How do I convert 16v AC to 12v DC?

[Ruston]

Evening all,

 

On my 4mm layout, Nant Y Mynydd,  I have used Dingham couplings with DG electro-magnets; the electro-magnets have to be powered by 6-12v DC.

 

My loco controller (a Gaugemaster hand-held thing) and the point solenoids (also Gaugemaster) are fed with 16v AC and for this I have a mains transformer that gives out 16V AC.

 

To work the uncoupling electro-magnets I have a seperate 12v lead acid battery but this is cumbersome and needs a pair of untidy wires going to it on the ground.

 

I would like to be able to run the uncouplers using the same 16v AC transformer as the loco controller and points use, so is there a gadget available that can convert 16v AC to 12v DC?

 

If not, what is needed to make something to do that?

 

Thanks

Edited April 15, 2017 by Ruston

[Stubby47]

A bridge rectifier will convert ac to dc, but I'd have thought energising the electromagnet whilst also driving the loco would cause the loco to at least stutter due to the increased load.

An option is a cheap Hornby controller which has a 12v dc output.

[2mmMark]

I'd strongly recommend not using a 12v lead-acid battery, they can deliver a lot of current if shorted out, which a low resistance electromagnet will effectively do. I have a scar on my wrist and a hole in a steel watch strap from when I accidentally shorted out my bike's 48ah battery to the frame with my watch.

 

It's perfectly OK to use AC for electromagnets, I've been doing it for years. There's less sparking inside the pushbutton and it avoids residual magnetism building up in the couplers. I also feel that the slight oscillation helps overcome stiction in my chosen DG couplers.

 

I use a seperate transformer for my magnets, as Stubby says, running the magnets from the same transformer, even if on a different winding, will interact with the controller.

 

You could try using neodymium permanent magnets, moving them into position mechanically, like this example:

http://www.rmweb.co.uk/community/index.php?/topic/52042-tucking-mill/page-8&do=findComment&comment=1725281

 

Mark

[Ruston]

A bridge rectifier will convert ac to dc, but I'd have thought energising the electromagnet whilst also driving the loco would cause the loco to at least stutter due to the increased load.

An option is a cheap Hornby controller which has a 12v dc output.

Thanks, Stu.

 

I hadn't thought about the loco stuttering. Would it though? Shunting with a tiny little Peckett which will be running on a lot less than the 12v maximum?

 

I have just typed "bridge rectifier" into Maplin's website http://www.maplin.co.uk/search?text=bridge+rectifier&x=16&y=21 There's loads of them! I clicked on the first one and it gives the specification but it's all Greek to me. What exactly do I need? Then I presume the output from that will need to be brought down to a maximum of 12v by a resistor? How many ohms resistor will that need?

 

The idea of a cheap controller defeats the whole object because it's another cable to another mains socket. I want just the one cable to the one mains socket and the one cable from the transformer to the layout.

[Ruston]

I'd strongly recommend not using a 12v lead-acid battery, they can deliver a lot of current if shorted out, which a low resistance electromagnet will effectively do. I have a scar on my wrist and a hole in a steel watch strap from when I accidentally shorted out my bike's 48ah battery to the frame with my watch.

 

It's perfectly OK to use AC for electromagnets, I've been doing it for years. There's less sparking inside the pushbutton and it avoids residual magnetism building up in the couplers. I also feel that the slight oscillation helps overcome stiction in my chosen DG couplers.

 

I use a seperate transformer for my magnets, as Stubby says, running the magnets from the same transformer, even if on a different winding, will interact with the controller.

 

You could try using neodymium permanent magnets, moving them into position mechanically, like this example:

http://www.rmweb.co.uk/community/index.php?/topic/52042-tucking-mill/page-8&do=findComment&comment=1725281

 

Mark

Nice idea, Mark. Trouble is I have already fitted the electromagnets and I couldn't fit your design in anyway because of the way the layout is.

 

I didn't know that the electromagnets can use AC. I guess that I'd just need a reisistor then?

[bertiedog]

This circuit will work, the bridge rectifier rating should be as many amps as possible say 5A and any voltage from 20 volts upwards, the Capacitor is to smooth the DC a bit, rated at at least 25 volts or more. The capacitor is a bit optional for your uses.

 

 

No need for a circuit board, build it on the back of the bridge rectifier. They are usually marked with the AC and DC contacts on the case.

