Using jumpers to quickly swap components

[Buriedcode]

The passive components in the photo that I included above are 0805s.  The holes in the board are 0.1" pitch so they fit how the same components would appear on stripboard.

I meant, with stripboard one tends to cut tracks over holes, with a drill bit, meaning the gap can be >3mm between copper, which can be too much for an 0805.  But I tend to cut tracks between holes anyway, so it doesn't matter.

[james_s]

I also prototype with SMT parts on perforated prototype boards. Originally I was inspired by Elm Chan, (Google if you haven't seen his projects) I can't match his neatness but I can make stuff that works.

When I prototype on blank copperclad I sometimes put polyamide tape down under pins I don't want to connect to the ground plane. For DIPs I bend the thin part of the legs I want to connect outward and solder the ground pins butted up to the board. I don't like to flip the chips upside down, I have too much trouble keeping track of what is what and which pin is which.

[HendriXML]

I also prototype with SMT parts on perforated prototype boards. Originally I was inspired by Elm Chan, (Google if you haven't seen his projects) I can't match his neatness but I can make stuff that works.

It is surely amazing.

[Siwastaja]

I think the best way is not to overthink it. You'll find the best method of ad-hoc prototyping by just doing whatever needs to be done, in that moment, when you need it, without thinking too much about whether this is "the right way". After all, lab prototyping is all about one-offs, and your workflow, so do it the way which is quickest to you. These prototypes often don't need to be robust.

[netdudeuk]

I think the best way is not to overthink it. You'll find the best method of ad-hoc prototyping by just doing whatever needs to be done, in that moment, when you need it, without thinking too much about whether this is "the right way". After all, lab prototyping is all about one-offs, and your workflow, so do it the way which is quickest to you. These prototypes often don't need to be robust.

I linked the Elm Chan on the previous page, as an example of what can be achieved.

However, I agree with the overthinking and his work seems like the returns may not be so rich.

I overthink myself and it definitely slows me down and even stops me completely.  Have a look at Mike's LCD hack to see an example of what can be achieved when you don't mind what the rig looks like

[tggzzz]

I spent a couple years making "useful stuff" like this. I think it's just a phase some of us go through. Prediction. In 2 years, you will be using mostly smd parts and swapping them out with a hot air station.

Just curious, but how does one prototype with smd parts without using designed pcb’s?

Breakout boards.

SMD friendly prototyping boards.  http://www.busboard.com/surfacemountpcbs https://github.com/electroniceel/protoboard http://www.talkingelectronics.com/EM/Matrix%20Boards/MatrixBoards.html

Manhattan techniques, especially for decoupling in RF circuits.

All those are better than solderless breadboards, but then anything is better than solderless breadboards.

[Zero999]

I mean did you come here to help, or just to show off how smart you are?

My help is for you to read the spec's in the datasheets then calculate what is needed to meet them. Also to buy quality components.
A model was mentioned, for a simulation? Most simulations simply use "typical" active parts, ignoring minimum and maximum spec's so some of the assembled circuits will not work. You can either throw away the ones that don't work or fiddle with them if all the spec's were ignored.

Yes some SPICE models are poor, especially at simulating the output stage of op-amps. I've had problems with the TL072 model showing it sinking current when the output is saturated to the positive rail and I remember someone saying they simulated an LM358 circuit and the model suggested it can sink 10mA, with the output at only 100mV, with respect to the op-amp's negative supply.  Datasheets are generally better than SPICE models but are still often imperfect, so it makes sense to prototype.

[Nusa]

I've been known to put components on protoboards (dead-bug style if they don't go directly) to create breakouts to header pins. Then I can use the square header pins as wire-wrap posts to connect things up. I've got some long headers for cases where I expect to need several wraps.

I'm also fine soldering mod wires when that's appropriate, but no reason to throw out techniques learned 45 years ago, and tools I already own, that still work.

[netdudeuk]

I've been known to put components on protoboards (dead-bug style if they don't go directly) to create breakouts to header pins. Then I can use the square header pins as wire-wrap posts to connect things up. I've got some long headers for cases where I expect to need several wraps.

I'm also fine soldering mod wires when that's appropriate, but no reason to throw out techniques learned 45 years ago, and tools I already own, that still work.

I've done something similar.