My First-Ever QSO -- 50 Years After Passing Ham Test

Discussion in 'Ham Radio Discussions' started by G3EDM, Aug 27, 2021.

ad: L-HROutlet
ad: l-rl
ad: Left-3
ad: abrind-2
ad: L-Geochron
ad: MessiPaoloni-1
ad: L-MFJ
ad: Left-2
ad: HRDLLC-2
  1. G3EDM

    G3EDM Ham Member QRZ Page

    A very successful day.

    The new Jackson capacitor arrived today. It is a 56pF, double-bearing model that I'd actually purchased as a candidate for my future VFO.

    It is probably wasted on a regenerative radio, but what the heck, I gave it a go.

    First of all, I wired it up in parallel with the existing, misbehaving main capacitor (a Hammarlund that I may well have damaged myself). I opened up the main capacitor all the way.


    Then I rotated the shaft of the Jackson capacitor. No scratchy noise at all.

    Then I opened up the Jackson all the way, and tuned with the Hammarlund. Scratchy sound is back.

    At this point I had not read the further advice from Jim, @GM3ZMA, about removing and cleaning the wiper of the Hammarlund.

    I went ahead an simply replaced the Hammarlund with the Jackson, and was 8 GBP poorer. (That's actually an extremely reasonable price. It was purchased from on the advice of Karl-Arne, @SM0AOM.


    It was not by any means a "drop in" replacement and ended up taking all day.

    The easy bit was sawing down a piece of wood as the support base, given that the shaft height above bottom was not the same as the Hammarlund.

    The hard part was that when I started working on this and removed the Hammarlund, I discovered that the main tuning dial (a Jackson 36:1 model) was very loose and "floppy". It turned out that when I first installed it six years ago, I had used a thin aluminium template provided with the dial. I could not find my micrometer, but eyeballing it, it was considerably less than 1mm thick.

    This had not been noticed before because the Hammarlund's rotor shaft had, in effect, become the stabilizing force and the dial itself, the "play" element.

    This was obviously not good at all, so I resolved to make the dial support as rigid as possible, and the capacitor support as rigid as possible, and use the very nice Jackson flexible coupling (shown in the above photo) to deal with any misalignment.

    What I did was take the Jackson aluminium dial template and ... use it as a template to drill a new dial support, using the thickest aluminium sheeting I could find. Long story, but there were already three layers of aluminium in that area of the front panel (one of them already added for stiffening) so with the additional sheet, we now have four thicknesses of metal:


    This was time-consuming, much more than the actual capacitor replacement.

    While I was in there I shortened and made neater the RF wiring in the general area of the main tuning capacitor. It was quite a rat's nest.


    Took the radio back to the shack, powered up, and: scratching sound completely gone. (As expected, of course.) Dial now completely mis-calibrated, because the new capacitor is 56pF max, the old one was 50pF: the 40m ham band starts considerably "up" the dial. Tomorrow I will remove a turn or two from the detector tank coil to get things back to normal.

    A post-scriptum: I discovered a new and completely different "noise" in the headphones. It was caused, not by turning the tuning dial, but by knocking the set or manipulating its switches. I quickly discovered it is caused by the long brass shaft that connects the Regeneration Throttle capacitor to the front panel. Even with a brass "panel bushing" or "panel bearing" this shaft is causing intermittent noise. So the plan is to replace the flexible coupling with a plastic one, and possibly replace the brass shaft with a plastic one too.

    For the moment I've simply removed the brass shaft and gone back to reaching behind the front panel to adjust the Regeneration Throttle.

    After making the above changes, a test. I tuned into a ham CW signal and pounded the shack table hard with my fist. Result: Nothing at all, not the slightest wobble in the headphones. Not bad, for homebrew gear!

    73 de Martin, G3EDM
    Last edited: Oct 11, 2021
    WD5GWY likes this.
  2. W0PV

    W0PV Ham Member QRZ Page


    Pondering your "scratchy" tuning problem and staring at the pics.

    I might be way off, but IF the tuning cap rotor is the side of the component connected to ground (?) is there not also a potential conductive path via the shaft, through the metal bearings of the vernier mechanism, and to the panel its mounted on (?)

    In the original smaller panel this looks to be electrically isolated by being mounted on the wooden base. However now it is mounted on the larger metal front panel, which looks like it might be connected to other metalwork, which may be grounded somewhere.

    If so perhaps the shaft / vernier mechanism is forming a parallel path or loop to ground, and debris or dirt is grinding within it causing electrical disturbance.

    Just a wild thought I had ...

    73, John, WØPV

    PS - nevermind, it seems you've solved the problem :D
    G3EDM likes this.
  3. G3EDM

    G3EDM Ham Member QRZ Page

    You're right, that might have had something to do with it. Hard to know, in retrospect. The front panel is explicitly grounded; and so was the smaller panel that the dial was mounted on, back in the "experimental/modular" phase of this radio.

    For a regenerative radio, this set is exceptionally stable. There are no hand-capacitance effects at all, i.e., you can touch or wave your hand around any part of the set, externally. Of course if you stick your hand in the middle of the detector circuitry inside the set, you get some mayhem. But that's to be expected.

    I ascribe at least some of the stability to the fact that although the "chassis" of this set is wood, the front panel is grounded, and so are other metallic elements such as the armatures of the big chokes.

    I also attribute the stability to the advice of Charles Kitchin, N1TEV, who said the detector should be located some distance behind the front panel and not (as is often the case with regenerative radios) immediately behind it. Unfortunately, physical layout plays an enormous role in the success or failure of a regenerative radio....

