Call Search

New to Ham Radio?
My Profile

Friends Remembered
Survey Question

DX Cluster Spots

Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement

donate to eham
   Home   Help Search  
Pages: [1] 2 Next   Go Down
Author Topic: homebrew receiver...  (Read 9270 times)

Posts: 773

« on: January 29, 2008, 09:47:09 AM »

I am working on building a homebrew CW rig, using tubes, and have a basic receiver design from my previous two homebrew receivers designed for hi fidelity AM reception.
Those receivers work real well, but dont have a product detector.
So I am looking for idea's for the product detector (tube type) for CW, and the agc circuit.
Most handbooks used an audio derived agc setup, how do they work?

I am looking at the 1967 ARRL handbook, HB-67 receiver for the product detector and agc setup, but want to know if there was a better design.

Thanks for any idea's.

Posts: 41

« Reply #1 on: January 30, 2008, 05:00:31 PM »

If you can gest a 7360 tube at a good price they make excellent produt detectors.

Common 12AU7 dual triodes can make good productors also.
You might want to down load the schematic of the famous Collins Radio 75-S3B from the Collins Collectors websight and steal their product detector.

Most RX's use RF derived AGC...that this is is amplification of a sample of the IF signal....turning it into a DC voltage with some form of rectifier andfeeding back  the DC control voltageto maybe the RF amplifier and/or IF amps to get a fairly steady output from the speaker.  

Audio deriver AGC is similar except audio from the detector is turned into a DC control voltage/

Any questions  ??  just ask

Clark Fishman  WA2UNN

Posts: 17483

« Reply #2 on: January 31, 2008, 08:04:30 PM »

One problem with using a BFO and AGC at the same time is that you
need to make sure the BFO energy doesn't get into the AGC system and
cause it to reduce the gain.  This is one reason why many receivers use
audio derived AGC, but you can also do it by extracting the AGC from
the first IF with the BFO at the second IF.

An excellent source of interesting tube circuits is the book, "Amateur
Radio Techniques" from the RSGB (by Pat Hawker, G3VA, who has been
editing the "Technical Topics" column in their monthly magazine for
almost 50 years!)  The latest edition is the 7th which came out in 1980
so it is out of print but you may be able to find a used copy somewhere.  
By 1980 many of the circuits used transistors and integrated
circuits, but valves still held their own, too.  While this includes several
7360 beam-deflection product detectors, there are also some twin-diode
versions, regenerative detectors, etc.  By the late 1960's, the
twin-diode product detector using a pair of 1N34 diodes appeared
in otherwise valved receivers, though you could use a 6H6 instead (or
even a 5Y3 if you want to be perverse).  Remember that a product
detector is simply a mixer, so other sorts of mixer circuits can be tried.

More recently the Technical Topics columns have been reprinted in
5 year groups as the "Technical Topics Scrapbooks".  While there aren't
as many tube circuits, they do still appear, including some of clandestine
radio equipment from WWII.   You might check some of the other
RSGB publications, as they tended to publishing valve circuits for a while
after the ARRL had mostly gone to solid state.  I have a copy of their
"VHF/UHF Manual" that has some interesting circuits in addition to
those in their handbook of some years back.

Oh, and you might browse JF1OZL's web site - he has some tube
receiver circuits there, as does Harry's Homebrew Homepage.

Have fun!

Posts: 773

« Reply #3 on: February 01, 2008, 06:54:46 AM »

I have lots of designs, there are some good ones
in old RSGB handbooks, some which work for am and ssb/cw,
and allow the use of AGC on both, separate IF stage to make the AGC,
audio derived circuits, dual triode circuits, solid state diode setups,
but what I was wondering is there any reason to use one over another
in practical terms for a CW receiver.

I want AGC and an S meter, and a nice sounding receiver, and wont know
if any designs work better then others unless I build a prototype I guess.
Maybe start with the simple easy designs first and try them?

I am also thinking about the BFO. The best thing seems to be a xtal osc
with 2 xtals selected by a switch, using 454 and 456 KHz xtals,
where would I get them?!

I found 455 KHz xtals, I use them in the current receivers without
product detectors, but 454 and 456 might be hard to find!

That or make a tunable bfo, but it is unlikely to be as stable
as the xtal setup.  I am not sure if I could build one stable enough
for a good cw receiver.

Well, this is the fun part of homebrew, and where you learn a lot,
that and the parts layout and so on....



Posts: 1562

« Reply #4 on: February 01, 2008, 05:37:26 PM »

6AR8 a sheet beam tube. Very findable at lower prices than 7360 and similar to 7360 in application.

