Power Transformer Design Questions.....

[Kid Squid]

OOPS.....

 

[Kid Squid]

Hiya Chaps,

Been doing an awful lot of reading lately, with regard to designing /DIY'ing some power transformers. Been reading up on the B.B. Babani - Coil design and construction manual / Wolperts's design manuals, and the Excellent Pat Turner site (although they are a mix of bloody imperial or Metric !!)http://www.turneraudio.com.au/education+diy.html

Anyway,
I've been drawing up my notes, And i will put them up for scrutiny,but firstly,there are a few questions I need help with answering.

here's what i'm after. ( briefly, i might add)

230V Pri ( 2x115V),
250 - 0 - 250 V @100ma (SEC1)
6.3V @ 2A (Sec2)
5v @2A.

Now,
My first question is , on the SEC1, do I design around the Babani way, which states - only half the sec HT winding, supplies current at one time, eg,
So my total wattage would be

250x0.1 = 25W
6.3x2 = 12.6W
5x2 = 10W Total VA = 42.6 , or,

500x0.1 = 50W
6.3x2 = 12.6W
5x2 = 10W Total VA = 72.6

obviously, this is before using,

** add 10% for losses **, and
**Use core with 33% higher rating ** ( Pat Turner)

The Babani book , is the only place I've read about this ( I can see that only half the winding will supply current at one time), but does it pay to be this 'tight' when designing transformers. ( maybe, the extra manufacturing costs)
I would have thought , that the transformer, would be designed around the 'whole' winding.

Question 2.

When measuring the magnetic length of the E/I stamping, is both windows taken into account ? eg.

if you can imagine the stamping, I've been working it out based on one 'window' of the E/I stamping, not both sides......if you can understand what I'm on about....

I've gone through a load of examples, but I'm still not quite sure on a few things......

Made a great contact last week, with regard to a small ' cottage' transformer manufacturer, and I got to call around there this week - So I'd just like a bit more knowledge, before i give them a call.

Also, what is the design criteria for C Cores ?, is it the same ?, it's just that i can't find out that much, with regard to DIYing them.

Cheers my Dears,
Steve :thumb:

 

[PRR]

> does it pay to be this 'tight' when designing transformers.

The average mass production purchasing manager will jump through flaming hoops to save a penny. None of this 33% round-up slop, all sharp pencil design.

The only reason you care about your HV winding being in two parts is sag and sag-related heat. The core VA is sized for the load. But the vacuum rectifier is cheaper if you double-wind your HV, accepting a more expensive and more lossy winding in return for not needing a 2-cathode diode.

And in today's world of $0.19 1N4007, your choice of a vacuum rectifier and the implied lossy winding means you are NOT looking to save every penny.

Read the PT section in Radiotron 3rd.

But in general, learning good PT design is a 20-year process. This forum, the web in general, is not going to teach you everything you need for good design. Sad, because a lot of the old-time designers are retiring and nobody in US/EUR is learning their craft.

Your specs can easily be met by a Hammond, or a Fender Deluxe PT, or your cottage winder can surely glance at the spec and pull the right core and wire by "intuition" (really a lifetime of study).

I applaud your DIY, but if you want to get-er-done, buy the design with the transformer.

 

[familyguy]

Diying transformers is not hard but unless you intend to make a heap of them the raw materials in the way of spools of wire will probably cost more than a ready made one. But if you must then Radiotron designers hand book ( I have ED. 4) ch5 sec.5 has a design example for a transformer with 300 0 300 sec winding able to deliver 60mA DC. Wattage is calculated by using the required dc output volts and current (after rectification), which seems to be about the same as Babani. I understand that with an EI core magnetic path length includes both sides/windows. Going through design examples is the easy part, the hard part is the actual winding and getting it all to fit.

