Homebrew 80/40m SSB/CW Rig -
Vložit
- čas přidán 20. 08. 2024
- Video looking at the construction of a SSB crystal filter using the July 1987 QST article "Designing and Building Simple Crystal Filters" by Wes Hayward. The video looks at the selection of the crystals, construction and testing. Please feel free to provide constructive feedback on any points I may have overlooked or incorrectly carried out. I'm here to learn!
73s
Charlie, ZL2CTM
hola charlie. felicitaciones por ese trabajo de cristal filtro. te cuento que esa configuración la realice hace unos 30 años atrás aproximadamente. pero en frecuencia de 10 mhz. use en esos años un monitor de estación a válvulas heathkit que tenia incorporado un oscilador de doble tono para ser usado es ssb calibración. me construí un qrp en banda lateral única y lo quería en LSB con cristal filtro . para encontrar el corte ideal no seleccione los cristales de cuarzo, todos fueron al azar, solo que los capacitores fijos entre ellos los remplace por capacitores variables y usando el doble tono encontré los valores correctos para una respuesta decente. fue tan bueno el resultado que cuando salía en 40 metros todos pensaban que era un equipo de fabrica .jajaja. fueron muchas noches de ensayo y error pero el resultado fue espectacular. tu video me hace recordar aquéllos años cuando era un bebe en electrónica pero lo aprendido en general de construir tu propio equipo es impagable. nuevamente felicitaciones y me entretiene mucho tus videos. mil gracias por publicar.
atte
Enrique Vidal
CE5LBM
Hi Charlie! Thank you for the great video! I just wanted to note that you could use the SDR for fine readings too. Set the SDR to SSB, then pipe the audio using VB-CABLE to an audio spectrum software. I would recommend using Friture. This way you can run FFT on a very narrow bandwidth, say 2400Hz which will allow you to get sub-Hz resolution on the spectrum in a matter of seconds. I have used this method with really good success. Hope it helps you too!
I love you when you talk about the basics. Keep doing them.
I'll try Moustapha, noting that I am no expert and a lot of what i do is the World according to me. I rely on constructive feedback to help me and the others reading the comments.
Hi Charlie,
I also built a crystal filter (12 Mhz) for a spectrum analyzer project from Ashaar Farhan. I used the Haywood articles and built a crystal tester from a Doug Demaw article from QST. This tester also had an comparison oscillator but allowed you to throw in a pad (on a seperate circuit )so you could measure the resistance of the crystals. Supposedly once you sort the crystals by frequency, you place them with the highest ohm crystals in the center and work down on both ends. I only tested the filter by
hand tuning the sig gen through the pass band and plotting the P-P scope output. The bandpass was supposed to be 1000
and with my imprecise measurements, it seems like about 600 to the 3 db points.
Andy KM4TRT
That's very interesting thanks Andy. I build a filter a while back where I too tried to characterize each crystal. That filter didn't work very well either. I suspect it was my method/equipment that was a fault. Pete N6QW has written quite a bit about his experiences too.
Great learning curve watching this video ,Thanks Charlie
It's not too bad for the cost of the crystals and the simple approach from Wes. I really do nee to make an analyzer at some stage!
Be nice see the whole finished device functioning
Good luck from Indonesia 73,
(Amatir Homebrew Handmade)👍👍👍
Thanks. It worked very well.
Some time ago I picked up some 400 pieces of 8.3886mhz crystals in H6C/u packages on ebay for a few bucks. These larger crystals supposedly have higher Q than the mini micro processor sized crystals, might make better filters. I've been meaning to try and build a filter out of some of them using your approach. I'd probably use an Si5351 as a signal generator to sweep the filter, with my Leader frequency counter and old school Tek scope.