Note using either requires a steel shield and carful placement away from transformers or other magnetic fields.


Posts: 17483

« Reply #5 on: February 01, 2008, 11:01:07 PM »

The 6JH8 is another beam deflection tube similar to the 7360 that may
be easier to find.

Choice of circuits often comes down to the parts you have available,
especially the tube types.  A dual-triode mixer can be built with a
12AU7, which is much more commonly available than the 7360.  If
you are trying to run the AGC from the IF stage along with the BFO
then a balanced circuit of some sort may help to reduce feedthrough
into the AGC amplifier.   (The common solution was to use disable the
AGC and use manual gain control on CW when the BFO was on.)

Regarding the BFO, if you are using 455kHz then an LC oscillator
probably has sufficient stability.  You can put a trimmer on the front
panel to adjust the BFO frequency above or below the IF.  Or you
can try one of the 455 ceramic resonators instead of a crystal - they
are cheaper unless you have the crystal on hand.  If you aren't planning
to receive SSB then you only need a crystal on one side of the passband.

Presuming you have most of a receiver already with an AM detector, it
shouldn't be hard to breadboard a BFO/product detector to go with it.
First I'd make sure that there is a good range of gain control available
for the AGC to drive.

There are circuits using a triode/pentode or a pentagrid converter
where the BFO and product detector can share a single tube, but if
you are going to experiment then it may be better to build the BFO as
a separate unit so it can drive various product detectors.   You can
build the AGC system into the AM receiver without the BFO or product
detector to make sure that works.  (Commonly the AGC voltage is
sampled off of the AM detector stage, and the BFO voltage is simply
applied to the diode detector.  This works, in that you can hear the
CW signals, but the BFO does mess up the AGC.)

If you are using audio derived AGC then full wave rectification is a
good idea, as it responds faster (especially to low audio frequencies.)
That probably is the easiest thing to add to an existing CW receiver to
avoid the BFO leakage problem.

In the end, it doesn't matter a lot which circuits you choose.  I'd suggest
something fairly simple for which you have the parts (or can get them
easily.)  Be sure to read the sections in the handbooks where they
discuss AGC, BFOs and product detection:  they should point out some
of the issues with each circuit.  And sometimes it is fun to choose a
circuit just because it is out of the ordinary!

Posts: 166

« Reply #6 on: February 02, 2008, 09:36:37 AM »

AM receivers tended to have a lot of gain so that the diode detector was operating linearlty with several volts of IF 9 often as much as 10 to 20) applied to it. If you use a pentode as BFO, with a Hartley circuit and the screen grounded as far as RF is concerned, you can get adequate conversion gain by applying the IF signal to the suppressor grid, and taking the AF from the plate. In many cases, the IF signal level is so high that it needs coupling into the suppressor through a small (2-5pF) capacitor. As a result, the BFO leakage back to the AGC detector is insufficient to produce AGC from the BFO signal, so you get the advantages of RF derived AGC (faster attack, less modulation ripple). This is just making the BFO into a self oscillating mixer.

Posts: 773

« Reply #7 on: February 03, 2008, 04:43:12 PM »

The receiver will be based on a design I built for my exsisting AM receiver, which worked so well I sold off all the other stuff (R390A, SX17, scott slrm, etc.).

I am thinking about an additional IF amp running into a rectifier for AGC.

For the product detector, a twin triode setup might work well, the BFO will likely be an L/C osc instead of hard to get xtals 700hz off 455 KHz.

My current homebrew receivers use a digital readout from almost all digital electronics, a 6be6 mixer, 6c4 osc tube, a kiwa filter board, into 2 stages (6ba6) of IF amplification, into a 6bj8 low distortion AM detector with agc.
Also on board is a 12au7 s meter amp, and a 6bh6 455KHz xtal osc with variable output (for zero beating AM signals).
The antenna input is made of high Q B+W coil stock, right into the mixer. The mixer and both IF amps get AGC.

Its a stable receiver, the digital dispaly is very accurite, noise is incredibly low, fidelity is very high.
It should make a good CW receiver if I can find a good narrow 455 KHz filter.



Posts: 340


« Reply #8 on: February 08, 2008, 12:13:50 PM »

The 7360 is indeed a wonderful product detector tube...but they ARE expensive these days....on the order of $60, last I checked.  But it should last a lifetime.

YOu need to be careful about the output should be bifilar wound and very carefully constructed.

