Hi all,

The strontium lab at the JQI is finalizing the computer control system for
our new experiment, and I am finally revisiting ARTIQ to figure out what
hardware we need.

Our current experiment uses the following:

* 48 digital outputs (8 are designated as triggers for other devices)
* 24 analog outputs (from NI's PCI-6733 DAC cards)
* several (~8?) homebuilt DDSs, which aren't phase synchronized but have
profile-select control for hopping frequencies/initiating simple frequency
ramps with ~ < 1ms latencies.
* Point Grey / PIXIS cameras for image acquisition, currently analyzed
"offline" between experimental shots
* For analog ramps, the fastest update rate we currently use is 10us/sample
(although that also has to do with our shitty labview control interface,
which can crash if your DAC sample buffer gets too long...)
* Eurocard 19" rack enclosures are already used for most of our custom
electronics; we like this form factor.

That all said, in the new system we would like

* A few phase synchronized DDS (likely operated at 30-50MHz), eg, for
fancier lattice AOM control, feed forward/feedback on narrowline clock
laser, ... other
* analog input
* upgraded analog output to move away from NI; some channels with PDQ-esque
AWG capability, although most of our analog control needs are relatively
slow.
* more analog/digital I/O for future expansion.

Given those requirements, here's what I'm thinking to buy in the short &
longer term. Since we're still building the vacuum chamber, many of the
"fancy" control requirements are many months away. I was hoping to get
input/feedback/suggestions as to whether this plan sounds good:

* For ARTIQ core device, we would ideally jump straight to using a Kasli,
but as that isn't likely to be done in the next few months, I was planning
to use a KC705 as the core.
* KC705 has 2x FMC headers, which would drive 1x VHDCI carrier card,
providing 8x IDC for EEMs. We would buy FMC -> VHDCI adapters for this
interconnect.

>From 8 EEM headers, we could get:

* 6x BNC EEMs -> 48 DIO
* 2x EEMs for... analog IO? Initially, we could continue to use the 6733
DACs, but longer term switch to Zotino/PDQ/...

It seems we will eventually want at least one Kasli expansion DRTIO module.
This would be connected on the KC705 SFP.

**Specific questions**:

* what limitations are there (latency/bandwidth/etc) on daisy-chaining
additional Kasili DRTIO modules off of the single KC705 SFP?
* What IO mapping needs to be done to support KC705 as a core driving EEMs
over the two FMC connectors? Is there any plan to allow the HPC connector
to drive an entire VHDCI adapter on it's own (thus expanding the KC705 to
drive 12 EEM modules, 1x from the HPC + 0.5x from the LPC), and is this
even desirable over, say, adding a Kasli over DRTIO?
* Is there an estimate on the timescale for finished Kasli?
* Should I reconsider any of this approach, or does it seem reasonable? Any
other suggestions/advice?

Thanks in advance for the input!
Neal
_______________________________________________
ARTIQ mailing list
https://ssl.serverraum.org/lists/listinfo/artiq

Reply via email to