Author Topic: Newbie - How to not get discouraged during a project?  (Read 888 times)

0 Members and 1 Guest are viewing this topic.

Offline mortrek

  • Contributor
  • Posts: 33
  • Country: us
Newbie - How to not get discouraged during a project?
« on: January 04, 2020, 07:22:33 pm »
I've been working on a design for a while and built it all up in breadboard form, then built it digitally and ordered PCBs. I get the PCBs back and realized I made some mistakes (such as I forgot a features I wanted), so I fixed them and ordered new PCBs and got them, and realized I made other mistakes (such as a header for a module is in a spot that causes the module to overlap with another tall part, which wasn't obvious until I tried assembling it)... it's not terribly expensive but I'm not exactly rich, and I'm new enough to this whole process that I am learning by doing, which means failing a lot because of stupid mistakes.

It's a little discouraging to keep making these stupid mistakes presumably because I don't have the experience to not make them, or maybe I'm overeager to move forward with the project, leading to impatience... do you just figure it's a learning experience, that the cost of ordering the PCB or extra parts is essentially tuition?

How do actual engineers go about these sorts of prototype projects? Do you really spend tons of time trying to perfect the project on paper, and only then do you move forward with printing PCBs?

I have poor short-term memory and a hard time holding a lot of data in my head at once, so I struggle with understanding my own project without breaking it down into small modular sections. When it came time to translate my breadboard to a digital design, it was difficult to not accidentally omit connections. I'm still waiting on a couple parts, so I don't even know if my current PCB version is at all functional...

I realize this is a bit of a ramble. I'm sure I'll keep working on this project, but it's a bit discouraging to make so many mistakes. Breadboarding the design was a fun challenge but if I made a mistake, I could just correct it easily. At this later stage of my project, I'm just getting frustrated.

Thank you for any advice you may have on this matter.

 

Offline ataradov

  • Super Contributor
  • ***
  • Posts: 6511
  • Country: us
    • Personal site
Re: Newbie - How to not get discouraged during a project?
« Reply #1 on: January 04, 2020, 07:42:11 pm »
It really depends on the complexity of the PCB. In general for simpler designs the best way is to just order the PCBs without overthinking it too much. If you make mistakes you can't fix by bodges, reorder the PCB.

I never prototype the whole design on breadboards. It makes little sense for boards with FPGAs or MCUs. I only prototype tricky analog parts. And even then, often simulation in LTSpice is sufficient.

If you expect to make mistakes from your previous history - design your PCBs so they are easy to modify.

The cost of PCBs fitting into PcbWay cheap prototyping size range is essentially negligible. Shipping can get expensive, but if you can wait, cheaper shipping methods are quite good.
Alex
 
The following users thanked this post: sokoloff

Offline ebastler

  • Super Contributor
  • ***
  • Posts: 3534
  • Country: de
Re: Newbie - How to not get discouraged during a project?
« Reply #2 on: January 04, 2020, 08:37:22 pm »
It's normal to still have iterations at that stage of the design process. I rarely get my PCBs right the first time round, unless they are just variations on a theme on which I already have a very good handle. Either something doesn't work (or doesn't work robustly), or I have another great (?) idea, e.g. for an imroved user interface, which results in a board revision.

One thing to keep in mind is that a new iteration doesn't always have to mean a new PCB spin. As ataradov said, ideally think ahead a bit and make provisions for modifications in your layout: Add test points and jumpers, maybe even add a little prototyping area (individual pads) on the side if you have room to spare.

But even if you couldn't plan ahead (and would it be an honest mistake if you had planned for it?  ;)) --
it's totally OK to "hack" the first PCB version at your heart's content and jury-rig whatever fixes and improvements come to mind: Solder plenty of bodge wires, cut some traces on the PCB, glue on an extra IC "dead bug style" and wire it with free-floating bodge wires... Anything goes!

