Laptop Power Supply

[RagingResistor]

Ok - so I have a train going around my room - this train has a semiautomatic uncoupler. The uncoupler is just a coil that when activated creates a magnetic field that pulls the knuckle couplers on the train apart thus disconnecting the cars. It needs 3 amps at 16-18VDC. I purchased a laptop power supply which provides 18VDC at 3 amps. I cut off the ferrite bead and hooked it up. It overheats the uncoupler immediately. I contacted the designer of the uncoupler and he said that it should not ever overheat at 3 amps 18VDC so he sent me a new coil and the same thing happened. There are no shorts - and everything is hooked up properly. I tested the supply with a multimeter and came up with 17.2VDC and varying amperage - it went all over the place from 2 to 6 to 1 to...... I thought the supply was bad so I tested my actual laptops power supply and it did the same thing - except I know this one is good because it charges my laptop everyday. I hooked the coil up to a train power supply that ran at 17VDC at 2 amps and the coil did not overheat but did not get enough power to pull the couplers apart. I thought maybe its the multimeter - so I tested the train power supply and got a steady 2.3 amps at 17.1VDC - so the meter works. Are laptop power supplies different that regular ones and if so what part of them creates this jumpy amperage so I can take it out?

ANY help WILL be greatly appreciated.

THANKS!

 

[shumifan50]

Is this a DCC train set?
Does the coupler get its power from the rails?
Is the coupler controlled with a DCC accessory controller?

I guess that somehow it is not DCC as then you would not need a separate power supply for the coupler. But how do you activate it?

If the coil is getting hot, it would seem like your laptop supply is not 18V DC. I suspect it is only half wave rectified(as my dell inspiron PS), so very 'bad' DC. Your train power supply, on the other hand, is full wave rectified and smoothed. What does the multimeter show if you put it on AC rather than DC.

You could try putting a very large electrolytic cap across the power supply and see if it improves the situation.

Unfortunately, here I am also out of my depth as I always use model train power supplies(I know they are expensive).

 

[(*steve*)]

Maybe you're supposed to supply a quick pulse and not continuously drive it with 18V.

 

[davenn]

I tested the supply with a multimeter and came up with 17.2VDC and varying amperage - it went all over the place from 2 to 6 to 1 to...... I thought the supply was bad so I tested my actual laptops power supply and it did the same thing - except I know this one is good because it charges my laptop everyday. I hooked the coil up to a train power supply that ran at 17VDC at 2 amps and the coil did not overheat but did not get enough power to pull the couplers apart. I thought maybe its the multimeter - so I tested the train power supply and got a steady 2.3 amps at 17.1VDC - so the meter works. Are laptop power supplies different that regular ones and if so what part of them creates this jumpy amperage so I can take it out?

and how exactly were you trying to measure the current ?

 

[(*steve*)]

I suspect it is only half wave rectified(as my dell inspiron PS), so very 'bad' DC

I suspect this is very likely to be bad information.

laptop power supplies are switchmode, and whilst they are half wave rectified (in a manner of speaking) this is fine for a number of reasons I'm not going into here.

 

[RagingResistor]

Is this a DCC train set?
Does the coupler get its power from the rails?
Is the coupler controlled with a DCC accessory controller?

I guess that somehow it is not DCC as then you would not need a separate power supply for the coupler. But how do you activate it?

If the coil is getting hot, it would seem like your laptop supply is not 18V DC. I suspect it is only half wave rectified(as my dell inspiron PS), so very 'bad' DC. Your train power supply, on the other hand, is full wave rectified and smoothed. What does the multimeter show if you put it on AC rather than DC.

You could try putting a very large electrolytic cap across the power supply and see if it improves the situation.

Unfortunately, here I am also out of my depth as I always use model train power supplies(I know they are expensive).

Yes its DCC
No it gets it from the laptop power supply
No
It is activated by a simple rocker switch

It shows as 0.0(AKA its DC)
Yeah I can try putting a capacitor there.

THANKS!

 

[RagingResistor]

Maybe you're supposed to supply a quick pulse and not continuously drive it with 18V.

Nope - the designer said it is meant for a continuous drive.

THANKS!

 

[RagingResistor]

and how exactly were you trying to measure the current ?

One probe on the negative one probe on the positive.

THANKS!

 

[davenn]

One probe on the negative one probe on the positive.

THANKS!

I suspected that
You CANNOT measure current across a PSU output like that

Current is measured with the meter in series with the load

cheers
Dave

 

[shumifan50]

Firstly, depending on your multimeter, setting it to measure current(amps) and connecting it directly to a power source will blow the fuse in multimeter(cheaper type of protection), damage the meter permanently(no protection cheapest meter) or cause it to disconnect. If the fuse is blown the meter will behave like you describe with values drifting around. NOTE I now because I have blown the fuse several times with measuring too high currents.

As I suspected, running AC through a DC solenoid will cause it to heat up. If it does not pull the rod in, it will heat up more quickly as it draws maximum current. Half wave DC will look like AC to the solenoid, so this might very well be your problem

See this
AC/DC solenoids

 

[(*steve*)]

As I suspected, running AC through a DC solenoid will cause it to heat up. If it does not pull the rod in, it will heat up more quickly as it draws maximum current. Half wave DC will look like AC to the solenoid, so this might very well be your problem

This would be excellent advice if he were not using a laptop power supply that produces pretty clean DC.

 

[shumifan50]

If you have a scope, could you have a look at the output from your laptop power supply, especially when connected to the solenoid?

There are only 3 possibilities for your problem that I can think of:
1. The voltage is DC but is not what you think and is too high driving too many amps through the solenoid.
2. The power supply is not outputting clean DC and is causing overheating as per my previous post.
3. The coil in the solenoid might react with the output stage of the power supply, but I don't know enough about this to comment.
@Steve: You are right, my netbook (Inspiron DUO) does provide quite nice DC viewed with a scope.

Just found an interesting bit of information regarding polarity and solenoids HERE:
So if your solenoid has a magnetic core, it will behave different based on the way around you connect the power supply.

 

[RagingResistor]

How can I find out its halfwave DC? My multimeter's fuse is good - I am measuring in series - and when I try to measure AC volts it comes up with 0.0. So how can I find out if its halfwave and if it is how can I fix it?

THANKS!

 

[shumifan50]

If your multimeter can measure up to 5A connect it in series with the switch set to measure current and check how much current is drawn when you engage the coupler (with the problem power supply). You might have to be quick to take the reading so you dont fry the solenoid.
Then also measure the voltage across the solenoid when you close the switch.Quick read again.
This way of testing is by no means desirable as it risks burning the solenoid, but will answer the current/voltage question.

The only way I know of determining if its halfwave rectification is with a scope. But as Steve said, even though I thought my netbook had halfwave rectification, it turned out to be nice DC. I already apologised for my misleading comment.