Author Topic: SMD prototyping techniques -- looking for advice  (Read 5737 times)

0 Members and 1 Guest are viewing this topic.

Online ebastlerTopic starter

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: de
SMD prototyping techniques -- looking for advice
« on: July 27, 2019, 10:53:52 am »
So far, I have still been building my prototypes mainly with through-hole components. (Mostly digital/µC, some audio-range analog, some analog up to a few MHz). I mainly use 2.54mm pitch protoboard with square pads, and solder on Kynar-wire connections. Where SMD ICs were required, I have either used little adapters to DIP footprint, or soldered SOICs directly onto the protoboard with every other pin bent up and wired free-floating.

That kind of works, but with more and more ICs becoming avaiable in SMD-only style, it has become more and more awkward. It feels like it's time to move to SMD as the "default" prototyping technique. I could use some advice here:

(a) What protoboards do people use? Is there an equivalent of the generic 2.54mm square grid -- something like pads without holes but in 1.27mm pitch? Does that even make sense, assuming I can largely stick with SOICs? What do you do if you need higher-density packages?

I am aware of the various SMD protoboards which carry an assortment of specific TSSOP etc. footprints. But these seem rather limiting, almost to the point where you need to know your designs's chip complement before you order the protoboard. If these are the way to go, how do you work with them? Or is it back to individual DIP adapters? if you need to mix and match various SMD packages in a circuit?

(b) What is your personal "sweet spot" size for passive? 0603, 0805? I would want to order E12 or E24 sets of resistors and capacitors, but only in a single mechanical size.

(c) I think I have the soldering under control, either pad by pad or using extra flux and dragging the tip. Haven't experimented with solder paste and oven/pan soldering yet, but that would be used for small series rather than early prototyping, right?

(d) Any other advice for SMD prototyping? Many thanks in advance for your help!
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: SMD prototyping techniques -- looking for advice
« Reply #1 on: July 27, 2019, 11:10:22 am »
Watch Dave’s last mailbag for SMD proto boards. I started with 0805 but proved concept on a breadboard therefore needing to use IC’s that are available in DIP and SMD.

Once you dive properly into SMD you’ll love it.

Oh, and get some SMD smart tweezers before too long.

Oh and something else, get some K style Hakko tips or similar, they're great for rework in that they reach both ends of passives so you can grab them with tweezers or sweep them off the pads.
« Last Edit: July 27, 2019, 11:32:26 am by tautech »
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 
The following users thanked this post: ebastler

Offline NivagSwerdna

  • Super Contributor
  • ***
  • Posts: 2507
  • Country: gb
Re: SMD prototyping techniques -- looking for advice
« Reply #2 on: July 27, 2019, 11:26:07 am »
I have experimented with SMD breakouts etc but it quickly becomes a major PIA so now I just go straight to PCB... sometimes creating a generic layout that can be part populated and also accept that v1.00 will be unlikely to be the final revision.
With DipTrace (or equivalent) and a super cheap 5 or 10-off PCB prototyping service Elecrow (or equivalent) (and a microscrope) it's a joy.
Have fun!
 
The following users thanked this post: ebastler

Offline KE5FX

  • Super Contributor
  • ***
  • Posts: 2096
  • Country: us
    • KE5FX.COM
Re: SMD prototyping techniques -- looking for advice
« Reply #3 on: July 27, 2019, 11:32:13 am »
Meh, everybody makes this 10x more complicated than it really is.  You don't need protoboards, just a bare piece of copper-clad board.  Flip the chips over, glue them down if you like, and treat their pins, lands, or balls like solder terminals. 

You will need a good microscope and some very fine wire.  It's good to have 30 AWG tinned copper and 40 AWG enameled copper handy.  I like to use 0402 parts, but 0603 is OK for most purposes too.  (Use the fine wire to provide some stress relief for the MLCCs, or they will break.)

Some people like to use gold-plated boards and Dremel tools and multilevel copper islands and other methods to make this technique 10x as complicated as it needs to be, but the electrons don't give a hoot how it looks.
 

