AD8108_bandwidth and power consumption

Document created by analog-archivist Employee on Feb 23, 2016
Version 1Show Document
  • View in full screen mode

1) In the datasheet, we can find the BW is 325MHz with 200mVpp and 140MHz with
2Vpp, but how about the condition of 1Vpp?
2) Will the BW reduce while the load increase? means that, if we route 1 input
to 8 output channels, is there any bad effect to BW?
    The customer said he found this kind of problem with our competitor's.
3) Because of the large power consumption, the part will be hot, so is there
any recommendation for the radiator product. For example, what do we use in
4) can we provide the sch and layout file of AD8108/9EVB to the customer?


1) We cannot specify the bandwidth for all voltages and the specs that we do
provide are typical. The bandwidth at a signal level of 1V will be of course be
in between the values for 200 mV and 2V. I would say that it will be a little
closer to the 2V specification, because it is the slew rate that limits the BW
at higher signal amplitudes. At 1V there is some slew-rate limitation of the BW.
If you want more BW using this part I can make a suggestion. Use smaller
signals through the part to get a higher BW. Then at the output, use a very
fast amplifier at a higher gain to make up for the lower signal amplitude. This
technique is often used to get higher bandwidth through the entire signal chain.
2) We know that some competitors parts have reduced bandwidth when more
outputs are driven by the same input. We do not have this problem at all. In
fact, we have seen that sometimes there is even slightly higher bandwidth with
more outputs. But we do not see any reduction in BW when driving more outputs.
3) We do not use any extra heat sinking on the AD8108/09 evaluation board.
This is not necessary to operate out in open air. The AD8108/09 will dissipate
about 500 mW (10V supply x 50 mA current), depending on the output load and the
currents that the outputs drive into the loads.
The theta J-A is specified as 48 degC per watt for still air. Therefore, the
maximum rise in junction temperature above the ambient is about 24 degC for
still air. For reasonable temperatures, this is no problem without any
heatsink. If there is moving air (from fans), then this problem is even less
severe. However, if there are very heavy output currents, the ambient
temperature is high, and the air flow is low, then some type of "stick-on" heat
sink can be added. This is usually a shaped piece of aluminum that has special
adhesive for attachment. Most heat-sink suppliers offer these products, and you
just have to find one that is about the right physical size t stick on top of
the package.
4) I will have to work on getting this. It was done a long time ago, and I
don't know what the current status is. However, if we have this, we have no
problem with sharing this with customers.