search

sinara-hw / Urukul

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

Consider adding TLT balun foot print #47

Closed hartytp closed 3 years ago

hartytp commented 4 years ago

Optional to consider for v1.5:

We've modified a couple boards to use the third Nyquist image. That works fine apart from the balun being out of its depth. It would be nice to have a alternative footprint for a TLT balun in that standard smaller case style.

jordens commented 4 years ago

TC1-1T-152X+ I think

dhslichter commented 4 years ago

+1 to this idea in general. @jordens is the reason for choosing this part just for pin compatibility with existing pads (i.e. no change in copper necessary)?

jordens commented 4 years ago

It's actually different pads than the current footprint. But that case and pinout is pretty generic from what I could tell and would allow going to a cheaper/more commonly available package for the standard 1-500 MHz balun style as well. That TLT part number is the only one that I could find that would work with the required biasing topology straight away (center tap).

dhslichter commented 4 years ago

p.s. the existing balun ADT1-1WT+ is a 75 ohm nominal balun...given that we're playing fast and loose with nominal impedances, has anyone tried dropping in an ADTL2-18+, which is pin compatible, nominal 50 ohm impedance but 2x transform, to see if the output is reasonable?

dhslichter commented 4 years ago

@jordens we need a center tap bias? I guess I missed that.

dhslichter commented 4 years ago

Or is that just for the 9912, seems that the 9910 sets its dc level using the load resistors, right?

jordens commented 4 years ago

Both are center tap grounded on the eval boards. See #40, #39, #43 for more info and questions on that topic.

jordens commented 4 years ago

Maybe there are other topologies that don't blow up at DC (i.e. when FTW=0) and still bias/terminate fine.

dhslichter commented 4 years ago

Both are center tap grounded on the eval boards.

Weird. Anyway, I am not trying to come in and blow things up, nor is there time for discussion on this given the urgency of putting at v1.5 release out today. I'm just a little curious/surprised at AD for the eval board choices for the reasons discussed in #40

gkasprow commented 4 years ago

dual footprint is not that trivial here. Let's drop it for v1.5 obraz

jordens commented 4 years ago

Ack

hartytp commented 4 years ago

Does anyone want to keep this open or shall we close?

jordens commented 4 years ago

We should keep this open.

gkasprow commented 3 years ago

I managed to slightly recuce the original balun pad width and placed the second TLT balun obraz