Online ebastlerTopic starter

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: de
Re: SMD prototyping techniques -- looking for advice
« Reply #4 on: July 27, 2019, 11:42:50 am »
Thank you all for the resonses so far! Thanks for the mailbag hint, I had missed that one.

In the meantime, I also found a post by forum member electronic_eel from early last year. His perfboard for combined through-hole and SMD prototypes looks quite plausible to me. (Pads with holes on a 2.54mm pitch, with smaller SMD-only pads fitted inbetween.) Anybody using it? How do you like it, with a year's worth of hindsight?
https://www.eevblog.com/forum/manufacture/improved-protoboardperfboard-for-direct-soic-mounting/

Meh, everybody makes this 10x more complicated than it really is.  You don't need protoboards, just a bare piece of copper-clad board.  Flip the chips over, glue them down if you like, and treat their pins, lands, or balls like solder terminals. 

Hmm, I have used that approach for single SMD chips on 2.54mm perfboard, but had not considered it as a general solution for SMD prototyping. How do you mount the passives? Always solder one end directly to a chip pin, have the other end floating?

 

Offline techman-001

  • Frequent Contributor
  • **
  • !
  • Posts: 748
  • Country: au
  • Electronics technician for the last 50 years
    • Mecrisp Stellaris Unofficial UserDoc
Re: SMD prototyping techniques -- looking for advice
« Reply #5 on: July 27, 2019, 12:46:53 pm »
Thank you all for the resonses so far! Thanks for the mailbag hint, I had missed that one.

In the meantime, I also found a post by forum member electronic_eel from early last year. His perfboard for combined through-hole and SMD prototypes looks quite plausible to me. (Pads with holes on a 2.54mm pitch, with smaller SMD-only pads fitted inbetween.) Anybody using it? How do you like it, with a year's worth of hindsight?
https://www.eevblog.com/forum/manufacture/improved-protoboardperfboard-for-direct-soic-mounting/

Meh, everybody makes this 10x more complicated than it really is.  You don't need protoboards, just a bare piece of copper-clad board.  Flip the chips over, glue them down if you like, and treat their pins, lands, or balls like solder terminals. 

Hmm, I have used that approach for single SMD chips on 2.54mm perfboard, but had not considered it as a general solution for SMD prototyping. How do you mount the passives? Always solder one end directly to a chip pin, have the other end floating?

For passives you can cut up small squares of single sided pcb and stick them on the blank copper devboard with their copper side up, making 'islands'. The passive then solders to two such squares.
 
The following users thanked this post: ebastler

Offline Buriedcode

  • Super Contributor
  • ***
  • Posts: 1718
  • Country: gb
Re: SMD prototyping techniques -- looking for advice
« Reply #6 on: July 27, 2019, 01:43:00 pm »
We live in a great age for electronics prototyping.  You can get a pack of SMD-DIP adapters on fleabay, or amazon, aliexpress etc.. for next to nothing.  Anything from the standard SOIC, to TSSOP, to QFN.

It is also the cheapest its ever been to get quality PCB's made, and delivered to your door in 3 days for $15.

That said, I still keep coming back to tri-pad and stripboard for prototyping.  SMD-DIP adapters for the IC's, and 0603 or 0805 on the copper side strips. Designed using eagle with the solder/bottom side for the strips, and component/topside for the links.  Soldering 0805/0603 is quicker than TH, although without the leads, you have no clippings to use as jumpers.  Also of course, the passives can only go between two adjacent strips.

A few years ago there was a company, Dip-micro I think, that made thin (1mm) protoboards, with the same layout as solderless breadboards, but with 0.05" pads in the centre for SOIC. These however, we somewhat limited.  I've used them a few times, but always go back to cheap stripboard.

Another option is a half measure.  Making modules for specific devices, that are more than adapters - having extra circuitry, passives, jumpers, sockets etc.. that can plug into 0.1" sockets, or solderless breadboards for prototyping.  And using the stripboard as a backplane to interconnection.

