Jump to content

Recommended Posts

4 hours ago, DavidCBroad said:

I have several computer PSU's stockpiled in my scrap drawer.    I am pretty cavallier about electrics but no way would I use anything like a 5 amp supply anywhere near model railway equipment. I have had wires glowing on a lot less than 5 amps and I wouldn't risk any more than a 1 amp power unit for accessories except possibly for throwing points with 1950s point motors and momentary contact.   Several separate 1 amp wall warts plugged into a proper 220/240 volt ring main properly installed by someone who knows what they are doing, (not necessarily a professional electrician as one nearly killed my son by wiring a metal patress box to a live feed) has to be the better solution. Even better use model railway power units with a decent 1 amp cut out not a modern 2.5p  1 amp Polyswitch.    If you must use a 5 amp monster then 0.25 amp Polyswitches on the individual feeds should delay the onset of the fault which burns the shed down.

I don’t know what you’d think of our big O gauge DCC layout with 2x10 A and 2x8A switchers feeding it  !! 

Edited by Junctionmad
Link to post
Share on other sites
  • RMweb Gold
4 hours ago, DavidCBroad said:

I have several computer PSU's stockpiled in my scrap drawer.    I am pretty cavallier about electrics but no way would I use anything like a 5 amp supply anywhere near model railway equipment. I have had wires glowing on a lot less than 5 amps and I wouldn't risk any more than a 1 amp power unit for accessories except possibly for throwing points with 1950s point motors and momentary contact.   Several separate 1 amp wall warts plugged into a proper 220/240 volt ring main properly installed by someone who knows what they are doing, (not necessarily a professional electrician as one nearly killed my son by wiring a metal patress box to a live feed) has to be the better solution. Even better use model railway power units with a decent 1 amp cut out not a modern 2.5p  1 amp Polyswitch.    If you must use a 5 amp monster then 0.25 amp Polyswitches on the individual feeds should delay the onset of the fault which burns the shed down.

Hi David, my plan was to use max 1 amp inline fuses to protect each of the I individual sets of accessories/branches. I have 5 control panels, 6 switch panels with 12v relays and 14 MERG 12v servo 4 boards, so using multiple wall warts sounds like it's getting very busy. Hence the concept of the ring main and branches off. 

 

Working back from my power source at Hinkley Point, I will have the following:

1. Normal household 13 amp fused supply with RCD

2. Railway shed has a separate modern consumer unit again with RCD. This is connected to the house by armoured cable.

3. PSU Mains plug has 3amp fuse

4. 2amp inline fuse on PSU 12 volt line

5. 0.8amp (TBC) inline fuse on individual branches

6. Shed can be electrically switched off from the house.

7. Plan is for the railway to be switched off after each running session.

 

 

 

 

 

Link to post
Share on other sites

 

Quote

There's no fancy circuitry about a computer PSU. They are just another switched mode power supply which are  readily available for all sorts of supply voltages

Modern PC low voltage psus are actually quite complex , as any perusal of their schematics will testify 

  • Agree 1
Link to post
Share on other sites
15 minutes ago, Nearholmer said:

Inside a PC I get the logic of doing it the way it’s done, and I would surmise that each of the sub-assemblies being fed incorporates local electronic over current (and possibly direct thermal) protection.

 

But, I thought this unit was being sold as a general purpose PSU for use outside of a PC too. Is that not the case?

I would hope they aren't being sold as general purpose PSUs.

I've had several from superceded PCs and I just dismantle them for useful parts.

I would never use one on a model railway, they can fail quite dramatically, then you precious kit is fried.

  • Informative/Useful 1
Link to post
Share on other sites
1 hour ago, ikcdab said:

Hi Andy, that's a good point, but I am using MERG servo4 drivers which need 12v. I know they are developing a 5v version, bit I already have 14 of the 12v versions ready to go. I have also installed 12v relays for polarity switching. So it's just seemed best to use 12v for everything.

 

As the voltage reduces, so voltage drop becomes more critical. For sensitive electronics it's better to distribute at a higher voltage and use local regulation, just as the MERG servo drivers do.

  • Agree 1
Link to post
Share on other sites
5 minutes ago, Junctionmad said:

 

Modern PC low voltage psus are actually quite complex , as any perusal of their schematics will testify 

Depends what you call complex I suppose, but they are as I said, just a switched mode PSU, they are complex because of the number of outputs, the "switch-moding" is nothing special these days.