Maplins supply all of these bits needed but so does Ebay etc. The reason the DC is lower than the AC is the waveform averages out at a lower RMS value when rectified, so no need for any resistors. To protect things you could add a car type 12 volt  3 amp fuse in the output leads. This should not blow with normal uses, but will protect from a serious fault.

 

Stephen

[Pacific231G]

I'd strongly recommend not using a 12v lead-acid battery, they can deliver a lot of current if shorted out, which a low resistance electromagnet will effectively do. I have a scar on my wrist and a hole in a steel watch strap from when I accidentally shorted out my bike's 48ah battery to the frame with my watch.

 

It's perfectly OK to use AC for electromagnets, I've been doing it for years. There's less sparking inside the pushbutton and it avoids residual magnetism building up in the couplers. I also feel that the slight oscillation helps overcome stiction in my chosen DG couplers.

 

Mark

Hi Mark

I've heard this before but never understood it. Why doesn't reversing the current at 50Hz constantly reverse the polarity of the electromagnet to give an average of nought?

I'm guessing that's maybe  what does happen and the relevant part of the coupler, being unmagnetised, gets attracted to the magnet independently of whether it's seeing a north or a south pole. This must though mean that the coupler itself has very low hysteresis.

It would be different if it was acting on another magnet and I'm interest because I'm looking at a coupler design from the 1970s that used a permanent magnet on each vehicle to keep it coupled to the next one. Uncoupling was carried out by an electromagnet acting on part of the coupler to give it the same polarity as that from the permanent magnet so causing a repulsion. The  electromagnet in this case presumably would have to be DC powered. 

 

This would also explain why Tri-angs block instrument set could be powered by AC or DC as it used a couple of electromagnets in each instrument to drive the pointer from line blocked/normal either way to line clear or train on line depending on the position of the control in the other instrument. 

 

I agree about having an inline fuse with any lead acid battery. They can deliver a lot of power (why turning the key to start your car works a solenoid rather than directly switching the battery to the starter motor.  

 

 

Very interesting

Edited April 16, 2017 by Pacific231G

[BR60103]

A couple of observations.

The bridge rectifier drops the voltage; each rectifier in a series drops about 1.5 V.

Magnetic things seem to work OK with alternating current; in fact a transformer requires it. The old Lionel uncouplers all ran on up to 16V AC.

[Stubby47]

Dave,

I've now slept and realised I already do on Tinner's Forge what you want to do.

 

I have an old H&M Clipper controller which has both 12v dc and 16v ac outputs, as well as the controlled 12v dc for the track.

 

I use the 16v ac to power my Gaugemaster walkabout and Seep point motors, and the 12v dc for the Dingham uncouplers.

 

As TF is only small, there is no chance I'll need to power a loco and a point motor at the same time.

There appears to be no hesitation o the loco when the uncouplers are energised either.

And of course, there is only one mains power lead. I've also wrapped the two 16v & 12v leads together to form one cable and used coloured phono plugs & sockets to ensure correct connections.

Stu

[dhjgreen]

...The reason the DC is lower than the AC is the waveform averages out at a lower RMS value when rectified...

 

Stephen

Stephen, I am sorry but that is the wrong way round.  16V AC will give 22V on the Capacitor as it charges to the peaks of the AC waveform. 16V AC is 16V RMS, 22V Peak

[dhjgreen]

A 12V 2A wall wart for a fiver would do it.

 

That's what I use with my Dinghams.

Edited April 16, 2017 by dhjgreen

[2mmMark]

Hi Mark

I've heard this before but never understood it. Why doesn't reversing the current at 50Hz constantly reverse the polarity of the electromagnet to give an average of nought?

I'm guessing that's maybe  what does happen and the relevant part of the coupler, being unmagnetised, gets attracted to the magnet independently of whether it's seeing a north or a south pole. This must though mean that the coupler itself has very low hysteresis.

It would be different if it was acting on another magnet and I'm interest because I'm looking at a coupler design from the 1970s that used a permanent magnet on each vehicle to keep it coupled to the next one. Uncoupling was carried out by an electromagnet acting on part of the coupler to give it the same polarity as that from the permanent magnet so causing a repulsion. The  electromagnet in this case presumably would have to be DC powered. 

 

This would also explain why Tri-angs block instrument set could be powered by AC or DC as it used a couple of electromagnets in each instrument to drive the pointer from line blocked/normal either way to line clear or train on line depending on the position of the control in the other instrument. 

