Building a transformer
in [Design]
|
Prev: Texas Instruments microcontroller
Next: potentiometer
From: John Woodgate on 11 Aug 2006 15:24 In message <44DCC2BD.A021338F(a)REMOVETHIS.hotmail.com>, dated Fri, 11 Aug 2006, Eeyore <rabbitsfriendsandrelations(a)REMOVETHIS.hotmail.com> writes > > >John Woodgate wrote: > >> In message <44dc500f$0$7951$db0fefd9(a)news.zen.co.uk>, dated Fri, 11 Aug >> 2006, Mark Fortune <mark(a)fortrex.co.uk> writes >> >> >aha, you've discovered one of my planned future hobbies :) >> >any money in it though? >> >> Yes, you'll spend quite a bit. >> >> Everybody and his dog makes power amplifiers. Try something else (which >> is small-signal and thus less likely to emit flames and smoke). For >> example, low-level active crossovers; 2-way, 3-way, 4-way, Butterworth, >> Linkwitz-Riley, Bessel, transitional Gaussian, Thiele.... >> >> Enough to keep you busy for 50 years. And there isn't a great deal of >> competition, at sensible prices, so you could sell a few on Ebay. There >> is other low-level stuff which, compared with power amps, is a doddle to >> get working if you master the design mathematics. >> >> And you only need a +/- 12 V power supply at 10 mA. (;-) > >10mA ? That'll power a pair of 5532s ! > How quaint. Try different op-amps. In any case, two 5532s will do four active filters. Sallen and Keyes are quite good enough, unless you want to do tricks with phase. -- OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk 2006 is YMMVI- Your mileage may vary immensely. John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK
From: John Woodgate on 11 Aug 2006 15:34 In message <eece9$44dcba2e$4213eb5a$1352(a)DIALUPUSA.NET>, dated Fri, 11 Aug 2006, Bud-- <remove.BudNews(a)isp.com> writes >I ran across info on rectifier design up through filtering (but not >regulation) at: >http://www.ieeta.pt/~alex/docs/ApplicationNotes/Rectifier%20Applications >%20Handbook.pdf Pages 93-97 contain assertions that are very much not true. Approach with great caution. The basic problem is that the difference between a rectifier resistively loaded and one with a filter capacitor is totally obscured, and in the later case, the impression is given that the rectifier still conducts for a full half cycle, whereas it conducts only for a brief period around peak supply voltage. As a consequence, the text on harmonics is rubbish. -- OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk 2006 is YMMVI- Your mileage may vary immensely. John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK
From: meow2222 on 13 Aug 2006 09:03 Mark Fortune wrote: > Greetings to the collective. > > For my next project I want to build a bench power supply and do away > with this old switch mode AT computer PSU that i'm currently using. > > The design I have in mind will be pretty beefy, giving a wide range of > fixed and variable output voltages (i'm thinking from -50v up to +50) > and deliver up to 5amps of current. if my estimates are right i'll need > a 600va+ (100v * 5a + overhead) transformer to do the job well. Now i've > had a look at some transformers in this range, and they're a little out > of my price range. so now i'm considering building my own. > > The specifications I need are as follows: > > primary: 0-230v @ 50Hz > Secondary: 60v-0-60v > secondary output current max: 5A Need to backtrack on the maths a bit first. To get 50v dc regulated you need 50v dc +5v for regulator drop + 3v for rectifier drop + say 15% to allow for mains sag. Thats 66.7v dc. Now to get 66.7v dc peak you need 66.7 / 1.414 = 47v ac. To get +-50v dc you need the same transformer, no reason to have 2 windings at all. So you just need one 50v winding. Now, current. If you play it within all specs you could use 5A x 1.6 = 8A transformer. But I built supplies for years when younger without using that 1.6 factor and never had any problems from the bridges or tfs, and since this is for personal hobby use I'd be tempted to do that. Or if you need something portable you could even go smaller and add fan cooling. So you need 50v 5A = 250w transformer. Toroidals are smaller, lighter, and more money. Microwave transformers should be avoided, they just need too many mods to be worth playing with. I'd be asking myself if I really needed +-50v at 5A, and would be answering no. I once had a massive all singing psu and almost never used it. A small portable bench supply is a lot easier to use than a meatlump. For those occasional jobs that need more power I'd use something much more basic. These days I'd buy one, or maybe look at modding a pair of pc psus if for some reaosn thats not an option. NT
From: Phil Allison on 13 Aug 2006 12:19 <meow2222(a)care2.com> wrote in message news:1155474226.768278.253900(a)75g2000cwc.googlegroups.com... ** Groper FROM HELL alert > > Need to backtrack on the maths a bit first. To get 50v dc regulated you > need 50v dc +5v for regulator drop + 3v for rectifier drop + say 15% to > allow for mains sag. Thats 66.7v dc. > > Now to get 66.7v dc peak you need 66.7 / 1.414 = 47v ac. > > To get +-50v dc you need the same transformer, no reason to have 2 > windings at all. So you just need one 50v winding. ** So a two diode voltage doubler ?? 50/60 Hz ripple ?? Bad idea. > Now, current. If you play it within all specs you could use 5A x 1.6 = > 8A transformer. But I built supplies for years when younger without > using that 1.6 factor and never had any problems from the bridges or > tfs, ** BOLLOCKS The power tranny will COOK at full load with a 60% over current. > So you need 50v 5A = 250w transformer. ** BOLLOCKS. The OP has specified +/- 50 volts at 5 amps = a 1000 watt DC supply. Means a 1500 VA tranny for continuous operation. > I'd be asking myself if I really needed +-50v at 5A, and would be > answering no. I once had a massive all singing psu and almost never > used it. A small portable bench supply is a lot easier to use than a > meatlump. For those occasional jobs that need more power I'd use > something much more basic. ** Finally, some sane advice. An unregulated +/- 50 volt DC supply operated off a Variac is all the OP needs. Just a 600 VA tranny with twin 40 volt windings plus bridge and two hefty electros. ....... Phil
From: Phil Allison on 13 Aug 2006 12:22
"Phil Allison" ** Correction: The OP has specified +/- 50 volts at 5 amps = a 500 watt DC supply. Means a 800 VA tranny for continuous operation. ...... Phil |