They can produce prodigious amounts of power from small spaces at high efficiency, which is very imporatant.

 

Back in the '80s I used to work on equipment which used SMPS and had to fix the (often) faulty ones.

In those days the switching transistors were TO3 devices and the 450v reservoir caps were enormous

They had a ceramic pot core about 35mm dia and could provide (IIRC) 40A @ 5v to supply TTL logic.

I have known a fibreglass PCB to be glowing red hot where the current had bridged faulty circuitry and carbonised the board.

 

Link to post
Share on other sites
  • RMweb Gold
31 minutes ago, melmerby said:

it's 5A fast blow but the cable is rated way above that.

It feeds the  LDT occupancy detectors & turntable decoder plus some other 15v accessories.

 

Hi Keith, thanks. Numpty question: if the fuse is rated at 5 amps, doesn't that mean that 4.9amps could flow before it blows? Isn't that the thing I've been warned about? If my servo drivers etc add up to (say) less than an amp, wouldn't I then use a 1amp fuse?

Link to post
Share on other sites
  • RMweb Gold
5 hours ago, AndyID said:

As for a ring, don't waste your time. It's only going to make it much more difficult to identify any problems. (Very few countries ever adopted the "ring mains", for good reason.)

Ok thanks. Like any ring, it won't be a single bit of cable, but shorter sections joining junction boxes. So I guess I'll work out from the supply, working in both directions, and see how it goes, seeing what voltage drop I get. Each leg shouldn't be more than 9m or 10m.

We'll see.

Ian

Link to post
Share on other sites
1 minute ago, ikcdab said:

Hi Keith, thanks. Numpty question: if the fuse is rated at 5 amps, doesn't that mean that 4.9amps could flow before it blows? Isn't that the thing I've been warned about? If my servo drivers etc add up to (say) less than an amp, wouldn't I then use a 1amp fuse?

Fuse technology is quite complex and diifferent types react differently.

There are three main types you might encounter.

 

A normal fuse should be able to carry it's rated current all day, It will blow when that is exceeded, to blow fairly quickly it would need up to 2 x it's rated current. lesser amounts will cause it to take longer to blow.

 

A fast blow fuse will blow much quicker once it's rated current is exceded.

 

Anti-surge fuses allow a large inrush of current (way over the rated value) for such devices that need it without blowing but will blow when a continuous overload happens.

Link to post
Share on other sites
  • RMweb Gold
18 minutes ago, melmerby said:

Fuse technology is quite complex and diifferent types react differently.

There are three main types you might encounter.

 

A normal fuse should be able to carry it's rated current all day, It will blow when that is exceeded, to blow fairly quickly it would need up to 2 x it's rated current. lesser amounts will cause it to take longer to blow.

 

A fast blow fuse will blow much quicker once it's rated current is exceded.

 

Anti-surge fuses allow a large inrush of current (way over the rated value) for such devices that need it without blowing but will blow when a continuous overload happens.

Hi Keith, thanks.  So your 5A fast blow would allow (say) 4.9A to happily flow round your circuit unmolested?

Ian

Link to post
Share on other sites
18 hours ago, ikcdab said:

Very wise,  thank you.  Now I don't know what size fuse.  Info online suggests my 9g servos consume 200ma running  and 6ma at idle and I'll have 30 or so. I only intend to activate max 5 at a time,  mostly 2s and 3s.  So 5 x 200ma max power?

My leds are also roughly 20ma, say there's 40 of those,  so 40 x 20ma.

So does that mean a  2amp fuse,  or have I got that wrong? 

Thanks

Ian

 

Never forget that the fuse is there to protect the cable, not the equipment hanging off of it!

 

It's easier to replace a fuse than it is to replace the cabling when it decides to take on the role of a fuse!

Link to post
Share on other sites
3 hours ago, ikcdab said:

Hi Keith, thanks.  So your 5A fast blow would allow (say) 4.9A to happily flow round your circuit unmolested?

Ian

Yes, but would blow quickly once over a pre-determined current figure as one of the fuses' characteristics.

It might say something like 20% overload in 5 seconds, 50% overload in 1 second, 100% overload in 100mS etc.

 

Fuses have spec sheets, just like other components, designers would normally choose a fuse specification based on thir requirements.