So I tried your oscillator circuit, but I put in 100pf caps and used a junk box surplus 2n2369A transistor and the stupid thing oscillated at about half the marked crystal frequency. Put 100pf caps in parallel with the existing ones (didn't have any 220pf in the junk box), and now it's oscillating at the marked crystal frequency. Weird! I'm running the oscillator at 8v because I had an LM340K-08 in the junk box to hang on the output of my 13.8v bench supply. I soldered two pins from a broken miniature tube socket (valve holder in your neck of the woods) to the lash up to make it easy to swap crystals. Now I can start to sort them out and find a set of 4-6 rocks close enough to try and make a filter.
Wow 400 crystals. You should be able to build several filters with that many! Hi hi. I'm keen to rebuild mine once I have the network analyser built.
Charlie
Hi Charlie, great video mate!
I ordered around 500 10.7MHz crystals from China a few months back (paid little over twenty quid for the lot), not had any time to experiment due to work and family commitments but hope to have a tinker in the coming weeks. I wasn't sure which way to go with regards to the design, Pete Juliano N6QW was not convinced the software that is available gave a close representation of the real world results so I was keen to take the grass roots approach which you have clearly done.
As it happens I have have a copy of that magazine from QST archives so will have a read over it later today, but what has inspired an offshoot from my current homebrew SSB radio projects is a homebrew semi-automated crystal tester using an WiFi enabled Arduino as both a Frequency Counter and a data logger, I can easily knock a .NET application up to receive the data and dump it into a Spreadsheet document and quickly match the crystals that way - a bit over the top you might say but the idea sounds cool :)
Anyway, keep these videos coming Charlie, big fan of your work and happy homebrewing.
73
Rob
2E1IIP
Cheers Rob. That crystal tester sounds interesting, and should speed up the selection process. Pete Juliano has another standard approach for making crystal filters, which I have used in the past. I thought I'd try something different this time. Seems to work ok at this stage.
As for the videos, you'll notice I like to take a simple approach. It's not that I don't like complex capable circuits/radio, just more that I prefer the simplicity. I'm pleased you find them useful.
73s
Charlie, ZL2CTM
I realize this video is now super old and you've done newer crystal filter videos, but it's the one I used to build my crystal filter back in the day before I bought a house and had kids. Finally coming back around to it. My filter response is uncanny in its similarity, except I used 10 MHz crystals. I have similar I.L. and bandwidth and even have the same droop on the upper end of the filter response. I wonder if it's coincidence or a result of the methodology.
Have you earthed the crystal cans - seems to make a difference to the skirts?I use Spectran and a RF noise source to show the response curve.Great work with your experiments mate..73
Thanks Tony. I have earthed the cans in previous filters, but neglected to for this one. I should have. Good spot!
Very nice 73 from iraq
Thanks. It will be interesting to see how it performs in the radio.
موضوع جيد جدا الفلاتر الكرستال شكرا
Hi Charlie, that is nice guidance video, but please explain how you calculate those four xtal and get 21hz, 35hz, 30hz, 25hz, 16hz... I am curious how we calculate them.. Thank you..
I've used the free AADE filter design software before with good results, but never tried it for crystal filters though. Might be worth a look......
Thanks Brian. I think I may have used that (or something similar) in the past. Whatever I did it was wrong (my end, not the instructions/tool) and the filter wasn't quite right.
I also have a NWT500 Sweep analyser from China, absolutely great for checking filters. Much easier than plotting it on graph paper!
That's interesting Brian. I might have to look at getting one of those :o)
Charlie Morris, great work! why not a poor ham scalar network analyzer? Google PHSNA!, you already known everything about Arduino and si5351, only need yo buy a lm8307 RF power detector and voila.. plot the numbers from Arduino USB serial info
Funny you should say that Alejandro as I made a antenna analyzer a while back using an Arduino, screen and an AD9850. It scan a user defined spectrum and plots the antenna SWR. I recently modified it to test filters and plot the response. It works reasonably well. Might have to dust off that project and do some more work it!