Working with one-pad-per-hole and kynar is fine, but it takes ages.  Especially when there's solderable enameled wire and a propelling pencil ala ELM-chan: http://elm-chan.org/docs/wire/wiring_e.html
Using that, with a scalpel and a pair of tweezers, it can be surprisingly quick to connect everything.
 

Offline ogden

  • Super Contributor
  • ***
  • Posts: 3731
  • Country: lv
Re: SMD prototyping techniques -- looking for advice
« Reply #7 on: July 27, 2019, 02:15:44 pm »
When signal speed is not critical and trough-hole components can do the job - I use perfboard with smd->dip adapters for IC's, through-hole passives with some occasional SMDs if needed. When high frequency is limiting factor, you have to plan layout of your prototype anyway which means - better just route & etch your prototype PCB. Usually I make PCB for just hi-freq RF parts of the circuit, the rest is either MCU devboard or perfboard. I use two-sided presensitized PCB, do not even peel protective film from bottom and obviously do not UV-expose/etch because it will be solid ground plane anyway. Etch just top (circuit & power rail) side, then drill &solder ground vias where necessary, then apply components &connectors.
 
The following users thanked this post: ebastler

Online ebastlerTopic starter

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: de
Re: SMD prototyping techniques -- looking for advice
« Reply #8 on: July 27, 2019, 02:34:54 pm »
Working with one-pad-per-hole and kynar is fine, but it takes ages.  Especially when there's solderable enameled wire and a propelling pencil ala ELM-chan: http://elm-chan.org/docs/wire/wiring_e.html
Using that, with a scalpel and a pair of tweezers, it can be surprisingly quick to connect everything.

Hey, thank you for that link. I had come across the web page before (and the video it links to for further illustration), but didn't recall where to look for it. The Kapton tape below the SMD chip packages seems like a nice alternative to the bottom-up mounting suggested by KE5FX -- I would feel more comfortable having the parts oriented like they show up in the datasheet, and being able to read print on the chip package.

Yes, wiring things up with Kynar wire can get tedious; and with fine-pitch SMD the room taken up by the insulation can be an issue too. But I have had very mixed success with solderable enamel wire: More often than not, the enamel melting point was higher than desirable, and when it finally melted (burnt?), it left nasty residue behind.  Is there a particular brand, or material designation for the "enamel", to look for?
« Last Edit: July 27, 2019, 02:36:30 pm by ebastler »
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: SMD prototyping techniques -- looking for advice
« Reply #9 on: July 27, 2019, 02:41:00 pm »
Don't be worried, it is easier than it looks. To my surprise I prefer it to through hole :)

Have a look at my experiences https://entertaininghacks.wordpress.com/category/homebrew-pcbs/
and techniques in
https://bristol.hackspace.org.uk/wiki/lib/exe/fetch.php?media=skillshare:solderlessbreadboards03-03.pdf

The only thing I'll add is that removing/replacing SMD components can be done by a solder station "hot air gun" and using polyimide tape to deflect the heat from surrounding components. I've replaced resistors and capacitors on Tek 2465 A5 boards that way. Using an iron can be a problem if you thermally shock a ceramic component by touching it with the bit.

(a) What protoboards do people use? Is there an equivalent of the generic 2.54mm square grid -- something like pads without holes but in 1.27mm pitch? Does that even make sense, assuming I can largely stick with SOICs? What do you do if you need higher-density packages?

I am aware of the various SMD protoboards which carry an assortment of specific TSSOP etc. footprints. But these seem rather limiting, almost to the point where you need to know your designs's chip complement before you order the protoboard. If these are the way to go, how do you work with them? Or is it back to individual DIP adapters? if you need to mix and match various SMD packages in a circuit?

I mix-and-match a range of techniques, depending on the components and layout.

If you can't use a "proper" PCB, there's a lot to be said for dead bug, manhattan, and a dremel with a ball like dentists use.

