Page 1 of 1

External clock input of 10MHz

Posted: Mon May 21, 2018 5:06 am
by vkkiisc
I would like to use 10 MHz external clock to bladerf from other sources(like GPSDO) to generate my waveforms.
I read the default external clock input to the bladerf is 38.4MHz. Now i would like to configure from 38.4MHz to 10MHz external clock input.
Is there any direct settings to change or i need to modify the main bladerf code or any other way to do this?
Please help me.

Re: External clock input of 10MHz

Posted: Wed May 23, 2018 11:35 am
by bglod
One way to do this is to use J71 and enable the VCTCXO tamer algorithm as shown in this blog post.

Re: External clock input of 10MHz

Posted: Thu May 24, 2018 12:20 am
by vkkiisc
Thank you so much bgold for the reply. I have done the same thing to solve my problem.

Re: External clock input of 10MHz

Posted: Thu May 24, 2018 10:46 am
by bglod
Great! Let us know if you have any issues with it.

Re: External clock input of 10MHz

Posted: Mon May 28, 2018 9:51 pm
by vkkiisc
Hi bgold

I have an issue.
I followed this link: https://www.nuand.com/blog/2016-01-rc1-release/ for VCTCXO Taming via 1 PPS or 10 MHz input.
I changed the trim dac settings with reference from CSAC GPSDO 10MHz clock. But i dint take the backup of previous dac settings.
Now am worried about the performance of Bladerf. How should i go back to default settings of DAC. Will this be a issue for my further work progress in generation of signals. How will this effect me if i dont go back to default settings?

Thank you in advance

Re: External clock input of 10MHz

Posted: Tue May 29, 2018 1:32 pm
by bglod
We have a factory calibration lookup page: http://www.nuand.com/calibration.php

If nothing shows up for your serial number, please send us an e-mail with your serial number and I'll see if we can get it from the factory.

Re: External clock input of 10MHz

Posted: Tue May 29, 2018 11:51 pm
by vkkiisc
Thank You Bgold once again...

Re: External clock input of 10MHz

Posted: Wed Oct 24, 2018 7:18 am
by floor51
Hi,

I have bought the Leo Bodnar GPSDO. It doesn't come with a cable to connect it to the bladeRF.

Can you advise me what components I need?

Based on the photo on the blog page (https://www.nuand.com/2016-01-rc1-release/) I'm guessing:

1xconnector to the GPSDO. A crimp plug? https://cpc.farnell.com/pro-signal/jr60 ... dp/AP01791

1xattentuator - I see 10dB and 50ohm

1xgrey wire sheath??

1xyellow thing?? for soldering the 2 wires from the circuit board connector

1x PC style connector to circuit board

1xRG58U coax cable inside the grey wire sheath? https://www.thunderpole.co.uk/cable/rg5 ... cable.html

Re: External clock input of 10MHz

Posted: Wed Oct 24, 2018 9:50 am
by bglod
It's not as complicated as that! The picture we have on the blog post is just a quick frankenstein lab setup with random parts we had on hand. The silver cable is just an SMA cable, no sheaths or anything like that. It doesn't even need to be SMA; you can stick with BNC the whole way. All you need is the attenuator to protect the FPGA input (6 dB or 10 dB is fine, make sure it's 50 ohms). Then you need a way to get from the attenuator to a square header socket while maintaining 50 ohms characteristic impedance for as long as possible before breaking off into the sockets. You can do this however you want.

They sell "test leads" that do the conversion for you: BNC(M)-to-square-socket, BNC(F)-to-square-socket (combine this with an ordinary BNC cable for more length if needed), BNC(M)-to-minigrabber, among many other options, including the home-built mess shown in the picture!

Cheers!

Re: External clock input of 10MHz

Posted: Wed Oct 24, 2018 2:58 pm
by floor51
Oh phew, thank you :D I like the frankenstyle but the test leads sound much better.

Found this attenuator that is 10db, 50ohms and is BNC both sides to go with the BNC to square socket connectors you posted:

https://www.minicircuits.com/WebStore/d ... =HAT-10%2B

Re: External clock input of 10MHz

Posted: Mon Oct 29, 2018 12:34 pm
by bglod
That's the very same attenuator we (now) use in the lab for this purpose, so that should work well! :)