Digital Discovery to collect 16 bit SPI data - multiple slaves with same CS, CLK, MOSI and individual MISO

[JAlaj]

Hello 

I am looking to use Digital Discovery to capture 16 bit SPI data on multiple slaves, all slaves will have CS, MOSI and CLK shorted, and MISO line will be individual for each slave. I want to capture continuous data for 24 hrs and dump it in a  text file

I also want to power up the part using same HW, so that i can capture SPI right after power up

Pls suggest, 

Thanks,

Jalaj

[attila]

Hi @JAlaj

The sampling rate Logic Analyzer should be more at least twice of the SPI clock frequency !

Only the SPI transfers in the loop function are packed !
The loop Rate should be less than SPI Frequency/ SPI bits in loop cycle ! above 8-9MHz for 16bit * 500kHz
Do you want to read or write or.. ?

Currently the Protocol only supports the 16 DIOs.
I could add support for DINs as input but the data streaming need for 500kHz rate (72MBps) is beyond the USB bandwidth (45MBps). At this rate the device memory is sufficient only for about one minute.

[attila]

Hi @JAlaj

When capturing up to DIN15 the USB bandwidth may be sufficient, 19*500kHz*32bit = 38MBps but I'm not sure if the data processing can keep up with it.
The resulting record to file after 24hours would be more than 8TB !

[JAlaj]

Thanks Attila

Is there a way to dump data in text file and capture simultaneously so that i never hit the memory limit, something like filling/emptying FIFO

Also, how do you dump the 15 MISOs data in text file

Regarding your question on read/write in my SPI commands....my part works on below SPI format 

bit15 , bit[14:0]

if bit 15 is 0, then its a read command for part

if bit 15 is 1, then its a write command for part...

so if give Write (16, 0x0055), then i am reading address 55

if i give Write (16, 0x0055) followed by another Write (16, 0x00AA)....here i am writing register 55 with data AA

 

Regards

JAlaj

[Shalin]

Thanks Atilla for sharing the latest version. It has resolved most of the things. I will let you know in case I find any bug related to the functionality of SPI.

Edited August 25, 2023 by Shalin

[Shalin]

Hello @attila

In Sensor tab, when the commands are mentioned outside the "function loop()", there is a huge delay between each command. The similar commands written in the "function loop() have smaller delay. Please maintain smaller delay between the commands written outside the "function loop ()". I have attached the screenshots below for your reference. 

Thanks and Regards,

Shalin

 

Edited August 28, 2023 by Shalin

[Shalin]

Thanks @attila . 

Further, we are planning to use Digital discovery for SPI interfacing in our upcoming project as we earlier did with SWD. 

I wanted to know that do you have the support of SPI interface with LabView. Previously, you had provided the same for SWD.

I will be grateful if you can provide me the LabVIEW support for SPI interfacing as well. 

Shalin

[attila]

Hi @Shalin

The VIs provide simple usage and for advanced usage all the API functions can be called from LabVIEW, just like from Python...
The newly added multi-miso option is only available in the WF application.

 

[Shalin]

Hello @attila

I tried calling API functions in LabView. However, I am not able to generate the correct pattern using API functions. In LabView, I am trying to write 0091 in hex. In the output, I am unable to see the clock and getting invalid MOSI and MISO response. I am attaching the labview VI which I have created and its output captured by logic analyzer.

Thanks,

Shalin

SPI_Read.vi

[Shalin]

Hello @attila

I have extensively checked the API functions of SPI in LabView. These developed API functions are not operating correctly. As an example, I have also attached the sample LabView VI for reference in my previous post. 

I would be grateful if this issue can be rectified. 

Also. let me know if you require some additional information from our end.

Thanks,

Shalin 

[attila]

Hi @Shalin

See the following example where 3 words of 16bit are set on TX and 4 words received on RX dwfvi.zip

 

Captured with an AD3:

 

I have LV 2022 installed so I can't open your project.

[Shalin]

Thanks @attila. I will try it.

Also, I want to know that how do you record the patterns in waveform GUI when you send command through LabView. I was trying the same but the waveforms GUI app says that the device Digilent digital discovery is being used by other application (labview). As a result, I am not able to record the labview pattern in waveforms GUI. 

[attila]

Hi @Shalin

A device can be controlled by one app/process at a time. In the earlier test I used two devices.
LabVIEW controlled Digital Discovery with loopback wire between MOSI-MISO, then connected the SPI signals to an Analog Discovery 3 controlled by WF app.