Quote
(b) What is your personal "sweet spot" size for passive? 0603, 0805? I would want to order E12 or E24 sets of resistors and capacitors, but only in a single mechanical size.

0603 and 0.05mm pitch is easy, provided you have steady hands, tweezers, and good magnification.

Quote
(c) I think I have the soldering under control, either pad by pad or using extra flux and dragging the tip. Haven't experimented with solder paste and oven/pan soldering yet, but that would be used for small series rather than early prototyping, right?

Even for one-offs, I successfully use PCB plus solder paste, and a saucepan with a few mm of sand in the bottom.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 
The following users thanked this post: ebastler

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: SMD prototyping techniques -- looking for advice
« Reply #10 on: July 27, 2019, 03:14:39 pm »
(a) What protoboards do people use? Is there an equivalent of the generic 2.54mm square grid -- something like pads without holes but in 1.27mm pitch? Does that even make sense, assuming I can largely stick with SOICs?

I always have a few 1.27mm, double sided proto boards available. Usually all square pads, with a few through-holes on one or all sides of the board so you can also solder connectors on or through-hole components, wires or whatever. You can find those easily. Not always for cheap though, but you can find cheaper ones on eBay/ Aliexpress that are good enough.

From experience, depending on the complexity of your circuit, it can actually take up more of your time hand wiring all that compared to designing a PCB and have it made for cheap, so most of the time, in my case the prototyping on such boards is limited to small circuits / evaluating some specific IC / etc.

What do you do if you need higher-density packages?

Depends. Either I can use dedicated (small PCBs) adapters with the required footprint, or I sometimes hand-solder tiny wires on the IC pads (very rare, but I have on occasion).

I tend to favor using ready-made breakout boards or evaluation boards though, if available for a reasonable price. A time saver. So I usually look that up first.

(b) What is your personal "sweet spot" size for passive? 0603, 0805? I would want to order E12 or E24 sets of resistors and capacitors, but only in a single mechanical size.

Depends. For proto boards, 0603, 0805 (or larger if required by power dissipation considerations). Anything smaller than 0603 for all-purpose SMD proto boards is a bit nasty.
For regular PCBs, there is no sweet spot, it will all depend on size requirements. But I'd say 0603 most often usually. 0402 or smaller if needed. Larger if needed.

(c) I think I have the soldering under control, either pad by pad or using extra flux and dragging the tip. Haven't experimented with solder paste and oven/pan soldering yet, but that would be used for small series rather than early prototyping, right?

Also depends, mostly on your habits, available equipment and obviously, components themselves. If a board can be soldered with an iron and/or hot air only, I will do that for the first prototype(s).
Now if you have to deal with a lot of components and/or more than a couple BGAs or QFNs, using an oven or pan is worth it.

(d) Any other advice for SMD prototyping? Many thanks in advance for your help!

A good soldering iron. Lots of flux in various forms: having flux pens, liquid flux and gel flux is all handy. A hot air station. Decent solder wire.

And lastly and again, do not shy away from designing proper PCBs and having them manufactured. That's quite cheap and fast these days, and unless you are in a big hurry, you'll have a better time overall.
 
The following users thanked this post: ebastler

Online ebastlerTopic starter

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: de
Re: SMD prototyping techniques -- looking for advice
« Reply #11 on: July 27, 2019, 03:47:39 pm »
Thanks again for a lot of good advice!

Specifically, thanks for the various votes to go straight to a PCB layout for anything with potentially critical HF/crosstalk properties. I have never etched my own PCBs, but have used various cheap Chinese PCB houses quite a bit, and agree that the cost for nicely made PCBs is not an issue anymore. (I have used SMD on some small board designs, and have enjoyed working with them.)

