I’ve been just using 2 x 4 header pins for these, but I downloaded a footprint and used it on 2 boards and discovered:
It has .7mm holes in the pads. These have header pins for 1mm holes. I solved the problem by removing the header pins from the modules and using wires to tie the module to the board.
Then, for grins, googled “Kicad nrf24L01” and found several footprints for these, some with nice, and some not-so-nice 3d models, and all those I looked at had .7mm holes.
So if you maintain a library that has an NRF24L01 module in it, check the pads and holes.
Ok. If I was nice, I’d research this and provide a list of right and wrong ones, but I’m too ASD for that
I guess I can be nice and include mine for anybody looking for one. This has 1mm holes. NRF24L01Module.kicad_mod (1.3 KB)
These counterfeit modules all over the internet are the reference design from the datasheet and a compatible chip. They wouldn’t be counterfeits if they were labeled with the chip they contain.
Always link to a datasheet or similar thing if you start a discussion about a footprint or symbol for a particular component.
For all we know this part is not meant to be soldered by use of pin headers but with something else. (Without a datasheet we can not determine if you misunderstood something or if everyone else misunderstood something.)
And i really tried to search but entering the string nrf24l01 will bring you to the datasheet of the chip not a module.
The datasheet includes a sample schematic and board layout. The internet is flooded with these little modules with this board layout for about a dollar each.
With no dimensions given (other than the overall size) This is because they do not sell it and therefore do not need to publish specifications for it (They simply document what they used to get the values contained in the datasheet.)
And from my experience without any documentation. Which is why i doubt you can ever use a module footprint made by somebody else as it is most likely made against an unspecified part.
After all to check if you can use a footprint by somebody else you at least need to check if the specification they used as a basis agrees with the specification of the part you intent to buy. (This is why every asset in the official library must contain a link to a specification that contains all the necessary information.)
Again: the specification of neither the footprint you used nor the specification of the module you used is known to us. We are still in the same boat.
It is for example still possible that the original footprint was designed for a board that had purpose made leads in it (I grant you it is unlikely for cheap knock offs but maybe sometime in the past there was a fully enclosed module that was made this way. Certainty can only come from a specification sheet.)
I probably have maybe twelve nRF24L01 modules. The IC itself is QFN so because I don’t have a microscope or a reflow oven I use the module.The pin spacing on the breakout module is 0.1" but it’s not breadboard friendly. I built a couple of adapters from perfboard so I could plug them into breadboards.
The footprint needs to be a double row 02 x 04 female connector with 0.1" pin spacing.
I’ve never successfully soldered those leadless parts TQFP are easy.
Next time I try the leadless parts, I’ll make the pads longer so they stick out from the chip a mm or so. I was trying a MEMS chip and you could see where the contacts were from the side of the chip, but I never got it to work. Oh, I think you don’t want to put anything under the chip so the chip is sitting directly on the solder pads.
Anyhow, these counterfeit NRF modules are so easy to come by, easy to use and work very nicely that I tend to use them instead of wires.
Oh. For soldering smd parts, I use a frypan. Electric skillet, actually. I took the thermostat apart and added a wire because the thermostat does not let it get quite hot enough.
So, looking through my magnifier, I use a toothpick to put a tiny dab of solder paste on each pad. It takes a little practice but there’s a big difference between not enough and too much, so it’s not hard to do. You’ll get your first board working. Then I stick the smd parts in place. Forget 0402 parts, they are pernicious. 0603 are manageable. Then, with the parts I want to put on all in their places, I set it on the cold frypan and plug it in. Watch carefully. In a couple minutes, you’ll see the dark gray solder paste turn bright silver, then slurp itself up into the pins. Just unplug the frypan and let it cool. Any bridges are easy to deal with using solder wick and your iron. For an unsoldered pin, I put the tiniest bit of solder on my freshly wiped tip, and apply it to the joint.
There are several pages on the 'net where this technique is described in more detail, with color photos, etc. I have been doing it successfully for almost two years, using a garden variety, bog-standard, stock, unmodified electric skillet from a second-hand shop. For more details, search this Forum for posts I made, that contain the word “skillet”.
My process is a little different. I preheat the board and skillet to around 150F - 175F (70C - 80C) and allow a minute or two for temperature to equalize. Then I start the actual soldering run.
A good lighted, desktop magnifier is essential. For me, the $20 magnifier lamps from the discount store are more frustration than help. The one I use cost about US$100, but it holds focus well across the whole visual field, and doesn’t have colored fringes around objects.
A binocular microscope, though not essential, is DEFINITELY a BIG help. About US$200 from Amscope.
I’d say that’s an understatement. Considering this approach in light of my life expectancy, I switched over to stencils and have no regrets.
