#2102

[@romancharak3675]

When making DIY boards with small ICs, I design the traces so that they extend out from the IC, extending enough that I can get a fine-tipped soldering iron on the the trace. Also, I don't cover the trace with mask. Lots of flux and you can get solder to wick up under the IC. Still tricky though!

[@andye2005]

The datasheet specifies that the chip uses a 5 bit code, 0-31 your code is using 0-64, so it is going to cycle twice which will explain the jumps in the display. I think the performance should be better than what you have found, so it might be a idea to simplify the design down to just a single chip and a more conventional inductor. and cap arrangement Sweep that and see what you get. I've had some odd results trying to use chip inductor and caps in filter circuits. As you have a spare board, tack some 0-10pf trimmers in place of the 64102's and see what happens, Andy

[@reedreamer9518]

"I don't have a use for it, but it was just for fun" -- LOL

[@LutzSchafer]

Could you put an LCR meter across the varicap? Just wondering....

[@analog_guy]

The filter response would likely be sharper and have less loss if the chip inductors were replaced with larger high-Q inductors. One turn of hookup wire about 3/8 inch diameter might be in the right ballpark for each inductor, oriented to minimize coupling between adjacent inductors.

[@Mr.Leeroy]

check if `int Cap_value` gets translated to correct type for `ShiftOut()`. Arduino may map `int` to e.g. int16_t depending on the target MCU core and you simply shift out multiple bytes. Better to specify in traditional C way what your code requires during variable initialization.

[@hangup23]

for small boards like these, i found those $10 usb-c hot plates from aliexpress very useful. i solder those types of ICs (and much smaller ones) manually all the time. when JLCPCB screws up my PCBA orders i often have to re-work those types of parts too. QFNs are not so bad, actually. the worst are the tiny little RF filters and diplexers and packages like PG-TSLP-7-1 which have a large ground paddle and if stencil isn't made correctly you get too much solder and have to re-solder them manually all the time. hot air station works too, but with a board like this when you have 4 of those packages and not much else, you might as well use the hot plate.

[@noggin73]

I got one of these cheap hotplates to learn SMD. The 603's soldered great. ATtiny SOIC-8 was also good. The USB connector was not so good. Well, I have another 8 boards to learn.

[@jjoeygold]

You could build a discrete switchable filter with caps and FETs on the GPIOs of mcus

[@jameshall5835]

Where did you get those SMA knobs ? 3D print them ? Love ‘em !

[@IMSAIGuy]

www.pcbway.com/project/shareproject/SMA_Thumbwheel_Wrench_b3dc35b6.html

[@jameshall5835]

@ Oh I’ll definitely get a bag of these ! If you have one for F connectors, I would buy 100 of them ! I hate F connectors !!

[@erichanson420]

So this is the sort of thing that could be useful in an analog synth setup or something?

[@humidbeing]

You have a reflow plate. It takes 5 seconds to redo the bad chip.

[@paultomlinson528]

If you edited the code to address each chip in turn that may help and would be interesting to see the behavior

[@mechanoid5739]

OK! What have you done to your finger! Can't help but notice a stitch!

[@IMSAIGuy]

DIY hazards

[@Peter-rt3tu]

Maybe if you put a delay command somewhere near the pulling of the buttons, it would be easier to set specific values.

[@staryduren]

Maybe try to add line: 68.5 Cap_value

[@mikefinn2101]

Nice project well explained 73s wb7qxu