My ADALM-Pluto transmits intermittently.

xugmu said:

Library version: 0.10 (git tag: v0.10)

Your version of libiio is pretty ancient. The current version is v0.23. Upgrading will make those errors go away.

xugmu said:

I am trying to transmit an 8Mhz signal with my 3 days ADALM-Pluto and gnuradio , if I set cyclic to true, the signal appears stable in the spectrum analyzer, but if I set it to false, the spectrum appears and disappears continuously. It is a dvb-t signal and the television analyzer does not hook up in either of the two ways

Since Pluto is USB2.0 based it will have limited transmit speed. To maximize the USB throughput I would use relatively large buffers and switch to a CDC mode: https://wiki.analog.com/university/tools/pluto/users/customizing#changing_the_usb_ethernet_compatibility_mode

-Travis

I've been dealing with this issue for several days and there's no way.

The only way to work with gnu radio is with ubuntu 18, synaptic and its corresponding packages gr-iio and libiio(0.10) and gnuradio 3.7.

In the rest of the occasions: gnuradio 3.8, 3.9, 3.10 , Ubuntu 20, I have problems, the most common is that gnuradio does not recognize the gr-iio package (soapy pluto source  appears but gnuradio tells me that it does not find the ADALM-Pluto)., other times it turns out that packets don't match.

With the hackrf I have no problems transmiting 10Mhz, usbtop gives me a stable 4 MB/s I think I remember. I understand that the adalm-pluto is twice the number of bits but I don't quite understand the limitation.

I would appreciate a stable way to get this device up and running.

Best regards

Actually what I was interested in knowing is if the ADALM-Pluto is capable of transmitting 8 (preferably 10) Mhz, logically with a signal that is continuous over time, not repetitive every 100 ms.

With 20 seconds I think it would be enough for the test equipment hook up and measure the signal.

Best regards

So 10 MHz is 40 MBytes/sec (4 bytes per complex sample). The theoretical max of USB2.0 is 60 MBytes/sec but realistically you will get ~20->30 MBytes/sec with real controllers.

There is a benchmark feature in the latest libiio master builds you can test with. See the -B flag.

-Travis