Phil VK2BDF from the Manly Warringah HAM Radio Society shows off his 1970's vintage home made 200MHz DIY spectrum analyser.
Forum: http://www.eevblog.com/forum/blog/eevblog-575-diy-1970s-spectrum-analyser/'>http://www.eevblog.com/forum/blog/eevblog-575-diy-1970s-spectrum-analyser/
EEVblog Main Web Site:
http://www.eevblog.com
EEVblog Amazon Store:
http://astore.amazon.com/eevblogstore-20
Donations:
http://www.eevblog.com/donations/
Projects:
http://www.eevblog.com/projects/
Electronics Info Wiki:
http://www.eevblog.com/wiki/
Forum: http://www.eevblog.com/forum/blog/eevblog-575-diy-1970s-spectrum-analyser/'>http://www.eevblog.com/forum/blog/eevblog-575-diy-1970s-spectrum-analyser/
EEVblog Main Web Site:
http://www.eevblog.com
EEVblog Amazon Store:
http://astore.amazon.com/eevblogstore-20
Donations:
http://www.eevblog.com/donations/
Projects:
http://www.eevblog.com/projects/
Electronics Info Wiki:
http://www.eevblog.com/wiki/
All right, I'm here with Uh Phil and your what's your call sign bk2 BDF K2 BDF Excellent and you're going to show us a really neat homebuilt Uh Spectrum analyzer SP analyzer. Tell us all about it! Very, very long time ago when the Radio Club was first building the 2 m repeater, we were having terrible trouble getting the cavities working right and we had very limited access to Uh RF test equipment. Mhm I worked in a RF laboratory and had access to a spectrum analyzer and I just wish I had one at home. So I decided there was no way I was ever going to be able to afford a Hiller Packard Spectrum analyzer in those days late 7s, early 80s.
So I decided to build one. Excellent, All right and this and this is the result and this is it. Wow, This this device has been used to get the cavities working for the Uh, the original 2m repeater for the Radio Club M and several times since. Uh for EMC testing of various pie of commercial equipment, right? Uh, probably the the most recent and valuable was uh, the traffic line controller I've been working on for many years had a a leakage of 160 Meg signal out of out of the microprocessors right? and you were able to track that down.
We took the we took the TR the Trav light controller up to Col to the open air test site. Yep, they failed it and they said that Signal's got to go down about 10 DBS Yep, I Bought it home and set up the Trev light controller in my carport. Used a spectrum analyzer with a dipole antenna MH and went through board by board till I found where the problem was. nice and fixed it.
Went back to Col and got it past. All right? Uh yeah, it's it's great. Um, all right so tell us how it, uh, tell us explain all the sections in here and tell us how a home Brw Spectrum analyzer works well. A Spectrum Aner is really just a radio receiver mhm.
where the the last detector stage is a logarithmic amplifier instead of a linear voice detector. M It's uh, it's very much like an AM receiver. There's uh, there's no, uh, there's no gain controls. it's just runs flat out all the time.
But the the last stage, the detector which is this last section here yep is a logarithmic amplifier right? that just takes a 10.7 Meg signal and rectifies it like an AM detector. But it's set up so that the the output of the detector is logarithmically proportional to the input, right? So it's it has we working out before. It has about 70 DBS of gain. MH So 60 DBS is a th000 times in voltage.
another 10 is another three. So about 3,00 times in voltage and that equates to the height of my stren. MH So these these signals here that that's about 40 DBS down from from the top of the screen. So the noise floor is What minus 70 or thereabouts is it? yeah, about about 70? Yep.
sure. where the other signals are and what? Uh, sweep range are we looking at there at the moment? That's that's the zero. Mark All right. Okay, up to 200 Megs 200 Megs So this is a 200 about it.
it tops out at about 200 Megs right? Some somewhere here will be the paging transfer. There's there's somebody transmitting right? somebody around us. Yeah, yeah. everyone's got a transmitter around here. Like on his belt there. who's got a handheld? Can you care up your handheld? Can you hand? Can you just te up your handheld? it? Any pregancy here? We go. here. we go.
Oh there we go. Look at that. Wow. You've got to be kidding.
You're allow on the air with thats thanks. Ben All right, it might have momentarily overloaded the front end. right? Anyway, that was. that was him there.
y yep so that's 146 Megs about there somewhere. that'll that'll be the 14 Meg paging transmitters and I'm not sure where 88 to8 gone. There was a that's that's all the FM broadcast transmitters in there. It's on narrow.
that's why. See there they are all there. Ah there they are. They're the FM transmitters.
Yep, they all the local FM stations 88 to 108. M Yep, that's them. Spot on. So can you tell us out all the sections starting from the antenna input well, antenna input yep comes down into front end mixer.
Yep. Now the mixer takes the RF signal, mixes it with the voltage controlled oscillator MH that that produces a 25 Meg If right? This this receiver actually converts up rather than converts down like a traditional receiver. This can only work this way right? Okay you because we're trying to tune everywhere from from zero frequency to 200 M You've got to take that and convert it up to and converted up to an if that's higher than than the signals you're trying to receive. So the the the first filter here m is it's somewhere around 205 Megs right? That comes down into homemade double balance diode mixer that does look like a double balance mixer.
You can see the four diodes in there there and the little toids and the little toids cuz that's all that's in a double balanced mixer. Yep. So this section here is just a crystal illat and multiplier Y and that that's mixing that 200 Migs down to about 50 migs, right? The frequencies are not terribly important and A based based on what crystals I had in my junk box at the time. Okay, the whole thing works that way.
So just a a filter and then a a band pass filter at Mhm I'll say 56 Megs right and then another Crystal oscillator. another double balance mixer Mhm and that that mixes down to 10.7 migs out through. uh yeah through that coax into into here Mhm and this is just a a 10.7 Meg if amplifier right? Um, ceramic band pass filter. There it is.