The idea is to implement as many bug fixes and bright new ideas as possible in this "Frankenstein" prototype, then cast them into a final (?) PCB revision once you are satisfied with the functionality. That saves on PCB and parts costs, and -- at least as important for me -- the annoying waiting times to receive a new PCB. After you are done, toss the Frankenstein board, or keep it as a memento of your design journey!   8)
 

Online jfiresto

  • Frequent Contributor
  • **
  • Posts: 307
  • Country: de
Re: Newbie - How to not get discouraged during a project?
« Reply #3 on: January 04, 2020, 09:06:35 pm »
Have you tried compiling check lists? Each time you make or notice a mistake, write down a rule that would catch it, then verify your board against the check list before you send it off. For example, after my latest goof, I might add: Verify the position of each daughterboard interconnect exactly matches the motherboard connector it plugs into. (I was a little too quick checking the daughter because I knew another spin would be cheap.)
 
The following users thanked this post: SilverSolder, tooki

Offline james_s

  • Super Contributor
  • ***
  • Posts: 12066
  • Country: us
Re: Newbie - How to not get discouraged during a project?
« Reply #4 on: January 04, 2020, 09:09:45 pm »
When there is any question of mechanical fitment I normally print the PCB out on paper actual size then cut it out and test fit it with the other mechanical bits to make sure everything lines up. It can be helpful in some cases to stick the layout to a piece of stiff cardboard cut to size. You can also test fit components to make sure you used the right footprint.

If you don't want to respin boards, many errors can be corrected by cutting traces, adding jumpers, drilling and cutting, whatever is needed. Back in the early days when PCBs were laid out by hand with drafting tape and the process of having boards made was very expensive it was common to find "bodge wires" on boards in commercial products. They would cut traces and add jumpers made of kynar wire to fix mistakes. I have a late 70s arcade game board where you can tell a trace was supposed to go on the other side of the board so it crosses through another trace. They cut it on each side of the unintended junction and soldered in a jumper wire.
 
The following users thanked this post: pepelevamp

Offline pepelevamp

  • Regular Contributor
  • *
  • Posts: 80
  • Country: nz
Re: Newbie - How to not get discouraged during a project?
« Reply #5 on: January 05, 2020, 05:45:49 am »
hey mate hang in there.

i learned a while back that some times projects can take serious time.

i fully agree with the waiting between PCBs and stuff being a pain. i am looking into doing the home made PCB trick where you laser print a circuit onto some paper, and then rub on silver adhesive & heat gun it dry.

no chemical etching etc, plus the pcb actually flexes. 3 ohm resistance traces for 30mm or so. something like that. but you can do multiple layers to get it down.

someone wise once told me 'tedium is part of the hobby'. its alright mate just try to enjoy your journey along the way & kinda expect ahead of time that yep something's gonna screw up. part of the frustration is being surprised by it. ya can reduce stress quite a bit by anticipating that bad shits already gonna happen.

plus you got mates on the forum here to help you out with stuff.

 

Offline djacobow

  • Super Contributor
  • ***
  • Posts: 1066
  • Country: us
  • takin' it apart since the 70's
Re: Newbie - How to not get discouraged during a project?
« Reply #6 on: January 05, 2020, 06:38:11 am »
You've gotten good advice already. I'll just restate two things that work well for me.

First, if the PCB is wrong, then bodge in as many fixes as you can. If you can get it working with the bad PCB, your next spin is likely to be the last. If youcan't, then the next spin will only fix as many problems as you already positively identified.

Secondly, when you've made a fix to a design and are about to click buy on new pcbs ... don't. Just go to bed. Resist the urge for two nights' sleep if you can. If you haven't thought of anything you don't like after that, *then* order the boards. There's nothing more annoying than knowing the board you just ordered is broken days before it even arrives.
 
The following users thanked this post: SilverSolder, ebastler, tooki, Jacon, aix

Offline ebastler

  • Super Contributor
  • ***
  • Posts: 3534
  • Country: de
Re: Newbie - How to not get discouraged during a project?
« Reply #7 on: January 05, 2020, 08:18:25 am »
Secondly, when you've made a fix to a design and are about to click buy on new pcbs ... don't. Just go to bed. Resist the urge for two nights' sleep if you can. If you haven't thought of anything you don't like after that, *then* order the boards. There's nothing more annoying than knowing the board you just ordered is broken days before it even arrives.