My main mental block there is around shipping costs, I guess. Using free standard shipping makes the turnaround time painfully slow. But paying more for expedited shipping than what the actual boards cost feels wrong and wasteful to me -- somehow even more so if it's for an early prototype which I know I will discard eventually... I guess I will just have to get over that!  ;)
 

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: SMD prototyping techniques -- looking for advice
« Reply #12 on: July 27, 2019, 03:56:10 pm »
My main mental block there is around shipping costs, I guess. Using free standard shipping makes the turnaround time painfully slow. But paying more for expedited shipping than what the actual boards cost feels wrong and wasteful to me -- somehow even more so if it's for an early prototype which I know I will discard eventually... I guess I will just have to get over that!  ;)

Yeah. For us Europeans, shipping from China is either atrociously slow or surprisingly expensive (beware of the various fees and taxes that do not appear upfront!)
Nothing much you can do about it... except grouping several PCBs if at all possible. Or use Eurocircuits... shipping not as expensive, but PCBs a lot more.
 

Offline NivagSwerdna

  • Super Contributor
  • ***
  • Posts: 2507
  • Country: gb
Re: SMD prototyping techniques -- looking for advice
« Reply #13 on: July 27, 2019, 06:44:21 pm »
My main mental block there is around shipping costs, I guess. Using free standard shipping makes the turnaround time painfully slow.
Different manufacturers use different shipping methods... that's once of the reasons I stick with Elecrow because they offer HK Post which is cheap and rapid to the UK at least.
 

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15797
  • Country: fr
Re: SMD prototyping techniques -- looking for advice
« Reply #14 on: July 27, 2019, 06:49:53 pm »
That's nice to know. And yeah I think I've also got good results with HK Post, unfortunately not many companies seem to offer it?

Another option if you're not in a hurry and want it cheap is to use OSHPark. At least for over here, you can select a free shipping option, which goes through the postal services. Count about 3 weeks from order to delivery (yes it's a lot, but not that bad compared to some free or very low cost shipping options from China).
 

Online Siwastaja

  • Super Contributor
  • ***
  • Posts: 9330
  • Country: fi
Re: SMD prototyping techniques -- looking for advice
« Reply #15 on: July 27, 2019, 07:27:24 pm »
1) Bare copper clad PCB for a ground plane.
2) Deadbug ICs
3) Air wires using thin, colored leads.
4) Decoupling caps, pulldown resistors etc. directly soldered to the ground plane, like poles.
5) Series resistors etc. directly soldered to the deadbugged component legs

If large non-GND planes are needed, make cuts to the ground plane copper clad using a Dremel like tool to isolate a section. Another option is to cut another piece and glue it on the top.

I don't tend to use specialized SMD protoboards.
 
The following users thanked this post: ebastler

Offline hagster

  • Frequent Contributor
  • **
  • Posts: 394
Re: SMD prototyping techniques -- looking for advice
« Reply #16 on: July 27, 2019, 07:38:27 pm »
Meh, everybody makes this 10x more complicated than it really is.  You don't need protoboards, just a bare piece of copper-clad board.  Flip the chips over, glue them down if you like, and treat their pins, lands, or balls like solder terminals. 

You will need a good microscope and some very fine wire.  It's good to have 30 AWG tinned copper and 40 AWG enameled copper handy.  I like to use 0402 parts, but 0603 is OK for most purposes too.  (Use the fine wire to provide some stress relief for the MLCCs, or they will break.)

Some people like to use gold-plated boards and Dremel tools and multilevel copper islands and other methods to make this technique 10x as complicated as it needs to be, but the electrons don't give a hoot how it looks.

We call this ’Dead-bug' style.

Personally, if it's anything mor than a small test for a circuit sub-system, I prefer just getting a cheap PCB designed. I have seen some amazing dead bug builds, but they generally aren't very reproducible once the first batch of changes are made.
 
The following users thanked this post: ebastler

Online Siwastaja

  • Super Contributor
  • ***
  • Posts: 9330
  • Country: fi
Re: SMD prototyping techniques -- looking for advice
« Reply #17 on: July 27, 2019, 08:04:50 pm »
< 50 total component legs: deadbug
> 50 total component legs: design and order a PCB.
> 50 total component legs, and you must have a proof of concept prototype finished tonight: deadbug, and good luck!

