Pages: 1 [2] 3 4 5   Go Down

Author Topic: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build  (Read 107148 times)

G3EDM

  • Member
  • Posts: 1456
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #15 on: June 08, 2013, 06:23:55 AM »

Prior to heading to the beach for the rest of this rainy weekend, I wired the two potentiometers that adjust regeneration. Surprise! I found a mistake in the ARRL instructions. I've now updated Chapter Five to include this detour, and the testing that I did to make absolutely sure that ARRL was wrong and not me, the novice ham.... Same link as before and if you've already seen the pictures of this problem, no need to go here again. If you haven't, here it is again (there is a series of new photos at the end of the gallery:

http://tinyurl.com/mubpkwk

73 de Martin, KB1WSY
Logged

G3EDM

  • Member
  • Posts: 1456
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #16 on: June 10, 2013, 05:47:42 AM »

The receiver's finished, and it works! Chapter Six, "Completion and Testing," is here:

http://tinyurl.com/orqta39

My overall comments on the entire project are in a separate text post immediately after this one.

73 de Martin, KB1WSY
« Last Edit: June 10, 2013, 06:28:01 AM by KB1WSY »
Logged

G3EDM

  • Member
  • Posts: 1456
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #17 on: June 10, 2013, 05:49:45 AM »

This post is long. I hope you won't mind. It's good to put all my post-construction observations in one place.

To recap, this project involved building "A Three-Transistor Receiver for the Beginner" from the ARRL book, "How to Become a Radio Amateur." This was the book that inspired me to get into ham radio when my parents gave it to me on my 12th birthday in 1969. Unfortunately, back then, my interest in ham radio only lasted a few years and I never got on the air although I did pass the British amateur radio exam (during this time I was living partly in the U.S. and partly in the U.K.).

Initial report on the performance of this set:

When it was first successfully powered up, there was an extremely loud AC hum in the headphones. Turning off all of the AC appliances in the shack made no difference. However, when I flicked "off" the red power switch on the (passive) APC surge protector through which all the equipment is powered, the hum vanished completely. Interesting.

To maximize sensitivity, I shorted out the "broadcast band filter" at the front end of the set. This raised sensitivity considerably and there was no problem with AM BC breakthrough, except on 160m where a very strong local Spanish station in the Boston area was audible throughout the entire rotation of the bandspread dial. So, I've installed an alligator clip lead back there and plan to short out the filter, except when tuning 160m stations.

I then tested the set for a couple of hours on 80m and 40m in the early daylight hours. For once, I was delighted to hear a group of OMs discussing the medical state of the universe in an SSB net on 75 meters! The audio is excellent and there is plenty of volume. The Jackson ball drive makes it easy to tune in these fellows and null out the "Donald Duck" effect. I would even say that the overall audio quality and ease of tuning is better than my modern Ramsey direct conversion receiver kit with its gazillion IC-mounted transistors, but this could be influenced by pride in my own radio "creation"!

Setting the regeneration is tricky, but I knew that would be the case and having both "Coarse" and "Fine" controls is excellent. You just have to get used to working both the tuning dials and the regen dials simultaneously as you tune through the bands.

On both 80 and 40, I heard plenty of CW, including both fast, experienced operators and much slower people who are sending more in my range (17 to 20wpm). I'm at the 18-character level in the Koch method, so can only partially copy these QSOs because of the gaps in my alphabet. This is providing strong motivation to assimilate the remaining characters!

Perhaps the biggest surprise is that, at least during the daylight hours which is when I did the testing, single-signal reception is entirely possible. I was expecting selectivity to be a major issue with this set but, if I choose the right time of day and the right overall conditions, I think it is capable of being used in serious QSOs. We won't find out for sure until I've built my transmitter and completed the Morse leaning process.

More general observations:

My strongest reaction after going through this intense homebrew experience is that it was an immense amount of work. That's also what made it immense *fun*: this is a hobby you can really sink your teeth into. This was coupled by amazement that, a half century ago, boys, girls, teenagers, adults, actually had the time and the motivation to build this equipment. When you see how over-scheduled many middle-class American kids are, with school and after-school activities, plus the much larger number of (electronic and other) distractions available, it is hard to imagine any but the most obsessive aspiring ham doing something like this nowadays.

Doing this project in the 21st century was a challenge. Finding the parts took nearly a year; the total cost was about $150 perhaps (I did not keep precise track). In real terms, that is actually a lot less than the $50 that the ARRL estimated for this project 45 years ago. Once I had amassed the parts and obtained a whole bunch of tools, the building phase took just over one week which was about 70 hours of my time: apart from my "day job" I did essentially nothing other than eat, build, and sleep. I was being extremely carefuly, because some of the parts in this project are virtually unobtainium nowadays and I didn't want to break them. Also, after about 40 years away from radio, I had to relearn a lot of building skills almost from scratch.

