• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Mixed Linear/Switched high power supply

Status
Not open for further replies.

Dalzani

New Member
Hello.

I´m build a linear power supply 13.8V / 60A for a transceiver radio.
The common approach will be use common regulators like 7812 or LM317 and a lot of power driver transistors, a big heat sink, big capacitors and of course a big transformer.

I´m just think about if is there some advantage changing ONLY the common regulator for a switched mode design using the same large transformer and same large capacitors.

At this link http://focus.tij.co.jp/jp/lit/an/slva001d/slva001d.pdf on page 27 there´s a circuit that explain exactly what I´m trying to do. Of course, the driver transistor must be replaced for a compatible project amperage (60A).

Is there some advantage doing this? Less power transistors maybe, better regulation...

Thanks.
 

MrAl

Well-Known Member
Most Helpful Member
Hi there,

It sounds like you are asking about using a switch mode power scheme rather
than a linear.
The advantage of a switch mode power supply is efficiency, but only when
there is a certain input output voltage differential, otherwise a linear may be
just as efficient.
Most apps for switching power supplies however do have large input output
differential and so are well suited for the switching type supply.
If you have very large input output differential then a switcher is definitely
a good idea over the linear because a huge savings in power will result.

You should also use a switcher when you have a wide variation in input voltage.
The problem with the linear is that at high line there will be very low efficiency.
The switcher wont have this problem.

You do have to realize that switchers are a bit more complex to build than
regular linears, but it sounds like your app could benefit from a switcher so
perhaps this would be a good idea. I say this because even smaller power supplies
that are linear require fairly large heat sinks, while switchers will always have
smaller ones which means less space too.

Since it is quite likely that you will be using a switcher, perhaps you should go
after the MOSFET transistor type design. The MOSFETs have less loss overall
as compared to bipolar when they are driven correctly, and that's always a
good idea.
 

Sceadwian

Banned
That's close to 1000 watts OuttoLunch, he's definitely a ham =)

MrAl one thing really bad about switchers is noise. Even if he uses a switcher he may need a linear regulator to get truly flat clean power. At the currents he's using if there's substantial ripple at full power it's gonna piss off anyone with 20 miles or more of his antenna, it may be more efficient to use an SMPS to provide just enough voltage over the dropout of a linear regulator. Most purely linear regulator setups use a transformer output well above the linear regulation in order to provide enough voltage under full load, an SMPS will be smarter about the voltage it delivers so you won't need quiet so extensive heatsinking on the Linear side of things. Exactly how much efficiency this will give you I'm not sure and it would be a complicated build.
 
Last edited:

Dalzani

New Member
Thanks for first replies.

I don´t know if I can explain exactly what I´m trying to do... After all living in south Brazil, English is not my natural language.

Sceadwian: Yes I´m a ham :D. I don´t really need 60A, I need 45A but until I now in this hobby a good amount of extra feeding power can solve a lot of problems (ripple, overheating...).

Well...

Doing this project using a typical linear approach we have:

A) A transformer from 220V AC to something like 25V/60A AC.
B) A rectifier to convert AC to DC.
C) Two huge filtering capacitors 2 x 60.000 uF.
D) A regulation stage using common ICs like LM78XX driving lots of power transistors.

What I´m asking is... Using all the same for A, B and C, there some advantage changing only part D to a switched regulation (13.8V) approach?
Can this give me a better regulation? Maybe less waste in heating energy. Maybe less power transistors are needed... I don´t know :(

If yes, I ´m thinking to use the circuit in page 27 http://focus.tij.co.jp/jp/lit/an/slva001d/slva001d.pdf and change only the power driver transistors and the 140uH inductor.

Thanks all.
 

MrAl

Well-Known Member
Most Helpful Member
Hi again,


If you are considering using a 25v winding to put out 14 volts, well
that's quite a waste of power. 14 times 25 is 630 watts out,
but with 25v in that's 1125 watts input meaning the regulator is
roughly 50 percent efficient (truely bad).

A switcher can give you 80 percent easy these days. With MOSFETs
even higher.

Contrary to the older designs, switchers can be built to put out very
very little ripple with the addition of a secondary filter stage. This
is a relatively new concept even though i am sure it has been known
for years. The secondary filter stage however is kept out of the
feedback loop for good stability.

Switchers are much harder to build and design however, and im not
sure how complex you want your power supply to be.

If you do go with a linear, do a few simulations so you can reduce
the transformers output voltage to a lower level and reduce wasted
power consumption.
 

dougy83

Well-Known Member
What I´m asking is... Using all the same for A, B and C, there some advantage changing only part D to a switched regulation (13.8V) approach?
Can this give me a better regulation? Maybe less waste in heating energy. Maybe less power transistors are needed... I don´t know :(
Yes. Sounds good to me. The regulation will not be better than linear, you will have ripple in the output - you can reduce this with a passive filter, or possibly a low dropout regulator (e.g. SMPS outputs 14.1V with 0.2Vripple, LDO reg drops it to 13.8, so LDO average power dissipation is 18W).

If yes, I ´m thinking to use the circuit in page 27 http://focus.tij.co.jp/jp/lit/an/slva001d/slva001d.pdf and change only the power driver transistors and the 140uH inductor.
You will want to use a MOSFET for the switch, and you'll want a MOSFET high-side driver if you're using the TL494.
 
Last edited:
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading
Top