ADALM-PLUTO SDR

[mrf184]

re: tuning range, I bought some AD9363 chips ($10 each from shenzhen), made a quick test board with a Spartan-6 FPGA and USB interface, and wrote some code to configure it over USB using the AD no-OS API. The API does not ask the chip whether it's a 9363 or 9361, but rather you specify this in your code. I specified AD9361 and it tunes the full frequency range. Here it is picking up some FM broadcasts:

[RoGeorge]

That's a very interesting project (especially given the price for AD9363), I would like to see more. 

[xaxaxa]

bootleg plutosdr (collab with mrf184):


It can do up to 14MS/s through the usb 2 interface.

Will post schematics and layouts.

[Kalvin]

Got mine today from digikey. Tuning hack and dual-core hack work just fine. After tuning hack I can receive FM stations around 95 MHz using the Gqrx 2.11.5. Installing the GNU Radio package and building the GNU Radio blocks for Pluto on Linux Mint 18.3 was easy without any problems.

[xaxaxa]

Here are the schematics and PCB layout (gEDA) of the simple ad9363 SDR. There is one errata which is that the RESET pin of the ad9363 should be wired to a FPGA i/o pin.

EDIT: also added gschem symbol library

The ad9363 is connected to a spartan 6 FPGA (xc6slx9) which interfaces to USB using the usb3343 phy. Attached is a VHDL core that packs the 12-bit I/Q samples (24 bit words) into 3 bytes and sends an occasional sync word, plus a gnuradio plugin (in c++) that goes along with it. To configure the ad9363 currently the fpga logic accepts bitbanging commands via usb, and code is added to platform.c (in AD's no-OS library) to bitbang spi.

There are 4 RF channels routed, 2 tx and 2 rx. One tx/rx pair is connected to a 800/900MHz frontend (antenna switch + acpm-7868 PA), and the other tx/rx pair is broken out directly (the tx has a wideband trf37a75 gain block for higher output power up to 15dBm).

Total cost including PCB prototyping is less than a real plutosdr 

[Hagrid]

Hello, has anybody got an pluto SDR recently (a few weeks or days ago) and tested if it still can be unlocked to the full frequency span (70MHz to 6GHz)?
It would be nice if someone has made some tests over the whole range with newer units too.

[tsman]

Hello, has anybody got an pluto SDR recently (a few weeks or days ago) and tested if it still can be unlocked to the full frequency span (70MHz to 6GHz)?

Same revision PCB. Same revision Zynq and AD9364. You can still unlock the second ARM core in the Zynq and still expand the tuning range of the AD9364.

[OwO]

Anyone know if it's possible to add an SD card or some other way to expand storage on the plutoSDR? As it stands there's no way to install a real Linux system because the onboard flash is way too small. I want to run something proper like Debian so that development isn't a pain.

[Hagrid]

Hello, has anybody got an pluto SDR recently (a few weeks or days ago) and tested if it still can be unlocked to the full frequency span (70MHz to 6GHz)?

Same revision PCB. Same revision Zynq and AD9364. You can still unlock the second ARM core in the Zynq and still expand the tuning range of the AD9364.

Thanks a lot! Will order today, thats a really nice price for such a unit.

[tsman]

Anyone know if it's possible to add an SD card or some other way to expand storage on the plutoSDR?

If you have micro USB OTG adapter then you can attach USB devices. A USB 3 gigabit Ethernet adapter is recommended even though the ADALM-PLUTO is only USB 2 as 100BASE-T will be a bottleneck.

[wolfp]

I received PLUTO in July 2018 and expanded it to 70MHz - 6GHz. Then I made some measurements using a calibrated R&S Generator and a R&S (broadband) power measuring system.
Result: Output power 50MHz to 6GHz more then 0 dBm, max. 9dBm at 500MHz. The output is non-sinusodial therefore the result (measured broadband) is influenced by the waveform.

I tested the receive-path using a -20dBm  signal from 70MHz to 6GHz. Pluto showed abt. -30dBfS with a maximum of -24dBfs at 1500MHz (i have to look closer for the reason). The usable dynamic-range seems to be between -80dBm (-90dBm under 3GHz) and -10dBm.

All attenuators set to 0dB, using 0.1m RG174 SMA-SMA (comes with PLUTO) and N-SMA adapter.

Wolfgang

[ploegmma]

Just a small post regarding my experience regarding crashes and lockups with the ADALM-PLUTO board (Rev. B):

My Pluto board sometimes seemed to stall when connected to the antenna - SDR# also crashed when such a stall happened. Using an USB-ETH-Interface the situation did not change at all - the board would still stall from time to time.

Some measurements later the problem was narrowed down to electric pulses due to switching of lamps and (capacitor-based) motors.
Those devices caused a spike for in the coax-shield (approx 60V, 5-20µs duration, 1-2 MHz freq.). 
To check if the device was frozen or just the USB connection I decided to connect an CP2104-MINIEK serial adaptor to the internal RX/TX/GND and see if the device continues to run when the connection is lost.... but the problem was gone - no chance to reproduce it while the serial cable was fully attached.
Well, that was a start - and according to the Analog wiki there were reports for EMC problems with the USB-OTG-detection circuit... so I decided to check out the grounds. 

Diagnosed problem:
Power-GND and GND have a common mode choke as only connection, but that choke does more harm than good when there is a sudden rise in the GND "level".

Solution:
My fix for the situation was simple: I added a bypass/bridge that directly connects GND and Power-GND. In addition I also bridged the 1k resistors to the USB-shields, but I suppose that measure was unnecessary.

Attached you can find the schematic.