 

1 hour ago, BokStein said:

 

Never forget that the fuse is there to protect the cable, not the equipment hanging off of it!

 

It's easier to replace a fuse than it is to replace the cabling when it decides to take on the role of a fuse!

Hence my 5A circuit using more than adequate 1.5mm (15A) stranded cable.

The individual devices hanging off the supply will fail in their own way long before the cable is affected, usually a low rated component that will go open circuit once too much current passes.

Link to post
Share on other sites
2 hours ago, BokStein said:

 

Never forget that the fuse is there to protect the cable, not the equipment hanging off of it!

 

It's easier to replace a fuse than it is to replace the cabling when it decides to take on the role of a fuse!

Or, the equipment is there to protect the fuse!

 

That's why choosing the protection ratings is so important. You don't want a fuse blowing all the time, otherwise eventually someone will replace it with one that won't blow.

 

A friend of was told to get some globes to protect his kit train controllers. He did, but got some 50 Watt QH ones, and wired them in. But the controllers only handled about 1.25 Amp. He was advised to change them for car brake light globes, but told us that he already had them, so was going to use them.

So we had weekly running nights and for a few weeks, he averaged one dead controller a night! Not a happy man, as he had to keep sending them to his friend to repair!

Link to post
Share on other sites

I suspect that the thing that attracts people to computer PSUs is that they are, effectively, free. Everyone loves free stuff. 

 

Given a sensible approach to overload protection on the output(s), I really don't see a problem with them in principle. But, then, I grew up reading old back issues of both modelling magazines and stuff like Practical Mechanics, and some of the electrical practices shown in those make pretty much anything not likely to be immediately lethal look pretty safe and responsible. 

Link to post
Share on other sites
8 hours ago, melmerby said:

There's no fancy circuitry about a computer PSU. They are just another switched mode power supply which are  readily available for all sorts of supply voltages

If you wanted just12v DC a dedicated 12v SMPS would be a lot smaller than a computer PSU.

 

But they have to be seperate wires in a PC as each one goes to a specific item via a suitable plug/socket combination, be it HDD, graphics card, motherboard etc. All have various combinations of supply voltage requirement

 

Well I regard it as fancy circuitry compared to a transformer and rectifier.  I use 12vdc for a lot of things so I use a 5A one  and it is very similar to a computer one. You have chosen to use a 16vAC 5A bus. I use a 12v 5 A one. I also use a 15V DC 3A one for DCC, just two. For lower voltages you can also get Switched mode regulators to fit on a PCB. 

 

David there is no great problem using a 5 A supply at 12v and the less power connections you have the better in my view as it is easy to leave one on when you have individual ones. It can also minimise the amount of wiring under the layout.  It is very helpful to have something powered by the 12v Bus that will show a lit led at all times so you don't forget to turn it off.

As for the ring main you are of course aware that a disconnection at one end of the ring will not show as a fault  but will leave a cable with a 32A fuse which is not rated to carry that current . With a radial circuit the fuse rating must be lower than the minimum rating of the cables used. 

I would consider leaving the soldering iron on probably constitues the biggest danger  especially those ones with a spring like wire holder that seem unable to hold the iron securely.   

 

 

Link to post
Share on other sites
7 minutes ago, Donw said:

 

I would consider leaving the soldering iron on probably constitues the biggest danger  especially those ones with a spring like wire holder that seem unable to hold the iron securely.   

 

Or the Maplin one where the plastic socket that holds the iron in place just slowly disintegrates because of the heat.

Or the cheapo market stall irons where the insulation resistance between the element and metal case is too low and trips any sort of RCD.

 

 

10 minutes ago, Donw said:

 

Well I regard it as fancy circuitry compared to a transformer and rectifier.  I use 12vdc for a lot of things so I use a 5A one  and it is very similar to a computer one. You have chosen to use a 16vAC 5A bus. I use a 12v 5 A one. I also use a 15V DC 3A one for DCC, just two. For lower voltages you can also get Switched mode regulators to fit on a PCB. 

 

My PSU consists of high grade toroidal transformers, which have better regulation than ordinary ones.

It is AC because it feeds a Lenz system (first output) & accessories (second output), nothing needs DC . I suppose I could use DC as they would all still work.

Link to post
Share on other sites
On 17/09/2020 at 09:49, melmerby said:

Depends what you call complex I suppose, but they are as I said, just a switched mode PSU, they are complex because of the number of outputs, the "switch-moding" is nothing special these days.

