> From: Peter Chadwick <Peter.Chadwick@gpsemi.com>
> To: amps <amps@contesting.com>
> Subject: [AMPS] Delrin,failures, etc
To: <amps@contesting.com>
> Date: Fri, 20 Jun 97 09:28:00 +0000
Hi Peter,
> Tom, I think that you're suggesting that the current limiting mechanism
> in the SB220 is the 'usual' one in a transformer. I was wondering if the
> transformer in the SB220 had been specially designed to have a high
> leakage reactance so that the inrush current was limited - that was done
> on a Marconi high power tx back in the '60s, ( I think it was pure
> serendipity, as there was 30Kv insulation between primary and secondary
> for other reasons) but it sounds as if leakage reactance is not the case
> for the SB220.
That's correct Pete, it is not a high leakage transformer. It has a
very compact winding on a short magnetic path core, and hence has
very little flux leakage.
By the way, I see many hams think core size limits current. That is
untrue. Core saturation occurs as primary voltage is raised, and the
magnetic field no longer can be "contained in the core". If you could
imaging raising the voltage while watching the current, eventually
you'd notice a point where the current skyrockets with a slight
voltage increase. THIS is core saturation.
Core saturation is NOT a result of secondary load impedance (with a
resistive load), since the secondary generates a counter flux that
opposes and subtracts from the magnetization flux. When that happens,
the primary draws more current, in the attempt to keep the flux level
the same.
The limit of current you can get in a small conventional transformer
is due to resistive losses, NOT core saturation. Core saturation
current limiting is an old wive's tale, unless you mean the current
is limited as you raise the primary voltage.
> Joe, W8IK, is re-iterating the point I made some time ago - what we can
> do in a home brew isn't always practicable in manufacture. A good rule
> of thumb is that you sell for twice what it costs to make, and it's a
> fair bet that labour (English spelling!) is half the cost of
> manufacture. So if it costs $20 in parts, that is likely to come out at
> $80 in selling price - you might shave it down a bit if things are
> really competitive, but I suspect that labour costs are reasonably high
Rich may not understand about manufacturing and statistical failure
analysis. I suspect it's because he has never designed or
manufactured a complex product. People in that position often have no
concept of how their ideas would affect the consumer, they
just look at one narrow perspective.
For example, if the amateur operates the filament too low in an 8877,
3CX800, or other metal oxide cathode tube, the result is cathode
contamination and ruining the tube. I've seen several "modified" PA's
where the customer lowered filament voltage to the point where they
ruined the tube. Some web pages even suggest people do this, and
spend more time bitching about manufacturers being stupid or
dishonest than warning people about potential problems.
Very few amateur PA's see anywhere near enough filament time to
justify the added component cost and failure rate, to say nothing of
potential operator error, of including a rheostat for the filament.
While it is true some PA's operate with way too much filament
voltage, that's a problem with the basic transformer spec. It isn't
because they "omitted a rheostat".
Since MOST commercial amps don't even have filament controls,
suggesting amateur PA's should smells like a personal grudge to me.
73, Tom W8JI
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|