Perhaps the biggest challenge was soldering. In this kind of point-to-point wiring you often have several components leads and wires being soldered to a single terminal lug. It is very hard to get the temperature high enough to make a good joint, without also ruining the components. Also, if I am not mistaken, modern solder doesn't seem to flow as easily or get hot as fast as the old stuff (note: I am *not* using lead-free solder). I'm using a Weller soldering station set to a very hot 850F, with a relatively wide screwdriver tip, but even then it seemed to take a surprising amount of time to get the joint hot enough for the solder to flow in that shiny, bubbly way that indicates a good connection. I did ruin several low-value components (mainly resistors) in the process of building this radio but fortunately I had spares. It did get quite a lot easier as the project progressed so I think I am gaining experience in this essential skill.

As it says in the ARRL book: "The construction details that follow are not the detailed, step-by-step, take-me-by-the-hand instructions you would get if you were building a commercial kit. Kits are fine, but after building one, schematically you might not be able to tell a resistor from a capacitor. But the main reason for building your own is to learn something about radio, and in order to construct this receiver your will need to learn to read circuit diagrams."

I hope this is only the beginning of a homebrew odyssey, taking me through building a simple transmitter and antenna to make my first QSOs and going on to build a more elaborate station and making more of my own design decisions rather than slavishly following the ARRL blueprints. I strongly recommend it to anyone who hasn't tried building their own stuff. It doesn't matter what you build or what technology/era you target, it's the building that counts!

73 de Martin, KB1WSY
« Last Edit: June 10, 2013, 06:33:59 AM by KB1WSY »
Logged

W1JKA

  • Member
  • Posts: 2099
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #18 on: June 10, 2013, 07:48:27 AM »

Congrats on your successful build and excellent construction pictures,from now on you will be considered as what most of us define as a REAL ham. ;)
Logged

KB1GMX

  • Member
  • Posts: 2252
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #19 on: June 11, 2013, 04:08:48 PM »

Congrats Martin.

Yes, that BCB station is about 4mi from me (I'm a bit west of you).

Allison/KB1GMX
Logged

K5DE

  • Member
  • Posts: 6
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #20 on: June 11, 2013, 07:01:30 PM »

Martin:

Congratulations on the finished, working project.  I very much enjoyed the photos and descriptions, and the workmanship looked top notch -- I would have gone blind on those coils.  An inspiration to move forward with some ideas of my own.

Rick
K5DE 
Logged

KQ6EA

  • Member
  • Posts: 611
    • homeURL
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #21 on: June 12, 2013, 10:47:05 AM »

Beautiful!

73, Jim
Logged

JAHAM2BE

  • Member
  • Posts: 335
    • qrp-gaijin
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #22 on: June 14, 2013, 08:35:01 AM »

Very nice build. One suggestion: for maximum coil Q, the coil should be located 3 coil diameters away from the metal front panel. This information comes from an article on regenerative receivers by N1TEV, who has spent some time optimizing different regen designs.

G3EDM

  • Member
  • Posts: 1456
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #23 on: June 14, 2013, 12:44:39 PM »

One suggestion: for maximum coil Q, the coil should be located 3 coil diameters away from the metal front panel. This information comes from an article on regenerative receivers by N1TEV, who has spent some time optimizing different regen designs.

I'll bear that in mind next time I build a regen (if that ever happens) ... a little late to alter the layout on this one!

By the way, the completed radio, including the batteries and headphones, weighs 2.3 kilos (5.2 pounds). The ARRL article warns of hand capacity effects but I have found no issues with stability. The variable capacitors are not only bolted to the chassis but also secured with shaft nuts to the front panel, which is itself also bolted additionally to the chassis with 8 screws -- that is not my design, it is the original ARRL template. The Jackson dial is huge and very smooth, with no backlash, and the radio is so heavy that it's not going to shift around on the desk when you operate the controls. It's a lot of iron for a 3-transistor regen with 3ma current draw. I tried to weigh the transistors but they don't even register on my letter weighing scale!

73 de Martin, KB1WSY
Logged

G3EDM

  • Member
  • Posts: 1456
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #24 on: June 15, 2013, 06:04:10 AM »

One of the issues with homebrewing analogue equipment is the knotty question of figuring out the frequency you are listening to, or transmitting on. With this little regenerative set I haven't had too much trouble finding the approximate band edges and "neighborhoods" (DX CW, QRP CW, SSB phone and so on) within the bands simply by tuning around and listening carefully, but still, it is laborious. (Welcome to the 1950s, I guess.)

