[USRP-users] Isolating the grounds of LFTX and LFRX in N200

Ralph A. Schmid, dk5ras ralph at schmid.xxx
Mon Oct 20 08:38:35 EDT 2014


Sure, ground is one path for stray RF, but it is not the main path in such a setup, inside a box, and anyway you can’t do very much against it in an existing device. Shielding will be the only way to get some more dB of isolation. Using the same frequency for both TX and RX in one box is a job for the few professionals in this field, and usually it is nearly impossible to modify a finished design to achieve this. Remember that we may talk here about 100dB up to almost 150dB (20 dBm TX out and -125 dBm RX sensitivity are possible with an USRP!), what is quite a lot :)

 

Ralph.

 

From: khalid.el-darymli [mailto:khalid.el-darymli at mun.ca] 
Sent: Monday, October 20, 2014 1:25 PM
To: Ralph A. Schmid, dk5ras
Cc: usrp-users at lists.ettus.com
Subject: Re: [USRP-users] Isolating the grounds of LFTX and LFRX in N200

 

Hi Ralph,

Yes, I understand that the proximity of both PCBs and the lack of shielding is a part of the problem, but I think the common ground too plays a role into this. As mentioned in my earlier email, I personally verified this for the case of two separate N200 devices.

Please see this,
http://web.mit.edu/~jhawk/tmp/p/EST016_Ground_Loops_handout.pdf



All the best,

Khalid

 

 

On Sat, Oct 18, 2014 at 2:31 AM, Ralph A. Schmid, dk5ras <ralph at schmid.xxx <mailto:ralph at schmid.xxx> > wrote:

Not the ground is the problem, but the proximity of both PCBs, and the lack of shielding. Guess it will be a bigger task to shield and isolate them, using separated devices is better.

 

Ralph.

 

From: USRP-users [mailto:usrp-users-bounces at lists.ettus.com <mailto:usrp-users-bounces at lists.ettus.com> ] On Behalf Of khalid.el-darymli via USRP-users
Sent: Friday, October 17, 2014 7:45 PM
To: usrp-users at lists.ettus.com <mailto:usrp-users at lists.ettus.com> 
Subject: [USRP-users] Isolating the grounds of LFTX and LFRX in N200

 

Hi,

I have two N200 devices. One of which is used as a Tx and Rx simultaneously and the other is only used as Rx. The LFRX and LFTX daughterboards are used for Rx and Tx, respectively. In my present test, I am using FMCW chirp with a BW of 50 KHz.

I noticed that if the grounds of the two N200 units are connected with each other (e.g., through using external LNAs with a common ground) I can see the FMCW sweeps with a power around 10 dB above the noise floor in the Rx only N200 unit; without having nothing connected to its Rx input. I managed to absolutely cancel out this effect through simply isolating the common grounds between the two N200 units.

The issue I am having now is with the N200 unit that has both the LFTX and LFRX daughterboards inside it. In this unit, I am seeing the FMCW sweeps at the Rx (without having nothing connected to Rx) with a power of around 20 dB above the noise floor.

It seems that since the LFTX and LFRX daughterboards are plugged-in to the same motherboard, they both share the same ground!

My question is, how do I isolate the grounds of the LFTX and LFRX daughterboards connected to the same N200 unit to minimize the coupling between two of them?

I will appreciate any help on this.

Thanks.

Best regards,

Khalid





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20141020/b29d6a9e/attachment-0002.html>


More information about the USRP-users mailing list