I do not think you need that second
setting. The application can regulate the current by itself by adjusting
that feedback or voltage setting signal. Less voltage means less current.
Unless you are talking about single
fault conditions; in which case you have a similar problem when using digital
communication.
Furthermore, I think you forget that
when using a secondary side controller, you still need something to kick
start the adapter. Since it will not be possible to get that start-up power
from the application, you need something on the primary side. Additionally
you need a current or power limit on the primary side, in order to comply
with safety in single fault conditions.
Finally, I have never seen a digital
core controller that is lower cost than a normal CMOS PWM controller. All
digital controllers are at least 4 times the price of a standard PWM controller
with a 1A gate drive output, running on 12V (<30V). This voltage needed,
to drive a MosFET, is something you may want to consider if you opt for
digital, where operation voltages are in general 5V or below (<0.5um
process).
Best regards,
Rene
"Paul Panepinto"
<paul@xxxxxxxxxxxx> Sent by: upamd@xxxxxxxx
There should be at least two pieces
of information that need communicating: (a) voltage level and (b) maximum
current limit. If you are able to use a 100W power adapter with a
15W load and something goes wrong where the load tries to draw more than
15W, it would be a very good idea if the power adapter is smart enough
to know that the load is not rated to accept the higher current/power and
refuse to provide more power than the load is rated to accept. Given
that at least 2 pieces of information need to be communicated between load
and power adapter (voltage and maximum output power,) an analog circuit
may not be simpler or lower cost than a digital processor.
Only one controller is all that
is needed (not two), as long as the proper isolation circuitry is in place
for both primary and secondary control.
I agree that we should consider
all good options at this point for analog and digital communications.
Regards,
Paul Panepinto
From: upamd@xxxxxxxx [mailto:upamd@xxxxxxxx]
On Behalf Of LEI / Rene Koch
Sent: Tuesday, August 17, 2010 1:04 AM
To: 'UPAMD'
Subject: Possibility for low cost voltage setting
I would still like to propose to keep the possibility open for simple low-cost
solutions.
Which means for example a simple analogue level of the communications connection
in order to set the required voltage of the load; any digital signal on
that communications connection would overrule the analogue signal. This
will of course not be possible if we choose to use a communication scheme
over the power lines.
General power adapters have a controller on the primary side, which uses
a feedback signal from the secondary (DC) side to adjust its operation
level. Mandating digital communications on the secondary side would mean
an additional controller on this secondary side, with additional requirements
of operating voltage (supplied by the adapter itself) and control components.
I believe the impact of this is just too large for general manufacturers
to justify; unless the requirements become mandatory by governments.
(+70% cost for a 20W adapter; +40% for a 60W and about 25% for a 100W adapter)
Since I do not believe that standard commodity products will come available
with a double power connection, I believe this has a major impact to the
success of the UPAMD.
This will also allow for a transition period, where mobile device manufacturers
will learn to appreciate our efforts and where costs of cabling and connectors
will be reduced due to large scale usage.
It does not have to interfere with or restrain anything that we are trying
to achieve; it can just be an additional requirement.