Incidentally, my first encounter with a 7360 was as an FM Stereo demodulator!  I think it was a popular Electronics article that had a project when FM stereo was new...(sorta shows my age!)


Posts: 773

« Reply #9 on: February 11, 2008, 06:53:20 AM »

I built a product detector using a dual triode, 12au7, and tacked it onto the homebrew receiver.
That circuit has the signals combined in the common cathode circuit.

It worked, but....

The bfo level seemed very critical, maybe its just because the bfo is unshielded and is running 455KHz, the same as the IF frequency.
CW sounded strange at low frequency beat notes, like it was 60 hz modulated, could this be because the bfo frequency is exactly 455 KHz and not offset some?
I may be able to pull the xtal freq with a cap to test this, its not ac hum as things sound good at normal pitches, ssb sounds good, and the DC is pure.

I may pitch the idea of IF based AGC though, the bfo gets into the IF chain easy, and I dont want any loss of sensitivity from bfo bleedthrough, but would need to test with a well shielded BFO.

The next experiment (after pulling the xtal off 455KHz) is the two diode mixer setup, I want ot see how well that works, and how critical the signal levels are.



Posts: 773

« Reply #10 on: February 12, 2008, 06:09:06 AM »

I pulled the bfo xtal off 455khz with a cap, and still had the rough sounding note on low frequencys, but found that was because of where I was taking the IF from.
The circuit works well but is more sensitive to BFO signal levels than I want.

I also tried the two diode mixer setup last night, using various types of diodes.
It worked like crap, the dual triode worked so much better. I tried all sorts of different ssignal levels of the IF and BFO, it was more touchy than the dual triode setup, more noise, lots less gain.
My next experiment is a mixer based setup using a 6be6.

Fun with radio!


Posts: 20

« Reply #11 on: February 12, 2008, 02:08:08 PM »

Ive been using fet's to build a receiver here, and its going great. I use to fet's in cascade to simulate mosfets because they are getting too expensive.

What I have found is a fet product detector works great and an LC bfo at 455 works just fine with a buffer after it.

To keep RF out of the AGC amp, pick the AGC off from the middle of a two stage IF amp. The signal is pretty clean there and works like a charm.

I am trying to find a decent SSB filter without relying on a mechanical. Ive tried usind ceramic filters in a ladder arrangement, but no luck yet...Good luck on your homebrew projects!

Posts: 773

« Reply #12 on: February 12, 2008, 02:54:38 PM »

For good filters, try Kiwa

I use their standard filter modules in the homebrew receiver (4.5 and 5.5 Kc) and they work like mechanical filters, but have no loss.

The price was right at $50.00 and you can get a 2 filter switch.
You used to be able to pick the bandwidth you want, but I dont know what they can make any more...

I need to find good CW filters at 455 khz, say 1000 hz...


Posts: 773

« Reply #13 on: February 12, 2008, 07:05:32 PM »

The 6be6 mixer works a treat!
As a matter of fact, even with the thing tacked together and on the operating table, hooked up with clip leads, it works well enough to use.
The bfo is not shielded and is in the radio, but very very little gets into the IF chain, even unshielded and with long jury rigged wires, I can use the AGC off the AM detector, the S meter works, with no antenna hooked up the receiver is silent, and the S meter only reads S1 from the bfo!

It would make a great CW receiver if I had a narrow filter for it.

The AGC actualy seems to work great on CW, the volume is quite steady, static crashes dont wipe things out, sensitivity is very good.

The 6be6 mixer seems to work the same with a wide range of BFO input signal, from a little, to a lot, and even a very little works fine on small signals or large.

Now to find a narrow filter....


Posts: 20

« Reply #14 on: February 13, 2008, 01:07:01 PM »

Im having the same awesome performance as you. I use a MPF102 audio preamp driving an LM380 power amp. I have the 10k pot between the two.

With it half way open, you cant hear any noise from the speaker, but then when you attatch an antenna, it blows you out of the room and you have to barely crack the volume.

Im using a product detector "mosfet style" with two fet's. It works also as a high sensitivity, low noise AM detector by simply killing the bfo voltage and hooking a 10uf bypass cap to ground on the source leg. To get product detector, simply energize the BFO and disconnect the capacitor lead from the source.

I do have seperate amps for the s-meter and AGC. This way, one doesnt load the other.

Its been a process of trial and error and I believe my receiver will outperform and sound better than many a kilo-buck rig.

You seem to be having fun with those tubes, maybe my next receiver will follow in your footsteps with hollow state. Im sure the performance will be even better.

Now to look into those filters...
Pages: [1] 2 Next   Go Up
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!