Electronics > Projects, Designs, and Technical Stuff

SMD prototyping techniques -- looking for advice

<< < (8/8)

Someone:

--- Quote from: ebastler on July 27, 2019, 08:25:03 pm ---Surprisingly, no votes for dedicated SMD protoboards, unless I overlooked something. That's great information -- exactly the kind of practical feedback I was hoping for.
--- End quote ---
The SMD pitch prototype boards are great but only if you have chips matching their spacing, for more complex prototypes you quickly end up with more than one pin pitch and it gets hard to do anything other than dead bug. But they are still a great starting base:
https://www.digikey.com/product-detail/en/vector-electronics/8021/V2017-ND/565946

Unixon:
I would suggest to use a totally different approach to SMD vs TH.
TH prototyping is efficient at component-level, while SMD prototyping is not, SMD becomes efficient at least at module-level.
Therefore, it is much better to organize the entire workflow in a different way:

1) Establish a continuous project queue/conveyor. Now you will have to design, assemble and test different boards at a time, forget about working on a single project.
2) Design simple modules that include an IC and all requisite parts to get it working. Buy SMD-TH adapters where appropriate.
3) Keep sending boards to manufacturing and don't rely on quick turnaround, just switch to next task.
4) Assemble project from these self-sustaining modules, not raw parts, this is no longer the way.
5) Think about a "standard" form-factor and interconnect for your modules.

You will feel much more comfortable working with SMD when you'll have enough basic building blocks to assemble all you want and all design stages filled with sub-projects.

nigelwright7557:
That pcb was routed pre making changes to my pcbcad software.
The problem was "off grid" SMD pins.
The software can now align track backwards to previous pin and forward to next pin.
This gives a much cleaner layout.
The pcb works absolutely fine at 200MHz, have three working systems.
Most of the i/o signals are low speed anyway.

The crystal and its caps and decoupling is close to the PIC on the pcb so thats major hurdles dealt with.
After that the I/o is simply 3 lines to a 74hc4051 and the rest to power rails.

nali:
A bit of a mixture for me. Normally I'll go straight to PCB as proto PCBs are cheap or quick enough now (unfortunately never both!). Putting 0R resistors in signal lines can be pretty handy as they offer scoping points and breakout pads if something went wrong and you need to do some wire bodges.

Anything I'm really not sure about I'll invest in a manufacturer's eval board.

Also, for low speed breadboard these save the faff of soldering to adapters
https://images.app.goo.gl/kAgmN6Xfw7uGsEcC9


nigelwright7557:

--- Quote from: nali on August 03, 2019, 12:40:02 pm ---A bit of a mixture for me. Normally I'll go straight to PCB as proto PCBs are cheap or quick enough now (unfortunately never both!). Putting 0R resistors in signal lines can be pretty handy as they offer scoping points and breakout pads if something went wrong and you need to do some wire bodges.

Anything I'm really not sure about I'll invest in a manufacturer's eval board.

Also, for low speed breadboard these save the faff of soldering to adapters
https://images.app.goo.gl/kAgmN6Xfw7uGsEcC9

--- End quote ---

I thought about using one of those but I needed decoupling and crystal right next to the processor.
The processor runs at 200MHz so is fussy about decoupling.

Navigation

[0] Message Index

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod