DS90UB954-Q1EVM: Power over Coaxial Filter

Hi Nick,

Thanks for the response.

What is the reasoning for the 150mA current limitation of the filter.

As far as I can tell, the big inductor (parallel with resistor) value is the biggest factor on the cutoff frequency, so as long as that has the power requirements then we should be fine.

Regarding the series ferrite beads, it looks like the recommendation is 3kOhm (2x ferrites) @1MHz to increase the attenuation of the high speed signals. So I was thinking as long as we can find ferrite beads that are able to handle a higher saturation current, then it should be fine in terms of the current limitation. I'm thinking we might have to put 3x ferrites to get the series resistance @1MHz in order to get a high enough saturation current for use of a higher current.

Alternatively, what about using a smaller value inductor in series instead of the ferrite beads? For example, using 2x 100nH inductors.

Do you have any thoughts on that?

Thanks,

Adam

Adam,

The limit is not based on max current limits of the components individually. It is based on the impedance over frequency that the network is providing to isolate the high speed line from noise in the power line. The network from the datasheet is using the inductors primarily to provide impedance in the back channel frequency range, while the FBs are providing impedance in the FC frequency range. The saturation characteristic of the FBs is such that with only ~20% of their rated current bias, the impedance at high frequency starts to drop off dramatically meaning they are no longer isolating the DC/AC path. 

We have spent quite some time creating these networks to supply different levels of current while still maintaining good RF characteristics for the channel. We have a few different types of networks we can share through private E2E message so please send Nick a message through E2E chat and include what devices you plan on pairing (SER/DES), along with chat current level you are looking to support. 

Best Regards,

Casey 

Hi Casey,

Thanks for the response. 

For the impedance over frequency, my attempt at that issue was to find a higher saturation current ferrite (which generally have lower impedance at high frequency), and just add more in series to reach a similar high frequency impendence. 

Also, thanks for the explanation on how the main inductor's function is for the (slower) back channel, whereas the ferrites are for the (faster) forward channel frequency range.

In any case, I'm attempting to send Nick a chat request, but in my messagebox it says "cannot message" when I try to add him as a recipient. I'm not exactly sure why, but I've added him as a friend in case that's what is blocking me from messaging him.

I'm definitely interested in seeing what other networks you have recommended for higher currents.

Thanks,

Adam