Discussion in 'Ham Radio Discussions' started by G3EDM, Aug 27, 2021.
Pssst: Consider the 1AG4 for power amplification. The 1AD4 is intended for small-signal use.
Ok I see that the 1AG4 is described as a "power pentode for use in output stages". Comparing the data sheet with the 1AD4 I'm not seeing the data that shows more power, but I'm very rusty at reading these sheets (Raytheon in both cases).
The push-pull 2nd AF uses two 1AD4s as you know. It has ample amplification to drive modern low-impedance 'phones but is anemic at driving the speaker. I don't often use the speaker, but sometimes it's useful to have the receiver on in the background while not operating and working on other tasks. Also it's the easiest way to "show off" one's equipment to other people, whether it is (a) incomprehending but very polite friends who visit and listen patiently, or (b) making videos to post on the Zed or elsewhere.
Another issue is the gain loss when the CW audio filter is switched in. Having more amplification behind it (or even ahead of it) might help in some cases. I've deliberately not discussed that filter much here yet -- it is a whole encyclopedia in itself. The time will come, but let's get rid of these parasitic oscillations first.
Would there be any advantage in using a 1AG4 for the first AF?? (Maybe not a good idea to mess with that circuit.)
I'll order a few 1AG4's and give them a spin! (They're a bit pricey, for a sub-mini....) Edited to add: I ordered them, from a U.S. supplier -- cannot find here. The postage cost as much as the tubes....
73 de Martin, G3EDM
Congratulations and fanstastic post!
If The Beast turns out to be a usable receiver for boosting my QSO count, and possibly scoring DX, I have already started thinking about how to make it less unwieldy. This receiver takes up most of my operating desk. There is barely room for a notepad.
The set can be made much less space-consuming by doing these things:
Unbolt the PSU's 2x4 wooden support at the back (the battery rack) and move the whole caboodle somewhere else, such as a high or low shelf in the shack closet, connected to the receiver with a multi-conductor power cable. Saw off the blank wooden plank where the PSU's 2x4 was bolted. The set is immediately 4 inches (10cm) less deep. In its current iteration it is 11 inches (28cm) deep.
Or alternatively, replace the batteries with a (much smaller) AC-powered PSU.
Then cut the receiver in half, right down the middle. (Either by literally sawing the bottom plank in half, or replacing it with two lighter-weight plywood panels and moving over the modules from the original set.) Stack the two halves on top of each other, with the RF/Detector/First AF on top and the CW filter and 2nd AF on the bottom. This still leaves full access to the complex set of controls in the RF-path stages. The CW Filter control, speaker, and headphone socket can go on the bottom level. In its current state the receiver is 25 inches (64cm) wide. The new configuration would be 15 inches (38cm) or possibly slightly less.
Purely for cosmetic reasons, add a wooden front panel covering the whole front. The top has to stay open for access to controls.
Ventilation is not an issue. Subminiature tubes pretty much don't need it. They were designed to go into guided missiles, that sort of thing....
This would work much better in my new shack here in the UK, which has a "vertical" configuration. In my much larger American shack, large gear was not a problem and there was masses of horizontal space as well as a much deeper desk.
This will not turn The Beast into The Beauty, but it would do a lot to make the operating position roomier. It's basically just a carpentry job and I can do that on the dining room table here (yes, that's what I've been using for all my carpentry projects recently, after protecting the table of course....).
But let's get this receiver working properly first....
73 de Martin, G3EDM
At the same time as planning essential fixes to the receiver, I'm also doing this for the transmitter. The aim is to reduce or eliminate chirp and "yoop".
Also to add a "spotting" function.
Looking at the transmitter, this is going to be hard, and exacting, because the wiring is very compact and relatively hard to alter post facto.
Among other things I'm going to be trying various different vallues for R1, the grid bias resistor on the 6C4 oscillator tube. For those not familiar with tube technology, the larger the resistance, the higher the voltage. The resistor, as built, is 47K (yellow-purple-orange) but I'll be testing lower values, in fact by replacing the resistor with a small potentiometer and adjusting to find out the best value. The idea is to reduce the stress on the crystals and reduce or eliminate chirp etc.
