Thanks for the recommendation. "PSU Designer" is a pretty neat program. I
downloaded it and also got the psu files from G3SEK's site and played with
them. I can see where this program could really help.
One thing confuses - I have never worked with the "series resistance" of
the capacitor. Is this a parameter I can measure on the bench? How? The
help file doesn't.
I presume your article on PSU Designer is not on the web. I wish it was.
None of what has been discussed in this thread relates to how hot the
transformer runs in service, but isn't that the _ultimate_ determinant of
its KVA capacity?
Thanks to all who helped.
Jerry W4UK
At 03:00 PM 12/15/2001 +0000, Ian White, G3SEK wrote:
>Jerry Flanders wrote:
>>
>>For "ordinary" old power transformers:
>>
>>Is there a rule of thumb to help estimate transformer KVA capacity from
>>the iron (or total) weight? If not, why?
>>
>>Prior to hypersil, weren't most transformers made from similar iron in
>>similar configuration? If so, wouldn't capacity be directly related to
>>core weight?
>
>Only indirectly. Weight is a rough indication of core cross-section, which
>is important but still only one factor in the transformer's overall
>performance.
>
>The rule of thumb was "Size of a six-pack = 500 watts, size of a 12-pack =
>1500 watts, too heavy to lift = 2500+ watts!" This of course is very
>variable, and older transformers tend to be bigger and heavier for their
>power rating than modern ones.
>
>However, I have been bitten several times in the past by building a whole
>PSU - or even a whole amp - around a transformer that passed the
>rule-of-thumb weight test but was quite unsuitable.
>
>A *much* better way is to measure the winding voltages and resistances and
>look at the electrical performance. Rich gives some useful electrical
>rules of thumb, but before you commit to building a whole power supply
>around a particular transformer, you'd do well to run it through the 'PSU
>Designer' software.
>
>'PSU Designer' is Windows freeware from an audio firm called Duncan
>Amplification and it's an easy way to put some real 21st-century accuracy
>into PSU design. Duncan's amps use relatively small PSUs by our standards,
>but the software works just fine for big PSUs too.
>
>I wrote a short article on it in my monthly magazine column, and there are
>links from: http://www.ifwtech.com/g3sek/in-prac/ Slide down to August
>2001, where there are links to the download site and some extra files to
>download.
>
>The extra files relevant to AMPS are the circuit files '3kV Bridge.psu'
>and '3kV Doubler.psu', and you'll also need the expanded RECTIFIERS.TXT
>file which contains a high-voltage rectifier.
>
>The two .psu files are a basis for modeling normal high voltage supplies.
>Double-click on any item in the schematic to alter its properties, and
>then click Simulate to get a simulation over a chosen time period. You can
>change the load current by altering the current sink connected to the output.
>
>Then click various items on the form below to see graphs over time. The
>most interesting item is probably V(C1) - the output voltage.
>
>As you get further into using PSU Designer, check the Help files for more
>information.
>
>A note about transformer modeling: PSU Designer only lets you model the
>secondary resistance of the transformer. You can include the effects of
>(primary + mains supply) resistance by multiplying those values by the
>square of the step-up turns ratio and adding the answer to the secondary
>resistance. There's more information if you double-click the transformer
>symbol, and then choose Help.
>
>This really is a great program! It takes some practice, but I wouldn't
>ever build another PSU without running the components through PSU Designer.
>
>Coming back to Jerry's original question about current ratings, if the
>winding resistances are low enough to give good voltage regulation at the
>currents you want, you pretty much have your answer about current ratings too.
>
>--
>73 from Ian G3SEK
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|