Hello Peter,

> Manu,
>
> The chip you have is from a very early batch.  The fault you are probably
> seeing is a race condition.  This problem was completely eliminated in June
> (before we launched mass production).  The sample you have is from a
> preproduction engineering batch (not for commercial sale).  The antidote is
> to obtain a new card from KNC1 with a later batch number (your IC probably
> has C1L marked at the top, get one with C2L to be sure or > MLT 22 624)...
>
> The symptoms are as follows.
>
> The IC performs perfectly in DVB-S modes (all modes, all symbol rates).
> When in DVB-S2 depending on temperature, supply voltage and carrier to noise
> ratio the device may lose either FEC lock OR the FEC may remain locked but
> the packet delineator (and thus transport stream) outputs errors.
> Heating or cooling the device (say with freezer spray / hair drier) will
> bring the device back to correct operation (you may need to relaunch an
> acquisition).
>
> If a blast of hot or cold does not solve your problems let me know.
>

KNC1 has 2 mails as a reply on the issue.


Regards,
Manu


thank You very much for Your information. We have checked the whole
e-mail communication.
Unfortunately we have not detected a clear description of the problems
you are having, but
 only an estimation of someone from ST Microelectronics.
As a matter of fact the demodulator is being driven in a complete
different environment
on our board than on any reference design from STM. We are using a
different power supply,
making it possible to supply the IC at the lowest possibly voltage at
very sharp tolerances.
Furthermore we are using a much more sensitive CAN-tuner-PLL compared
to the rather primitive
silicon tuner to be found on competitors designs. This makes it
possible to choose a lower
Clock frequency of 90MHz instead of 99MHz as we are using in our
Windows-software. We have
never encountered any themal issue onto our boards and have not been
able to get a wrong
demodulator back to work by cooling it down further.
In our opinion it is required to either tell us your problem clearly
or to send the board back
to us without further delay and without stressing stressing it
unnecessarily. You could easily
damage the board completely through excessive temperature stress or by
too much humidity caused
by the freezer spray. Especially the high current voltage regulator
nearby the demodulator will
not widthstand any fluids onto the pcb. You could also cause
inacceptable frequency shifts of
the external crystal oscillator or terrible noise on the tuning
voltage and thus come to wrong results.
We can not take any responsibility for damages caused by violations of
the environmental conditions
and insufficient examinations by the customer.
Please be so kind to send the board back to us or to ask Your local
dealer for returning it.
If you want to sent it back to us please use the following RMA Number
and this adress.