 

I agree about having an inline fuse with any lead acid battery. They can deliver a lot of power (why turning the key to start your car works a solenoid rather than directly switching the battery to the starter motor.  

 

 

Very interesting

 

I've never got into the theory, when I was frequently replacing electromagnet pushbuttons due to internal arcing, I simply swapped to using AC and fewer issues. Completely removed by using a higher rated pushbutton and packing it with Vaseline.

 

I'm currently using "PK" electromagnets, which have a low internal resistance. It would be better to have ones with a higher resistance. Coils from the old Post Office 3000 relays are very good but are somewhat bulky. These had 5 & 30 ohm options.

 

Shorting out the 48ah battery gave me a valuable reminder how batteries work!

 

Mark

[PatB]

I would have thought that thee electromagnets would be equally happy on DC or AC. However, as Stubby47 has said, a bridge rectifier will get you DC and, thanks to the inherent voltage drop across a silicon diode, will get the voltage down to 15.3 V or so. Assuming the magnet coils draw about 1A, a small resistor (10 Ohm or so should do it) in series will do the rest.

[dhjgreen]

I've never got into the theory, when I was frequently replacing electromagnet pushbuttons due to internal arcing...

 

Mark

You need a Back EMF diode across the supply, + to + and - to - much reduced arcing.  I forgot to mention that's what I use with my wall wart.

Edited April 16, 2017 by dhjgreen

[dhjgreen]

I would have thought that thee electromagnets would be equally happy on DC or AC. However, as Stubby47 has said, a bridge rectifier will get you DC and, thanks to the inherent voltage drop across a silicon diode, will get the voltage down to 15.3 V or so. Assuming the magnet coils draw about 1A, a small resistor (10 Ohm or so should do it) in series will do the rest.

That 15.3 V is still RMS so still nearly 22 V peak.  10 ohms at 1 Amp is a 10 Volt drop so would get you to about 12V peak; for safety's sake would need to be rated at 10 Watts though as that is the peak load.

[bertiedog]

With a plain transformer into a bridge rectifier the voltage output is lower than the AC input . The frequency is doubled, but the output is always lower than the peak AC voltage at an RMS value, plus loss through the pair of diodes in parallel, in practice about 12 volts DC.

This assumes no load.

Stephen

[bertiedog]

Stephen, I am sorry but that is the wrong way round.  16V AC will give 22V on the Capacitor as it charges to the peaks of the AC waveform. 16V AC is 16V RMS, 22V Peak

The example was not for a 22 volt AC input , but for 16 VAC peak.  Even if 22volt peak is assumed the output will be 14 Volts DC.

To be sure I have just measured an "HM16VAC" on a scope and it is 15 VAC no load, and after rectification reads 11 volts DC.

The circuit given will produce 12 volts from a HM unit or equivalent.

The peak voltage does charge the capacitor, but the charge stops on each peak and the load stops the capacitor reaching full voltage, this is the ripple voltage, and depends on the load. There is no stabilisation in the circuit, it needs a regulator to do that.

Sorry to go sideways from the post.

[cliff park]

Dhjgreen is correct. The peak value is 1.414 times the RMS. RMS is the normal quoted figure for ac devices. So when the op spoke of a '16v ac ' source this will have a peak of around 22v. Rectified AND SMOOTHED, no load, this will be the dc voltage.

 

EDITED to add:- If you rectify it, and don't smooth it, it is very hard to quantify the voltage. It is a pulsed waveform. An analogue meter would probably give you somewhere near an average reading, a digital voltmeter could give anything at all since it samples the voltage at intervals, and the instantaneous voltage could be anywhere between 0 and peak

Edited April 16, 2017 by cliff park

[Nick G]

I use these to power my 12v stuff from my 16cAC transformer.

 

16v AC to 12v DC Smoothed.

 

 

http://www.expressmodels.co.uk/acatalog/Regulated_Power_Supplies.html

[John ks]

I just checked a very old H&M Clipper Probably from the 60's or 70's

It uses a selenium plate rectifier

 

The AC Volts was measured at 17.22V AC

 

The DC side was 14.99V DC With no load 

 

These measurements are fairly consistent with Bertiedog,s measurements in post 17

 

I seem to remember measuring a Tri-ang controller with a capacitor across the DC terminals having an output voltage of about 20V DC 

This memory is consistent with Dhjgreen's statement in post 10

 

John

[bertiedog]

I seem to be pushed into a corner here, I am a trained electrical engineer with long experience and I can measure AC on a scope quite accurately. If I published the exact formula here it would be a massive turn off! RMS values are approximations to save calculations.