    73 de Martin, G3EDM
    W0PV likes this.
  4. KP4SX

    KP4SX Premium Subscriber QRZ Page

    That's better than a lot of the commercially made rigs of that era.
    G3EDM likes this.
  5. G3EDM

    G3EDM Ham Member QRZ Page

    I finally found my micrometer and measured the thickness of the front panel at the dial area as it is now: four pieces of aluminium, plus a thin paper dial sheet.

    The result: 0.269" or 6.84mm. That is the kind of rigidity (about a quarter of an inch of metal thickness) that is supporting the main tuning dial. This is connected to the capacitor, which is screwed into substantial planks of wood which are themselves bound to the front panel with long self-tapping screws.

    The four sheets of aluminium are bound together in a small area around the dial by four 4-40 screws greatly tightened and with lock washers of course.

    The Jackson capacitor came with its own plastic base, which is a substantial object in itself, and I used four self-tapping mounting screws inserted into the wood. Although perfect shaft alignment between the front panel (dial) and the tuning capacitor is almost impossible, there is an excellent Jackson flexible coupler in between.

    As for the rest of the "RF deck" a lot of stiffening has now been deployed, mainly, angle brackets (aluminium) and extra wood. Soon: Additional shielding in strategic places.

    A regenerative set will never be particularly stable, but on the key metric of the stability of the main tuning circuit, this set is now pretty good. I now realise it was sub-par before, but you live and learn....

    Edited to add: I just weighed the radio. It is 7 kilos (15 pounds).

    The experience of building and improving this set will stand me in good stead when I build, in the future, (1) a good superhet receiver and (2) a VFO.

    73 de Martin, G3EDM
    Last edited: Oct 11, 2021
  6. G3EDM

    G3EDM Ham Member QRZ Page

    Have been calling CQ on 40m for a couple of hours before dentist appt (aargh).

    CONDX seem pretty terrible both on the transmission and reception side, and N0NBH data confirm that.

    The receiver needs re-calibrating now that the max capacitance of the tuning dial is 56pF not 50pF. I was originally going to remove a turn or two from the tank coil to shift the 40m back to the correct portion of the dial. Instead, I'm going to try changing the value of the shunt capacitor across the main cap from its current 100pF to 94pF (achieved by using two 47pF caps in parallel). That fixed capacitance is there to improve linearity of the dial, but lowering it slightly should not have a major impact on that, and should simply bring the bottom of the 40m band back to the bottom of the dial. I've ordered a couple of 47pF silver micas.

    73 de Martin, G3EDM
  7. G3EDM

    G3EDM Ham Member QRZ Page

    After a four-day dry patch, finally, a QSO, with a ham near Liverpool. Distance, 362 km (225 miles).

    Phew. Back to the shack.

    73 de Martin, G3EDM
    W0PV likes this.
  8. G3EDM

    G3EDM Ham Member QRZ Page

    Today (Wednesday) was strange.

    Got up early to try to score some DX. But within minutes, I was too frustrated by the long-running issue of key clicks on the sidetone.

    These clicks are only on the sidetone from the transmitter's sidetone oscillator (they are absent from the RF signal radiated on-air). They have been there for the whole of the six or seven weeks that I've been on the air. To start with, I whined about it, then I decided to ignore the issue and just get on with scoring QSOs. It is purely an "operator comfort" issue but a huge one. The clicks are incredibly loud, so loud that I actually worried about hearing loss. There is a volume control on the sidetone generator, but the clicks are always the same deafening level regardless of how loud, or soft, the tone is.

    The explanation for this is relatively simple. With cathode keying, there is an appreciable voltage on the key (I think about 30V in my transmitter). This creates small sparks at the key contacts, both on "make" and "break". This, in turn, generates broadband RF: much like a spark transmitter in Ye Olden Days, but on a tiny scale. This RF doesn't travel more than a few metres, which is why it is not heard in the transmitted signal. But it can, and does, affect sidetone generators under certain circumstances.

    So I gave up on the idea of getting more QSOs today and decided to troubleshoot this issue, after putting it off for a long time.

    Next post: Trying to stop the problem right at the key.

    73 de Martin, G3EDM
  9. G3EDM

    G3EDM Ham Member QRZ Page

    The problem of "local" key clicks (heard only in your own headphones, not on the transmitted signal) was well known half a century ago.

    In the ARRL handbook for 1963, there was a suggested circuit to add, right at the key, to (a) stop the spark-generated RF from travelling up the keying line, and (b) reduce the intensity of the sparks.

    It consists of adding a 1 mH to 2mH choke in series with the line, and a capacitor across the line (to reduce travel of the RF). To reduce sparking, you add a small resistance (up to 100 ohms) in series with the capacitor.

    Anyway, I had tried this before, with no success but on a "breadboard" experimental basis. This time I decided to do a properly shielded version.




    Also proper lugs to make firm contact with the key's screw terminals:


    Result? Nada, absolutely nothing. The key clicks are just as loud as before.

    Next up: Shielding all the interconnect cables.

    73 de Martin, G3EDM
    Last edited: Oct 13, 2021
  10. G3EDM

    G3EDM Ham Member QRZ Page

    The rig that I built is the one described in a 1968 ARRL book, "How to Become a Radio Amateur".

    In that version, all the audio interconnect cables between the transmitter and the receiver are unshielded twisted-pairs.


    These are high-impedance circuits and neither side is grounded. So simply using shielded cable for the two conductors does not help in any way. There is an obvious issue that these cables could easily be picking up the RF generated by the key sparks.

    So what I did today was use modern twin-shielded cables typically used for stereo connections in audio gear. Each of the two conductors has its own shield. I grounded both shields, and at both ends (receiver and transmitter).




    Result: Nada. Key clicks exactly the same as before.

    Next up: Add an RC filter to the sidetone generator to filter out the RF.

    73 de Martin, G3EDM
    Last edited: Oct 13, 2021

Share This Page