Reducing track voltage

[Ray H]

Our group's O gauge layout uses a couple of controllers that take their supplies jointly from a separate mains powered transformer rectifier - 2 controllers from one transformer rectifier. We also have other controllers on the same layout with a more normal (?) output voltage of around 12v. There is no noticeable performance difference between the two.

 

However, we've discovered that the maximum track voltage being output from the initially mentioned controllers is in the region of 24 volts when measured without a loco on the track. None of the O gauge DC locos have any problems with this. We're less certain of whether any decoder fitted locos (running on the DC parts of the layout) will be quite so receptive especially as they may be fitted with the less power tolerant decoders normally found in OO gauge locos. We did kill one such sound decoder some while ago although we don't know where on the layout or when it actually happened.

 

We'd like to try reducing the maximum voltage that can be supplied to the track via the initially mentioned controllers without modifying those controllers so that we can be sure that the high voltage isn't to blame for any similar failures in the future.

 

I've a vague idea that this should be possible using a zener diode + resistor in series across the output from the transformer rectifier without needing to be overly concerned about the current that one or both of the controllers may draw. Is this correct? I'm thinking of using a 15v zener. What value resistor should I use?

 

Is there any other arrangement that could be used that doesn't impact on the current that can be supplied to the track?

[PatB]

A zener diode sounds feasible. However, how sure are you that you need to drop the voltage? The important number is not what you get when the rails are open circuit but what maximum you will actually see across the rails when there is a loco in operation to provide a load. This is likely to be much lower. 

[smokebox]

Whilst I agree that the on load voltage will probably be much lower than the open circuit voltage, it may be a good idea to have a voltage regulator anyway.

 

What I would try is a series regulator like this regulator.

 

The zener is then not doing all the hard work and getting hot but the transistor will possibly get hot and can more easily be fitted to a heatsink. Sorry, I have no idea of what component values you would need to use.

[cliff park]

Also bear in mind any zener would need to be before any reversing switch, if that is inside the controller then you would have to fit the zener in there.

[dhjgreen]

Something like this is easiest, the data sheet shows how to wire it. It will need a heatsink.

 

http://uk.rs-online.com/web/p/linear-voltage-regulators/7147783/

Edit to add, assuming 2 Amps is enough for 0 gauge

Edited August 20, 2017 by dhjgreen

[Crosland]

I've a vague idea that this should be possible using a zener diode + resistor in series across the output from the transformer rectifier without needing to be overly concerned about the current that one or both of the controllers may draw. Is this correct? I'm thinking of using a 15v zener. What value resistor should I use?

Not really practical with the kind of currents you want to supply for an O gauge loco.

 

For example, to drop 12V at 1 Amp the resistor would need top be rated at 12 Watts. You then need to calculate what happens to the zener when there's no load. It's trying to clamp the voltage at 12V resulting in the same 1 Amp through the resistor and Zener. The Zener also needs to be rated at 12 Watt.

 

Zener diodes are only suitable for low power applications (e.g. regulating the voltage to a higher power transistor) or where the current does not vary.

[dhjgreen]

Not really practical with the kind of currents you want to supply for an O gauge loco.For example, to drop 12V at 1 Amp the resistor would need top be rated at 12 Watts. You then need to calculate what happens to the zener when there's no load. It's trying to clamp the voltage at 12V resulting in the same 1 Amp through the resistor and Zener. The Zener also needs to be rated at 12 Watt.Zener diodes are only suitable for low power applications (e.g. regulating the voltage to a higher power transistor) or where the current does not vary.

See post 5

Edit thinking about it it needs a series resistor to dissipate some of the power.

Edited August 20, 2017 by dhjgreen

[meil]

See post 5

Edit thinking about it it needs a series resistor to dissipate some of the power.

Not sure what you are getting at there. Crosland was referring to the Zener diode suggestion. You are referring to a linear voltage regulator.

[soony]

Stop thinking of how to codge it together , get a proper controller , You spend hundreds of pounds on each loco , don't know how many thousands on a layout , then try and get it working with suspect circuitry in the hope of saving pennies.

[kevinlms]

Our group's O gauge layout uses a couple of controllers that take their supplies jointly from a separate mains powered transformer rectifier - 2 controllers from one transformer rectifier. We also have other controllers on the same layout with a more normal (?) output voltage of around 12v. There is no noticeable performance difference between the two.

 

Not sure that I understand what you have here.

 

You state that you have 2 controllers on one transformer rectifier. Well that is bad practice for starters. For DC operation, you should have a separate power source for each controller & train.

 

You also state that you have other controllers on the same layout, with a 'more normal 12 volt'. Well surely the answer is there, replicate what these other controllers are powered from.

 

Like others have recommended, the layout has taken a lot of money and effort to build, so why not do the job properly and spend some many on proper power sources?