I'm glad I found this thread. I think I have the same issue. I'm using the Pluto with a network adapter over OTG and that seems to work great. The whole point of this was to get the Pluto off my desk and install it near my discone where I have UTP also. But then the problems started. It stopped working and I spent hours on searching for a cause (fiddling with the network, cable, switch, adapter, power, etc.). At a certain moment I dragged a long cable with me from my desk (where things worked) to the antenna location where it stopped working each time. Then (when it still stopped working at that location) I figured that the only difference was the antenna. I just had to contact the antanna ground to the Pluto (antenna input) ground for a while and the network activity LED on the network adapter stopped blinking!
To figure out the rest of the story was a bit outside my knowledge area. But when I read your post it all made sense... So of course I want to make this modification. Perhaps you could make a close-up photo of it (just to be sure). And did you use enameled wire? What gauge (if that matters)?

Thnx, Majodi

[ploegmma]

Regarding any added shielding: I experienced some RF-reflection problems - in my case I tried adding copper planes of single sided PCBs (cut to fit the enclosure) ... such shielding didn't help with the OTG-loss but caused the noise-level to worsen significantly - especially when the shield was close to the RF-input, chip and the chokes in between both. It's not necessarily bad to radiate some unwanted RF away - I didn't have RF-absorbant material around, but even then I'd expect bad HF voodoo to happen when adding anything with a dielectric constant near any RF part

PS: What also helped (a bit) was just creating a really good GND-connection (I used alligator-clips and lab cables) from the SMA port to my local grid's earth conductor - this reduced the GND-interference conducted by the coax-shield arriving at my pluto by providing an alternative path to earth. For obvious reasons this just helped with mild interference as it just weakened the incoming stray signals (at risk of ground-loops).
Choking with a clip-on ferrite didn't help as the frequency is quite low and the intensity too strong to simply choke away.

This I would like to test also. I made a shielded box to fit the Pluto, hoping it would create the best possible environment for it. Didn't think it could have a negative side to it. Where (frequency) and what exactly do you see with these RF reflections?

Majodi

[a59d1]

re: tuning range, I bought some AD9363 chips ($10 each from shenzhen), made a quick test board with a Spartan-6 FPGA and USB interface, and wrote some code to configure it over USB using the AD no-OS API. The API does not ask the chip whether it's a 9363 or 9361, but rather you specify this in your code. I specified AD9361 and it tunes the full frequency range. Here it is picking up some FM broadcasts:

where tf in shenzhen are you getting a $120 chip for $10?

[OwO]

https://item.taobao.com/item.htm?spm=a230r.1.14.59.48bc2db7Ybdwv7&id=576472782264&ns=1&abbucket=9#detail

https://item.taobao.com/item.htm?spm=a230r.1.14.17.48bc2db7Ybdwv7&id=539437082795&ns=1&abbucket=9#detail

[a59d1]

https://item.taobao.com/item.htm?spm=a230r.1.14.59.48bc2db7Ybdwv7&id=576472782264&ns=1&abbucket=9#detail

https://item.taobao.com/item.htm?spm=a230r.1.14.17.48bc2db7Ybdwv7&id=539437082795&ns=1&abbucket=9#detail

Wow. Surely these are factory rejects, right?

[radiolistener]

Wow. Surely these are factory rejects, right?

I think these are Chinese clones, they are not genuine

[OwO]

I think these are Chinese clones, they are not genuine

I would be impressed if the Chinese can actually clone a chip of this complexity.

There is a peculiarity with the taobao AD9363 which is they all have the BBCZ suffix while distributors all only stock the ABCZ suffix. Arrow lists AD9363BBCZ but without a price and not available for order. The datasheet also only lists ABCZ.

EDIT: https://ez.analog.com/wide-band-rf-transceivers/design-support/f/q-a/80027/what-is-difference-of-ad9363-abcz-and-bbcz

BBCZ is not an open market release part, Kindly request customer to contact ADI directly for such queries.

FYI.

AD9363ABCZ Band: 325 MHz to 3.8 GHz

AD9363BBCZ Band: 650 MHz to 2.7 GHz

These are the ones that went into DJI drones and now have hit the market.

[radiolistener]

I would be impressed if the Chinese can actually clone a chip of this complexity.

then, how they do it cheaper? Factory rejects?

[2N3055]

 
there is only AD9361BBCZ in AD datasheet....
https://www.analog.com/media/en/technical-documentation/data-sheets/AD9361.pdf

[RoGeorge]

I don't think the AD9361 price is about chip complexity.

Saying that because, judging by the current selling price, it's hard to believe the AD chip is on par with, let's say, a CPU like AMD or Intel, which are about the same price, and running at about the same frequencies, yet the CPUs are way more complex and huge in area when compared with an AD9361.

[radiolistener]

the CPUs are way more complex and huge in area when compared with an AD9361.

the CPUs produced in large quantities, so their price is distributed over quantity.
But this is not the case for RF Transceiver chip.

[tsman]

https://zeptobars.com/en/read/AD9361-SDR-Analog-Devices-DAC-ADC-65nm

[vik007]

Hello Gentlemen,

A few days ago I bought an Adalm Pluto SDR, I did the first Test of reception and it works very well.

I tried to make a change after this tutorial:     I did not succeed, I probably did not do something right or it was not the software.

After this failed operation, plutoSDR does not do anything, no longer sees any COM port, the computer no longer recognizes it as a storage unit.

When connecting to USB sees only the Device Manager PlutoSDR DFU.   

Has anyone else happened to this wonderful little box?

Please help me,

Thank you  (Victor - yo3dex)

[mayor]

Hi,

hopefully, this article will be of help:

https://wiki.analog.com/university/tools/pluto/users/firmware#dfu_update