Also in the photo: the cathode keying line. This is buried under a bunch of other wiring. I will be "lifting" that line and instead, routing it to a relay independent of the keying circuit. Exactly where to put the relay, physically, is a bit of a challenge.
Bottom-center is R3, connected between grid of the PA and ground. It's currently 10K (brown-black-orange) but later published versions of the same circuit changed it to 22K and that is what the tube specs call for. Essentially, an ARRL mistake that was eventually corrected, but not before hundreds of hams built it with the wrong value. That included me, more than 50 years later, because I used a relatively early publication of this circuit. Changing that resistor to 22K will be a bit tricky because of the tight space and the other wiring overlaid on top.
In the centre of the following picture: the unfinished circuit for bandswitching between 40m and 80m. I will be installing a relay to break that jumper (U-shaped black wire) and instead connect it to the left-hand end of that capacitor, which is currently just floating nowhere.
Finally, possibly the hardest issue of all, which is the unusually high voltages I am measuring in the transmitter. I will have to troubleshoot the power supply. Maybe some resistors in the voltage divider have drifted? Maybe I will have to add a Voltage Regulator tube? To be continued.
73 de Martin, G3EDM
Martin, you have some of the nicest construction technique I've ever seen in homebrew gear. You really do a very nice job - especially for RF circuitry where short leads rule - you still make things look so tidy! Like right out of a book.
That second photo, where the RF choke is in full view, is pure art.
Thank you so much! It was very hard work at the time!
73 de Martin, G3EDM
Unfortunately the technical news is getting worse.
We are probably on the way to fixing the issues with my currently installed receiver, nicknamed The Beast. It is a weird and interesting radio, with which I have already had some QSO success.
However, this evening I did another run of tests of all of my crystals on my "Boosted Pierce" W1TS transmitter. They now basically all sound terrible, with yoop/chirp and some of them drifting, even the handful (four!) that provided a rock-solid signal until now. This is not a "charming vintage signal with a hint of chirp", it is full-blown and nasty instability in tone and in some cases, frequency.
So I must reluctantly get off the air for a while until this is properly sorted out. It should not take too long because I have a good list of fixes for the transmitter's faults. Also, I have ordered the required components and they will be arriving in the next few days. Some of them have arrived already.
The only case in which this pause could last more than a short time is if, despite all the troubleshooting, the transmitter never gets fixed to my satisfaction. It would not be the first time that hams "gave up" on the "Boosted Pierce" design and, now that I know more about the issues, I don't blame them. I am optimistic, however, and got a lot of great help here especially from @N2EY and @W9BRD. In particular, the key modification of running the oscillator continuously on TRANSMIT and keying only the PA looks promising (it was my idea, but actually a classic technique for better keying stability).
The chirp/yoop that I am hearing is mostly on the "attack" of the pulse. In theory this should be partially or entirely solved by only keying the PA. The modification is relatively complex, partly because of changes to the station's control circuitry (the T/R switch), but it should not take too long.
If the TX cannot be fixed, I will be building a new (65-watt!) transmitter, which could take quite a while to complete, because of limited workshop facilities. But let's hope it does not come to that.
I will continue to follow this thread and post here even while off the air.
The main thing of course is that after 50 years I Got On The Air, almost three weeks ago now. Think of this as a minor and temporary setback while we QRT and regroup.
73 de Martin, G3EDM
Forgot to say: When you build "tight" and very direct "point to point" for good RF technique, it is really quite hard to alter later! That TX is very compact. The wiring and components are "layered". It will be a challenge.
Another factor is my (absurd!) insistence on maintaining reasonable fidelity with the "look" of the ARRL original, while still making improvements to make it ... well, usable.
I suppose, 50 years ago, thousands of Novices built this thing and no one cared about the chirp/yoop, all the Novices were in the same boat?
73 de Martin, G3EDM
I don't know that "thousands" of Novices built the W1TS transmitter. Or even hundreds.
I do know that, at the time, the vast majority of US amateurs using crystal control were using genuine FT-243 crystals - or something even bigger. I have an adapter that allows a DC-34 or DC-35 crystal to plug into an FT-243 style socket.
73 de Jim, N2EY