What you needed to do was not to put the flags elsewhere but to connect the +12V pins and the GND pins together.
1 Like
They are connected, just ugly
On my pcb the power connecter for 12v is going two ways I don’t know why I didn’t notice, just talking about power circuits lol. Thinking about it though, isn’t this really only bypassing the ferrite bead (another thing I will be changing)
2 Likes
Ah, I see now. Sorry for the misunderstanding.
2 Likes
Your way is much nicer
1 Like
Not entirely sure what I’m seeing on the PCB but in the schematics, the capacitors are still connected to +12V and GND but the resistor has the same net on both sides so won’t do much; the current will take the route with less resistance. The PCB has an “FB” instead, with the capacitors on the other side and a “C1 - Pad1” track going downwards (I’d assume it meets up with +12V somewhere?)
2 Likes
I think you need to remove these to fix it.
2 Likes
This is probably my issue if bypassing the ferrite is no problem. Swapped out the bc850s for 3904 but looks like they have different pin out.
3 Likes
Lol why they gotta do that
2 Likes
I burned 4 regulators, because in the datasheet it was drawn from below and I did not notice!! xD
4 Likes
Speaking of which
5 Likes
No Way!!!
I would never have even questioned that!
3 Likes
Now add LM317 and LM337 to the mix 
(ok, 79xx and 337 have the same pinout so that’s only one more permutation)
(I have a supply board that can take all four in the same position, if you pay a bit of attention when populating it…)
(On the other hand, I just soldered in a BC547 instead of a BC557 on another board, so maybe I should stop doing things like that…)
4 Likes
Definitely what messed me up! Doh!
Probably going to waste 8 more 3904s and do a funky placement to see if everything else works
3 Likes
OK, today was the next episode in the ongoing saga of fredrik & the opamps: turns out that if you need your opamp to work as a non-inverting amplifier, you shouldn’t connect the feedback to the positive input, because that turns it into something else 
(luckily I have plenty of PCBs where I didn’t mess up the opamps, otherwise you’d think someone was trying to tell me something 
)
(also, it was an old PCB that I just hadn’t gotten around to test. I think I’ve gotten more careful…)
8 Likes
It makes me feel a lot better that even @fredrik makes mistakes
7 Likes
Oh, I make a lot of mistakes but opamps seem to be my true nemesis.
Let’s see, same signal to both inputs, misconfigured feedback loop (salvagable), swapped power lines (salvagable!), and now this.
(All these mistakes were made in EasyEDA and when using stock components, so have switched to my own footprints 
. I also wonder if the extreme ease of use of a browser-based EDA hasn’t been a factor here – at least a couple of these were drafted on some random laptop that I couldn’t have run Eagle/KiCad on, and only turned into a PCB later, in the “do I have something else that could go in this PCB order” phase…)
4 Likes
A few of my mistakes have been down to KiCad doing odd things. Starting to get used to checking everyhting in the schematic multiple times followed by the same in PCBNEW…
4 Likes
Left column: EasyEDA’s standard 1N4007 diode with SMD and DO-41 footprints. Right column: EasyEDA’s standard 1N4148 diode with SMD and DO-35 footprints.
That could explain at least one case of dumbassery
5 Likes
Its things like this that keep me from committing to doing proper pcb fabs (not just front panel stuff). The footprints on these things are just crazy wrong sometimes.
3 Likes