 

The H&M unit is marked 16 VAC and is very near that, but calculation off the scope showed 15 volts to be the effective voltage, and the output voltage from the bridge rectifier was 11.5 volts DC, measured on the scope and on a digital meter. There is ripple that reduces the effective no load voltage further, which rises with the capacitor in circuit, again no load. The final load reduces the voltage proportionately with current.

Unless the H&M I have are oddities, I have never seen 22 volts peak from one, it does vary a bit across the three units I have, but not that high.

__________________________________________________

 

On electromagnets AC works fine, the poles alternate at the frequency of the AC wave, and have a phase lag due to the build up on each pulse and the collapse of the field, till the reverse wave takes over. An iron core helps keep the magnet steady, as it gets saturated by the field, and maintains a pull despite the alternating wave form. AC is a bit more efficient at maintaining the field against pure DC which being constant is producing a bit more heat, and produces a saturated iron core with steady poles. When shut off the DC one will produce a back EMF kick, and if the core is iron or steel it retains some magnetism.

 

With AC, chance comes into shutdown, the wave form may be anything between 0 and full voltage, but the saturated field is still there, so on shutdown you get the back emf but it will vary in effect, and usually leaves the iron un-magnetised or with a weak field.

 

Generally if you are attracting steel to an electromagnet, then AC or DC will both work, and if you are attracting a magnet then DC is better, as AC fields may demagnetise the magnet. With DC you have the choice to attract or repel the magnet by reversing the coil.

 

What is a game changer are Neodymium magnets, which resist quite strong AC fields trying to demagnetise them. Simpler Alnico will suffer in the same field. An AC solenoid will over come the Neo magnet and attract it, despite the poles. Also a DC solonoid will overcome the repulsion of a Neo if they are close enough.  But it is safer to use DC to work with magnets, no risk to the magnets and repulsion or attraction works.

 

The idea of using magnets as couplers in pairs and then using an AC field to release them sounds interesting, but may have a flaw in that the alternating field would at any one wave only affect one magnet, and the other maintains the grip.

 

However if one wagon has a magnet on the end and the next has steel instead, then they would part in an AC field. They would not part in a DC field as all that happens is pole reversal, and steel will still be attracted.

 

So tiny neo magnets on one end of a wagon, and steel on the other in an alternating rake, sounds like a good experiment to try out. No way to get advanced uncoupling, but you could use a hand held electromagnet to wave over the coupling to release it for remote situations.

 

Stephen

Edited April 16, 2017 by bertiedog

[Ruston]

I use these to power my 12v stuff from my 16cAC transformer.

 

16v AC to 12v DC Smoothed.

 

 

http://www.expressmodels.co.uk/acatalog/Regulated_Power_Supplies.html

That's more like it! That's what I meant when I said "gadget". I didn't expect the Spanish Inquistion...

 

You see, I'm one of those people who is quite happy to use electricity but has no idea how it works.

[dhjgreen]

That's more like it! That's what I meant when I said "gadget". I didn't expect the Spanish Inquistion...

 

You see, I'm one of those people who is quite happy to use electricity but has no idea how it works.

A wall socket 12 V supply would be cheaper and self contained, as I said earler.

 

David Green, Diploma in Engineering (Electrical and Electronic)

Edited April 16, 2017 by dhjgreen

[Dagworth]

I just checked a very old H&M Clipper Probably from the 60's or 70's

It uses a selenium plate rectifier

 

The AC Volts was measured at 17.22V AC

 

The DC side was 14.99V DC With no load 

 

These measurements are fairly consistent with Bertiedog,s measurements in post 17

 

I seem to remember measuring a Tri-ang controller with a capacitor across the DC terminals having an output voltage of about 20V DC 

This memory is consistent with Dhjgreen's statement in post 10

 

John

There is no smoothing cap in the clipper. The cap will push the DC voltage off load to the PEAK voltage of the AC. (less diode voltage drop) Stephen's drawing shows the cap, therefore his values on the drawing are incorrect. 

 

Put a smoothing cap on the DC output of your clipper and measure again and see what result you get

 

Andi

[Suzie]

You just need a bridge rectifier. Only if you get a rattle from the magnets or couplings do you need to worry about adding the capacitor and resistor.