Re: LCD refresh rates - don't understand

The message <VA.000004ca.0334d587@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>
from Roderick Stewart <rjfs@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>
contains these words:

In article <313030303737303648B4517211@xxxxxxxxxxxxxxxx>, Johnny B Good
wrote:
The voltage available from the PSU during quieter passages could quite
easily rise 50% or more above that during RMS test signal conditions for
maximum undistorted power from all channels. This would allow double the
rms power for a few tens of milliseconds or double that again for the
'Peak Power' on the crest of the waveform.

50% regulation? That sounds like what I'd call a truly *** power supply.
Even I could design a better one than that.

Well, believe it or not, that's quite typical of even the more
expensive 100 watt rms per channel Hi-Fi amps (although the pre-driver
stages will normally enjoy the benefit of a regulated lower voltage
supply).

When I was commisioned to design a 100 watt per channel PA amp by a
friend some 30 odd years back, I was forced to design a strictly
regulated 45v PSU on account the large collection of surplus TO3 power
transistors I'd elected to use didn't have the luxury of the 80volt
rating of the 2N3055s normally used in similarly rated amplifier
designs.

The limited voltage meant I had to use a bridged output, but this led
to the benefit of reduced peak currents on the supply and facilitated a
balanced/unbalanced input, and allowed me to use a cunning method of
overload protection against otherwise fatal short cicuits on the speaker
leads.

I was also able to add clip limit LED indicators[1] to show the onset
of clipping before it became audibly obvious. These indicators also
doubled up to show when a shorting fault had occurred by permanently
glowing and an accompanying loss of sound, removal of the fault
condition would automatically reset the electronic circuit breaker used
to provide said overload protection[2].

In the end, the amplifier I produced could deliver 100 watts rms per
channel into 8 ohm loads or 200 watt rms per channel into 4 ohm loads...
simultaneously, i.e. a total rms power output of 400 watts or 800 watts
total PMPO even with 1 KHz square waves - a _sustained_ PMPO rating
rather than the transient PMPO rating of conventional designs using a
non-regulated PSU.

An amplifier with a total PMPO rating of just four times the continous
rms wattage per channel rating is a dead giveaway for a design that uses
a strictly regulated PSU. The more usual factor of 8 to 10 times is a
sure sign that the PSU is _not_ regulated.

[1] My philosophy regarding output level indicators on a PA amplifier is
that all you need to know is whether it is starting to be be overdriven
into clipping, so that you can throttle the level back if the maximum
undistorted output is actually required to generate the required sound
level for the venue concerned.

If a venue demands just another 3db more, you're screwed anyway, so
you're best avoiding the gross clipping distortion that would result in
trying to achieve the impossible (doubling the power output). As long as
it sounds loud enough, it doesn't matter whether the peaks are a mere 20
watts or at the maximum limit of 200 watts.

[2] My philosophy regarding overload protection (whether it be short cct
faults or 'awkward' speaker impedance problems) is that it should be an
all or nothing shutdown mechanism rather than the namby pamby drive
limiting 'protection' so oft employed by a lot of the idiot commercial
designers of home hi-fi equipment.

The benefit being that you _know_, if it doesn't embarrasingly go
silent for a second or two, that the protection isn't adding distortion
due to an 'awkward' speaker load (unlike the namby pamby drive limiting
method which will simply add distortion under these conditions and leave
the listener guessing as to why it doesn't sound quite right).

I might add that such clip indications don't demand the use of a
regulated PSU, so, provided the output devices can handle the extra
voltage (and current), the greater transient PMPO performance benefit of
an unregulated PSU can still be employed.

--
Regards, John.

Please remove the "ohggcyht" before replying.
The address has been munged to reject Spam-bots.

.