Hi guys,
I have a transformer that is reading 28.1 volts AC output from secondary. When I rectify this through a bridge I am getting 38.4 volts DC across the 5000uf capacitor.

As a somewhat newbie, this doesn't seems correct to me. I am thinking that I should be getting about 27 volts DC.

Could someone please help me understand this.

Thanks

 

With no load, the capacitor is charging to the PEAK value of your transformer's 28.1 RMS voltage. Conversion from RMS to PEAK is RMS X 1.414 which, in your case, should be 39.73 volts. Since you are running a bridge rectifier, we will assume silicone diodes and each diode will have a voltage drop of approximately 0.7 volts. Two diodes conduct at a time in a bridge, so total bridge loss will be approximately 1.4 volts. 39.7 - 1.4 =38.3 volts. Pretty close to your 38.4 volt reading.
Hope this helps.

 

Thanks Bill3857,
I did get that calculation myself but it seemed too high.

As my output transformer will give me 30+amps, should I look at using a voltage regulator to stabilise the voltage rather than depend upon the output load to vary the output voltage?

Once again thanks

 

If you wanted 27 Volts with over 30 Amps that will be a very serious regulation project.

11 Volts dropping 30 Amps is about 330 Watts. It will be a real space heater.

The simplest way to drop those volts might be 30 Amps worth of Automotive lightbulbs to soak up the power all attached on the collector of a Power Transistor Linear Regulator.

That 330W is a horrible waste so you will be told to use a Switching Mode Power supply. Using a Coil to drop the voltage will let you get much of the energy from that drop back because the coil just stores the power in a magnetic field flux that it gives back after the power is switched off again.
It is transformer action even though it often uses a single coil.

Your transformer for 30Amps is massive because it works at 60 Hz line frequency. The Switchmode Coil to handle 30 Amps will be much smaller and cheaper but it will only work at much higher frequencies where its reactance is high enough to control the current.

A switch mode has a reputation for being tricky. Get a good toroidal coil, the right Driver IC, use more than enough capacity in Switching Transistor, heat sink things so well that they don't even start to heat up, wiring it all carefully so that the high frequency switching is not impeded or creating noise. That might seem like a lot but it really isn't.

A Linear Regulator is a peice of cake to set up in comparison but because of all the wasted power it can have reliability problems. I suggested that you put about 360 Watts of bulbs on top of any linear regulator as a way to radiate that waste power and keep it from making the regulator a nightmare to keep cool enough so that the transistor would not smoke. You would still be talking a much more expensive transistor than for switcher.

 

Hi there Potato Pudding,

Lots of information there. I picked this transformer up on ebay for $30. It seemed like a good buy!

I was hoping to use it to make a 12 and 24 volts power supply. Now I have too much energy. As you say, I should run a heater! I am mucking around with DC motors so as they can be hungry it seemed like a good buy.

I shall have to investigate switch mode power supplies. I have only heard of them.

Thanks and regards

 

If you are simply running motors, have you thought of using 1/2 wave rectification instead of the bridge? That will give you a lower effective output voltage but more ripple.

 

$30 is not bad for that size of transformer but it should be heavy enough that the shipping alone would be about $20.

 

One way around dissipation problems is to run a preregulator so the linear pass transistor can run very close to saturation - http://www.seekic.com/forum/22_circuit_diagram/17620_SCR_PREREGULATOR_FITS_ANY_POWER_SUPPLY.html

 

I like that. That could be an excellent system especially if big capacitors are being used and the 5000uF that the OP mentioned would qualify. SMPS are meant to be so responsive with their high frequency that they don't need a big capacitor. If he has the parts and doesn't mind the bulky build then this should be a good brute force option.

It seems simpler than a SMPS and would allow a Linear Regulator without too much waste power.

The ripple voltage being so high could shorten the life of the capacitors.

 

Hi guys,

This is great stuff. With Beenthere's circuit am I correct in assuming that the preregulator is placed before the input to the Primary winding? If so, would my current remain the same on the secondary? this is a complicated circuit for me to work out at the moment.

Hi Potato Pudding, it gets better with the transformer, it was a local pick-up! How often does that occur!

Regards

Edit, wrong above, I can see that it follows the output so what percentage of power is available?

 

am I correct in assuming that the preregulator is placed before the input to the Primary winding?

No. The preregulator is used to control the voltage on the filter capacitors. The circuit would replace the full wave or bridge rectifier.

 

I believe many of you are overlooking something, a transformer driven power supply cicuit will come out as the peak voltage when rectified (less the voltage drop of the rectifiers as someone noted) when you aren't drawing any power from it, yet once you put a decent load on the circuit the voltage will drop towards the actual transformer rating.

There are formulas out there for estimating this, no time to seach at present as I've got to get to work, but for example I use a lot of 120 to 24 VAC transformers in various project power supplies. Rectified and filtered I get the expected ~34V but once I draw at least half the current rating of the secondary it drops down much closer to the 24V you would expect.

This of course depends a bit on how large a filter cap you use, the larger the closer to the rated output you need to pull.

