ADALM-PLUTO (new DATV SDR)

[G3GJA]

My Pluto arrived yesterday; getting good results with bit rates up to 3MS/s but 4MS/s causes the output to 'blip' as seen on the SA and it refuses to lock the Minitioune.

16APSK works with 2Ms/s but not 3MS/s, 8PSK just OK with 2MS/s

I've run the software change to allow a wider range of frequencies but if I try to go below 325MHz the Express transmitter crashes. Not sure whether that's because my Pluto can't be forced to go below 325MHz or if there's a problem with the app.

Thanks for your efforts Charles

Clive G3GJA

[g0mjw]

Did you do the below 325MHz hack? It won't work below 325MHz otherwise. There is a link to it further up the thread.

Mike

[g0mjw]

They are all AD9363s and should work if you follow the software mod details.

We are speculating that AD9363s are AD9364s that did not meet the full performance
when tested and were downgraded.

The mod is reversible

- Charles

Think you might be right Charles - it may be the software but in SDR# mine doesn't work first time but does later, like its taking too long to achieve lock or it has to warm up maybe. Might be worth putting it in the fridge and in the sun to see if the 9393s work across the temperature range and also looking at the phase noise, sensitivity and output compared to spec. and if there are any holes in the frequency coverage. It doesn't make sense to have two designs like this with identical performance but reduced coverage. It would make sense if the AD9363 had better performance than the AD9694 within its coverage range, but its doesn't appear to have according to the data sheet. Unless of course, export controls are being applied, but then I would expect the limits to be hard coded.

Mike

[on8ge]

Hello,

Mine arrived yesterday, I am waiting for the corrected software to try it.
Thanks a lot Charles for the big work !!!!!

73

Pierre

[G4GUO]

I will address a few issues.

The crashing below 325 MHz is due to the fact the extended frequency mode has not been done.
Analog Devices libiio which is the interface between my code and the board destroys a channel
if an out of range frequency is selected. That is what causes the crash, there is nothing I can do about that.
I only check the frequency you have requested is between 48 MHz and 6 GHz.

Poor symbol rate performance, unfortunately libiio does not seem to efficiently use the USB2 bandwidth.
It is also machine dependant. On my i7 machine I can get 5 MS/s max performance on my old dual core
machine I can just about achieve 3 MS/s. DATV-Express hardware can achieve 10 MS/s on both machines.

Jerky video, due to the very large buffers required by libiio to maintain USB2 bandwidth the software
currently injects too many NULL packets into the transport stream. It is a bit like trying to talk to
someone on the phone with a long delay, the conversation tends to become very jerky. I am trying to
find an inventive way to get around this problem. Analog Devices has said that later versions of
their library may provide metadata that would enable me better control over NULL insertions.

As you can see these problems are mostly outside my control. I am sure that over time the PLUTO
firmware will improve.

- Charles

[G4GUO]

My Pluto arrived yesterday; getting good results with bit rates up to 3MS/s but 4MS/s causes the output to 'blip' as seen on the SA and it refuses to lock the Minitioune.

16APSK works with 2Ms/s but not 3MS/s, 8PSK just OK with 2MS/s

I've run the software change to allow a wider range of frequencies but if I try to go below 325MHz the Express transmitter crashes. Not sure whether that's because my Pluto can't be forced to go below 325MHz or if there's a problem with the app.

Clive G3GJA

Clive I don't think you have done the mod correctly, check that

# fw_printenv attr_name
# fw_printenv attr_val

return the right values someone else couldn't get the extended range to work and it turned out
fw_printenv attr_val was returning "ad9364" when it should have been ad9364 without quotes

- Charles

[M0SKM]

When I hacked mine I made the mistake of leaving the quotes in; it still works down to 48MHz but I can't yet measure above 2400MHz.

Software is much more stable now; although it still transmits momentarily when the software opens.

Good work Charles.

Off I go to take the quotes out....

[G4GUO]

Give you some idea of what a pain this thing is to program
below is a very small excerpt from the device info table
the bold item was the one needed to suppress the transmitter.

- Charles

iio:device1: ad9361-phy
9 channels found:
temp0: (input)
1 channel-specific attributes found:
attr 0: input value: 28947
altvoltage1: TX_LO (output)
7 channel-specific attributes found:
attr 0: fastlock_store value: 0
attr 1: frequency value: 1249000000
attr 2: fastlock_recall ERROR: Invalid argument (-22)
attr 3: fastlock_load value: 0
attr 4: fastlock_save value: 0 126,126,126,126,126,87,126,126,126,126,126,126,244,126,126,157
attr 5: powerdown value: 0
attr 6: external value: 0
voltage2: (output)
6 channel-specific attributes found:

[M5TXJ]

For info;
Currently Digikey are showing no stock on these units, Mouser UK is the same but they are showing an expected delivery shipping 29/8/17. Price is £76.17 + vat vs £76.08 + vat from Digikey.
73 Dave

[G4GUO]

Bit of fun, Plutos being tested. The Raspberry Pi seems to get everywhere.

https://www.youtube.com/watch?v=xSO-cFeyDFs
https://wiki.analog.com/university/tool ... rs/testing

- Charles