ad: Amateur-1

Oscillator/PA Cathode Keying, "Spotting" Your Frequency, and Chirp

Discussion in 'Homebrew and Kit Projects' started by G3EDM, Sep 5, 2021.

ad: L-HROutlet
ad: l-rl
ad: MessiPaoloni-1
ad: Left-2
ad: Left-3
ad: L-Geochron
ad: HRDLLC-2
ad: abrind-2
ad: L-MFJ
  1. KX4OM

    KX4OM Ham Member QRZ Page

    That was the first transmitter I ever built. The "Novice Special." Same form factor, the pilot lamp, the long skinny chassis.

    Carry on. Very interesting thread, Martin.

    Ted, KX4OM
    G3EDM likes this.
  2. G3EDM

    G3EDM Ham Member QRZ Page

    You're right, I go all over the place actually. Average is probably around 10wpm. When I really slow down to below 10wpm, I slow down the characters too, so that the Farnsworth is less evident.

    But you are the expert, with decades of CW under your belt.

    I tend to send most of my CQs QRS. I speed up a bit during the QSOs if conditions are OK. But the best assessments of my fist, and my speed, are the ones from other ops who listen to my code, and a detailed and (constructively) critical assessment would be great.

    I am always very critical of my own sending on the straight key. Also, there is a huge difference between what you hear as you are sending, and what you hear if someone plays a recording of the same thing back to you later. It's a little bit like hearing the sound of your own voice when talking, vs. listening to a tape of your voice later. Quite different! So the worst analyzers of CW skills are probably those who try to analyze their own code, even something as simple as estimating the speed they are sending.

    The skimmers routinely copy my CQs at between 9wpm and 12wpm for what it's worth.

    One reason to send CQs slowly is that I'm QRP so I had been assuming my signal will be faint at the other end. That actually has not really turned out to be the case. Some OMs are even giving me 599, while I am giving them 549 or something much worse, because ... my receiver.

    73 de Martin, G3EDM
  3. G3EDM

    G3EDM Ham Member QRZ Page

    Thanks. BTW my pilot lamp died a few days ago. It's a pain, because that's a neon. Those were still easy to find when I built the TX six years ago, so I hope that is still the case. I did not know that neons could die!

    73 de Martin, G3EDM
  4. G3EDM

    G3EDM Ham Member QRZ Page

    Overnight I received a report that not only is there a little bit of chirp on the signal, but the carrier frequency is also drifting, enough to be annoying and require constant adjustment on the receiver RIT at the other end. When I check RBN, that also seems to be the case: the drift in the space of less than an hour was 300Hz and at one point in the day, 500Hz, compared to the nominal frequency that the crystal normally oscillates at. I think, but cannot confirm for sure, that this is a new phenomenon (i.e. it was not there two weeks ago when I first started using the TX on-air) and therefore a subject of concern.

    This is a more serious issue than chirp, but it is, in my opinion, worth discussing here in the same thread. I have just measured the key-down voltages and they are higher than they should be. The oscillator plate circuit is 9% higher (218V instead of the 200V on the schematic) and the PA screen is 30% higher (254V instead of the 195V on the schematic). Here's the schematic. The voltages in pencil are the ones I measured today.


    A couple of things:
    • This is quite frustrating because when I first bench-tested the transmitter upon completion six or seven years ago, the voltages were also higher than they should have been. There were also indications of instability, and there was a pronounced chirp on the signal (much worse than what we have today). At the time, N2EY helped me out by designing a voltage divider circuit in the power supply, replacing the simple dropper arrangement from the original PSU. This essentially solved the chirp back then, or at least that is what I thought at the time. It also brought all the voltages back to their design values. I wonder what happened since? Resistance value drift?
    • The 200V design voltage in the plate circuit of the 6C4 is intended for "original" vintage FT-243 crystals. So there is an argument, is there not, for lowering that voltage somewhat, to make the circuit kinder to the "modern crystals in a vintage FT-243 holder" that I have lots of. All of my crystals seem to be oscillating really vigorously, so I would have thought we could safely bring down that voltage and still have a functioning oscillator.
    Here is the schematic of the voltage divider devised back then in cooperation with N2EY. The circled red section is the bleeders from the original ARRL schematic, which we eliminated because the HT can now leak through the divider instead, upon power-down. The red voltages are the ones I measured before the divider was installed and you can see that they were very high. The green ones are the ARRL design values, which is what we ended up with (or very close to) after the voltage divider was installed.


