has a long cable run. If so what sort of interface issues between both should
be aware of?
I see from an old AD9889 reference circuit that I found, the CM2020 (California
micro devices) was used for a similar purpose.
Also on page 15, fig 30 of the AD8195 datasheet, the typical application
circuit shows an optional ESD protection. He asks if this is required to comply
with HDMI standards or if the internal ESD protection of the AD8195 is
adequate. I presume as stated in the datasheet the additional ESD protection
should certainly be considered if the AD8195 connects directly to a HDMI output
I don't think there is anything too difficult about connecting the AD8195
outputs to the AD9889B inputs.
The AD8195 has a fixed input EQ which works over the full range of cable
lengths. Therefore, nothing has to be changed in this circuit.
The AD8195 has a pre-emphasis option on the output. If the distance from the
AD8195 output to the AD9889B input is not too great, then no PE will be
required and the PE_EN pin (29) should be at logic low.
Similarly, I don’t know about the input EQ situation for the AD9889B. But if
the distance is short, it should be turned off or set to minimum.
ESD protection is not mandatory for HDMI compliance. But if it is used, the
device must be compliant with the ESD device present. This can affect the
capacitance of the traces and the electrical performance, if it has a big
It is the customer's choice as to whether to use extra ESD protection. Some
customers have told us that the HDMI connector has a full ground ring around
the circuit pins, and this can discharge any harmful ES-voltage before the
signal pins make contact. Thus, they don't feel it necessary to add the extra
cost of a protection device. It is not mandatory, the decision is governed by
how robust the manufacturer wants the end-product to be.