Solder bridges, and loose solder balls jammed between pins, pretty much vanished when I started using stencils and calibrated my wrist to spread the solder paste thin, and evenly. I occasionally get a totally dry joint, but the really insidious ones are where the component appears to have floated on top of the solder and flux, without sinking down and wetting the contact with solder. I’ve heard this is a characteristic of old, expired, solder paste.
With SMD, I discovered that a good joint takes a lot less solder paste than you expect.
Yeah, I use 0805 and 1206 passives. 0603 is usable. For IC’s and active devices, I can (efficiently) work with basic pitches down to about 25 mils (0.65mm).
Once, I had just moved and couldn’t find my soldering skillet. So I walked up to the local second-hand junk shop, figuring he’d have one for a dollar or two. He didn’t, but he DID have a 3d printer?!? I still use it.
Yea, I should probably start ordering stencils for my boards. When I started using SMD, the stencil was and extra 200 bucks.
I have noticed a few of the quick-turn board fabricators now offer inexpensive plastic stencils also. There are some stand-alone stencil cutters, such as OSH Stencils and Pololu . Could be time to resurrect and update the thread at Cheap stencils? Where to get?.
I used to spend hours squeezing solderpaste though a syringe because a the extra cost of a stencil added too much to the cost for a small board run. However, the cost of stencils has come down dramatically. Last stencil I ordered from JLCPCB was about $6 I think. This is for a laser cut stainless steel stencil. Hardly worth worrying about at those sort of prices and it makes doing small pitched stuff so much easier. You pay more for a frame and electro polishing but for small runs this isn’t essential. I expect other Chinese fabs have similar offers. (No affiliation)
I’ve seen people using some sort of laser cutter akin to a printer to make them. I forget what these things are meant specifically to do so I can’t look one up as an example. Some kind of thing meant for crafts if I remember right so not terrible expensive.
Silhouette or Cameo type craft cutters are what you are thinking of, I think. I did borrow one of these and didn’t have much luck as the driver software was closed source but I think others have been more successful. I doubt if the stencils will be as good as laser cut steel. The demo cuts in Mylar that I tried suffered from ‘hanging chads’. (And you know, these can be a real problem …). I have used laser cut Mylar too but the steel stencils are rather better. Obviously, if you need the rapid turn around this might be worthwhile.
Wow. I’ve got several QFP chips and a tube of solder paste that I’ve had for more than six years, all untouched. I wonder if ‘partial’ stencils are available. I would only need stencils for specific IC’s
Well, forget using all your might to try to squeeze the paste out of the tiny needle. Instead, unscrew the needle from the syringe and squeeze a dab out onto something that doesn’t matter. Then you can easily pick up little bits of it with a tooth pick.
I’m ordering a stencil next time I order boards, and I think I’m going to quit home etching altogether. I already order out anything non-trivial. I’ll start ordering out the trivial things too.
(I used to tell programmers, "Use SQL for all things non-trivial, and most things trivial.)
Over time, the paste gets thicker - and MUCH more difficult to squeeze! (That’s different from my wife, who has also gotten thicker over time but is still quite squeezable.) I suspect the change in the solder paste is mainly due to evaporation of the solvents in the flux. This change in consistency affects not only the squeeze-ability, but also how well the solder paste “prints” through stencil openings, and how well the solder paste sticks to bare copper once you get it into place. When you manually populate a board, and visually observe the soldering process, you have a greater ability to compensate for these effects than if you were using the solder paste in a fully-automated soldering line.
Manufacturers of solder paste specify a shelf-life for their products, typically ranging from 30 days to 6 months, and usually based on refrigerating the material to 40F (5C) or so. I have no idea how much of the shelf-life rating is based on solvent evaporation, and how much is due to chemical degradation of the active ingredients in the flux.
I have resurrected some very thick solder paste, kept in my desk drawer for over 18 months, by squirting some LIQUID FLUX into the paste and mixing until the consistency “seemed about right”. There are a few discussions about this on the web (e.g., HERE and HERE) with no clear consensus about the best procedure.
My “revived” solder paste, using the electric skillet as a reflow oven, seems to flow, wet, and adhere as well as it ever did. It looks like I can get another hundred boards or more from the 150 gram jar of solder paste I bought over two years ago. The same technique should work with solder paste in a syringe, though you must figure out how to get it out of the syringe. Your results will undoubtedly vary.
I think the consistency of the solder paste is more important when using the syringe than using the stencil. With the syringe, my limit for efficiently achieving reliable results was about 50 mil (1.2mm) pitch - SOP IC’s, 0805 passives, etc. SOT-23 packages with more than 3 leads were challenging. At those component sizes I could populate and solder boards with very few defects on some days . . . . but spent a lot of time with the solder wick, removing shorts and bridges, on other days. After switching to stencils and acquiring a “feel” for how to use them, those same boards come out defect-free more often than not. (Three-lead SOT-23 transistors are still a challenge!)
