Slow reprogramming of Novatech DDS boards

[Rohit Prasad Bhatt]

Hi all,

We are using the NovaTechDDS9M board in our experiment. In order to avoid the 100 us timing limitation for all other channels (described here : https://groups.google.com/g/labscriptsuite/c/QdW6gUGNwQ0/m/AJTUufvbAAAJ), we are using the NovaTechDDS9M with a prawn blaster secondary pseudoclock. The master clock is a pulseblaster.

We have noticed that if a lot of instructions need to be written in the memory of the DDS, it takes a long time of several tens of seconds. The number of instructions are less than half the maximum available set of instructions. We were able to improve the speed by selecting highest baud_rate of 115200, but the reprogramming still takes about 25 seconds.

We also observe that the reprogramming time decreases if we reduced the amount of instructions. This is expected, but we don't understand what could be the main limiting factor for the slow reprogramming of the NovaTechDDS9M.

Does anyone have a clue about this?

Regards,
Rohit Prasad Bhatt

[Philip Starkey]

Hi,

How is the serial port attached to the PC? (USB/PCI/Ethernet/onboard)

Generally whenever I've seen slow Novatech programming times it's been due to the serial port. For example, we once tried using a lightweight ethernet to serial adapter that really slowed down the programming rate. I believe some USB to serial adapter's suffer from the same issue.

As such, I would see if you can reproduce the slow programming outside of BLACS. That will indicate if it's a labscript suite issue or not.

Let us know what you find. Quite a few people use them so if it is a serial port issue, it should be solvable as long as it's not being limited by the processing time of the Novatech microcontroller.

Cheers,

Phil

[Rohit Prasad Bhatt]

Hi Phil,

The serial port of Novatech is attached to the PC via a USB to RS-232 adapter cable. This plugs into a USB port of the PC. As you mentioned, the cable itself could be an issue because this is a pretty cheap one. I also tried to install the drivers in the recommended way as per the vendor website but that did not help either.

If the situation improves with another cable, I will mention it here. Thanks for your suggestions!

Regards,
Rohit Prasad Bhatt

--
You received this message because you are subscribed to the Google Groups "the labscript suite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to labscriptsuit...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/labscriptsuite/8fe6d6f7-1deb-4fda-a989-26b78dc6d848n%40googlegroups.com.

[dihm....@gmail.com]

Rohit,

I will say that my personal experience is about the same as yours so I don't think it is going to be your adapter. I have a standard sweep sequence with 3000 points and it takes about 15s to program them all fresh at 115200 baud. Unfortunately, all of the commands are sent as ascii text, each step requires ~2x29 bytes (two channels at each step), and 115200 baud is only 11520 bytes/s. That comes out to just shy of 200 table steps per second.

Kind-of a shameless plug, but a student and I have been working on an alternative that greatly limits the number of commands that need to be sent if most of your points are being consumed by linear ramps of a single parameter. Code for the device is here. Haven't gotten around to the labscript driver yet but it hopefully won't be too hard to write.

-David

[Ian B. Spielman]

This is also in line with our experience with the Novatech devices. In practice we make heavy use of the smart programming option for which the novatech usually is only an issue during the first shot of a sequence of similar files.

— Ian

Ian B. Spielman

Fellow, Joint Quantum Institute
National Institute of Standards and Technology and the University of Maryland

----- WEB -----
http://ultracold.jqi.umd.edu

----- EMAIL -----
spie...@jqi.umd.edu

----- ZOOM -----
https://umd.zoom.us/j/7984811536

----- PHONE -----
(301) 246-2482

----- MAIL -----
UMD:
2207 Computer & Space Sciences Bldg.
College Park, MD 20742

NIST:
100 Bureau Drive, Stop 8424
Gaithersburg, MD 20899-8424 USA

----- OFFICE -----
UMD: Physical Sciences Complex, Room 2153
NIST: Building 216, Room B131

> To view this discussion on the web, visit https://groups.google.com/d/msgid/labscriptsuite/ff6c0a5b-6ba4-495b-aa16-31a5a147088bn%40googlegroups.com.