Відео

Thanks for the comment John. As a matter of fact, I published the original project in QEX, May/June 2010, the device in the video is an evolution of the original project.

@@vk5fe943 ah, very cool, will have to look it up on the archive. I work in the ARRL lab actually and the test equipment you've built is really cool.

Pls contact vk5fe@wia.org.au.

Thanks Juan.

73 Marko. I make them myself, there are other videos showing how I machine them. The purpose of the channel is to show that we certainly can homebrew our own microwave stuff, avoiding, for instance, expensive PTFE boards but still building useful modules for our experiments. In the case of this amplifier, the data presented shows about 10% gain difference (5 to 18 GHz) from an ideal amplifier (data sheet). Thanks for the comment. cheers.

Are you sure? I could just invert the panel to make you happy.

@@vk5fe943 100% sure. every graph goes from t=0 on the left to increasing time to the right. signal travels forward in time - from left to right

The cheapest, FR4 0.8 mm thick.

@@vk5fe943 from my experience FR-4 behaves very poorly on frequencies above 6 GHz. Very big losses

​@@andrsam3682 FR4 cost $5, PTFE $300. This is just another way to do inexpensive modules with performances not so far from professional units, that's the reason why coaxial cables were used.

@@vk5fe943 agree!

​@@andrsam3682 if you look carefully you can see that the rf is mostly not on the pcb anyway but in the coax cable, that should help limit the effects of poor pcb material choice too

The account belonged to someone else and now all the information that was once there is lost, sorry about that. If there's something related to amateur radio you would like to access pls let me know, maybe I still have the original files. regards.

Thanks for the comment - good tip, I'll try it.

Im thinking that is why the plots are so different. the number of insertions though is definitely having an affect on the Chinese ones and that appears to be independent of the cable length. Having said that I don't know how much the change due to insertion really has when installed. I use a lot of Chinese engineering stuff and for the price its fine for non industrial stuff I think

...plus one... thanx, vk3ola

Plus 1

Plus

+1

Thanks Juan. Yes, it's the DSI unit - I don't regret buying this device, a fraction of the price of a full microwave generator but, as I said, very capable.Cheers.

Thanks for the comment Zacki, I appreciate it. In fact I am not aware of groups sharing information on microwaves experimenting, except groups dedicated to amateur radio bands and how to put together modules to establish a communication link. The fact is that I started the channel just because of what you said, apparently there's a void in this area on youtube. I try to get some information from academic publications, instrument manufacturers publications, watching good youtube channels like The Signal Path, Microwaves and Machinig, etc and so on. I am currently studying a way of expanding the range of my 18 GHz spectrum analyzer and frequency counter to 30 and hopefully 40 GHz and I expect to be able to make a vídeo covering this subject in the future. Again, thanks and feel free to contact me (vk5fe @ wia.org.au)

it's the AVA183A, from Minicircuits.

@@vk5fe943 thank you for answer.

Exactly, I didn't expect that as well.

True, you can have better tones definition changing the timebase, my bad. However, the dynamic range is about 30 dB better using the spectrum analyzer (at least with my instruments) and it's easier to measure. Thanks for the input.

Thanks for the comment. The noise contribution of one attenuator pad is about 0.0203 uVrms (-140dBm) for a BW of 500 Hz (span chosen), so doesn't impact much on the measurement, also because there's also the noise contributon of the generator and amplifiers. The timebase phase noise is about -150dBm@100Hz. The attenuators were introduced mainly to adapt the levels, but the second attenuator has also the function of providing a consistent 50 Ohm resistive load to the mixer, minimizing reflections.

If you have any question (for topics not clear in the video), just let me know . cheers

Thanks Wagner !

Yes you are right, what matters is the impedance, but I normally prefer using inductors with an SRF not far bellow the maximum frequency used. The SRF, in fact, was given just as an additional information. Thanks for the comment.

Thanks for the comment, Dima. CNC is really a fantastic tool.

Valeu Edu

Hi Pavel. I used this one: www.sherline.com/product/carbide-tool-left/#description

@@vk5fe943 Thanks Rubens. I have watched your videos with great interest.

Thanks Ruel.

Thanks.

No, it's a commercial equipment produced by DS Instruments. Thanks for the comment.

Yes, it's a commercial equipment produced by DS Instruments. Thanks for the comment.

They are 18 GHz Amphenol connectors and cost around USD10

Valeu, PVT !

I prefer using warm white bulbs, instead of cold white, and depending of the time of the day there's no natural light from the window hence that yellowish appearance. I would like very much to have the opportunity of machining gold but I am afraid that's not gonna happen... 😀

I first used this solution in despair, when I realized that for a FR4 board this size it would cost $5 and PTFE $300, 😊. Thanks for the comment.

As it was only one unit, there are not schematics and gerber files are not available anymore, sorry about that. But it was a simple design, just additional line filtering for the commercial 24VDC switching power supply and a power split at the output. Anyway, if you want additional pictures, just drop me a line.

@@vk5fe943 thank you very much, I want some additioanl pictures, how to contact you? Leave my email address in here?

Just send an email to VK5FE@wia.org.au

I wish I were... I bought the R&S pair directly from them, when the AUD was 10% stronger than the USD (very unusual, I must say) and the spectrum analyser was a demo unit, so I paid about half the price of a new unit. Sadly that generator, with that nice form factor, is not manufactured anymore. Thanks for yout nice comment, btw.

Thanks Ruel, for this nice comment.

Indeed...

Thanks, mate.

Suggestion (if I didn't miss anything): On the MAX2870 power up the fractional mode is set, integer division is 250, fractional division is zero and the doubler and dividers in the input path are disabled. If you apply 10 MHz at the reference frequency input you should have 2.5 GHz at the output A (B is disabled) and the lock signal should be ON. If the output is too high for you to measure, modify Register 4 accordingly. If it doesn't work, consider checking soldering problems. I hope it helps.

@@vk5fe943 thank you! I Have 100mhz refrence now on board.. . The lock Led not work when power up. Only manual set to on off is work. Ref 10mhz gen now travel to me on post. I Will try it, or may be soldering first...

@@vk5fe943 it was work OK with external REFIN, problem was in internal REF gen

@@vk5fe943 you are using a simply scheme for NAX2871 - can you share it for me? (may be CAD files) i want to create simply gen device with MAX2870, but always seen more complex schemas, but i need simply one freq signal, how on your device. Thank you!