BTW: The reason I use so many of these is I can pick up a 120V - 24V @ 2A transformer for about $5 from several of our local heating and air parts suppliers, they use them as thermostat drivers. You just can't beat $5 and no shipping.

Fully rectified and under no load your output will be right on the edge of what a 78xx series IC will take for the input, and if you need to drop way down to 12 or 5 VDC you'll need a big heat sink on the regulator IC. I tend to use two regulators, the first one being an L7824CV (24V @ 1.5A) to take some of the edge off then whatever 78xx I need after that.

Another trick is to stick one or more 5W sener diodes in series with the unregulated supply line before you hit the regulator but it's advisable to add a resistor to ground right after the diodes to make sure at least the minimum zener current is being drawn through them should the following components possibly subject them to a no load situation.

While I do prefer to use a PWM controlled MOSFET regulator setup in many of my circuits I build so many things that I just want done quick, easy and cheap.

 

I don't really disagree with anything you said marshall but take another look please.

The drop down towards RMS under load was not really overlooked.

It is just going to take a lot to drop down a 30Amp transformer. If it were a 300 millamps transformer - you could just shunt it down. I just don't think that is going to be a good option at 30 Amps especially not with 5000uF filter capacitors.

The OP even asked if he could count on it dropping down under load. You would almost have to drop a crowbar across it so natural drop is not an option.

I think you missed the current capacity or else you would have specced that the 78xx would need a parallel Power Transistor and probably some series power resistors (another option to the lightbulbs). Getting a 78xx to regulate for 30Amps is more complicated than putting it between the Vplus and the load. You can't more than a few watts out of most 78xx regulators no matter how well you heat sink, and this is a case where several hundred watts that need to be switched or lost.

 

Small monitor and questionable eyesight here. I read the original post then just scanned the rest, bet I thought he was dealing with 3.0A.

A linear regulator could be built using several MJE3055 transistors (+separate emitter resistors and a huge heat sink) but unless you really need that 30A for something it's kind of overkill starting with a transformer that big. If he's making a variable benchtop power supply it might not hurt to keep on the lookout for an inexpensive variac to use on the primary side. Took me a couple of months of bidding but I finally snagged a brand new 10A variac off eBay for $40 a few years back, needed it to trim the voltage output of a neon sign transformer to drive a homebrew laser.

Lacking that I would think the only logical solution to be a switching regullator circuit, plenty of options for ICs driving power MOSFETs and your output filtering wouldn't take massive caps.

 

Good idea to mention the Variac.

He would only need about 10 Amps on the Variac connected in front of the Primary. I think the OP likes auctions.



28 Volts x 30 Amps /120 Volts = 7 Amps so a fairly common 10 Amp Variac should work.

I would derate the output to about 70% so maybe 20 Amps at 24 Volts with this setup. Should be enough for most uses.
I would drop it to 50% if you are using it for a 12 Volt output and again 15 amps is still ample for plenty of uses.


A Variac for preregulation and you should be able to power up to half horsepower motors fairly easily at 24 Volts and quarter horsepowers at 12 Volts.

 

In addition to the auctions another good source for variacs would be any old TV repair shop, in the old days when solid state horizontal drive was just starting out they were very commonly used to "soft power up" a set. Since very little is repairable anymore there's bound to be quite a few of them gathering dust if you can find the right place and the owner is hungry.

 

Good idea to mention the Variac.
I think the OP likes auctions.

Hi there, I certainly do! Sometimes you can be really lucky. However, sometimes it pays to shop at the real price for the guarantee.

Back to the power supply. I am a bit at sixes and sevens now with all the options. I like the variac, and, have been looking for an "auction" one. Need a min of 240 volts with 5 amps input. Not an elegant solution but practical.

I have started to look at the switched mode power supplies. Lots of work here for me. The circuit from "Beenthere" looks very clever and I wonder if it will give me the power control that I might want, especially as woltage regulator chips hit the high price once they get over a few amps. And I still have to add that to the design.

Regards

 

Also note the max current you can draw then using a typical rectifier/AC filter setup. This will give you some insight

http://www.hammondmfg.com/pdf/5c007.pdf

 

Bobbie, there are a lot of switching regulator circuits which are far more simple, I'll try to look a few up for you when I get home tonight.

 

This may look complex but really isn't, these type ICs often have a lot of extra pins such that they can be interfaced to control circuitry. This looks like it could be simplified and meet your requirements, best thing is if you can "act" like a company Maxim is pretty good at handing out free samples. If you go through their website you'll find many options, same with TI that makes the "Simple Switcher" ICs although you've got to do a bit more of the design on your own.

I like this one as it generates its own Vcc for the IC thus you only need one input supply if you're regulating a higher voltage. It's accurate, it's adjustable and depending on the FETs you choose for the output stage it should take about anything you can throw at it.

The output voltage is determined by the ratio of R1/R2. As with most switching IC regulators you should keep the resistive divider that forms the FB voltage under 100K total (R1 + R2) to reduce noise susceptability.

http://datasheets.maxim-ic.com/en/ds/MAX15046-MAX15046B.pdf