[Ray H]

Thank you all for your various responses.

 

We have no reason to replace the controllers that have as they perform the task for which they were originally designed satisfactorily. I understand that they are all proprietary brands.

 

Our only reason to investigate the voltage limitation is because we are uncertain whether DCC chipped locos set up to run on DC will be adversely affected by the higher voltages that a couple of the controllers output.

 

A voltage regulator solution was my original thought but I recalled having seen a zener diode and resistor used in some other circuitry and wondered if it could be used in this situation.

[dhjgreen]

Thank you all for your various responses.

 

We have no reason to replace the controllers that have as they perform the task for which they were originally designed satisfactorily. I understand that they are all proprietary brands.

 

Our only reason to investigate the voltage limitation is because we are uncertain whether DCC chipped locos set up to run on DC will be adversely affected by the higher voltages that a couple of the controllers output.

 

A voltage regulator solution was my original thought but I recalled having seen a zener diode and resistor used in some other circuitry and wondered if it could be used in this situation.

I think Crossland has answered that fully. You could use a string of rectifier diodes to drop the voltage, you would need getting on for 17, but they cost next to nothing.

[Ray H]

I think Crossland has answered that fully. You could use a string of rectifier diodes to drop the voltage, you would need getting on for 17, but they cost next to nothing.

 

Indeed he has. I should have acknowledged that in my response, Apologies.

[dhjgreen]

The string of rectifier diodes is serious suggestion, there are people who use this method with a rotary swictch as a speed controller.

[Ray H]

The string of rectifier diodes is serious suggestion, there are people who use this method with a rotary swictch as a speed controller.

 

Indeed we've used that arrangement as well for some other circuitry. Thanks.

Edited August 20, 2017 by Ray H

[DCB]

If you are desperate to keep off load voltage down you could use an off the shelf Lucas Motorcycle Zener Diode and heat sink, as used on 1971 to 1976 BSA and Triumph Motorcycles with the unregulated permanent magnet alternators. These used 10 amp/hr batteries or for off road a capacitor in place of the battery, a capacitor about the size of those used on a typical CDU.  These Zeners are capable of keeping voltage down to around 14.5 volts from 2000 to 7000rpm whether the lights are on (80 watts) or off.  The Zeners come in - ve earth and + ve earth versions and the heat sinks get hot!

 

On the other hand a decent Morley Vector or similar will keep the off load voltage down if that is an issue but to be honest I can't see 24 volts DC making any difference to a DCC decoder. [Edit,] as NMRA standards require decoders to be OK on27 volt DC

Edited August 21, 2017 by DavidCBroad

[Ray H]

. . . .  but to be honest I can't see 24 volts DC making any difference to a DCC decoder.

 

That's pleasing to hear. We couldn't find any other reason why the sound decoder blew earlier so we put it down to the high voltage level - luckily we got it replaced fora nominal fee.

 

It sounds as though we can leave things be.

 

Thanks.

[uax6]

Not sure that I understand what you have here.

 

You state that you have 2 controllers on one transformer rectifier. Well that is bad practice for starters. For DC operation, you should have a separate power source for each controller & train.

 

You also state that you have other controllers on the same layout, with a 'more normal 12 volt'. Well surely the answer is there, replicate what these other controllers are powered from.

 

Like others have recommended, the layout has taken a lot of money and effort to build, so why not do the job properly and spend some many on proper power sources?

 

Can you explain to me why it is bad practise to have two controllers off one rectifier unit please? I cannot see any problem with this as long as the rectifier unit provides enough current so that both controllers can provide maximum output without overloading the rectifier.

 

If its such bad practise maybe I should stop using my 35 Amp 50V rectifier unit that provides the 50v for my telephone exchange and provide each relay and selector with individual 50v supplies.

 

Andy G

[Crosland]

That's pleasing to hear. We couldn't find any other reason why the sound decoder blew earlier so we put it down to the high voltage level - luckily we got it replaced fora nominal fee.

 

It sounds as though we can leave things be.

 

Thanks.

No, you have to remember that Mr Broad does not use DCC and has displayed a lack of understanding in the past. There are some brands of sound decoders (including sound) that are known not to conform to the NMRA standard of being able to withstand at least 27V track voltage.

 

Always check the documentation and be wary of anything fitted with Stay alive capacitors. Check the voltage ratings of the capacitors.

Edited August 20, 2017 by Crosland

[DCB]

Just been checking the cost of Lucas "12 volt" Zener diodes on ebay, anything from £3 to £1200!  High power zeners seem to be old hat and low power zeners as part of an electronic regulator seem to be the modern way.  No reason you couldn't use such a regulator on the outputs provided you had a bridge rectifier in circuit to protect it from reverse current.

I think Crossland has answered that fully. You could use a string of rectifier diodes to drop the voltage, you would need getting on for 17, but they cost next to nothing.

