ad: M2Ant-1

3-500Z Base Seal temp and protection.

Discussion in 'Amateur Radio Amplifiers' started by N4ZAW, May 5, 2017.

ad: L-HROutlet
ad: l-rl
ad: Subscribe
ad: L-MFJ
ad: Left-2
ad: Left-3
ad: abrind-2
  1. N4ZAW

    N4ZAW Subscriber QRZ Page

    thermalswitch.jpg The Eimac datasheet shows the max base seal temp rating to be 200C. Proper cooling across the pins is an issue in some amplifiers, such as the SB220, with this limit being exceeded in rare cases, especially when the amp is over-driven, operating AM, or CW by long-winded ops. This isn't normally an issue for me, as I mostly listen, and am partial to SSB operation on HF. But if a tube gets stuck "on" in a welded-contact relay failure, (or a cat napping on the TX key), the seal temperature theoretically would soon be breached, not to mention, the stresses the entire amplifier's components are subjected to during such an 'event'.

    So, I've been kicking-around the idea of securing a N/C thermal switch, such as the one in the pic above (below?), to one of the grid pins of each tube, with a tripping temp rating of 180C . It would connected in-series with the amp's keying relay circuit.
    My theory is that, in the event the pin seals ever reach 180C, this simple addition would force the amp into STBY until the tubes cooled, and the thermal switch resets (which is around 110C for the ones I have).

    I'm wondering if I should have concerns over stray RF interfering with this addition, or with the way the plate interacts with the grid, and the switch being that close to the two fields.
    Also, maybe a lower temp-trip rating on the switch could be better, or worse?
    Do you think any of these concerns are valid, or do you think I should just try it?
    I have no immediate plans to attempt this, but if/when I have one of my 3-500Z amps on the bench for repair, I've been planning to.

    What are your thoughts / concerns / suggestions?
    Last edited: May 5, 2017
  2. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    I don't see any "pic below" (maybe you forgot to add it?) but in general, any monitoring of pin seal temps should be done at the filament/cathode pins, not the grid pins.

    The filament pins run much hotter and those are the two pins on each socket that usually cause socket failures.
    N4ZAW likes this.
  3. N4ZAW

    N4ZAW Subscriber QRZ Page

    It reported "error" first attempt at uploading. I hope that the 2nd try fixed it.
    A VERY Good point! And will plan to move over to the filament pins. Thanks.
  4. N2EY

    N2EY XML Subscriber QRZ Page

    The main concern would be RF. W8JI and others recommend that the grids be solidly grounded, so that concern is minimized.
    N4ZAW likes this.
  5. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    I think the right pins to monitor would be the filament pins, anyway, and they do indeed have RF applied to them in a GG amplifier. However they're pretty low impedance and sticking something to them may not impact anything -- I'd have to try it, to see.
    N4ZAW likes this.
  6. WA7PRC

    WA7PRC Ham Member QRZ Page

    Heath's cooling method works as long as it's not compromised. If you want to measure the filament pins temperature, a non-contact method (such as an infrared sensor) would be best.
    KA0HCP likes this.
  7. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    Sure it would, but that's not so easily done when the amp is live and transmitting.

    I have a bunch of nearly zero-mass thermocouples (twisted wires) that can be tack soldered to almost anything that will take solder, along with a digital 8-station thermometer that works with them and reads out temp in F or C degrees (all purchased at the local swap meet for about $50). I'd consider tack soldering those to the filament contacts on the sockets, which should be very close to the same temp as the tube pins are, and lead the tiny wires out through a hole in the cabinet.

    Since the filament pins are driven by RF, might have to use a decent RF choke in each line between the socket terminals and the rest of the thermocouple wire.

    I built an SB-220 45 years ago and owned another one since then and don't recall this really being a problem unless someone runs full power on RTTY or any kind of power on AM. The CW and SSB modes don't get anything hot enough to fail, assuming the fan is working normally -- even in contests where you do a lot of transmitting.
    N4ZAW likes this.
  8. N4ZAW

    N4ZAW Subscriber QRZ Page

    True, or if the amp is abused, or a 'parasitic event' (other than a visit from my step daughter) attacks it. The SB220, or most other GG amateur amps weren't really designed for continuous duty anyway, but I've seen the damage first-hand on tube pins caused by overheating.
    Those come in handy for sure. My concern, though, is with a tube overheating in normal service, case-on, which isn't always conducive to checking pin temps, as Steven stated above.
    I've just been kicking-around this possible solution for a while. It won't help in the event of an amp going into a parasitic oscillation when connected in-series with the TX relay as I'm proposing, but if it were to be used to trigger a high-current relay in the 110/220VAC service, or the soft-start, it theoretically could. As that would shut the whole amp down, as opposed to just disabling the TX relay, maybe.
    BTW, I appreciate you guys chiming-in on this. I have no plans to implement this just yet. That is, unless one of my personal amps happens to need repair all-of-a-sudden.
  9. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    The amp itself (SB-220) will tell you when the pins overheat, as they'll disconnect and stop working.

    That's not normally a catastrophic event, but a common problem is overheating the socket contacts and then the contacts or the sockets require replacement.

    I've also seen 3-500Zs literally have the solder which bonds the tube element wires to the pins melt and reflow. Usually that isn't catastrophic and they can be re-soldered.

    I don't think I've ever seen a 3-500Z actually fail due to a glass seal leak caused by overtemperature. It might happen, but I don't think I've seen it.

    I've seen them fail from gas leaks over time...just a function of aging, after many years.
  10. WA7PRC

    WA7PRC Ham Member QRZ Page

    Yeah, they suck (air). It's been said (by Carl KM1H and others) the later Eimacs produced in Salt Lake City from '86-'89 (link) had issues with poor wire-glass seals. And, early Chinese-made toobs may not have been pumped down as well. My Chinese-made 3-500ZGs lasted only 6 years before flashing over. The Eimacs that came w/ my SB-220 lasted 35 years.

Share This Page