This is a cool circuit idea! Thanks for sharing it. I have salvaged numerous CB radios over the years, and used them in all kinds of "cool" projects. One idea that I have had kicking around in the back of my mind, is a spectrum analyzer! Most CB's have a very narrow IF, and are dual-conversion. (10.7 Mhz 1'st IF down to 455Khz 2'nd IF)

They already include a nice VCO (controlled by the PLL chip) which can EASILY be modified to run via an analog control voltage. (I have, in fact, used this method to convert several CB's into receivers for the following ham bands: 75, 20, 10 Meters. I have also made several into AM Broadcast receivers. In all cases, they work extremely well.

Injecting a sawtooth into the VCO would sweep the frequency, and the IF will be the filter. Your circuit has inspired me to look into this a bit further. :) I have even thought of using/making a triple-conversion system, with the 1'st IF being somewhere in the 110Mhz range. The advantage would be the ability to sweep the entire HF spectrum with the LO only having to sweep from ~110.1Mhz to 140!

Hi nice project, I like the simplicity.

Could you post more info on the discrete version instead of the 555?

some schematic maybe?

Thanks

Hello, I am fans of your site. Some day if I have time I will build your 555 Spectrum Analyzer. Please upload a better ( whole ) schematic diagram, and L1 - L3 coil construction explanation, I am beginners in electronics. Thanks.

Randy,

Glad you like my little 555 SA project. :)

The mixer source resistor is indeed 1.5 kR, it isn't too critical, if you use different JFETs you will probably need to change it a bit.

A blaking signal can be taken from the pin-3 output of the timebase 555, it is driven low during the retrace, so something as simple as a diode from there to the Y-output (after the unlabelled - 1 kR - resistor) to pull it down might do the trick... However the diode has a finite forward drop, so it won't quite pull it right down to zero volts - probably not a major problem. If your CRO has a Z-input you might also use this signal to drive it, but you'll probably need an inverter/level-shifter to match what the CRO wants.

Try it first without the retrace blanking, you may not need it, especially if you are just triggering on the X output, rather than using XY-mode.

Regards,

Alan

Alan

Thanks for developing such a interesting and useful project. In the process of breadboarding your design I'm having difficulty determing the J310 source resistor value in the mixer stage (1K5 ? I think). Also if I might gain some additional insight regarding adding a blanking

transistor to the Y-output stage.

Thanks and Regards

Randy Lauderbaugh

What an absolutely fascinating project!

I was teaching a group of my students at the University in Jakarta, before returning to the UK. We were looking at Radio Astronomy.

I was using an ATTEN AT5011 commercial analizer to monitor the 17 to 22 Mhz section of the shortwave band where there are various frequencies used in Radio Astronomy for noise and interference. I found the Youtube video by accident!

I was somewhat cynical at the start, so curious, we built one over a few days. The results are really impressive to say the least. It does everything the students need to see in this band. Within a couple of weeks, there were 5 or 6 versions on the go. One enterprising lad, has hooked up a TV tuner that covers 48 to 800 Mhz, the spectrum analiser is now a fixed IF and the ramp sweeps the tuner, a switch and relay switch this out to restore the analizer to the HF end of the spectrum.

Ive built the poor mans spectrum analiser and thought that was as simple as you can get. I dont know about winning an award for the most novel use of a 555, this certainly has to reap an award for the most simple (and useful) SA ever!

As for a computer interface, why not keep things simple like the analizer? Use a 12C Pic Micro, a small 8 pin device, use the A/D to sample, and the RS232 to feed the data back to the PC, food for thought maybe?

Keep up the good work

Dayle,

There was an entry in the 555 contest for a similar hacked-receiver spectrum analyser in addition to mine. The marker system you describe is a great idea. You really miss that, even with excellent selectivity and a comb source it is easy to get lost.

One of these days I'll probably sit down and build a SA using an MCU and a superregen log detector, probably the xtal one for the selectivity. I've been toying with the idea for years now, and I've always wanted to make a web-based HF-watcher that collected 2D spectrograms of HF.

Regards,

Alan

Davy, thanks mate, there is an RSS feed you can follow for my projects. I've been a bit slow of late, busy at work etc. Should be back to posting stuff more regularly soon.

Thank you Rubens. :)

Alan,

What a very interesting project! I am wondering was the goal to design a simple SA, or to prove the log detector idea?

