Make your own very crude Walkie-talkie with an Arduino 

Hi GreatScott, as always, great job done 😀 I have also tried to built that walkie talkie and with tips given by Andreas seems to be work.

Hey Great Scott! You didn't answer to his question, wouldn't you like to meet him and test the fixed system on a motor bicycle tour with him?!

ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-QhPSdN2hogU.html Please explain about this video

Neat. Cool it got fixed. :) I was thinking about doing this. You really don't need any large data rate to transmit voice. If you build in a vocoder, it will allow you to extend the range significantly, and get the transmission rates down to around 6kbps per channel. If you split it with fewer bands, even lower datarates. That's how a lot of walkie talkies work, and it's how GSM/LTE systems transmit voice. Half rate GSM is like 5Kbps per channel, and full rates I believe anywhere from 12/16Kbps dual. It also allows you to put basic encryption on the signal. Cool project. Nice video. 👍

GreatScott! I love your videos and have a mission for you. Please please please make a video about the low power library. You will be able to simplify it so well! Thanks!

bic !

This video is sponsored by Schwan-Stabilo. Fine liners and text markers in a variety of colors starting under $2.

So true

Why do you think he is not sponsored by them already?

Watching him write schematics and data is really enjoyable. I'm very neat with how I write, but he does it with so much ease.

:-)

0:30 around 30?

That is my guess too... a simple rc filter could help a lot. I also think the word smoofly should be added to the English language :D

Agreed, I did some similar. Output was crap. Made a crude class A amplifier using LM317 IC and output was clearer albeit the IC required a heat sink because of poor design

@@SimranjitSinghMehra How did you make a amplifier out of a voltage regulator?

I would have thought you would need a voltage divider on one side of the speaker so there would be an equivalent of half voltage either side of the PWM signal to get the speaker to go in AND out.... It worked for me when I tryed a similar circuit years ago ( pwm to sound out )

And he needs a 8 kHz low pass filter before the ADC and another 8 kHz low pass filter before the speaker/speaker amplifier. 16 kHz sample frequency is only capable of 8 kHz audio reproduction (Shannon Nyquist sampling theorem). By not having those low pass filters, all of the PWM noise, high frequency PWM harmonics, and clipping (high frequency pops) is drowning out the useable audio signal. Signal to noise ratio is really bad without filtering out the unusable frequencies.

I think that rectangular signal is not the best wave form for audio output to speaker. Perhaps if the wave form is more sin like, the quality can be improved a little bit.

Like in a Class D Amp To filter the PWM Signal

I just was going to mention this too. The PWM signal is just poor and results into that kind of artifacts without a proper filter. For an even better quality, I rather would go for a dedicated DAC though.

Correct ADC requires DAC on other end to rebuild non quantized signal, that's why audio is clipping

Illegal to operate, but would be working with a few more steps. First of all you of course will have to kind of filter the PWM signal to get some decent audio quality. OK - one could argue the mechanical slowness of the membrane of your speaker could do that for you. But what is by far worse you did drive the speaker with DC BIAS! That means, the membrane of your speaker will already be pulled or pushed against one mechanical limits of the full stroke length, that is why the system did sound so badly! Use a proper dimensioned C to block DC Voltages and you most probably will have a working system, if you want to do it the proper way then you'd have to have a proper filter (lowpass filter and buffer amp) then to drive the speaker. Apart from the fact that you will not be allowed to use such a system legally in most states all around the world since you then would operate a 2.4GHz transmitter without any EMC safety tests being applied. And maybe not even being respecting the bandplan or maximum allowed transmitting power of your country. And maybe not being respecting minimum allowed distance for transmitting devices/ antennas from any kind of human livings and some of their best parts (well, just in case their best parts may also include their brain and are not only their reproductional systems which I don't want to decide here). So, nice idea and it could easily work if it was done properly!

