> >I would like to ask the proponents of series rectifier diode strings if
> >they have ever measured the developed bias between the minimum and full
> >power conditions?
>
> Yes.
>
> >I've done it with a Henry 3K Classic X, using and 8877,
> >and the bias soared about 6V from zero to full signal.
>
> 6V sounded a bit high, so this morning I tested a 1n5408/3A diode for fwd
> V-drop using 8877-sized currents, assuming a ZSAC of 0.1A and a signal
> max Ik of c. 1.2A. At 0.1A, the V-drop was 0.734v. At 1.2A, the V-drop
> was 0.813V. Thus, with a 10-diode string, the change in bias V for an
> 8877 would be 0.79V from zero-sig to max-sig. For improved IMD, and
> less chance of overdrive with more than 78w drive, add a few ohms in
> series with the diode-string to increase V-drop at max sig.
>
> //That Vf/If curve, e.g., 734mV/100mA and 813mV/1.2A, may be true for the
> 1N5408 tested, but isn't necessarily representative of all 3A rectifiers.
> In fact, Vf is normally only specified as a "max" at full current, and
> nowhere else, and can run all over the place. I've seen lots of 3A
> rectifiers having <<500mV @ 100mA, and >>0.8V @ 1.2A. The "Vf max" of
> most mil-spec 3A rectifiers, of which the 1N5408 is not (but the very
> similar 1N5554 is), is 1.0V @ 3A, and of course it has a negative tempco,
> meaning the hotter it gets, the lower the Vf will be. If each rectifier
> had a 250mV delta Vf, and ten were used in series, this is a 2.5V delta
> Vf. Could be quite a lot, in some cases.
>
> -WB2WIK/6
>
>
>
>
|