If you are serious take a trip to the local tip and take home any old junk that has a transformer in it(stay away from the microwave ovens though). Pull the tranny, wire it up and plug it in making sure you observe BASIC ELECTRICITY SAFETY RULES, put a dummy load on it measure the volts out, calc the current with the dummy load, leave it on for a while and see how hot it gets. When you've made all the measurements you can think of pull it apart unwind it, count the turns, measure the wire size and the core size, calculate the turns per volt, note it all down. If you have a variac wire it up backwards, monitor the input and output voltages and SLOWLY wind up the variac and find the point where the core saturates noting the voltages to see how close to saturation it is in normal service. When you're done you can throw the bits away and try another one. You will find that most small asian made transformers run close to saturation with a current density through the windings of 5amps per square mm or more where 3amps per square mm is generally regarded as a safe figure for a continuous duty transformer.

The good thing about winding your own is that you can include extra windings. Just finished a guitar amp for my son using a diy power transformer, it has separate bias supply winding, center tapped filament winding is wound bifilar ensuring that the center tap is truly in the center, separate 12v winding for channel switching relays, separate 6.3v non center tap winding for output valve filaments and another 12v winding to provide a regulated 9v dc socket on the back of the amp for the tuner and any pedals if he ever decides to use them.

 

[Yves]

If you're able to understand few french words, look at:

http://www.dissident-audio.com/Pst/Pst.zip

Be aware that, even if similar looking, european core references ARE NOT the equivalent of imperial ones.

And, heu ! my brain is hardwired for metric only, sorry :green:

Yves.

 

[Guest]

Output transformers may be handcrafted if you know secrets of sectioning and interleaving and have time to wind turn by turn manually, but for power transformers I would definitely go to edcor.com

 

[PRR]

> pull it apart unwind it, count the turns, measure the wire size and the core size, calculate the turns per volt

If you know how to energize it, and there is space between winding and iron, no need to disassemble.

Wind one turn. You can even use solder. Bring the primary to rated voltage. Measure the AC Volts across the one turn.

Sanity-check: should be near 0.1V. 0.05V, 0.15V, depending on size and design. On "normal" size iron, anything way off from 0.1V/turn is dubious.

For better precision, wind 10 turns.

From this, you can figure the turns in the original windings.

And eyeballing the mean length of turn will tell you the approximate length of wire in each winding.

De-energize and measure resistance. Resistance and length and a wire-table tells you the gauge used.

If for some reason, you need a low-volt small-current winding, you can try to cram a few dozen turns around the existing winding.

 

[familyguy]

Horses for Courses I guess. I believe when that when you're starting out and haven't even wound one transformer it does no harm and probably a lot of good to pull some apart even if its just to see how its done. They are after all junked transformers, might be different if the tranny was some exotic vintage iron. Reverse engineering by measuring usually only gets you ALMOST all the way, its nice to be able to pull it apart just to be sure.

I'm about to make another Steve Bench RAT valve tester and plan to incorporate all the things that I should have done first time like a multi tapped filament tranny, most taps I've done to date is only about 3 or 4, guess I'll see how it turns out when I get there.

 

[PRR]

> it does no harm and probably a lot of good to pull some apart

Oh, sure, I have no objection to that. I love to rip things apart.

Just noting another way to get some key data. An awful lot can be learned in sneaky ways.

Might be handy when you have something near-right, but you are not sure enough for a tear-apart. Or you just want some cross-checks against similar transformers.

 

[Kid Squid]

Cheers Chaps,

Thanks for all the kind words !
I'm a sucker for a challenge ........

Anyway, I've designed a power transformer that gives,
2x 120v Pri Windings,
250-0-250 V @110mA,
6.3v C/T @ 2.2A
5V @2A.
I'm in the process of getting my wire ( sorted), Lams + bobbin (sorted),
Insulation / glass tape.( not sorted ......yet)
any ideas as to where i can get some different sizes of insulation, Mylar etc..etc....

I'm writing it up at the moment, so I'll post it when i'm done.

would be really apreciated, if you kind chaps, could have a look over it for me, just in case i've overlooked / or miscalculated something.

Regards,
Steve :thumb:

 

[CJ]

i have liquid opium for you.
what is your shpping addy?
inned sipo addy for kubi and jens
please help