That's an excellent addition to the prior advice! I have certainly fallen into the trap of sending off another set of Gerber files prematurely, because I wanted to "make progress" on a project...
 
The following users thanked this post: tooki, mcovington

Online TERRA Operative

  • Frequent Contributor
  • **
  • Posts: 793
  • Country: jp
  • Voider of warranties
    • Near Far Media Youtube
Re: Newbie - How to not get discouraged during a project?
« Reply #8 on: January 05, 2020, 08:29:53 am »
Also, don't go watching a youtube video and go away thinking they finished that project as quick as it seemed in the video.
Often a short video is actually documenting months of work.

You can always do what I do though, just have so many projects on the go at once that you can always put one away when you are bored or angry at it, and pick up another from the shelf that has been there so long that you have forgotten you were even upset at it months ago.... :D
Any circuit design must contain at least one part which is obsolete, two parts which are unobtainable, and three parts which are still under development.

https://www.youtube.com/user/NearFarMedia/
 
The following users thanked this post: tooki

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 11842
  • Country: gb
    • Having fun doing more, with less
Re: Newbie - How to not get discouraged during a project?
« Reply #9 on: January 05, 2020, 10:15:14 am »
Experienced engineers make mistakes.

Having inappropriate mechanical dimensions is a common problem,  and large companies spend large amounts of money to reduce the probability of mistakes.

Going from the breadboard to PCB, did you draw the schematic in the tool and the create thepcb from that schematic? If so, all connections in the schematic should be on the PCB tracks. Use the tools DRC design rule checking to verify that.

Ensuring all the components and connection are in the schematic is a matter of repeated crosschecking with what's on the breadboard.

Personally I always draw the schematic first, then created the prototype from that. Any changes I ake in the prototype I back annotate into the schematic.

Overall a key technique is to think of several different ways to crosscheck the design before sending off to be manufactured. And then check again the next day!
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 mortrek

  • Contributor
  • Posts: 33
  • Country: us
Re: Newbie - How to not get discouraged during a project?
« Reply #10 on: January 05, 2020, 04:00:42 pm »
Thank you for the advice everyone. I've bodged in a serial connection directly to the micro pins. That saved me a lot of headaches. The layout is still almost unusable for the extra module, but that can wait for a future iteration to fix. In the meantime, I'll make sure the basic electrical functionality is there, and the module will just have to stick out awkwardly. I will resist the urge to order new boards until I get this one working as perfectly as possible.

I've been going the easiest route for this, from what I can tell. It's just EasyEDA+JLCPCB. Probably not ideal or how any pro would do it, but it seems to work.

The rule/checklist suggestion is also a good one. Hopefully I won't need many more revisions. It's a simple enough board, with 21 components including a micro, 4 channel optoisolator, couple relays, bunch of resistors and diodes and headers to go off to other stuff/modules. I'm hoping I can pull this off without too much more trouble, but building a checklist should help, especially with future projects, to reduce waste and frustration.

The fitment issue was just a stupid oversight that I should have thought about but somehow totally ignored. That won't happen again.

I'll definitely be holding off as long as possible on a new PCB. It's frustrating to have a non-optimal one to work with, but I'd rather not rush to it and order one that's bad before it even gets here.

Part of my problem has also been that I've been living somewhere temporary for the last 3 months, so working on a project like this is very challenging. I make messes on the dining room table for an hour or so and then have to put everything away. The prototype is being kept in a large cardboard box so that I don't have to try to carry it without disconnecting wires and such. Soon I should have my own house again, and ample space for project messes.

Thank you again for the support and advice. I feel a lot better about this and future projects. Just need to work smarter and keep going.
 

Online TERRA Operative

  • Frequent Contributor
  • **
  • Posts: 793
  • Country: jp
  • Voider of warranties
    • Near Far Media Youtube
Re: Newbie - How to not get discouraged during a project?
« Reply #11 on: January 05, 2020, 04:48:45 pm »
Does EasyEDA do 3D models etc? I've never used it...