Manual prototyping without a custom PCB is painfully slow, no matter what methods and tools you have. It's used when actual calendar time is more important than work hours. Otherwise, designing and ordering a PCB tends to be less painful.
 
The following users thanked this post: ebastler, NivagSwerdna

Online ebastlerTopic starter

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: de
Re: SMD prototyping techniques -- looking for advice
« Reply #18 on: July 27, 2019, 08:25:03 pm »
Thanks all for the input so far! In summary, the recommendations seem to fall into three categories:

  • SMD chips on DIP adapters, everything else on 0.1" perfboard using through-hole components. Fine for non-critical (low speed) circuits.
  • Mount SMD ICs dead-bug style or with Kapton tape under the pins; solder wires (preferaly thin, enamelled) directly to the pins. Should allow for higher speed, due to shorter wires, SMD passives, and availability of a ground plane.
  • Do a proper PCB layout right away.

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.

Approach (1) is what I have mostly used so far. Sometimes I have used (3), but only when I felt reasonably sure of my design, and was hoping to actually use it in my project. So approach (2) is what I will try!

I have a preference for the "sunny side up + Kapton tape" style, I think, rather than using dead-bug mounting. And while I see the point of a ground plane, I like holes and solder pads for the occasional though-hole component and connector. Come to think of it -- I used to have some plated-through perfboards with square pads on one side, and very small pads and a ground plane on the back. Haven't seen those in a while, but they must still be available?
 

Offline magic

  • Super Contributor
  • ***
  • Posts: 7453
  • Country: pl
Re: SMD prototyping techniques -- looking for advice
« Reply #19 on: July 27, 2019, 08:26:09 pm »
In the meantime, I also found a post by forum member electronic_eel from early last year. His perfboard for combined through-hole and SMD prototypes looks quite plausible to me. (Pads with holes on a 2.54mm pitch, with smaller SMD-only pads fitted inbetween.) Anybody using it? How do you like it, with a year's worth of hindsight?
https://www.eevblog.com/forum/manufacture/improved-protoboardperfboard-for-direct-soic-mounting/
That's neat. I sometimes mount SMD passives and SOT23 transistors on normal perfboards similar way, it makes the circuit smaller than using THT stuff.
0805 size is just right for mounting on 2.54mm pitch perfboard. 0603 works too. I prefer resistors in 0805 because markings are easily visible. I have good eyesight and can read that 0603 stuff, but 0805 really make it easier. Higher power too and less likely to get lost.

OTOH, I have 100nF capacitors in every size down to 0201, for soldering directly to IC pins when power is next to ground.

A quick and dirty option for simple one-off PCBs is to draw them by hand with a felt pen and etch.
« Last Edit: July 27, 2019, 08:30:27 pm by magic »
 
The following users thanked this post: ebastler

Offline Miti

  • Super Contributor
  • ***
  • Posts: 1405
  • Country: ca
Re: SMD prototyping techniques -- looking for advice
« Reply #20 on: July 27, 2019, 10:09:19 pm »
a) I always design the board and order it at www.pcbway.com. At $5 for 10 pcs plus shipping you can't go wrong.
b) For hand solder, I don't recommend going smaller than 0603. In the oven, 0402 is possible.
c) I only hand solder the boards with very little parts count. You can order a cheap stencil together with the board. I love my toaster oven.
d) Yes, forget TH. Dive into SMD and never look back.
Fear does not stop death, it stops life.
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: SMD prototyping techniques -- looking for advice
« Reply #21 on: July 27, 2019, 11:09:46 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.

See the very usable techniques in https://entertaininghacks.wordpress.com/2020/07/22/prototyping-circuits-easy-cheap-fast-reliable-techniques/

Note that they aren't used exclusively, but are used where beneficial and in conjunction with other techniques.
« Last Edit: May 14, 2021, 03:40:09 pm by tggzzz »
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline KL27x

  • Super Contributor
  • ***
  • Posts: 4108
  • Country: us
Re: SMD prototyping techniques -- looking for advice
« Reply #22 on: July 28, 2019, 01:08:29 am »
Quote
Anybody using it? How do you like it, with a year's worth of hindsight?
I have not used electric eel's version, but I have tried my own version.

