search

sinara-hw / Urukul

4 channel 1GS/s DDS (AD9910 or AD9912 variant)
14 stars 8 forks source link

Depopulate screening clips? #10

Closed hartytp closed 6 years ago

hartytp commented 6 years ago

I'm happy to leave them in, but removing them saves a BOM line.

We've tested Urukul quite extensively at this point, including in a rack with Kasli as an EMI source. We don't see any evidence of cross-talk or pickup at a level that bothers us (which shows that @gkasprow's routing is good!).

gkasprow commented 6 years ago

we can make them DNP

hartytp commented 6 years ago

any objections? @jordens

jordens commented 6 years ago

OK. But I'd still like to see a measurement of the crosstalk (kasli to urukul, urukul to urukul, and channel to channel) with the cans mounted on one of the existing boards. The crosstalk channel channel to channel with the cans not mounted. was done in December.

hartytp commented 6 years ago

"OK" as in "okay, let's DNP them" or as in "okay in principle, but let's get some data first"?

Would be an interesting measurement to make, but we'd need to get some screening cans made first.

jordens commented 6 years ago

DNP them now, get the cans made for an existing board, and measure. Same goes for Zotino and Sampler.

hartytp commented 6 years ago

:+1:

dnadlinger commented 6 years ago

Ack. The one caveat I would add is that it can be quite difficult to provoke EMI problems without proper test equipment (although I guess the main worry for Urukul would just be picking up single RF tones at inconvenient frequencies, which we could easily simulate using a loop antenna and the Frankonia amp).

jordens commented 6 years ago

Yes. But I think that's only difficult if you need to split the attenuation into emitted and immitted. I didn't know Frankonia also builds amplifiers...