You may be interested in my SIX component SA... with a little cheating of course.

A sweeper at 3 Hz was made from a unijunction and 2 resistors, 1 capacitor, the ramp then fed into the Vtune line of a preloved Sangean world band radio, the existing Vtune VCO line to the PLL PD being cut. This sweeps the radio from 150Kc to 40 odd megs. The scope vert input is connected to the AGC line, the time constant here reduced by judicious removal of the filter caps. The scope ext trigger is connected to B2 of the unijunction initiating a synchronised display.

Here is the clever bit.

The "pip" produced at the "lock" output of the PLL PD, generally used as an audio mute, as the VCO sweeps through the capture range of the PLL coincides with whatever frequency is programmed into the radio via the keyboard. This "pip" signal is interfaced with a transistor and resistor collector load... 2 more components and ANDed with the AGC output, giving an accurate frequency marker on the display.

So, what do we have?

A dual conversion spectrum analyser, bandwidth of around 8 Kcs, reasonable image rejection, with accurate programmable frequency markers, with a sensitivity of a microvolt or two using just 6 added components!

Improvements:

A digitally programmed D/A would linearlise the display, offsetting the varicap curve, the same MPU could also reprogram the radio keyboard ahead of the sweep, to allow any number of frequency markers on the display. A log detector and accurate RF attenuator would make a fairly high quality and useful instrument.

Thank you for such an interesting and thoughtful design.

Hello!

Very very very nice project!

I'm a French fan of your blog Alan...

Please continue to share your project with us !

73's

Davy

Alan, it's a lecture on ingenuity. Congratulations. Rubens/VK5FE

Andrew,

Nice website you have there.

The 7555 is indeed an Intersil ICM7555. I used the CMOS version of the 555 here to minimise the switching transients, as it is operating at ~40 kHz. I have no doubt a normal NE555 would work just fine, but you might need a larger capacitor across the rails near the chip.

The varactor diodes are BB521 (roughly 2-12 pF) units. I got them from Rockby some time ago. They are speced to low UHF, probably for TV tuner use, and are relatively linear.

The unlabelled resistor is 1K, just to protect the op-amp from accidental shorts/drives external to the circuit.

Regards,

Alan

Very tidy, elegant and original piece of work, Alan. Two questions: First, is the chip you noted on the schematic as a "7555" actually a regular 555? The two 555s you used look the same in the photo, and 7555 chips are somewhat harder to find than the regular garden-variety 555. Secondly, what did you use for the varicap? Obviously not critical, but just nice to know which one you used.

Drone,

I too saw that posted in the comments on N9ZIA's rather impressive SA build that made Hack A Day this morning.

I wonder how effective that impulse integration method is in compensating the varactor and tuning characteristics to achieve linear tuning?

Regards,

Alan

Link to an older (73 Feb 1990) SA article using a 555 for sweep signal generation with access to an existing receiver IF (.pdf):

Arv,

Thanks mate, I am quite proud of it. I think it is one of my more useful hacks in a while...

In particular I think it is a great project for people that want to cheaply build an SA for fun and the learning experience. Maybe they want to learn how SAs work? Maybe they are thinking of investing in a commercial one and aren't sure if they would use it enough? Maybe they just can't afford one at all? It teaches you the basics, and its limitations are there for you to learn about and improve upon. It isn't a complete toy and is easy to extend with a tracking generator, marker, noise source, step attenuator, etc. Simply it is a true hackers minimal SA project.

Many of us have the parts in their junkbox and could dead-bug this thing in a few hours. It doesn't use anything hard or expensive to get. No obsolete or expensive ICs, just vanilla multi-sourced parts. (The J310s can be replaced with MPF102 or BF245 or any RF N-JFET really). The front-end filter inductors and source choke in the detector are the only slightly exotic things using commercial toroidal cores, but they can be solenoidal-wound by hand to avoid that. A commercial 10 uH choke will likely work just fine for the detector source choke.

If there was enough interest I'd consider designing a PCB and kitting it. Publication wise, as always, just email me and we can work something out.

Regards,

Alan

Well done Alan !

Because of the high cost of conventional spectrum analyzers and the apparent simplicity & effectiveness of this project, your efforts should be popular reading, and may start a series of ham-built simple spectrum analyzers. This deserves further attention, development, and publication.

Thanks for showing the rest of us how to do it.