Then today it ocurred to me that a regenerative set oscillates at exactly the same frequency as the one you are listening to. So I pulled out the only digital radio I have (a little Sony SW-1 from the 1990s) and sure enough, was able to find the oscillations of the regen set immediately, and determine the tuned frequency of the regen set to within a few KHz. It more or less confirmed the band markers that I had already surmised from simply listening to the bands, but still, it's pretty nifty and much easier than with a superhet where you'd have to listen to the output from the heterodyne oscillator, then add or subtract an offset for the IF of the set.

This has got me interested in obtaining a used digital frequency counter from the same era as the stuff I am homebrewing (mid to late 1960s), solid-state but with a nixie tube display. I figure that it still fits in with my "all homebrewed station" because I can rationalize it as a piece of test equipment, rather than part of the station per se. It will also liberate me from ever having to calibrate a dial, or even from having any numbers on the dial at all. This would be both practical, and more accurate than a dial anyway. I'm a hopeless perfectionist and if I start marking up dials, I can imagine an infinite process of fine-tuning the calibration ... better to just leave the dial unmarked and use an external reference such as a frequency counter.

I'm actually thinking of getting two beat-up old nixie-tube counters, taking them apart and mounting them on a "period" hombrew chassis with aluminum panel. One of them would be for the transmit frequency, the other one for receive. I could even have a third nixie display, on the same panel, displaying UTC.

Part of my motivation for moving in this direction is that, although I have a collection of four NOS dials from the 1950s/1960s that I acquired in a (probably unwise) frenzy a year ago, I am very unimpressed by them. The National dials are stiff and the transparent fascia is not glass but very thin, buckly perspex (this is something that is not obvious from the beautiful photos of National-equipped homebrew stuff that you see on the Web). I have a big Millen dial but it is very stiff (this issue was known at the time and various solutions were peddled). I have an Eddystone 598 which is the best of the small dials: beautifully smooth and light, with just the right amount of resistance when you couple it to a capacitor. But its backplate is very thin, the dial itself has yellowed, the black crinkle paint is chipped, and there would be a lot of restoration work involved. The only one I am truly impressed with is my big Eddystone 898, which I am saving for a future "super receiver" to be built many years down the road.

OTOH I am thrilled with the Jackson Bros 4489 ball drive that I mounted on that little regen set. They are still available new, from the successor company in England, and their only "problem" is that there is no easy provision for writing/printing letters and numbers on the dial. You can turn the existing plate over and engrave on the blank back side, but there's not much room (suitable for one or two bands only). But by using a frequency counter, one could simply eliminate the need ever to mark up the dial.

73 de Martin, KB1WSY
Logged

G3EDM

  • Member
  • Posts: 1456
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #25 on: June 15, 2013, 04:12:33 PM »

Now we get to Chapter Seven: the inevitable experiments to improve the set (I have been told that a homebrew project never ends)! Today, I built a new coil to get better bandspread on 40m CW and made some unexpected discoveries along the way. Photos available here:

http://tinyurl.com/kebfdwn

73 de Martin, KB1WSY
Logged

KE4JOY

  • Member
  • Posts: 1428
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #26 on: June 15, 2013, 04:21:30 PM »

A really nice build article and superb workmanship. Brings back alot of memorys. Well done  :D
Logged

G3EDM

  • Member
  • Posts: 1456
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #27 on: June 18, 2013, 01:57:48 PM »

Am answering my own question about where to find Q-dope. My earlier searches had turned up some things on the Internet but it wasn't clear whether it was the "real McCoy." However I have now tentatively confirmed that "the" coil dope favored in the last century for radio work was made by a company called General Cement. It is still available here, and presumably elsewhere too:

http://www.suburban-electronics.com/display/10-3702/Q-Dope

Wow, $9 for 2 fluid ounces. I am ordering some and will do some testing. My local electronics store also has coil dope, but it is colored bright red. By all accounts, the "real" Q-dope has no color.

73 de Martin, KB1WSY
Logged

KQ6EA

  • Member
  • Posts: 611
    • homeURL
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #28 on: June 18, 2013, 02:02:11 PM »

Yep, the "real" Q-Dope is crystal clear.

IIRC it's just acrylic or styrene plastic in solvent.

I've got several bottles of it with $1 price tags on them, so it's something else that's really gone up in price over the years.

Jim
Logged

G3RZP

  • Member
  • Posts: 2254
RE: "A Three-Transistor Receiver for the Beginner" ARRL 1968 -- Build
« Reply #29 on: June 19, 2013, 04:44:17 AM »

I've found that clear nail varnish is very good.
Logged
Pages: 1 [2] 3 4 5   Go Up