjust it for output of 13-14 volts under a typical load.  Then you should be able to get 12 volts to your tracks.

Recommendation 2:

     Skip all the electronics and use a 10-15 ohm rheostat and a DBDT reversing switch between the rectifier and the track (in the remote). This should also deliver about 12 volts to the track.  The wires to and from the remote will have to carry all the current, so I would use something like lamp cord to give both flexibility and current-carrying capacity.

  

 

[fiatfan]

 After some correspondence off line with Fred Wright, I have been experimenting with the original circuit.  IT WORKS!!!!!!!

After trying a variety of connections, I put the circuit back the way it was originally drawn and reversed the positive and negative feeds. I'm getting ~13.5 volts to the track and everything works as it is supposed to.

Thank you, Fred for all your help, time, and guidance on this. 

To all the others who contributed, thank you also for taking the time to put up with me.

 

Tom

[fwright]

 I'm glad to hear it works! 

 Fred W