    Trouble is, if I measure the resistors while they are in-circuit, I could get a false reading of course.

    Do you gurus agree that the resistors are the suspects here? A bit weird, as they were brand-new six years ago (not NOS). I did use (new) carbon composition rather than metal film. The 5W is a KOOLOHM.

    I do think that before going off on all sorts of other tangents to troubleshoot the chirp, and the drift, we should firstly get the TX back to its design voltages (and in the case of the 6C4, actually try to lower it even below its design value, to be kinder to modern crystals).

    By the way, all these voltages were measured with a $2 Harbor Freight DVM. (For non-American readers, this is a chain of cheap-tool stores.) Back when I had a proper workshop, the voltages were measured with a VTVM. But still, the difference in voltage is big enough to assume that something is amiss, no?

    Photo of the voltage divider during construction. It looks pretty awful, but according to my notes, most of these joints were re-soldered before completion of construction. I did look it over carefully today and could see no sign of heat damage or any wiring problems.


    Over to you....

    73 de Martin, G3EDM
    Last edited: Sep 9, 2021
  5. W1VT

    W1VT Ham Member QRZ Page

    These days, local noise issues have just as much to do with how loud signals are at someone's station as the power level someone is using. 100W vs 5W is about two S-units, assuming 6 dB/S-unit.
    I saw the noise floor on my 80M vertical drop by 20 dB when they changed out the power pole in front of my house. I am very lucky to have neighbors that are relatively far away, living in the suburbs of Connecticut that was built in the 1950s.

    Zak W1VT
  6. N2EY

    N2EY XML Subscriber QRZ Page


    Two thoughts:

    1) Line voltages in the USA have crept up over the decades. We're at a point now where 127 volts is common.

    But your UK line voltage may not be double that, resulting in lower B+.

    2) Inexpensive meters are notoriously inaccurate.

    73 de Jim, N2EY
  7. G3EDM

    G3EDM Ham Member QRZ Page

    Yes I wanted to check that. Here is my confession. I plugged the meter probe into the wrong socket (the current socket, not the voltage socket). When I tried to measure the voltage by putting the probes across the mains, there was a lovely big blue spark and it tripped the circuit breaker for all the sockets in the house. It also blew the fuse in the power strip where I was trying to measure the voltage. (In this country, just about everything is fused. There is also a fuse in every AC plug, including for instance the one on the transmitter's power cord.)

    I have actually done that before, back in America about a decade ago. You would think you would learn from your mistakes....

    After that, when I tried to measure the house voltage with the probe plugged into the voltage socket, the AC scale no longer seemed to work. Or, perhaps, I still don't understand how to use a DVM, even a dirt-cheap $2 thingy from Harbor Freight. (Give me my analogue meters from the storage unit any day!)

    73 de Martin, G3EDM
  8. AD5HR

    AD5HR Ham Member QRZ Page

    Open the DVM up, there's usually a fuse inside for just that mistake.
    My cheap Radio Shack DVM even had a spare fuse inside the case.

    Please don't ask how I found this out, stuff happens.
  9. G3EDM

    G3EDM Ham Member QRZ Page

    Is it worth trying to retrieve my VTVM from storage? Or perhaps buy a good modern DVM like a Fluke? Don't want to tear my TX apart if the true voltage readings are not what they seem on my cheap DVM.

    I can power my Eico model 232 through this:


    (I've been wanting a Simpson analogue meter for years -- they still make them but they are pricey. Here in UK the famous meter is the vintage Avo-meter which many restorers of vintage gear swear by.)

    73 de Martin, G3EDM
    Last edited: Sep 9, 2021
  10. G3EDM

    G3EDM Ham Member QRZ Page

    Yes I know, I think I found that out last time it happened. I think the fuse cost more than just buying another $2 throwaway meter....

    73 de Martin, G3EDM
    AD5HR likes this.

Share This Page