- Posts: 58
Single-Board Universal Module
- octagon
- Offline
Nice work on that PCB!
I also made a version. My home demo Eagle does not allow me to change mid-layers, so they are intact, but they look OK.
I added a filter to the DC 3.3V switcher, which may knock down HF noise some ~60dB, according to LTspice. I added an 0603 inductor (to save space) and another 0402 1uF. Inductor is low cost, but I hate to add parts to the BOM. An 0805 may fit but its quite close to the edge. I put the "gate bleeders" on layer 1.
My lib has the original footprint for the switcher, it looks like yours been improved(?).
The layouts are not otherwise very different. Use as you see fit.
Please Log in or Create an account to join the conversation.
- PhracturedBlue
- Topic Author
- Offline
- Posts: 4402
Please Log in or Create an account to join the conversation.
- octagon
- Offline
- Posts: 58
octagon wrote: The lower cost Murata filter is pretty much identical. I'll see if I can run a sim on the discrete filter.
RFsim99 did not give happy results for the discrete circuit. The integrated filter is a safer bet.
Nice work on that PCB!
I also made a version. My home demo Eagle does not allow me to change mid-layers, so they are intact, but they look OK.
I added a filter to the DC 3.3V switcher, which may knock down HF noise some ~60dB, according to LTspice. I added an 0603 inductor (to save space) and another 0402 1uF. Inductor is low cost, but I hate to add parts to the BOM. An 0805 may fit but its quite close to the edge. I put the "gate bleeders" on layer 1.
My lib has the original footprint for the switcher, it looks like yours been improved(?).
The layouts are not otherwise very different. Use as you see fit.
Please Log in or Create an account to join the conversation.
- octagon
- Offline
- Posts: 58
445-3606-1-ND
Please Log in or Create an account to join the conversation.
- PhracturedBlue
- Topic Author
- Offline
- Posts: 4402
Please Log in or Create an account to join the conversation.
- PhracturedBlue
- Topic Author
- Offline
- Posts: 4402
Please Log in or Create an account to join the conversation.
- Fernandez
- Offline
- Posts: 983
One more comment I would rotate the antenna out ipx connector and route it straight to bpf, with shortest path to the bandpass, so remove the unwanted "stripline".
Extra Via´s could be added in the ground plane areas and RF spots, the more the better.
Please Log in or Create an account to join the conversation.
- PhracturedBlue
- Topic Author
- Offline
- Posts: 4402
Please Log in or Create an account to join the conversation.
- Fernandez
- Offline
- Posts: 983
I think there are still some cosmetic optimalisations possible on the digital lanes, the HF looks neat. Such as near TP8, whuy not go straight under the resonator?
And for instance trace under YA_1, also 90degree angles, I would avoid just for the looks.
But great job!! My full credits for all your time spend!
Please Log in or Create an account to join the conversation.
- PhracturedBlue
- Topic Author
- Offline
- Posts: 4402
Edit:
FYI, the reason I snake the PA control lines around the PA rather than going straight alongside YY_1 is that there isn't quite enough room for the routes between the pad vias and the resonators, and layout is so tight, I don't want to try moving any components. They are low frequency digital lines, so snaking around shouldn't cause any issues even if it isn't pretty.
Please Log in or Create an account to join the conversation.
- octagon
- Offline
- Posts: 58
Your layout is fine, the two DC output coils may couple but 90° will reduce this. Some coils are shielded, like the 0603 I suggested.
Increased space between coils helps of course too. It looks like there is room. I would not want to see dc/dc sidebands in the 2.4GHz output. A test program with no frequency hopping or spreading (direct carrier only) would easier show if the four RF sources pass sidebands from DC supply.
I would remove the restrict (layer 41, 42) on ground pour for the bypass caps around the micro-controller, I'd keep these inductance’s as low as possible even if not at problem at 40MHz.
Please Log in or Create an account to join the conversation.
- Fernandez
- Offline
- Posts: 983
Please Log in or Create an account to join the conversation.
- octagon
- Offline
- Posts: 58
Please Log in or Create an account to join the conversation.
- octagon
- Offline
- Posts: 58
- 445-172392-6-ND
- SRP2010-2R2MDKR-ND
- 535-12278-6-ND
You could get a couple each to see if the conducted noise is less with the shielded types.
Please Log in or Create an account to join the conversation.
- PhracturedBlue
- Topic Author
- Offline
- Posts: 4402
Possibilities:
1) the solder joints aren't ok somewhere, and I just can't see it
2) the NRF chip is bad
3) I installed the wrong component somewhere and am loading down the RF pin(s)
4) ???
Please Log in or Create an account to join the conversation.
- hexfet
- Away
- Posts: 1876
Please Log in or Create an account to join the conversation.
- octagon
- Offline
- Posts: 58
Do you have DC voltage on pins 12, 13?
You may be able to pick the LO with a inductive probe, like 5 turns of thin magnet- or wire-wrap wire, soldered to the end of a thin flexible coax, connected to spectrum analyzer. A 3mm coil held over the center of the chip may pick it up, it does on the Cypress chip (800MHz below output).
Please Log in or Create an account to join the conversation.
- PhracturedBlue
- Topic Author
- Offline
- Posts: 4402
Please Log in or Create an account to join the conversation.
- vlad_vy
- Offline
- Posts: 3333
Please Log in or Create an account to join the conversation.
- moss
- Offline
- Posts: 40
www.seeedstudio.com/depot/RFEM-WSUB3G-Module-p-1264.html
www.amazon.com/RF-Explorer-RFEM-WSUB3G-Module/dp/B00IF913O0
I'm getting one.
If the NRF chip is getting the power it needs and you cannot sniff any RF on the output pins,(some should remain even if "shorted") then maybe either the chip is toast, or it is not getting the code it needs.
Please Log in or Create an account to join the conversation.
- Home
- Forum
- Development
- Development
- Single-Board Universal Module