Charlie
Where did you get the crystals from? I've been looking for a source the order. I'm in the United States. Thank you for your videos. I like how how you explained the circuits and show the math behind it.
Hi. If I recall they came form AliExpress. You get what you pay for, but they were very cost effective.
would earthing the crystal cans improve the filter great video thanks
I have done both, and I must admit I didn't see much of a change. I have earthed them in more recent rigs. Just be careful not to over heat them.
Hey Charlie,
Just curious, for measuring the approximate frequency of the crystals, why did you elect to use the SDR instead of the frequency counter?
Hi. At the time that was the best option. Not perfect, but good enough to get a relative measurement.
A few questions: what model number is your sig gen and are you happy with it? I have a scope arriving later this week and now need a sig gen. Is passband ripple affected by terminating resistance? I got fed up trying to use the RTL-SDR dongle + noise generator to characterize my crystal filter and I brought the filter to work and used a VNA to measure it. I had 3 dB of ripple over the passband which seemed excessive but the design was supposed to produce minimal ripple. I didn't bother to properly terminate either end of the filter and directly connected to the VNA, hence 50 ohms. I seem to have a firm grasp on how to vary the bandwidth of my filter, but not the ripple.
Gidday. The sig gen is a generic FY3200S model from any number of online sources (Amazon, AliExpress etc). For what I paid for it, I think it works well. You cannot set the output to absolute zero (even though the display says so) and the case is not shielded which means the output signals leaks. Not really a biggy though. I did open it up and desoldered the speaker that clicked for every button selection and during rotary encoder rotation. It was quite load and a bit annoying. Having a frequency sweep function would be nice, but not necessary.
As for the filters, yes, the terminating resistance did have an impact on the passband ripple. I tried several values and settled on the one that resulted in the least ripple. I have had success feeding white noise into a mic and using the RTL-SDR to monitor the filter output. You can really see the passband and how well it's performing. The plan is to build a VNA after this. Justing waiting on the log amp to arrive.
I have two questions: which kidle of oscillator do you prefer: colpits or hartley?
Why didn't you put a coil as a tank circuit in this oscillator?
73:)
Hi Emerson. I only tend to use the Si5351 for my radios. In the past I have tried to use Colpits oscillators. This oscillator is crystal controlled (as oppsed to a non-crystal controlled LC osc), and is using the split capacitors to provide positive feedback (i.e., feed back to the base an in-phase signal, hence why is was taken from the emitter) in order to meet Barkhausen's criterion.
Hi Charlie,
That article you based the design on, could that be downloaded somewhere? If not, could you place a copy on your website. Thank you in advance, 73
I'll have to take a look. In the meantime, check out N6QW's blog as he covers it there too.
Hey charlie! I reached out to you awhile back asking if you'd be willing to give me some help and advice on a dual conversion superhet I'm working on at the moment. I'm having some problems understanding a certain aspect of it, and from what I've seen in your videos you have an excellent knowledge on the topic. Last time you told me your email was on your QRZ but since I don't have my license yet I'm unable to view your profile for your email.. I can provide you my email here if you'd be willing to shoot me an email and I can ask you some questions! Thanks in advance for reading this, have a great day! 73s
"nice tight skirts"! 🤣
Thanks Lawrence. It certainly worked well.
4:02 so up on the screen here we have "Max Headroom"! 😅
czcams.com/video/2LlYs7WbezI/video.html
he talks funny
Yi1hxh rafat 73 from baghdad
Lol it s like guitar tuning
Love your videos! Very well explained. Will definitely try some of these things myself. ON4CKM
Thanks for the feedback. I'm glad you like them. Charlie ZL2CTM
Feedback? Oof, man do I hate feedback, most annoying sound ever that means some noobasaurus rex turned the mic gain up WAY too loud!😏 I forget who's video it was, but have you seen how you can pop open the crystal case and make the frequency higher with a "light" amount of sanding on the quartz?
Tesla "black box"