Dropping voltage with diodes.

[davelectronic]

I'm putting together a linear power supply, I expect to get 13.80 Volts at a power rating of 10 Amps. I've used a toroidal transformer with an 18 Volt secondary, the problem is I've put the AC transformer output through a bridge rectifier and filter capacitors. But now I have 27.00 Volts AC reading from the multimeter. I think this is to high to put into a linear regulator, and need to drop that voltage some what lower. So could I use rectifier diodes to achieve this ? The power supply is built now with regards hardware installed. Or is dropping voltage this way not recommended. Thanks for reading, any thoughts appreciated.

[Psi]

Keep in mind that 27V will drop back a bit once you draw some current. You're charging the capacitors up to the peak voltage but the peak doesn't have much energy.

What linear regulator are you using for 10A at 13.8V?    27V isn't that high for a regulator to handle voltage wise.
But you are going to burn off a lot of heat to take 27V down to 13.8V.   It won't be the full (27 - 13.8 ) * 10A = 130W
but it probably will be 50W or so.  We're talking a pretty huge heatsink or a fan.

Dropping voltage with diodes isn't all that practical most of the time.
It not all that accurate as the Vdrop changes with current and you need lots of diodes if the drop is large, and it doesn't save you anything because the same heat just ends up in the diodes instead of the regulator.

One option is to un-wrap the toroidal transformer and remove some turns from the secondary until you get the voltage a bit lower.
This is usually pretty easy to do with toroidal transformers, compared to other types.

[davelectronic]

Thanks for replying, yes it's a fairly big drop to 13.80 Volts. I wanted to close that gap. The linear regulator is an LM317 and I was going to use fixed value resistors to achieve the 13.80 Volts output. The transformer is a toroidal type with twin secondary outputs I have wired in parallel. I'm not sure how to remove which windings and how many turns. I have an extra toroidal transformer with the same ratings I could experiment with. Just wouldn't want to remove to many turns. I know I have to close that voltage gap to bring down the waisted power and heat. I should mention I am using current boosting transistors to obtain the 10 Amps at 13.80 Volts. Most of the time I will be drawing about 8 Amps.

[soldar]

Why not just a power resistor?

[langwadt]

One option is to un-wrap the toroidal transformer and remove some turns from the secondary until you get the voltage a bit lower.
This is usually pretty easy to do with toroidal transformers, compared to other types.

can also add windings and connect them in reverse

[Solder_Junkie]

Don’t forget to add over Voltage protection to the output, when linear regulator transistors fail it is usually short circuit.

Originally a crowbar circuit was used, more recently MOSFETs with a monitor IC, such as the LTC-4368.

SJ

[JohanH]

Why not just a power resistor?

Not a bad idea in general. An RC-filter will always lower the ripple.

[nali]

You also need to consider the *minimum* voltage troughs which need to be greater than 13.8 + the minimum dropout voltage for your reg which is 1.7V in the case of a LM317 (not considering any current bypassing scheme) so >15.5V.

What size filter caps are you using?

[Ian.M]

The voltage on-load will drop considerably, maybe as much as by a third.

There's also mains voltage variation to take into account, which will proportionately affect your output voltage - legally the UK mains voltage is 230V +10%,-6% due to EU harmonisation.  In practice none of the nom. 240V equipment was replaced, and the national grid seems to prefer to run their adjustment taps a bit high so they've got more margin as loads come on, so although in urban areas the UK mains voltage tends to be reasonably close to 240V, in lightly populated rural areas you can see significant variation.

You need to measure the supply voltage, do a full load test and do the math before you decide on any dropper, as you may have far far less voltage headroom than you think. 

Also don't forget that the RMS current in a secondary feeding a bridge rectifier + reservoir capacitor is approx. 1.5 times the average DC load current, so your transformer needs to be at least 270VA for continuous 24/7 operation.  250VA would probably be acceptable if  its never going to be loaded at the full ten amps for longer than half an hour or so, but the no-load/full-load unreg DC voltage swing will be even greater.

[langwadt]

The voltage on-load will drop considerably, maybe as much as by a third.

There's also mains voltage variation to take into account, which will proportionately affect your output voltage - legally the UK mains voltage is 230V +10%,-6% due to EU harmonisation. 

I believe the EU standard was changed to 230V +/-10% in 2009, but some suppliers might chose to have tighter limits

[iMo]

What is your DC voltage at the capacitor (wired after the rectifier) when loaded with your 10Amps (or 8Amps)??

[DavidAlfa]

I believe the EU standard was changed to 230V +/-10% in 2009, but some suppliers might chose to have tighter limits

AFAIK the standard is 240 now. There's 236V in mine!

[BeBuLamar]

The LM317 is good for 37V but yes dropping down to 13.8V does waste a lot of energy.

[soldar]

AFAIK the standard is 240 now. There's 236V in mine!