[Shalin]

On 9/7/2023 at 7:49 PM, attila said:

Hi @Shalin

See the following example where 3 words of 16bit are set on TX and 4 words received on RX dwfvi.zip

 

 

Captured with an AD3:

have LV 2022 installed so I can't open your project.

Hello @attila

The API functions of the latest waveform version 3.20.31 is different in comparison to the API functions listed in the dwfvi.zip file.  One of the functions listed in the dwfvi.zip file with name "F DWF digital SPI write read 16" has got array unsigned type of "rgTX" parameter whereas the same function listed in the latest SDK version 3.20.31 does not have array unsigned type of "rgTX" parameter. Could you please make it array type in the latest SDK versions API as well ?   

Most importantly, I am sending a long sequence using "read write 16"  API function listed in dwf.vi. However, the output captured in LA is inappropriate. Similarly, I am receiving dissimilar output captured by MISO LA and MISO rgRX. I am attaching the VI below for your reference and some screenshots of LA. 

Thanks,

Shalin

 

 

 

dwfvi.vi

Edited September 18, 2023 by Shalin
This is a new query @Attila

[Shalin]

On 9/1/2023 at 9:41 PM, attila said:

Hi @Shalin

The VIs provide simple usage and for advanced usage all the API functions can be called from LabVIEW, just like from Python...
The newly added multi-miso option is only available in the WF application.

 

Hello @attila

In the labview, I am not able to receive MISO data for the input data. I am only able to see the data written through MOSI (rgTX) but the output obtained on the MISO channel (rgRX) is incorrect. On the other hand, I have compared the same by using waveform GUI application where you can see that for 0100, the output in MISO channel is 00D0.  I am attaching some screenshots below for your reference.  

Please optimize the API functions so that they work and deliver similar output as the waveforms GUI application. 

Thanks,

Shalin

[attila]

Hi @Shalin

See the example I attached earlier.
In C code a pointer can point to one value or to an array. In LabVIEW make sure to use the proper type.

[Shalin]

Hello @attila

I am facing an issue while I probe signal of MISO line on logic analyzer. In the  labview, when I write 0100, the MISO output (rgRX) reads value 00D0. Further, when I probe this output on logic analyzer, I am able to see the input (0100) written on the MOSI whereas I am not able to see 00D0 on the MISO channel. A wrong value FF is shown there. Please help regarding the same. 

Thanks,

Shalin

[attila]

Hi @Shalin

You have changed the MISO to be #4 DIO28
With a wire between MOSI-DIO26 and MISO the capture is correct:

 

[Shalin]

Hello @attila

My query is about when I see 00D0 in rgRX LabVIEW capture, why I am not able to see the same in LA capture on MISO pin. In LA capture, I am seeing FFFF instead of the response 00D0. 

Thanks,

Shalin

[attila]

Hi @Shalin

Make sure you are looking at the correct MISO wire.

2 hours ago, attila said:

You have changed the MISO to be #4 DIO28

 

[Shalin]

Hello @attila

I have checked the connections of the MISO pin connecting to the logic analyzer. The connections are perfectly fine. The labview is showing correct response of 00D0 whereas in LA it is showing FFFF. I am attaching the screenshot of Labview response and LA output below. 

Thanks,

Shalin

[attila]

Hi @Shalin

Are you using the latest software version ?

 

[Shalin]

Yes @attila, I am using the latest version 3.20.31.

[attila]

Hi @Shalin

For me you project returns the loopback data correctly.
Have you changed anything else in the subVIs ?

On 9/22/2023 at 1:07 PM, attila said:

You have changed the MISO to be #4 DIO28
With a wire between MOSI-DIO26 and MISO the capture is correct:

 

[attila]

Hi @Shalin

In your screenshot the MISO does not look right, it is very noisy, normally this should be stable for a clock period.
Such noisy MISO can not be received, interpreted correctly.

7 hours ago, Shalin said:

 

 

[Shalin]

Hello @attila,

The noise was within the Logic analyzer. I have switched to a new LA and the problem is resolved now.

Further, I am trying to write a command 0000. I am not able to receive any response against this command in the rgRX. Could you please help, I am attaching the developed VI below.  

Moreover, is it possible to write the 4 words in a single row rather than writing them individually in a separate row ? I am attaching the screenshot below to understand better. 

Thanks,

Shalin

 

 

SPI_read_3.vi

Edited September 27, 2023 by Shalin
attached VI