You could use a string of diodes to drop the voltage as I do with a 12 position rotary switch and diodes connected between the contacts so as to give full voltage, 10 progressively reducing voltage steps and an of position.  However these drop the incoming voltage, they don't limit the track voltage.  Using this switch in series with a resistance controller gives some very weird characteristics so wouldn't recommend it

 

 

Can you explain to me why it is bad practise to have two controllers off one rectifier unit please? I cannot see any problem with this as long as the rectifier unit provides enough current so that both controllers can provide maximum output without overloading the rectifier.

 

If its such bad practise maybe I should stop using my 35 Amp 50V rectifier unit that provides the 50v for my telephone exchange and provide each relay and selector with individual 50v supplies.

 

Andy G

The usual place you find this set up is old Analogue Scalextric sets and the two controllers off one power unit affect each other so much that backing off your cars power abruptly when your competitor is part way round a band will send them spinning off.  I use separate transformers for each controllers to avoid this.  This can also happen when some circuit controllers are piggy backed off transformer controllers as you speed one train up and the other slows.  Can also cause some same side dead shorts and I think if used on common return the wiring would simply short it out unless both trains were moving in the same direction.

Edited August 21, 2017 by DavidCBroad

[PatB]

TBH the use of zener diodes for motorcycle voltage regulation wasn't particularly brilliant. Not terribly reliable and needing to be mounted out in the airstream to keep the temperature under control. Not sure how one would go under a baseboard .

[Crosland]

Can you explain to me why it is bad practise to have two controllers off one rectifier unit please? I cannot see any problem with this as long as the rectifier unit provides enough current so that both controllers can provide maximum output without overloading the rectifier.

 

If its such bad practise maybe I should stop using my 35 Amp 50V rectifier unit that provides the 50v for my telephone exchange and provide each relay and selector with individual 50v supplies.

Your example has nothing whatsoever in common with a model railway layout.

 

If you run two controllers off one transformer winding then there is no isolation between the two controllers.

 

It can get interesting if there is a short between the outputs of the two controllers. Even a short across one controller may affect the other controller, depending on the nature of the "controller".

[DCB]

TBH the use of zener diodes for motorcycle voltage regulation wasn't particularly brilliant. Not terribly reliable and needing to be mounted out in the airstream to keep the temperature under control. Not sure how one would go under a baseboard .

The zener was actually totally reliable, unlike every other system of voltage regulation I came across it did not have any moving parts and did not restrict the amount of current below its operating voltage. I converted a lot of 1960s British 6 volt motorbikes to zener diode and capacitor 12 volt systems with decent resin encapsulated rectifiers in the 70s and 80s with a big weight saving dumping the battery and alternator secondary output switching and no problems.    

The only real issues with a Lucas Zener in train controller voltage limiting situations are how you connect it, before the reversing switch or through a rectifier after the reversing switch,  and the cost, new  the damned things are now very expensive.

 It should not get too hot on a normal 2 amp controller even on full output off load as they shunt 80 watts continuously in motorcycle applications,

 

Edit,  The problems of excess voltage spikes damaging motorcycle electrics is mentioned in a motorcycle magazine this month, I think it is something about sports bikes with a Blue Kawasaki H2 two stroke triple on the cover.  It mentions making a peak voltage tester using a rectifier (for AC) capacitor, resistor and DC scale on the Multimeter.  The Lucas Zener system minimised these spikes.

Edited August 23, 2017 by DavidCBroad

[kevinlms]

Can you explain to me why it is bad practise to have two controllers off one rectifier unit please? I cannot see any problem with this as long as the rectifier unit provides enough current so that both controllers can provide maximum output without overloading the rectifier.

 

If its such bad practise maybe I should stop using my 35 Amp 50V rectifier unit that provides the 50v for my telephone exchange and provide each relay and selector with individual 50v supplies.

 

Andy G

Sorry, you are totally wrong with your phone exchange example. I'm a phone tech here is Oz, so I know your argument is nonsense. Why, because such circuits have protection built in, so when one circuit (extension) goes faulty, it doesn't take down the rest. Those devices are usually called relays, and have relatively high resistance for that reason. 3000 type relays anyone? I wonder what 3000 could possibly stand for?

 

However as Crosland has adequately explained, if the same power supply is used, for two controllers, then a fault on one circuit will effect the other, the exact nature depends on the controller type in use. This is why Common Return cannot be used on a single power supply and is not recommended for for non common return. But perhaps you like mystery faults?

[AndyID]

 

This is why Common Return cannot be used on a single power supply and is not recommended for for non common return.

 

Sorry, but that's not correct. If both controllers are powered from a common transformer secondary winding, as soon as the controllers are switched for opposite directions, diodes in the bridge rectifiers attempt to short-circuit the transformer. Typically the diodes lose.