Your belief is very mistaken. All of Europe, including the UK and Spain, mandate 230V. In Spain since 2002.

Real Decreto 842/2002, de 2 de agosto, por el que se aprueba el Reglamento electrotécnico para baja tensión.
Publicado en:     «BOE» núm. 224, de 18/09/2002.

[Ian.M]

OTOH as I mentioned earlier, the various supply companies like to set the transformer taps to run the final distribution lines voltage biassed towards the high end of the tolerance range when lightly loaded, because it gives them more margin for droop at full load, and the consumers pay for power, not current, so the I²R distribution losses are reduced per KW delivered.

[DavidAlfa]

Aha, it's 230V with 7% margin so legally anywhere between 214-246V

[davelectronic]

I haven't loaded it down yet, after the rectifier and filter capacitors. It's a 300 VA toroidal transformer, rated at 18.00 Volts. I have previously used two 300 VA toroidal transformers for another power supply, they where rated at 12.00 Volts. They sagged very little with a 200 watt load. I don't have an electronic load at the moment. All I know is I need 13.80 Volts at 10 Amps maximum, although the actual current drawn will be closer to 8 Amps when running HF radio equipment. So it won't be a continuous load, more like a 50% duty cycle. When I have measured the mains where I live, many times over the years, I've never seen a reading lower than 240 Volts. This is a single very high power transistor, it's a Darlington rated at 50 Amps 300 watts. I wanted to see if I can get a linear power supply, to produce 10 Amps at 13.80 Volts with a single transistor. I'm not sure what to do with this 300VA 18.00 Volt transformer. I hoped to avoid removing windings, it never as tidy after doing that. I suppose I must find a load equivalent to 10 Amps to see what that rectified and filtered 27.00 Volts secondary drops to with a 10 Amps load. I can't use lamps unless I connect them in series, and then in parallel. Just parallel would blow them. The power resistor sounds like an interesting way of dropping power, although I imagine it would produce loads of heat. And would have to be a metal clad resistor bolted to a suitable heat sink. Now I think of it in terms of the voltage drop under a 10 Amp load, it might be ok, as you say, that 27.00 volts should drop considerably.

[iMo]

BTW. - the ham radio transmitters have around 50% efficiency, so you would need around 15-17A off the 13.8V in order to run 100Watts output TX power. I do not know what your rigs are like, but that is how it works..

Also it depends on the modes - a 100W FT8 or FM is not the same as a 100W SSB when talking the PSU's current..

PS: I've seen 100-200W linear PSUs in our local ham radio club (built in 70ties-80ties), and all what you need to be happy is a pretty large heatsink with several beefy transistors on it

[TimFox]

Have you considered adding a (much smaller) low-voltage secondary transformer in series with the primary of your high-power transformer to reduce its secondary voltage?
(Usually called a "buck transformer", popular in power distribution systems to obtain a desired line voltage.)

[davelectronic]

Now that's very interesting, I had not though of adding a smaller primary side transformer. It's something I will look closer at, although might it make the toroidal transformer secondary a bit droppy for want of a better word. I'm drawing 7.8 Amps at the moment on 11 meters FM, and ssb would be a lot more. I'm only driving the amplifier with 4 watts approximately. The PSU powering my main radio is a retail unit, it's a 30 Amp switching PSU. Yes I am aware of the power requirements for the radio equipment I have. I think I am putting out 75 to 80 watts roughly. Just a picture of the heat sink and hardware, I've still have to do heat shrink tube, don't have any that size left, so ordered some.

[soldar]

I haven't loaded it down yet, after the rectifier and filter capacitors.

You need to do that before you go any further.  Until you do that you are just spinning your wheels in search of a solution for a problem which is not known and may not exist.

[langwadt]

why linear?

[schmitt trigger]

I wonder how come nobody has asked exactly how will you deliver 10 amps from a LM317??
Are you using auxiliary series-pass transistors?

But back to your topic: if you would consider additional complexity, you could add a pre-regulator. The most efficient would be a buck topology with external switching Fets.

[davelectronic]

Yes I don't exactly know the voltage drop, but if the 300VA 12.00 Volts transformer (same manufacturer different voltage) it won't drop by much. The current is coming from an MJ11033 PNP Darlington transistor. I'm hopeful it can supply 10 Amps, as for a pre regulator, I wouldn't know where to start with something like that. And it sounds overkill for a basic linear power supply. Oh yes capacitors are 10000uf 63 volts electrolytic.

[langwadt]

Yes I don't exactly know the voltage drop, but if the 300VA 12.00 Volts transformer (same manufacturer different voltage) it won't drop by much. The current is coming from an MJ11033 PNP Darlington transistor. I'm hopeful it can supply 10 Amps, as for a pre regulator, I wouldn't know where to start with something like that. And it sounds overkill for a basic linear power supply. Oh yes capacitors are 10000uf 63 volts electrolytic.

https://www.ebay.com/itm/334836996334