💬 Just RFM gateway

[scalz]

i was sure, great work! i like your too :)
thx for precision ;)
yep atsam is native usb so no need of ttl converter, but need to burn the bootloader first through swd

[NeverDie]

@Koresh said:

@NeverDie
Can you list some of standard rfm modules shortcomings that you know? This board was designed strictly according to the reference design and manufacturer recommendations.

The two that I'm aware of are:

1. The RF switch that HopeRF uses on the RFM69HW is, allegedly, only rated down to 2.4v instead of the 1.8v minimum voltage that the Semtech SX1231h chip supports.
2. The RFM69HW can't cover the full Tx power range that, in theory, a proper implementation could. The RFM69HW's minimum Tx power is greater than for the RFM69W, and obviously the RFM69W's max Tx power is less than the RFM69HW. In theory, it should be possible for one module (or, in your case, radio on a PCB) to cover the entire range.

[kalina]

Great project! :+1: Can you share the scheme of your project? I want to
familiarize with your device more closely.

[Cliff Karlsson]

If you have time maybe it is not so much more work to also release a similar sensor board? Like a sensebender with rfm radio

[Koresh]

@kalina said:

Great project! :+1: Can you share the scheme of your project? I want to
familiarize with your device more closely.

Thanks! I will share schematic after applying some fixes.

[Koresh]

@Cliff-Karlsson See PM :smirk:

[Koresh]

@NeverDie
Thanks for the known shortcomings. Some thoughts about solutions.

1. This board is not a battery device, but I'm sure it is possible to solve this problem for battery devices. I can't find a spdt switch with recommended control voltage lower than 3v (2.5v is absolute minimum). But all of them require very low control current. So it is possible to use charge-pump converters (inductorless) as 3.3-5v low power supply (controllable by a controller like REG710, so it is possible to shut it down). This power supply can be used to control spdt switch.
2. Is this problem widespread? I can suppose it is caused by omitting some components in some modules.

[Koresh]

It works! :smile:
Have uploaded some photos. Will share project sources soon :wink:

[sundberg84]

@Koresh - woaw, RF PCB design - 4 layer or what do you use?
Via stitching for analog ground? How do you connect analog/digital groundplane together? F. Bead?
Man... so much to learn!
Please release the design... Im sure I have alot to learn from it.

[Samuel235]

I'm also here for some design files to learn from. It looks awesome, not too sure how applicable it would be, how do you plan on utilising this, is the usb port there for data or just power? If it is data, are you having it connected to your home automation server constantly?

[Koresh]

@sundberg84 Do not worry. It's not a very difficult design. :wink: ATmega is a relatively slow chip (I mean slow edges) and we have not long high speed data lines. So we do not need impedance control production and it is just 2-layer board (I have experience in multilayer design but it's really not needed in this case ;)). RF69 chip has no separate analog ground and even in reference boards all gnd pins are connected directly. RF ground planes are just decoupled without ferrite beads but may be I will add them, not sure yet if they are needed.

@Samuel235 This board contains usb UART-TTL converter (CH340, see bottom layer of the board and schematic which I will upload soon) so of course you can use usb to connect gate to controller. I have been using this device only for one day today and I find it really awesome. Very convenient :satisfied:

[Samuel235]

@Koresh I do like it, i just worry for people that use their main PC as a controller. Easily knockable/damageable unless they use a rear socket.