 

I don't know about the latest spec but I'm pretty sure that early PC power supplies were designed down to a price and had regulation on only one output voltage, which required a minimum load (5V in those days). So long as that output was in spec, the other outputs would be too.

Link to post
Share on other sites
2 hours ago, Crosland said:

 

I don't know about the latest spec but I'm pretty sure that early PC power supplies were designed down to a price and had regulation on only one output voltage, which required a minimum load (5V in those days). So long as that output was in spec, the other outputs would be too.

Seems logical

The only SMPS I worked on (communication test equipment) regulated the 5v by varying the mark/space ratio of the switching waveform and the other voltages fell into place as they were just extra secondary windings on the same transformer.

Link to post
Share on other sites

Hello Team, a question for the collective, if I may.  I also have been thinking about a ring circuit, purely for locomotive operating purposes, with any accessories on a purely separate feed or circuit. 

 

This proposed ring circuit would have droppers to maintain connectivity, and all soldered joints.  The proposed extra cost element is of little concern; I would like to achieve good operability. 

 

Have I got this right? Any insights or observations are, as useful, gratefully received. 

 

Cheers,

Ian.

Link to post
Share on other sites
On 18/09/2020 at 16:34, Crosland said:

 

I don't know about the latest spec but I'm pretty sure that early PC power supplies were designed down to a price and had regulation on only one output voltage, which required a minimum load (5V in those days). So long as that output was in spec, the other outputs would be too.

This must be why a lot of DIY instructions for using a PC PSU as a power supply suggest a high wattage resistor with a heat sink connected across the +5v and 0v rails.

 

Link to post
Share on other sites
3 hours ago, tomparryharry said:

Hello Team, a question for the collective, if I may.  I also have been thinking about a ring circuit, purely for locomotive operating purposes, with any accessories on a purely separate feed or circuit. 

 

This proposed ring circuit would have droppers to maintain connectivity, and all soldered joints.  The proposed extra cost element is of little concern; I would like to achieve good operability. 

 

Have I got this right? Any insights or observations are, as useful, gratefully received. 

 

Cheers,

Ian.

 

A ring circuit works fine unless you want to have switchaable sections. If you are using DC Cab control the ring will only be useable if the section switches are placed locally and can be between the droppers and the ring.

For DCC you would expect to run a bus anyway. I cannot see an issue with connecting the ends of the bus to make a ring.

 

For some network types such as CAN where terminating  resistors are used  it is normall to replace the two ends ones with a single one of half the value.

Do make sure the ring can carry the full load current without excessive volt drop.

 

Don.

  • Thanks 1
Link to post
Share on other sites
9 hours ago, tomparryharry said:

Hello Team, a question for the collective, if I may.  I also have been thinking about a ring circuit, purely for locomotive operating purposes, with any accessories on a purely separate feed or circuit. 

 

This proposed ring circuit would have droppers to maintain connectivity, and all soldered joints.  The proposed extra cost element is of little concern; I would like to achieve good operability. 

 

Have I got this right? Any insights or observations are, as useful, gratefully received. 

 

Cheers,

Ian.

 

Hi Ian,

 

As long as the wire you use is rated for the maximum current your power supply can produce you should not have a problem but if you use wire that is not rated for the total maximum current and you rely on the ring to split the current between two conductors you could have a problem if there is an unknown break somewhere in the ring. In that situation you could be exceeding the wire's maximum current rating.

 

Incidentally, that's why ring-mains systems are only used in a few countries.

 

Cheers,

Andy

  • Agree 1
  • Thanks 1
Link to post
Share on other sites
  • 3 weeks later...
On 17/09/2020 at 18:12, Nearholmer said:

But, I thought this unit was being sold as a general purpose PSU for use outside of a PC too. Is that not the case?

Hi, the ATX power supply is sold as a replacement, upgrade or repair for a PC.  It is not intended as a general power supply and does not work unless wired correctly to simulate the internal switch and power good connections on a PC mother board.  The internet has loads of instructions on how to convert them for general use.  Some are dangerous even opening the box and installing switches etc inside the case.  Fascia panels with sockets and external connectors are also commercially available to do this conversion.  Some also suggest combining the +12V and -12V outputs to get 24V - (not sure how this works with the different amperage).

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.