#avr Logs

Dec 08 2021

07:09 AM specing_ is now known as specing
08:31 AM PublicWiFi is now known as SantasWiFi
07:10 PM specing_ is now known as specing
09:05 PM WormFood: https://infocenter.nordicsemi.com/pdf/nRF24L01P_PS_v1.0.pdf <-- am I reading this right? pg 14. It says it draws more current on receive than transmit? Did I overlook something? That doesn't seem sane....however, transmitters are simpler than receivers, so maybe it is right.
09:23 PM cehteh: transmitters mostly need current for signal strength/HF amp ... while a receiver may fire up some serious DSP load for getting weak signals out of the noise background, so for short distance radios that could be pretty reasonable
09:26 PM WormFood: I seriously doubt they're doing any dsp stuff with this chip, but I could be wrong.
09:28 PM WormFood: however, I do understand how traditional superheterodyne receiver works, and that does take a bit of power. And if the transmitter is low power, and it is, then it makes sense that it's possible.
09:29 PM WormFood: however, I'd like to run some real world tests on it. I should be able to see the power dip when it transmits.
09:29 PM WormFood: AND, if that is also the case, then transmitting a lot would use less power.
09:31 PM cehteh: decoding digital signals almost always needs some dsp
09:33 PM cehteh: doesnt necessary be some programmable dsp, could be hardwrired/rom masked
09:34 PM cehteh: its not only that you need a receiver, descoding the signal and do error correction is important
09:36 PM WormFood: I see no reference to dsp in their block diagram, but that is beside the point.
09:36 PM WormFood: I agree, receiving signal, requires more power than most people might suspect.
09:37 PM WormFood: It does not need to be anything fancy, to be a functional receiver/decoder
09:37 PM cehteh: GFSK Demudulator could include it
09:37 PM WormFood: I think it is unlikely
09:38 PM cehteh: yeah no need to be fancy, but some stuff is certainly needed
09:38 PM WormFood: gfsk isn't anything fancy
09:38 PM cehteh: i didnt looked at the details
09:39 PM WormFood: Gaussian frequency-shift keying
09:40 PM WormFood: I know FSK and PSK, but I don't remember seeing, or at least looking up Gfsk
09:41 PM cehteh: i think these nrf things are mostly black boxes .. they packet, error correct and retransmit data, you dont need to do anything on yourself and can expect that when you get a packet then its data is valid right?
09:42 PM cehteh: dunno if the even do retransmissions or just drop packets when not correctable
09:43 PM cehteh: but they should have some more or less layer for error correction, otherwise i doubt you get reliable data
09:43 PM WormFood: well, it depends on how you fling around your data
09:43 PM cehteh: its 2.4ghz .. you can always expect some crap in the air
09:44 PM WormFood: it'd be wise, to have some way to figure out what the response is to, if you're sending data back in an ack packet
09:44 PM WormFood: I have some channels here, which are worthless
09:44 PM cehteh: as they have TX / RX mode i tihnk they dont by themself acknoledge packets and retransmit (but maybe i am wrong)
09:45 PM WormFood: Like 90% packet loss....probably higher, because each failed packet is retried 15 times, before it fails....and that's at max power, 3 feet apart.
09:45 PM cehteh: but that would mean they need to put some emphasis on getting data through in the first try
09:45 PM WormFood: with antennas
09:45 PM cehteh: yep
09:45 PM WormFood: I discovered that increasing the power sometimes hurt.
09:46 PM cehteh: yes more reflections can damage signal quality
09:46 PM WormFood: it was more reliable at power setting 1 (of 0-3)
09:46 PM WormFood: well, I think it may have been overloading the receiver. I should do some tests on a "clear" channel.
09:47 PM cehteh: could be as well, more power means more blurred signal
09:47 PM WormFood: I have test software setup, where I can change the channel, speed, power, and read settings from the serial port.
09:47 PM WormFood: turn the lna on and off
09:47 PM WormFood: I want to find out what exactly these things can do
09:50 PM WormFood: I've also discovered, that if you do a big channel change, and don't transmit, it frequently will not lock onto the new freq
09:50 PM WormFood: if you go from channel 76 to 0, the receiver won't work. But, transmit a single packet, and then it will work.
09:51 PM WormFood: if you go from 76 to 66, then it will be ok.
09:51 PM WormFood: fuckin' crazy shit.
09:52 PM WormFood: sometimes a jump of 50 channels will work, and sometimes not. Sometimes slow to lock. And sometimes going down 50 will work, but going up 50 won't. There is a pattern, but I'm not interested enough to discover it, but I do find it interesting.
09:53 PM cehteh: hf is close to magic
09:53 PM WormFood: it's "fm"
09:53 PM WormFood: that's "Fuckin' Magic"
09:54 PM WormFood: I just heard that the FCC just authorized FM on the CB bands.
09:54 PM WormFood: but, interestingly enough, if you know how, you can listen to an FM transmission, with an AM receiver
09:55 PM WormFood: most CBs can't do it, but some can.
09:56 PM WormFood: Just tune slightly above or below the center of the FM transmission, and you'll be able to hear the audio....now THAT IS magic!