Electronics tutorial - The Comb Generator

#98 In this video I look at a basic high frequency signal generator - the comb generator. In its most basic form in can be built with less than 10 components but to fully take advantage of it, first in must be understood. I look at the basic concepts behind this type of circuit and I try building one.
Datasheets:
www.tme.eu/Document/04fa0ef07...
www.vishay.com/docs/81857/1n4...
media.digikey.com/pdf/Data%20...
Spice model: assets.nexperia.com/documents...
Schematic: drive.google.com/file/d/1kMBg...
Special Thanks to all my supporters on Patreon! Especially @afiskon and Ralf B.!
If you liked this video be sure to check out my other videos and you can also subscribe to be up to date with all the new ones!
If you want to support the creation of more and better videos please consider checking out: / feszelectronics

[@YSPACElabs]

Decently cursed (using a reverse biased diode to make a sharp pulse) and pretty simple to get GHz frequencies. I like it

[@BrianSiegelwax]

You earned my "subscribe" with this one, which is actually the first video of yours I've seen. I'm clicking on your channel next to see what else you've got.

[@FesZElectronics]

Thank you! Hope you will find more useful things!

[@MJ-vt1kp]

Very good explanation of working principle for comb generator, thanks.

[@antoine___________________907]

excellent with very clear explanation

[@radio655]

Very well explained!

[@Parirash123]

Thank you for sharing

[@ufohunter3688]

Love your videos. I think Tesla thought an abrupt pulse was the key to unleashing his radiant energy. The sharper the pulse, according to him, the higher the energy release. He used all sorts of techniques to squelch the spark and make it transition as fast as possible. A few operators at Niagara falls station died when they threw the switch to off position. The sparks that came out of the contact point, killed a few men instantly. Maybe they were idiots! Who knows!

[@YSPACElabs]

I think softer switching is actually the key to better wireless energy transmission in a Tesla coil. That way, you get a nice sinousoidial primary current that lets the secondary resonate smoothly instead of building up voltage so fast that it creates an arc that wastes energy. However, wireless power transmission with a Tesla coil or a similar device is extremely impractical and inefficient (but definitely really cool). I was only able to get a few mW from a small solid state tesla coil powered with 24 watts.

[@KamleshGadhvana]

thank you nice one

[@grigbarbulescu4218]

Yet another nice and educative video! Congratulations! I like your spectrum analyzer, if you do not want to explicitly promote it, how do you suggest me to search for it? Thanks

[@FesZElectronics]

You can google or search on ebay for "LTDZ 35-4400MHZ spectrum analyzer" and you should find it. But I think there is a better board available, so maybe you can find something else by simply searching for "USB spectrum analyzer".

[@thomhpl]

First of all, thank you for posting such good quality videos, your hands on approach is really great! Could you release the gerber files for this pcb? I have an old (and probably not properly calibrated) HP 141 with the 18GHz spectrum analyser plugin and I'd like to test this. I would like to have the same gerber si that stray inductances and capacitances should be close to yours...

[@FesZElectronics]

I'm not sure how important the Gerbers are to get similar results - I did not apply any special routing rules, not even proper impedance calculation for the traces. Also its just a homemade 2 layer board done with an iron so if I where to make 2 pieces, they would turn out differently. Regardless, I release project related files only for a certain tier of Patron subscribers as a way to thank them for supporting my efforts.

[@thomhpl]

@@FesZElectronics I didn't know that and I totally understand. I'll try to reproduce the circuit with my cheap CNC and I'll let you know how it went if you're interested. For your information I also did some experimentation with the BFU768 silicon germanium npn transistor. It has a ft of 110GHz and it gives a nice comb pattern until roughly 4GHz, I can even still see some peaks until 8GHz thanks to the -80dBm noise floor of the analyzer. I never subscribed on patreon but I'll definitely check your page, your videos are amazing.

[@magtazeum4071]

Hi, What is it meant by the size of a comb generator? for example, 0.1 m or 2m comb generator, does it refer to the physical size?

[@FesZElectronics]

Hello. I guess I was comparing it to a proper signal generator. The CG can be built really small.

[@jamessimpson1506]

There are many components to learning electronics. One place I found that succeeds in merging these is the Gregs Electro Blog (check it out on google) without a doubt the most useful course that I've heard of. Check out this interesting site.

[@schwinn434]

Nice explanation, just wondering if you could have used the FFT analysis of the oscilloscope, in order to show all the harmonics in the physical board's circuit, which you built; I have watched several of your videos, but can't remember if you have used FFT analysis in your 70M scope; or, even if FFT is available in your scope, the 70M bandwidth limitation of your scope would limit an FFT analysis, in someway? I know you have you have used the FFT analysis in LTspice to analyze some circuits .

[@FesZElectronics]

My oscilloscope does have the FFT function, but it has 2 major limitations. On the one side the oscilloscope bandwidth is only 70MHz, so that will have an impact in the FFT results also; and secondly the FFT is done over single oscilloscope captures of 4000 data points points - so the FFT is not very useful to look over large frequency intervals. For these 2 reasons a spectrum analyzer is the better choice.

[@schwinn434]

@@FesZElectronics Thanks for the informative reply.

[@pbaemedan]

Fesz, good video, thank you. I truly enjoy your videos. You are a brilliant young Engineer. I noticed you used a model for you circuit, was that for the 1N4148 diode? Was there a parameter you wanted to investigate? Say the junction capacitance?

[@FesZElectronics]

Thank you! The model was for the 1N4148; I got it from NXP. The main reason for using that was to have a more accurate behavior than the default model that LTspice comes with for this component. Of course, any of the parameters inside the model can be edited afterwards to see how that influences the behavior.

[@pbaemedan]

@@FesZElectronics thank you for the information. Keep up the interesting uses of LTSPICE in the real world.

[@FPiorski]

@@FesZElectronics You probably already know this, but you can use subcircuits with the built-in schematic symbols www.analog.com/en/technical-articles/ltspice-using-an-intrinsic-symbol-for-a-third-party-model.html

[@FesZElectronics]

@@FPiorski I did not know of this. Thanks for pointing this out!

[@p_mouse8676]

You know that you can upgrade that oscilloscope for free to 100Mhz or even 200Mhz?

[@FesZElectronics]

I know. I just didn't want to risk braking it...