Hi folks - I’m a beginner and I have a circuit I’ve designed with a NodeMCU and a 433mhz transmitter. The transmitter has an antenna I’ve soldered onto it.
Am designing a PCB and wondering if I should try to add a trace PCB to the board instead of having an external antenna soldered on. I’ve read this is extremely complex - should I just stay away? If not, any pointers for how to do it? I have a CS but not an EE background.
Read all you can find in the internet about PIFA antennas. Calculate the antenna is quite difficult and more difficult to test it.
You can also copy the design of a 433mhz antenna from a proven circuit.
I would say go for it. Find a whitepaper with some dimensions of a working antenna and use that.
Check that the design that you are copying is the same thickness PCB, it matters but more so at higher frequencies than this.
At 433 MHz any track antenna is electrically tiny and not very efficient
You’ll be astonished how crappy those el cheapo external antennas actually are.
I hooked one to an VNA and quickly decided to pay triple the price to get an antenna that is tuned to the frequency required.
So, take a given design (someone gave a link) and you won’t be too far off.
But if you really want to make a perfect PCB antenna, you’ll have to invest a few 100 bucks in equipment. Plus time and failed PCBs.
The higher the frequency, the more of a black art this becomes. And I’m not at all one of those magicians!
Nick
You can do length tuning for simple antennas with a vector impedance analyzer/meter (~150$ in ebay). But that tells nothing about the radiation pattern in 3d space.
One option could be to use a chip antenna. Some manufacturers of those antennas offer service where they tune your design for best performance. But, again, the radiation pattern is a different thing.
Or you could learn to use for example openems, see e.g. https://fosdem.org/2019/schedule/event/openems/attachments/slides/3367/export/events/attachments/openems/slides/3367/openEMS_FOSDEM19.pdf and https://www.youtube.com/watch?v=byId9t0CBEo , but that will be a long path…
The TI app note design files link is broken
This page is better
http://www.ti.com/lit/an/swra351a/swra351a.pdf
If you don’t have an EE background, you can expect to find antenna design baffling and frustrating.
Its already one of the more difficult design tasks for someone who does have an EE background.
Cookbook designs are all very well, but without the proper test gear and a good understanding of what you are doing, you will likely struggle to get good results.
Save yourself the time and money and stick with the off-the-shelf external antenna. Unless you plan on making a million units?
Or space concerns.
But even of board antennas would need measurement equipment for getting the matching perfekt.
The main question to ask is however the following. Does it even matter? My experience on the sub GHz ISM bands says that it is quite hard to get it completely wrong. So if you do not need high data rates nor high range then just go for it. Worst case is you can only get a few meters and need to resent your packages quite often. This is especially true if this is a system designed for your own education. (Meaning you do not plan to sell it.)
If space is constrained, PCB antenna will be very problematic. Especially 433MHz PCB antenna which requires certain amount of space. IMHO best bet is to use one of the off-the-shelf helical antennas (like https://www.alibaba.com/product-detail/433Mhz-spring-antenna-OEM-433Mhz-helical_62014586792.html ) and use proper matching network to tune it
Without using field solvers and expensive measurement instruments, you’ll not get good results from PCB antenna.
That’s exactly that kind of utter crap I talked about earlier on.
You can take any screw driver and wind some stiff copper around the shank and get better at the first try. If not, snip of a quarter and try again. Then heat shrink and paint red for maximum power output.
Nick
Well as long as they’re not something like 315mhz sold as 433, they should be OK. Maybe you had a bad batch or something.
I have experience and use similar 868 parts; have matched them with a VNA (S11 below -15dB) and get more than satisfying results. Defnitely more stable and predictable than ceramic antennas I’ve been playing before. I don’t expect a 433mhz part will be that much worse. But of course ymmv.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.