I have 4 limeSDRs that I would like to have daisy chained on the clock. I’m using a python code to set it up and was calling limeSPI for writing the fpga register according to LimeSDR-Mini v2.4 - external clock
But from what I can tell, you cannot pass the serialnr of the device you want to set? Is there a possibility to send the serialnr or should I use something else? Maybe the soapy python bindings for writing registers?
You can select which device to operate by the limeSPI --device argument, it will do a text search among available device handles, and will operate the one that matches, if multiple devices match, a warning about ambiguous selection will be shown, so the argument needs to be more specific to match only one device.
For example limeDevice lists all available device handles:
limeDevice
Found 1 device(s) :
0: LimeSDR Mini, addr=0403:601f, serial=1D9EFA3E84B944
You can specify a part of serial number limeSPI --device 1D9EFA3E --chip FPGA read -s 00cf
Or you can specify the entire device handle limeSPI --device "LimeSDR Mini, addr=0403:601f, serial=1D9EFA3E84B944" --chip FPGA read -s 00cf
Without a distributor, each node in the chain will be creating more and more phase delay and jitter and all kinds of stuff that you generally don’t want. Plus, a 40 MHz ref has a wavelength of about 7.5 meters (25 feet) and so you’ll really need to pay attention to your overall cable lengths if you’re wanting to do phase coherent sampling or something weird.
But if you’re just wanting them to all be on the same central frequency, then I guess the daisy chaining would probably work.
Let us know how the daisy chaining actually works out – I’m curious about it myself.
Thanks for your input! I am using them to create a disturbance environment and I am not using the RX at all so far. I might be doing that later, so it’s good to know about eventual pitfalls.
I have them set up and so far it looks like it is working