Yeah then it it comes through into this section and then it's two ways with with a Diad switch it either goes straight through for for wide band, wide band being 250 khz yep or it goes across through those Crystal filters for narrow band which is 15 khz and that's the switch on the front. that's this one here. Narrow, wide, wide and narrow band. Yep uh then just into some more amplifiers which then feeds back into that logarithmic amplifier at the end. Yep and that's your final out. The output of the logi Aire comes out into the crow. yep as the as the vertical input or video input to the crow the the oscillator is actually being swept from the soloscope. the horizontal time base oh the hor.
you're taking the horizontal time base out out ah fed back into ah into this amplifier nice and that that drives the volage controled oscillator. Got it so that so that it can synchronize with the frequency is actually being swep swept on the with with the very clever, very clever. but that's how spect. Yeah.
But to integrate it with a scope like that is yeah is neat. but yeah, I could afford to buy the scope but it's not as spe to analy it got it. And these are all hand wound inductors. Yeah, little hand wound chokes.
yeah but yeah, the things full of handwound coils. Yep, nice. It's all built on. The base is a piece of little print circuit board Y and then copper and copper.
Shield sh in around so it in on on top of it. Y and you don't need to what? Shield the tops of them H Well, well okay, you've got shielded plates. Okay, y the lids, not not all the boxes ever got. Shield Okay, so the performance is still fairly adequate, even without the shield.
Yeah, we we were actually looking at a signal before. Oh yeah, there we go that that is both that one and that one. That's that's probably the Crystal and that's yep. that looks like feeding back.
So there there's there's picking up near the crystal. Yep, so that. but anyway. they.
they look like they're double the frequency to me. So that's the that's the second local oscillator being being picked up by the system itself. Got it? But if you if you disconnect the antenna completely and put the shields on, that's your noise floor. Yeah, that's not there.
There's not. There's not much left. Yeah. and you know that could actually be a signal.
It could be sneaking in. Yeah, yeah, there's a. There's a couple down here, but yeah, there's there's not much there until you put the antenna on. Sure, Very nice when I when I connect the back analyzer up to my beam I'm down at Beacon Hill Yep, down in the Gully I can I can get the receiver about S9 m I put the Spectrum analyzer on the beam, key up the repeater and actually see the the repeater.
come back. you know, just pops up here as a yep, it's just popping out of the noise. but got it. That's that's the sort of sensitivity that the device has.
Mhm it. It doesn't go down to you microvolts like a normal hand radio. W um not not sure what the yeah what the smaller signal is but yeah, 10 or 20 mic volt. probably something like that.
so would you if you were building a spectrum anal a discreet Spectrum analyzer. Would you do it the same now? or would you similar style? Pretty much yeah. similar components when I when I first looked at it, the voltage control oscillator was always going to be a problem right? and I I rang up um I won't say who. Okay the place up at Hornsby that sells all the fancy RF Equipment from mini circuits and I priced What's called a yig oscillator? Yig! now it's an acronym for something and I can't remember what it is. A y oscillator is a magnetically tuned sphere oscillator. nice and it's the heart of a real Spectrum analyzer. M and I wanted one that tuned yeah, 2,000 to 3,000 Megs Yep and it was $25,000 $25,000 This was back when. Uh long, early early 80s, late 70s, early ' 80s? Yep, no, there's just no way I was going to spend that much money on it.
A little oscillator in a box? Well nowadays everyone's doing it with the software defined radio these days. It's just not the same Is it? Yeah, it's not. It's not the same. There was there was an spectum analyzer kits that came out years after this where they actually used a TV tuner I remember that electronically tuned TV CH not not turret Chers y that they they had enough range that yeah, you could build a half reasonable spect analizer out of a okay a TV tun right? But would it have a decent noise floor in it? Well you had to build all the back end of it right? Okay, yeah, but yeah, $700 you can buy spectr analyzer.
The work better than this of course of course. Yep. but still that is very nice. Um sort of.
uh, dead bug style construction or as the Yanks call it man Manhattan style construction and uh, it's just beautiful Phil That is awesome. Well done. thank you very much. You're welcome.
Um.
This guy is a BAMF
he reminds me of an old engineer I used to work with and really admire
This is classic D.I.Y. Excellent!!!!
Amazing! Only
golden content – super cool
Hello from Frank, KB2VNG
Nice 🔧😀
I was wondering where my man cards went, but now I see that Phil has them all.
Thanks for helping me solve that mystery!
There are a lot of things to learn from Phil VK2BDF. Universities should employ creators like Phil to inspire students… Pure respect!
What a very educational interview love it Dave, Its like 1970 porn ooyeah bit rough around the edges but does the job so cool . Love your interviews Dave keep it up.
Is this guy smart!!!!!
Wow Phil !
That's a very interesting and impressive build , nice to see what one can achieve if you put your your brains to work.
Greetings from Holland
Where can I get used spectrum analyzers? I need one!! Anyone??Any ideas?
The prototypes made in HP's lab probably never looked better than this.
So…. Where can i learn this stuff?
Excellent job! Thank you Dave for sharing with us.
Best regards to you and Vy 73 to Phil de CT7AFR, Emmanuel.
A true hacker!
Has he ever come up with a comercial version of this?? i'm certainly interested!
Really clever.
Very nice. I even decided to build an analog one like this. Just using horizontal trigger to synchronize signal to a oscilloscope. However, I just cannot find a good analog bandpass filter schematic to have enough resolution for audio band, which should be accurate to 100Hz. For RF band, it is okey to do that. I am still learning that.
That was very impressive.
cool, love guys like that.
Just awesome