Andreas Spiess (the guy with the Swiss accent) made a video on debugging the project: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-YJ25eQRbhaQ.html

Same thing i thought, it just sound like a saturated sound. A quick view with the oscilloscope should make the problem clear...i would go for a simple dual stage amplifier and a low pass filter at the output.

He uses a potentiometer in the feedback, so he can tune it till it's good... I think this shouldn't be a problem.

upvote this so he can see

Yea probably from area 51 😂

I bet if he added a cross cut saw blade and a Speak and Spell®. He could have transmitted past the planet MELMAC!

Area 51 😂😂😂😂😂

@@highkicker11 agreed it sounded like a lot of clipping.

@@kreynolds1123 Yeah, i noticed when he was looking at the signal in the oscilloscope that it was clipping

Same, it worked nicely for a basic intercom for me.

Congrats :-)

Yeah...RF in all its analog glory is one heck of a abusive bitch.

2 plastic cups and a string :)

Basic FM transmitters are dead simple to build and there are super simple demodulator chips available for the receive side if you don't want to design that half. Just find out what frequencies are legal to use in your country without a license and pick one that's easy to design for.

@@defaultuser000 27Mhz band seems to be ok in most countries. Give it three more years and we won't be able to blink an led without calling 5 api's on our IoT enabled NeXtDuino's. This is getting out of hand. Not everything needs a microcontroller.

@@defaultuser000 It is one thing to use general communications frequencies with commercially available equipment. But using non-commercial home-brew transceivers is not legal unless you are a licensed amateur radio operator and use it in those bands.

Choke ? And what speaker is ? ;o)

@@jankomuzykant1844 Not one that small with a PWM that slow. The speaker isn't sufficient.

@@rich1051414 You forgot mechanical inertia. I think it's first episode - with the problem. Now I'm waiting for GreatScott's fix ;o)

@@jankomuzykant1844 At a high PWM frequency, you can get away with no choke. Also, a larger speaker with more mechanical inertia also helps. Here, with the PWM frequency being so low, there is NO WAY that little speaker can do anything substantial to the signal on it's own. A Class-D amplifiers needs AT LEAST a 40khz signal to IDEALLY produce no audible distortion from 20 to 20khz, and that is assuming there is enough inductance to 'connect the dots' on the pwm signal. It's best to not rely on the speaker, because you have no idea what present or future type of speaker will be hooked up, or any speaker at all for that matter.

@@rich1051414 Nah, its okay. You can PWM into earphones and you won't hear the distortion at 16kHz. Been there, done that. This is something else.

Well you can get such a FM transmitter with stereo generator for at least I have one of those for about $3

That is a lot power. Electrical safety should be the first priority. Be careful. However, if you want to do something similar. You need an IGBT H-bridge configuration. Transfer DC to AC. The modulation will be SPWM. That show the same parameters of sine wave. Important: dead time between IGBT switch. You need to implement diferent reference voltage to IGBT. Finally, you need to desing the protection systems. If you are more interesting the 3 phase speed control motors do that, but you can do that for monophase.

Maybe using the other op-amp in the LM358 as a Sellen key filter. Your on right path with this Big B

I would have liked to have seen him use the lm856 instead. On the audio I couldn’t help but think the microphone could be some of the problem.

Exactly what I was thinking I don't understand how PWM generate the required analog audio signal and why not use DAC

@@danuuu101 Variable duty cycle that's tied to the ADC value. You'll get an audio signal after you apply an LC low pass filter. Look up sinusoidal PWM or PWM DACs. I think GreatScott did a video on those, actually.

I do not see any in your plan

DAC is the solution, but a LC filter may work out.

@@lazar2175 LC filter builds a DAC basically. I would use an RC filter in combination with a LM386 amplifier to keep it small and simple. The more important thing which Arduino cant handle properly is high frequency sound signals. I would use a proper high sample rate 8 bit or even 10-12 bit ADC.

@@tunahankaratay1523 No, no it does not build a DAC. Like it's name suggests, it filters the pwm signal. The dac converts the pwm to a AC analog signal suitable for playing back in the speakers. He has a video on it, go watch it.

@@lazar2175 I would say an RC filter instead and LM386 ... but why at first place Arduino? This is a very simple task for 2 transistors on transmitter and 2 on receiver and you can beat audio quality and range with a whole magnitude

Hahahah OMG that brings back some memories hahah

Impossible Mission!

not even close :D

Great comment!👍

I wouldn't use FM for this

Esp8266 range is low

its a little bit harder to do bi directionnal walkie talkie in analog tho.

@@pascha4527 plenty of 27Mhz analog walkie talkie circuits floating around on-line. Doesn't seem to be rocket science :-)

Too many engineers here....wrong place for me😂😂😂

I was wondering the same thing, that’s why I was reading the comments to see if anybody else thought that.

It was the sponsor of the video. And honestly PCBs are like a couple of euros (even at that timr already) so no big deal..

It's still much cheaper than other manufacturers

​@Pilot16H It's more like $18 for 5 boards but still $3.60 per board is a great deal. Also that price is the same for sizes up to 10x10cm.

Isn't the first order free shipping? I never ordered off of them, but i don't see it as a big problem if it is free shiping for first order, just make new accounts.

@@lazar2175 You'd need to move a lot, as I'm sure they track the shipping address. ... I think the free shipping bit has changed. All the shill plugs have not mentioned this in awhile. I screw up too much for these services anyways. I'd hate to design, wait 2 weeks, then get stuck with 5 junk boards, redesign, wait 2 weeks.... I can etch a small single sided simple layout in 1-2 hours with toner transfer without standing up or leaving my bench. I can do a large double sided photoresist, with soldermask, and an aluminum can solder paste stencil in a day even as a partially disabled gimp. Heck, a manhattan style prototype or waterproof marker design is even faster. If all that is too much, I'll just use a razor blade and have a small board in 15 min. I don't generally like subscription services, but perhaps they could come up with something cheap that makes sense? I wouldn't mind an ultra striped down option for prototyping. If I could order a special one-off single board prototype to do iterative designs reasonably quickly and seamlessly, then order a general production version of 5-10 like once a month, or two, or even quarterly, I might consider it. I mean I don't care about soldermask, silkscreen, tinning, and I'd even give up through hole plating, for basic prototypes. That's how I design anyways already. If they had a service that was really cheap, somewhat flexible, and just sent me some cool basic designs, or useful generic SMD breakouts because I didn't order the full subscription boards allotment I might consider it. If it were possible to easily order prototypes or small production runs where I could get one or two BGA or QFN chips attached without hassles or coordinating with multiple parties, they could persuade me much more easily. Heck even attaching a TQFP with more than 64 pins can be a pain at times. I usually need to have 3 chips per board just in case I need them if it's a new design. I don't even have a BGA setup yet. That's probably the only way they could get me to sign on and really use the service. This cost per board thing is a funny concept compared to DIY. My entire dual style etching setup was less than a bill, including the 10x pack of single sided A4 sized CEM-1 from ABCfab. I still have 2 sheets of that laying around in addition to all the other stuff I've collected. My cost per board is just time. I wouldn't even carry the pocket change equivalent and call it money. ...and I believe the OP mocked me for my lack of BGA experience in the past, so I don't think I'm alone here :-) -Jake

@@UpcycleElectronics Quite obviously it is superior to make your own, in fact i've never ordered off of JLC because my country has quite insane customs so it would take well over 60 days for me to recieve the boards. I either "borrow" large pieces of unetched boards off of my work, or i order it through AliEx, they are 5 pieces of A4 size for 2€. I also can't get FeCl3 so i have to use HCl and add 17% H2O2 to make my etchant. I don't own a laser printer so i have to go to the city and have someone print it for me, mostly losing time. But i can make boards for about 0.30€ per board without soldermask and other shenenigans. For smaller projects i still use the stripboard, and for projects that require traces smaller than 0.2mm i order it locally.

Congrats :-)