rohitksingh has quit [Read error: Connection reset by peer]
rohitksingh has joined #m-labs
rohitksingh has quit [Read error: Connection reset by peer]
rohitksingh has joined #m-labs
<rjo>
hartyp: yes. you also did excellent work on the spec. we should try formalize/streamline/accelerate the review process for the other/next projects.
<GitHub30>
ionpak/master a39c95e Sebastien Bourdeauducq: HV PID control (untested)
sb0 has quit [Quit: Leaving]
hartytp has joined #m-labs
hartytp has quit [Ping timeout: 260 seconds]
sb0 has joined #m-labs
rohitksingh has quit [Quit: Leaving.]
hartytp has joined #m-labs
<hartytp>
rjo: thanks. Agreed though, while we're getting to a good result in the end, the current specification/review process is a bit ad hoc and inefficient
<hartytp>
Can you remind me:
<hartytp>
1. Is ARTIQ support for the FMC on Sayma already funded (e.g. to use with the FMC to VHDCI + VHDCI carrier)?
<sb0>
what kind of "support" would that be?
<hartytp>
2. What is the status of Kasli ARTIQ support? DRTIO slave only? Or, is using it as the master also funded?
<hartytp>
e.g. can I run a Zotino off Sayma?
<hartytp>
We discussed how this would work before, but I wasn't clear if it was actually funded yet
<sb0>
well, just "running a Zotino off Sayma" is basically adding the pins and core and recompiling the gateware (in theory at least, assuming vivado doesn't crash etc.)
<sb0>
for schemes that would not require gateware recompilation there are three ways:
<sb0>
1) find a set of peripherals in a single gateware config that would satisfy everyone
<sb0>
2) do some soft-muxing of the FPGA pins e.g. between SPI and TTL
<sb0>
3) do gateware partial reconfiguration with something like recobus
<sb0>
#1 might not really require funding, #2 may be funded via opticlock (rjo?), #3 is complicated, requires fpga reverse engineering, and is not funded
<sb0>
hartytp, by the way, have you seen ionpak? you commented about this project a few months ago
<hartytp>
sb0, right. I though we agreed on option (2). See #164
<sb0>
is there something that I don't know about peltier drivers?
<sb0>
I'd just do that with a H bridge + thermistor + MCU, but I've seen all sorts of complicated designs
<sb0>
or just a mosfet instead of the H bridge if you only want to cool
<hartytp>
That's about all the Thorlabs chip is. Just depends how much one wants to do oneself versus buying it in a convenient package. In our case, while the electronics is fun, we're pretty short on time
stekern has quit [Ping timeout: 245 seconds]
<sb0>
if you're already adding a MCU for ethernet, you might as well just add mosfets instead of this proprietary thorlabs thing
stekern has joined #m-labs
<sb0>
hartytp, oh and probably calcium, since the laser diodes are relatively easy to procure
<sb0>
hartytp, for ethernet on the mcu, maybe you can just reuse the ionpak code
<hartytp>
sb0: from #148 I get the impression that the default bitstream for Sayma AMC will have the FMC configured for muxed TTL/SPI phys, and so should be fine for Zotino
<hartytp>
re temp controller: I did think about that. If I have time, I would be interested in having a play with that. We'll see though if/when I find the time...