I use Diptrace and find the 3D display of the board helpful (Most decent PCB software does 3D these days), also the export to STEP file really helpful for importing into my 3D software of choice (Rhinoceros 3D) to then check fitment in the enclosure with all the other bits.

Like the images attached for example. It's saved a lot of frustration in making things fit in tight spots, but I still make stupid mistakes all the time regardless.... :D
« Last Edit: January 05, 2020, 04:50:43 pm by TERRA Operative »
Any circuit design must contain at least one part which is obsolete, two parts which are unobtainable, and three parts which are still under development.

https://www.youtube.com/user/NearFarMedia/
 

Offline ataradov

  • Super Contributor
  • ***
  • Posts: 6511
  • Country: us
    • Personal site
Re: Newbie - How to not get discouraged during a project?
« Reply #12 on: January 05, 2020, 05:45:47 pm »
I use different approach to "waiting". When I have an idea for some project, I quickly draw a schematic, do a board layout and sent it out for manufacturing the same day. And after that I switch back to what I was doing before. There is no "waiting". Quite the opposite, sometimes the boards arrive and it takes quite a while to get to them. This allows me to select cheaper, but slower shipping rate and not stress about the time it takes.
Alex
 
The following users thanked this post: SilverSolder, unitedatoms

Online unitedatoms

  • Frequent Contributor
  • **
  • Posts: 305
  • Country: us
  • LCR/ESR Impedance meter UA320 (in progress)
    • Unitedatoms
Re: Newbie - How to not get discouraged during a project?
« Reply #13 on: January 09, 2020, 03:25:38 pm »
@ataradov. Interesting. I never thought of having more than one project at a time. I mean not having, but actually proceeding in parallel, when money spending is involved.
Interested in all design related projects no matter how simple, or complicated, slow going or fast, failures or successes
 

Offline ataradov

  • Super Contributor
  • ***
  • Posts: 6511
  • Country: us
    • Personal site
Re: Newbie - How to not get discouraged during a project?
« Reply #14 on: January 09, 2020, 05:53:13 pm »
It really depends on the final goals. Having the freedom to try things really accelerates learning. You can spend a lot of time thinking if some part or idea will work, but ultimately this gives you nothing. You have to try it.

So money spent on projects for me is entertainment and education at the same time, and I think I'm getting a good value for the money.

Plus over time you build up the stock of components. So now if I need to try a new MCU, all I need to order is the actual MCU and the board. I don't need to worry about connectors, LEDs, resistors, capacitors, voltage regulators, etc.
Alex
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 12066
  • Country: us
Re: Newbie - How to not get discouraged during a project?
« Reply #15 on: January 11, 2020, 01:33:26 am »
@ataradov. Interesting. I never thought of having more than one project at a time. I mean not having, but actually proceeding in parallel, when money spending is involved.

I don't even know how many projects I have going on in parallel, usually there are at least 2 or 3 being actively worked on and then dozens of back burner projects filed away in boxes. I'm at a point in life where I am far more limited by free time and mental energy to work on them than by money to buy parts but I remember when it was the opposite of that. These days there are lots of cool things that can be built for very little money though.
 

Offline nigelwright7557

  • Frequent Contributor
  • **
  • Posts: 255
  • Country: gb
    • Murton-Pike Systems
Re: Newbie - How to not get discouraged during a project?
« Reply #16 on: January 11, 2020, 05:12:04 am »
Take your time and keep going over it until you are sure its right.
Be careful with your PCBCAD package that outlines are spot on so any overlaps stick out like sore thumb.

Rushing out pcb's is a mistake although the yare pretty cheap these days.
I remember a pcb that seemed right, when it came back and I built one up the git had 1VRMS output hum with inputs shorted !!!!!
I had mixed up power supply ground with audio ground and the hum was being amplified.
The charging impulses into the smoothing caps was modulating the ground line.
On the next pcb I kept the ground seperate and only joined at pcb connector and it was fine.
On another I somehow forgot to name a net "GND" to connect to all ground lines.
It sounded terrible until I twigged what was going on.
A simple link fixed it.


PCBCAD51/PCBCAD360/PCBCAD720 PCB design software https://www.murtonpikesystems.co.uk
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf