UPDATE: pyadf435x now has a firmware for $1.75 STM32F103 boards. kzfaq.info/get/bejne/j66pmqmonLbQko0.htmlm56s

How in the heck am I just now finding your channel? Seriously informative and well produced

I really appreciate the quality of your videos and the in-depth explanation. Great stuff!

Genius ! Learned a lot about sigrok, fx2, Linux (learned the "

This is one of the most thorough videos I've watched about a single IC. I'm surprised you don't have more subs.

I'm only quarter away from the start but already filled so many new things I must try. Your channel is really a great example of putting out videos much much beyond rich content. I'm going out of my mind learning many cool tools.

Slowly working my way through your videos. The content you've got here is brilliant!

The tools and techniques which you described are very useful. You have done a great job of clearly and concisely explaining how to use this chip and the software to control it. The significant time and effort you must have put into this project is greatly appreciated.

Awesome, wish I'd found this channel a long time ago. Bought a clone 24 MHz LA for $11. I'm sure it will fit most, if not all, of my needs. Pulseview is awesome. Love your videos.

Thanks for the comprehensive video. Well worth watching the whole clip!

Brilliant Video ! The amount of information sharing in a single go is unbelievable.

Fantastic video. Lots of hidden gems. I would say it would have been easier for people to find the various topics if you had split the video in multiple parts. Each one of the topics warrants it's own video. The PLL explanation for example could very well stand on it's own.

This channel is fantastic - please don't compromise on the length because it is appropriate for this kind of content but a time breakdown in the video description would be valuable when returning to it for reference. I would love to see you tackle an antenna analyzer for eg ISM bands that would be suitable for IoT, not just ham radio etc. LoRa at 433MHz and 868MHz for example, wifi or Bluetooth cantennae etc. I know that I could follow the existing projects for Arduino based antenna analysers with lower frequency signal generators but I know you would be able to convey the pros and cons of the this kind of system really well and perhaps mitigate some of the weaknesses through design. For example, you've already identified the lack of frequency calibration and temperature stability, and the harmonics from the square wave. Do they matter in an ISM band antenna analyser? Would love to see this as a blog post or video. Thank you for your contribution to the community! Subscribed!

Dude, keep up the good work... Love your videos.

Thanks a bunch. Without your detailed and knowledgeable video I wouldn't know what are the limitations of the ADF4351 and what I need to do to correct them.

Very interesting - thanks for taking us all through it.

what a nice channel, very informative -- keep it up!

I do have to say that your videos are very informative! Thank you for taking the time and effort in making them!

Great video! Very informative. Really looking forward to enjoy coming videos in this interesting field. Once again, - great job😃

Love the Details, Attention to detail is everything. Thank you very informative and motivating

Thanks a lot for this great channel! The quality of the content, clean production/edit, the in-depth explanation en great mix of theory and practice make this really stand out compared to others on KZfaq. Did you consider breaking up the video into a small series (much like chapters in a book), each focusing on a certain topic (e.g. the hardware overview, logic analyses, USB reverse engineering, pll theory etc.) as this will make the content easier to find (SEO) and probably get you more views. Suggest to even keep the full-length versions as you have today for the die-hards, and provide a playlist of same video but then cut into logical segments (+ corresponding title page/trailer linking to your content). Success and looking for more great stuff!

My second time watching this video. I understand the SIgrok stuff and the USB stuff much better. Still definitely need a third time! Your videos are awesome. Please do take the longest roundabout approach next time too.

This is a very good video. Thanks! Most likely I will watch it again at some point.

Very, very good video! Thanks for all the work you put in. Long but extremely informative, well edited and researched. Excellent stuff.

This is wonderful work and very useful insight in various tools and techniques to learn from, thank you!

This is fantastic content, thank you. I believe I can answer the questions about the chipset and the cost. For cost, they probably see the hobbyist as a nuisance market, where $150-$200 for an eval board is nothing to a large corporation. I think that's a mistake, too, because a lot of small companies have restricted budgets, too. For the chipset, I'd bet money that the designer of the board had code available for that chipset already written, that they could easily modify for their part. I'm not familiar with that microcontroller, but doing SPI in a RISK based microcontroller is pretty easy. Once you have the routines it's easier to reuse the reinvent.

Very good video. Useful and informative.

Excellent. Really, well done. Thank you.

You just saved the day, very much thanks to you.

I love your videos. I learned a lot of new and interesting things.

Amazing! So much info in just one video! Thank you very much!

Very nice description of PLL here.

Thanks for the nice summary on the ADF4351. The close in spurs are really well shown. FYI A squarewave output contains high levels of odd harmonics, so 3rd at 300MHz 5th at 500MHz etc. Hopefully these are easy to remove with filters.

I learned a lot from this video. Thank you.

Great thanks, very useful and informative, well put together video, helped me to think about moving away from bulky large lab-based equipment for smaller "in the field" projects, in less than benign operating conditions. Thanks again...

Thank you for a ‘Tour de Force’ romp through PLLs, Linux, USB, logic analysis, python, reverse engineering, etc., etc. It’s my second time through the video. If someone from ADI, TI, or the like does not hire you soonest, they are crazy.

This was epic. Well done!!!!!!

Great work!!! Thank you very much.

Great material and presentation! The only negative from my perspective is that there is no need to go down a rabbit hole with a rant about processor vs application. As miffed as you were about that, I was equally miffed about the time you spent on it!

Thank you, this video saved my day.

You guys with tech skills are so cool ;)

thanks, great job!

impressive work - looking forward your playing around with that boards. thank very much. maybe you can tell something about filtering the huge harmonics in one of your next videos? it would be great!

Great video!

Well explained!

Excellent video! Thankyou! You've inspired me to get some of these boards. Now to put a GPSDO/OCXO ahead of it. Oh, and perhaps tweak some of the voltage regulation etc. Cheers! de ZL1RWB

Very good video. The content on here is brilliant. It's the only video I could find that had all of the information I needed for my current project. Having said that, I was aware of this video for quite a while, but it was the last one I viewed because of the length (1.5 hours!). I think you could get a lot more views with some changes. I really want you to get more views because the content is so good. Firstly, the length of the video! I think most people these days use KZfaq to find things quickly, so I doubt many people would click to see a video that is over 1 hour long. 10-15 mins is my maximum, so I put off watching your video and I ended up watching all of the of the shorter (inferior) ones first. This video contains too much information of a broad nature. I think it would be better if you broke it down to several videos, with each one more specific and concentrating on one thing at a time. The open source lecture is also a bit irritating. Much of the video is spent explaining how to get windows programs to work in Linux, using Wine or some virtual machine etc. That right there is not a good advert for Linus. I use Linux at my work, just because I have no choice, and to be honest it is fairly crap and I end up having to use a Windows virtual machine for a lot of things things that I can't do using Linux. If I am looking to learn about ADF4351, I don't want to get into the pros and cons of Windows versus Linux, and be lectured about it, I just want to know the simplest way to get it to work. So in my opinion the open/closed source argument should be kept out of this video, as it is irrelevant. Besides my few negative comments, this is one of the best technical videos I have seen on KZfaq. There is some really very good educational material here. I appreciate your work and I look forward to seeing more!

thanks, I had never understood how a PLL could increase a clock speed.

Given that any money made from monetizing these videos will be reinvested in channel content, what's the verdict on the ad-break at 57:14? Thumbs up or thumbs down?

Very extensive explanation, but every minute is to be watched. I had success by using the CP2130 under Linux, with libusb. It has its caveats, but nonetheless, a reliable chip. Did many projects complete with software, thanks to the CP2130 and libusb. On the other hand, the FX2LP seems interesting for fast data acquisition, provided it does achieve USB 2.0 speeds (480Mbps). I have a similar board stored somewhere, but I'll use it if there is an app note containing the registers for USB configuration (or I may just use Wireshark to sniff packets going to a Windows VM and go from there).

It is Sept. 12, 2021 and these Analog Device Eval Boards @5:49 are $253.39 on Digikey with a 15 Week Lead Time. That is a significant price increase($163) from the date of this video (March 13, 2017).

Thanks for the great video! I'm wondering whether there could be more details about the very cool STL SDR spectrum analyzer you've shown. Is that from some open-source project or you made it on your own? I checked upon some similar designs and none has that very nice recording video and zoom-in feature in your video where you used to observe the settling time. Thanks! Would appreciate it if there is some source/guide on how to obtain one of that for hobbyists!

За одно видео рассказал больше чем за 5 лет учебы в универе! Nicely done! More explanation video pls. From Ukraine with love!

awesome

Fantastic video! Just a comment on the harmonic number: The 1st harmonic is the same as the base harmonic or 100 MHz in your test. That is, the 200 MHz peak is the 2nd harmonic, not the 1st, 300MHz is the 3rd and so on. Anyway: Excellent work!

Like another commenter mentioned... Analog devices and others don't have all the time in the world for their engineers to talk on the phone to hobbyists. The price is set at a barrier to entry. Any company or lab won't think twice to drop 200, 500, $5,000 on an evaluation and it devkit. Remember when companies used to give out boxes of samples and chips and all the latest everything five or ten years ago and then they quit... There's a good reason behind it in most cases.

Hi, When checking the harmonics up to the 10th, Why doesn't the phase noise appear broader and broader?

Did you disable the output during the frequency transitions and was the PLL lock on or of during those transitions? Other that that a very clear presentation and an nice explanation of some reverse engineering tricks, thank you.

Great video.. I was going to suggest a STM32F103C8 as a cheap micro-controller to replace the cypress, but I see you've already done that ;-) Looking on eBay, I've noticed that there are also ADF5355 boards available which claim to have a frequency range from 54MHz to 13.6Ghz, but they are around 4 times the price of the ADF4351 They look physically identical to the ADF4351 board, and the only difference appears to be the AD chip its self. But I personally, I don't have a need to generate frequencies above about 1GHz, so I'll probably buy the cheaper ADF4351 Aside from that, I think a nice project to build would be the STM32F103C8 paired with an ILI9341 touch screen display and the ADF4355, as all the components could be bought for well under $50 and using the Arduino Core (shameless plug here) to build a nifty self contained sig gen. The other cool thing that its probably possible to do with the AD4351 etc is to module the signal, by constantly sending commands to change its frequency. I did some experiments with the old AD9850, and it was easy to create a FM audio signal, so I presume that the same would be possible with the AD4355 etc

Great video. Can that eval board be programmed to have a step size of 100 hertz? Thinking about using one to replace a pll unit in an Icom vhf ssb radio.

1:17:16 This looks like the almost perfect square wave output, but this is completely different to what I've seen on the oscilloscope on the begining of this video. And in my opinion there is some reason for this. Look at the thing, that radioamateurs typically call S-meter. It indicates 0dBFS, which is too much. At this point I should explain what dBFS exactly means. dB indicates a proportion of power of two signals (e.g. output vs input, reflected vs forwarded etc) in a logarithmic scale, or, when some suffix is added, it indicates the ratio to some constant power level, like dBm indicates power versus 1mW level etc. FS means "Full Scale" and it's typically used together with A/D converters, where "full scale" means clipping. In fact, A/D converter is not able to give you anything bigger than the clipping level, so there is no possibility how to get a positive dBFS reading, they are negative or zero. So if you get steady 0dBFS reading, you can be almost sure that the device is overloaded, there is a lot of clipping, which makes almost perfect square wave from almost anything on the input. And that changing random noise is nothing more than a product of AGC dumping which slightly shifts the gain of the input up and down (this is how RTL SDRs software drivers behaves), which then varies the clipping point, which affects how much noise is displayed even if the noise level of the DUT doesn't change. If that board gives you a 0dBm at the output (and imo this is possible), then you need at least 60dB attenuator on the input of that RTL thing. I don't know where is the saturation level of that RTL device, but when it comes to receivers, the -60dBm is still relatively strong signal. In fact, this is going to be tricky, because these RTL SDRs are known for its poor dynamic range, so it's not good idea to attenuate it too much, but the -30dB is definitely not enough. You should get something like -3dBFS to -10dBFS reading to obtain reasonable results. But, at the end of the day, nice job, I absolutely love this.

Thanks for the good video. I am trying to follow this tutorial for my science project, and I am having hard time on wiring. I got everything I need, can you show me how to wire those up? Thank you.

It could be useful for testing low - mid range oscilloscope frontends if you have a high end oscilloscope to compare it to.

Could you recommend please good connection wires and bread-boards from AliExpress??

could this be able to replay a cap, exprimental offcource

What's the frequency resolution of this chip? Can I set the frequency at the KHz level even at 4.4GHz fundamental?

I am a real noob so excuse me for maybe asking a really stupid question but is it so that when you register a frequency, you can plug of the generator off your pc and whenever the generator is powered up it will automatically transmit that frequency? If not how to do so..

Thanks for sharing this. What a great channel you have. Very informative videos! Great that you include both the HW and the SW perspective... self.subscribe=True

Awesome channel. I enjoy learned a load. I wonder why you say you're scope is 200Mhz bw. Can you explain please. Thank you for your work.

Hi guys, forgive me for a silly question. can I use ADF4351 for PLL with an external RF source? Basically, I want to yield RFin from outside and get RFout at the same frequency?

Hi, Blowing air is a bad way to show the the temperature is shifting the frequency, since basically you are inputting capacitance into the circuit. still thought really interesting review.

Brother, with this, can the signal of WiFi router be made long range?

I am not understanding how you got the signal to display on your monitor? How is it that your able to run the frequency synthesizer through a TV receiver and just plug it in? I think a video on that would be much appreciated!

How did you get 200MHz out of a 1054Z oscilloscope? I thought the 1000Z series only goes up to 100MHz.

Thumbs up for linux user! :)

Talking about synthesizers - have you seen the Arachnid labs "Tsunami"...? It's definitely not meant for or capable of RF domain work, but it can both generate and measure stuff up to a few MHz with DC offsets, it has a TCXO, and it's controlled by an on-board Arduino. I'm not related to the vendor, just happen to like it... :)

que bueno que hubiera tenido subtítulos en español ...

That demo board comes with s/w (which needs to be specified, designed, written, tested, configuration managed, released, etc) , cables, CD, is professionally laid out for RF (again, the requirements are specified, documented, the schematic are made, the professional layout is made and configuration managed) , the board design is tested before release, there are manuals (probably written in English, by someone who does manuals for a living) , Analog Devices provides access to tech support, in fact, Analog Devices maintains a web site (which means some is paid to do that). That all adds up. Copying the board and selling w/o support is cheap. There, mystery solved as to why the real deal is "so expensive". Full disclosure: I don't work for Analog Devices (never did).

Please can i get some help i cant make the bin file for the firmware for the STM32F103 could you build one and put it on your github please :)

Why bother with a microcontroller if there are plenty of options to drive SPI lines? (CH341A breakout boards with SPI or CH340G/FT232R/CP2102/etc with bitbanging.)

So, this is the VCO ?

maybe that double sided board helps make a big ground plane?

My guess on the price of evaluation boards is that they comes with reference manual, all the schematics and ready to use programs. You effectively paying for the documentation and development time that you don't have to do yourself.

Not enough info for the beginners getting into synthesizer's, VCO's etc. It's easy for us advanced EE, but should be made a little more facile. More emphasis on the better quality boards would have been nice just to illustrate the difference in a quality waveform (and maybe you could have built a filter on the o/p to reduce all those awful harmonics)!!. I don't touch Chinese products but I suppose they are good enough for beginners getting into EE. Overall, good stuff and I look forward to viewing your other videos. Thank you.

Надо учить английский язык! Видосу лайк и подписка на канал.

Why doesn't he post anymore?

At time stamp 16:16, you say "I think it's ok to use the Analog Devices S/W (paid for by the sale of the demo board) to "get going" and then proceed to learn a tone of things from this s/w for your own purposes. So I wonder why you think it's ok "to get going", but not ok to, "keep going" (I guess). Just curious about the thinking there.

Ho-Hum!!!

Surely you could just set the scope to be a 50R load ? No reflections then.

Hello Joel, thank you very much for your very detailed video. It contains a lot of very useful side informations for me. Also the notes on your wiki are very useful. Question: is it possible to use an Arduino Uno (SPI 3.3V) for the ADF4531 control? There is a nice LCD board with an encoder and joy stick for $11 from elecfreaks.com: www.elecfreaks.com/store/lcd-key-shield-p-688.html There is already software to make a universal DDS-generator with this shield: www.kh-gps.de/freak_dds.htm Regards, Rudolf DL5FA

On board oscillator is rated at 50ppm which isn't terrible

Dev boards are priced that way intentionally to keep hobbyists out. The second you buy anything, from anyone, there's going to be a certain amount of expectation for support, even if you blast NOSUPPORT NO NO NO SUPPORT. There's always going to be one thick-skulled guy who's going to solicit support, and that's real dangerous at semis, where you can get engineers on the line pretty quickly. It's an intentional price signal. If a demo board gets super popular for..i don't know..glitching something.. and it's featured on ..Linus Tech Tips...do you really want to be that company who is getting barraged with support queries? Or the manager in charge of that product line (cause u gon' get fired at the end of the quarter) it's not like they're locking out the customers who are going to making a 2,500 unit production run. Every company I've ever worked at had a 'fuckit budget', usually anything up to $500 or $1k is no-manager-auth-required. AD can easily give those away (and they will-- use your company email, pick out a random figure of merit from the spec sheet and ask "Hey, we really like this product, but and we're thinking about using it instead of the MAX___ for our next gen, but can you tell me how you made the linearity measurements on ___ because they look a little *too* good."). Most companies are great about it actually, except Vishay, for some reason those dudes are bitches

Great video! BTW, for PC (Windows/Linux tested, not sure about Mac) to talk SPI (or I2C, UART, GPIO), you can also use FTDI FT232H with this Python package: github.com/Wei1234c/Bridges/blob/master/notebooks/readme/Bridges_readme_en.md

Anybody who actually needs the dev kit can afford it. If they can't then they're just a hobbyist and that's not the company's problem. They can't give away thousands of engineer hours on the telephone and customer service answering calls to hobbyist who could probably find the answer on KZfaq. It is irrational to expect companies like that to cater to some guy in his basement. I've been obvious myself for a long time in electrical engineering stuff but I also work in the field, and I think it is unfair to expect a company like that to cater to every single guy who wants a free chip.

The 2nd harmonic is twice the fundamental, 3rd is 3 times the fundamental, etc. There is no first harmonic!

Jesus, OCXOs are like £60 minimum.

wow that soldering looks terrible

The evaluation board prices are so high because the intended customers arent poor hobbyists like you and me. Their target audience wouldnt bat an eye at picking up a couple dozen at that price...

Hmm.. yeah thx for the video but I'm not sure if your clock or your mind drifted while making this video but all the things your were talking about in the beginning .. like checking this 35m 4.4ghz with your Rigol scope and where you would see what it could manage and show indication of bandwidth limits not least on the 200mhz bench scope.. now I'm more than 1 hour into the video, and your more or less did anything else under the sun in displaying your proud kung-fu programming skills and now we finally back to hardware and the original topic and now you say you have replaced the scope for some USB spectrum one, that you seem to say is for TVs and a brief glance on the rest of the thumbnail video, does not look like it wil be used.. but maybe it will come, will continue with the rest 20minute. // No IM gonna change my rating, simply drift from the intro topic and you don't even list the software you use. like this GQRX and then I got it and search for it etc and downloaded it and unpacked, just to realize that it is a challenge to get it to run for 88% of the people as solely for Mac OSX users (9%) and Linux (2%) and likely no windows support. (simple for who the 2% Linux users?) 1:08:58 c'mon' basic vital info, one would expect to inform in the video while displaying and promoting it (common sense). Lacking the basic aspect to relate to your own content from a viewer perspective, is not ideal' all that money begging info, like PayPal and crypto-currency Bitcoin links.. fair enough but relevant info in regards to video content' should take priority, so people don't waste their time..

I thought the max front end of the 1054 CRO is 100Mhz??? How did you get it to 200MHz???