There are major problems with this, in practice.

Yes, you can bridge pads to make a circuit. You can even go out of your way to use as few jumperwires as possible. In some cases, you can even make simple circuits with zero wires, which can look great.

The major problem is it is very difficult to debug. It is much easier to follow jumper wires than it is to see these solder tracks. When the empty pads are tinned (either factory plated or the fact they will inevitably get tinned during building/reworking), it is very difficult to follow half pitch (0.05") solder tracks with the naked eye. All you can see is a sea of tin/solder. And even under mag, where you can see perfectly, it's like being in the suburbs, looking for the tan house with the white fence. But all the houses are tan with a white fence, so you are constantly double-checking your row/column. When looking at a PCB, each trace has a unique shape and you have landmarks. Jumperwires, even if they are all the same color, each one has unique bends and angles and whatnot.

I occasionally use a piece of 0.05" small piece of veroboard, mostly for hacking/adding one component to something. In practice this usually means cutting just a tiny piece barely bigger than the component, just to give a place to put an IC and add jumperwires and small passives. It's usually easier to make PCB for the starting point, and just use these stuffs for modifying that.
« Last Edit: July 28, 2019, 01:30:31 am by KL27x »
 

Offline techman-001

  • Frequent Contributor
  • **
  • !
  • Posts: 748
  • Country: au
  • Electronics technician for the last 50 years
    • Mecrisp Stellaris Unofficial UserDoc
Re: SMD prototyping techniques -- looking for advice
« Reply #23 on: July 28, 2019, 02:06:42 am »
Thanks all for the input so far! In summary, the recommendations seem to fall into three categories:

  • SMD chips on DIP adapters, everything else on 0.1" perfboard using through-hole components. Fine for non-critical (low speed) circuits.
  • Mount SMD ICs dead-bug style or with Kapton tape under the pins; solder wires (preferaly thin, enamelled) directly to the pins. Should allow for higher speed, due to shorter wires, SMD passives, and availability of a ground plane.
  • Do a proper PCB layout right away.

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.

Approach (1) is what I have mostly used so far. Sometimes I have used (3), but only when I felt reasonably sure of my design, and was hoping to actually use it in my project. So approach (2) is what I will try!

I have a preference for the "sunny side up + Kapton tape" style, I think, rather than using dead-bug mounting. And while I see the point of a ground plane, I like holes and solder pads for the occasional though-hole component and connector. Come to think of it -- I used to have some plated-through perfboards with square pads on one side, and very small pads and a ground plane on the back. Haven't seen those in a while, but they must still be available?

This is a deadbug technique I'm experimenting with. I have more pictures and details on my doc page at:
 https://mecrisp-stellaris-folkdoc.sourceforge.io/prototyping.html#prototyping

 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29810
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: SMD prototyping techniques -- looking for advice
« Reply #24 on: July 28, 2019, 02:34:51 am »
Surprisingly, no votes for dedicated SMD protoboards, unless I overlooked something.
Maybe there's just not a good selection.
Here's Dave's mailbag vid where he shows off a selection posted in:

https://youtu.be/MwL3qgMNYLc?t=1743

Yes, forget TH. Dive into SMD and never look back.
Well yes and no.
You'll see jumpers used in modern SMD design when it's nuts to add another layer and so increase PCB cost when sometimes the use of a TH passive or 2 can open the routing pathways required so that even a jumper might not even be required. Otherwise to keep the design SMD get a selection of 1206 or 1210 passives to allow for a wider footprint to route traces under.
Depends a bit on how much TH stuff you have and want to hold onto but if you're doing any repairs you really should hold onto your TH components.

Each to their own I guess and I've found the occasional use of TH passives can restrict projects to just 1 or 2 layers.
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf