Unknown Farnell Power Supply Questions

[tvtech]

No Sweat

Regards,
tvtech

 

[throbscottle]

Hey tvtech - I never ignore your posts - they are too entertaining! Seriously though, you are a guy with a world of experience we can all learn from - certainly doesn't go unappreciated.

I used to regularly raid a skip (aka dumpster) at the back of a TV rental place where I grew up and take home various goodies to strip for parts, and there was a particular TV (don't know make or anything) that had a SCR mounted on a heatsink made of a big but fairly thin piece of aluminium, with just one bend in it. I figured out it was part of the psu, but never found out how it worked (and never bothered to trace the circuit so I could find out, either). Wish I had, now.

I even managed to use one of those SCR's I collected in a flash/strobe I designed. No internet in those days so limited for sources of circuit ideas and people to put me right where I went wrong. Internet Luuurrrvvvvlleeeyyyy

Hoiler - it's just a fact of life, anything technical you generally need at least a basic understanding of it to be able to modify it (unless you are just following some instructions of course), or at least an understanding of the bit you want to modify. I'd been hoping you'd pick up the baton on that one sooner, my friend

 

[Hoiler]

I guess so. I was just hoping that there would be a way to change the voltage out without tinkering about inside the supply.

The eBay user got back to me, and said just attach a 50k variable resistor to terminals labelled 'prog' and '+'. this allows me to adjust the voltage out.

I'm not sure though if this will allow me to draw the full 600W at 12V. i imagination not. you mentioned the possibility of a 48V to 12V converter and that there are plans around for it, could you point me on the direction of one please?

 

[throbscottle]

I would try the 50K pot idea first - no reason I know of why it should not give full power if it works at all. I hope it does

It wasn't me that mentioned 48 to 12v converter - what I said was you could build a linear regulator with that supply as it's input (maybe I didn't phrase it too well). This is a traditional inefficient design though, you would be dropping 36v across your pass element, which is going to need some very serious heatsinking if run anywhere remotely near the current this baby can deliver - it would be chucking away 3 times the amount of energy you are actually using. Ok it was a rubbish suggestion, but like many suggestions it seemed a good idea when I suggested it!

600W is an awful lot of power - what did you have in mind that might draw this much, anyway? At 12V that's 50A!

 

[Hoiler]

i have just tried it with the variable resistor - i had a small 50K one, with that i get about 2V to 50V. Very pleased with that. However it is way to small. I popped down to Maplin and bought a 47K one, (that can be mounted onto the side of the case) , the closest thing to 50K they had, and i get a range of about 4V to 40V. Interesting where the missing 40-50V go but after all i don't' really mind as 50V is way to high for anything i would want to use it for.

As you said, 600W is alot to play with. At 12V that is 50A. The reason i want so much is i want to use the supply in a sorta DIY UPS backup system. Not entirely sure how i will set it up, but i have a 600W 12V to 240V inverter, the idea being that the power supply charges several 12V batteries all connected in parallel, and off them runs an inverter. When mains is on the power supply runs the inverter. When mains fails, the batteries take over. As the inverter is rated at 600W that means it will draw up to 50A into it. not that i think i will ever use the 600W fully.

This is all just a rough idea that i have, but in the end after all what is there not to like about a variable 0-50V DC Power supply rated at 600W? i'm going to have fun with this i can see it now, cranking up LEDs until they go bang.. hm..

The one thing i wonder about is that if the front output terminals can handle 50A or not.

this is how the 12V version looks. a bit more beefy huh?

Anyways i dont' think i will be drawing 50A for long periods anyway.

 

[throbscottle]

Hooray! All without having to understand what goes on inside! Still, a learning experience anyway, eh?

You are right, the output terminals are only rated for the current of the supply at 48V, and whatever wiring internally goes to them will be similarly smaller, also, whatever components are feeding into that will also be de-rated in the same proportion. So be very careful when you drop the voltage, if indeed it can sustain the same amount of power. You don't want to fry anything, power semiconductors are expensive, and obsolete semiconductors may be impossible to replace. Same goes for the power transformer too.

You should look up wire ampacity on the net. Very educational subject.

Your psu/battery/inverter scheme doesn't require a beefy psu. The charging current for a lead-acid battery is only a few amps, and you can limit the current anyway. They can deliver hundreds of amps, which is what a starter motor takes to crank an engine, so no problem powering your inverter there. With suitable switching you can pass the mains straight through for normal operation, though I suspect it's rather more complicated than it first appears - at first thought you have a simple switch to kick in the inverter when the mains fails, but actually that still causes a momentary power loss. If you wanted a truly uninterrupted supply you would need the inverter running all the time, but with it's output in perfect phase with the mains, so the voltage would be present from it but the current draw very low - it would ramp up when the mains failed, making it much quicker than having to be switched on. I'm sure the commercial units have a much better solution...

There are some very good threads here on eto about charging of batteries, though the discussions concerning lead-acid mainly refer to the sealed type.

You would do well to read here: https://batteryuniversity.com/

 

[Hoiler]

the idea for the power inverter thingy was the inverter was always powered, always being used for electricity.

but that doesn't matter really now as the ebay user got back to me, and he says that they can only handle the max current 12.5 A no mater the voltage. So yeah. Not bad though. a 0-50 DC power supply offering 12A. lovely.

Thanks for your help on this one!

Now - my next project - a Microwave Oven Transformer. Don't' worry i don't' intend to be messing about with 2KV or anything dangerous like that. i've already started striping off the secondaries. Firstly, i want to have a go and rewind it for some extreme high current fun - like 600Amps at 1V or something. Then after that i will rewind it again for a 12V supply at crazy amps like 50A i guess. but thats for a later project, so what do you think?

 

[throbscottle]

Yeah that's still a lot of juice - I really am very envious.

Microwave transformers. Look on instructables for loads of projects like that. I made myself a spot welder using a MOT, though other people have obtained better turns ratios I think I figured out the best way to wind the high current secondary, like this:

https://www.instructables.com/id/Spot-Welder-from-a-microwave-oven-transformer-and-/?ALLSTEPS

You see some where people are melting nails and stuff... Mine was too high a voltage at nearly 5v - should have been closer to 3v.

Oh if you're doing really high currents, you really do need to read up on ampacity - this is a start: https://wiki.xtronics.com/index.php/Wire-Gauge_Ampacity#Load_Carrying_Capacities_or_Ampacities.

I think it's a good exercise to rewind it to give you a voltage you can actually use for a linear supply - you'd get really good regulation because it's so big (someone please correct me if I've got that wrong!) Oh but wait you've already got a HUGE psu anyway. Hmmm

If you want to know how to work out how many turns you need, just wind a known number of turns (like 10) of ordinary wire onto the core, power it up and measure the voltage. Divide it by the number of turns and you will have the number of volts per turn. (Probably a little more than 1). Round it to 1 or 2 dp's to get a useful number. Don't forget the DC voltage you will ultimately get from rectifying and smoothing the output will be around 1.4 times the rms voltage (ie, the peak voltage), less the diode drops of the rectifier, which will be between 0.6 and 0.7 volts per diode (double this for a bridge rectifier), but then when a load is applied the voltage will drop due to a load of stuff I never learnt, but this is where the "regulation" of the transformer comes into play.

Don't fry yourself by accidentally touching the mains side. I was careless once, nearly killed myself, being alone in my damp basement flat...

Since you'd be able to draw quite a lot of current you might want to think about connecting rectifiers in parallel - if you do this you need low value resistor in series with each one. Same goes for pass transistors - you can connect them in parallel, just put a low value resistor in series with the emitter of each one.

You also have the opportunity to create a tapped winding with all the fancy voltages you could wish for. Make a centre tapped winding and you only need 2 rectifiers and only have 1 diode drop.

Read this superb tutorial: https://www.electro-tech-online.com/threads/power-supply-design-for-beginners.113613/ explains it far better than I can.

If you know the exact mains voltage you can then work out the number of turns on the primary, so you would then be able know the exact turns ratio. I have no idea why you would want to know this however... I'm ranting now...

Good luck, have fun and be safe

 

[Hoiler]

finaly, after lots of hammering and hacking the old secondaries are out. Unfortunately the transformer nearly fell apart, and i scratched the primaries in several places However i cleaned up where the primaries were scratched, made sure there were no shorts and covered with epoxy. Then glued the transformer back together. i just hope there was no lasting damage - i hesitate on plugging it in just yet because i think it will just trip the MCB immediately, and i need to wait for the glue to dry anyway. What would happen, exactly if i plugged the primaries in to mains without any secondaries?

 

[tvtech]

Hi Hoiler

Much the same as me Today......except without your additional expense where parts blow...and you canno't replace them.

I would sit back and wait a while. No rush. I will help where I can. And speak when I know

Regards.
tvtech

 

[tvtech]

Hi Hoiler

Much the same as me Today......except without your additional expense where parts blow...and you canno't replace them.

I would sit back and wait a while. No rush. I will help where I can. And speak when I know

Regards.
tvtech

 

[throbscottle]

If there is no secondary nothing will happen, except maybe it'll hum a bit. I think the laminations are usually welded together on 2 sides - hard to damage! Just hope, as you say, you haven't made a short on the secondary. I masked mine with cardboard and wrapped it with tape to protect it, still scratched it in one place.

In what way did the transformer nearly fall apart? If it's a couple of metal slugs from between the windings, they are magnetic shunts, they are to protect the transformer from going into saturation if it's overloaded, and not strictly necessary. I left them out of mine, but there again it's only for seconds at a time.

 

[Hoiler]

I didn't short the secondaries! i tested with a [vintage] 200W light bulb and the transformer made a satisfying 50hz humming noise. i'm not quite sure what happened to mine to make it fall apart but a bit of epoxy and some claps overnight and it looks fine now.

I've got some really thick cable now, you know, the stuff that goes from the Meter to the fuse box. i've Wound it first with a slightly thinner earth cable and had a play with shorting it out. I'm going to need a better clamp meter - mine only measures up to 400A and each time it maxed it out. i measure about 2.5 V on the secondaries unloaded. Nice instructable by the way Original idea to use the copper from a water tank.

ill let you know how it goes

 

[throbscottle]

Sorry, typo - meant primary. Glad it's going well. Someone else used braided cable of the sort that's used to connect big earth terminals, or lightning conductors. It's more flexible.