The ESP-8266 packs a lot of networking power into a small package. Some would say too small, which is why they often come on a slightly larger carrier PCB. The PCB is usually little more than a breakout with an optional 3.3V regulator. [Frazer Barnes] went one step further: he put an equally tiny USB to serial bridge, an oscillator, and some power management on an ESP-8266 breakout board.
You can program the ESP-8266 via the serial port, so having a built-in USB port is handy. Of course, you might not need it in the final product, but with the board being 25x30mm, you can probably cram it into most projects. [Frazer] posted a bit about the project on Hackaday.io, and has a GitHub project, although right now the upload of the design files is pending.
There’s no shortage of ESP-8266 projects. We saw a small Zigbee to ESP8266 board last year, and also the antidote for a tiny carrier board that includes an LCD, switches, and more. We also have tons of breakouts on Hackaday.io: here’s one with all the bells and whistles, and a similar, stripped-down version. All open-everything, and ready to go.
Am I blind or the RF traces do have no obvious impedance control?
Can’t put too much engineering effort into it, otherwise someone might call it “not a hack”.
I think when you also see right angle traces its a beginner at work, however congrats for making this and putting yourself one step further away from being a beginner :-)
You’re not blind. This board design is absolutely ridiculous. Is there any word if this thing even works?
Maybe it’s secretly a 4-layer stack-up with 0.1mm prepreg between the outside and inner layers, such that those tiny 6-mil traces work out to 50 ohm. You never know :-)
Judging by the number of shorts on that QFN, it does not work…
Tested a breadboard-ed version of an ESP8266, deadbug style and was almost working. Keep in mind that some of the commercially available modules(especially ESP-03) route the antenna trace through vias.
I agree with you that the number of solder bridges is pretty damn high, but the project is an attempt to create something useful. In time, it will be refined either by feedback or fails. Seems that the antenna trace has some kind of matching network.
It’s “random impedance”, it’s the new thing, come on, get with it…
(Yeah, from a 30-year RF engineer… the creator could use some teaching on microstriplines…)
Would be nice to have an onboard SMA connector for more serious antennas.
If you look at the schematic there is support for an external antenna it’s just not populated on the board. I assume it,s the surface mount pads in the top right.
Another ESP board? There are already dozens of different designs on the market, and most of them are better designed. And whoever designed and built it clearly has no idea, how such things are done. And those shorts? I’d rather not show my design, if it looked like this…
I was trying to find your hackaday.io page so I could see your ESP8266 dev board that you invested hours of work into but I couldn’t find it.
Maybe it looks like the picture… He did say he’d rather not share if…
Point still remains. Is this a Fail of the Week, or does it actually work?
If I ever needed ESP2866 board, I’d buy one. Here buying Chinese modules is cheaper than ordering custom boards, buying parts and soldering everything together. And I don’t need to waste time on RF design (which must be done properly), which I could put to better use, like writing better firmware for device…
I’m sure you’ve never criticised something that you yourself hadn’t already done better right? You’ve never complained about a car breaking down, a phone not living up to your taste or a multi million dollar movie that you just didn’t find that funny. Right?
Yes. But this has some features:
– Random impedance
– Solder bridges everywhere
(function not included)
Easy folks. It sounds like some of you are implying that the editor just picked this one from the io page randomly without wetting it even a tiny bit. Besides, who cares if it is a good design or if it works? The guy who designed it put some effort into this and he even has a board made. That’s what matters.
Yes, every participant gets a ribbon…
More importantly everyone has to start somewhere. Quite a few of my first board designs never worked, but I learned a lot from my mistakes. Try offering suggestions instead of criticism.
Flux is your friend when trying to fix solder bridges:
http://amzn.to/1RVJRsE
Sparkfun has some good tutorials on soldering smd parts
https://www.sparkfun.com/tutorials/96
And RF is something you have to read a lot about to start to understand.
https://www.jlab.org/accel/eecad/pdf/050rfdesign.pdf
Why HaD? Is this a paid placement? There are plenty of these simple breakout boards out there already. Are you clueless, starved for content, or groveling for a buck?
Hey, the connectors on the edges are castellated! It can’t be all bad!
Tough crowd here. Jeez.
They have valid points though. But good job on getting something like this together. Especially for RF, it will be interesting to see how much of a performance improvement you can make.
Some tips: study the reference design if there is one.
A big thing that affect RF performance: impedance of the strip line, and the keepout (or lack thereof) around the ceramic antenna.
You can find calculator for this. Read the datasheet for the ceramic antenna as well…
Nice to read a comment that wasn’t just arrogant about the mistakes made during the design process. I learned something from this.
Those other comments say the same, only in Gordon Ramsey style.
Hi everyone
I’m Peter and I work on this project with Frazer.
Thanks for all your feedback, it has been noted. :)
The board does include pads for a SMA connector, we just didn’t have it in stock to fit it yet. The same is for inductors to control signal impedance, they’re on the rear of the board and not placed currently.
I’ll be happy to answer any other questions you have :)
Just a thought… based on the designs on the two other boards I see at https://developer.engineer (the site is on the board featured) could the hacker not have asked for advice from the engineer(s) who laid out those boards? I know the wrong kinds of engineers at my work, but they are usually egotistical enough to answer most questions and, at the very least, tell you what not to do.
Rather than a blurb about the board (I’ll avoid the issues I see with it), I’d like to see the work and thought process that went into it.
On a side note, given this article is a bit lame, anybody heard anything more about the ESP32? I’d like to get hold of one, but I might need to wait for them to port the Arduino platform again!