Grounding the Grids vs. Cap/RFC Grounding

Discussion in 'Amateur Radio Amplifiers' started by W4LAC, Jun 13, 2019.

ad: L-HROutlet
ad: l-rl
ad: Left-3
ad: L-MFJ
ad: HRDLLC-2
ad: L-Geochron
ad: Left-2
ad: MessiPaoloni-1
ad: abrind-2
  1. KM1H

    KM1H Ham Member QRZ Page

    Both were introduced in September 1939 and described together in the same article a month later in RCA Ham-Tips. The high mu 811 was slated to be the modulator and the low mu 812 as the Class C RF amp as was the practice at the time. They both introduced the new RCA developed Zirconium coating on the anode for increased temperature ability with a decent increase in plate dissipation over other tubes of the time. ICAS ratings were introduced in the same issue for several tubes.

    The 812 and 812A (6-29-48 intro) is seldom seen these days doing anything. I have a 60's unit using 3 A's for RF sputtering/welding. Nice HD power supply and desk cabinet for someone, make offer and pick it up; looks almost new.

  2. KM1H

    KM1H Ham Member QRZ Page

    Jeremy, I dont remember if I copied you on this article in the past so here it is again for your info and possible amusement by the "designer" of the AL-811 and 811H.

  3. G0HZU

    G0HZU QRZ Member

    Are you saying these AL811 and AL811H amplifiers designed by W8JI? Does/did he work for Ameritron then?
    He doesn't seem to be impressed with the design of the Collins 30L1 when I read your link! He also mentioned a Dentron Clipperton L amplifier so I could have a look at that one at some point.

    I think I need to go back and study the 811 and 811H system models again. Maybe I'm working from an obsolete schematic but I did get both from the Ameritron website. As I said a few posts back, sometimes K can dip below 1 really aggressively but the operator might not be able to create the required complex impedances at the input and output to create instability. None of the RCA-811A based amplifiers I've simulated manage to get K above 1 everywhere but the 30L1 and the K6SNO are impressively close. By contrast, the Ameritron amplifiers seem to be designed with a total disregard for K factor at upper HF and into VHF. Maybe something is missing from the schematic in each case.
  4. KM1H

    KM1H Ham Member QRZ Page

    Yes, he has been there for years after leaving other companies as a basic bench tech which included Dentron who didnt care to listen to his nonsense nor did Heathkit when he was rejected for a job. Due to attrition from low paid real engineers at Ameritron he got the nod awhile after MFJ bought Ameritron and moved it to Mississippi, the lowest of all 50 states in most categories including education and poverty level.

    He will never give anyone else credit and has stolen circuit ideas from many, including myself. After so many AL-811 series complaints Mr MFJ himself canned him and has hired some real engineers as consultants and possibly employees....Ive not kept track since the problems went away with JI. Magic I suppose:eek::rolleyes:

    The Clipperton L is a 4 x 572B 160-10M amp that was an instant success and is still in demand on the used market since it is simple and low cost to service. JI wont say so but that amp was designed when the 1000W DC input FCC rule was in effect. They often get damaged when the 1970's era filter caps break down and take out the transformer. The original parasitic suppressors have the identical problems as all others of the era. Some try to run them at 1200-1500W OUTPUT and cry like babies when s..t happens. I rebuilt one of those messed up amps I picked up free, made a few simple changes and run it almost exclusively as an AM linear, even on 10M with no instability or other problems. Just another example of my KISS philosophy. Dentron used only USA Cetron tubes and when those finally got weak the rush of Chinese (3 manufacturers reduced to one) garbage caused their share of damage. That has pretty much been resolved by RF Parts Company and Shuguang.

    The designer couldnt disregard something he had no clue about. He was a master at copying others and missing many if not most of the fine points of amplifier design and it wasnt the first time. Just look inside the AL-811's; there are plenty to see via Google.
  5. N2EY

    N2EY XML Subscriber QRZ Page

    RCA Ham Tips (1938-1970) are available online free for the download at a number of websites.

    The 811 and 812 are described in Vol 2 #5 and #6. The 812A first appears in Vol 8 #2 and the 811A in Vol 9 #4.

    The early ones have a sort of "STOP THE PRESSES!!!" tone to them.


    are two sources.

    General Electric also put out a similar thing (GE Ham News): (1946-1963):

    It is interesting to see the tubes and circuits that were presented as The New Thing back-when, and what endured and what didn't.
  6. G0HZU

    G0HZU QRZ Member

    Wow, lots of info in those RCA links, thanks!

    I googled for W8JI and AL811 and found this 'field update' page here

    The standby stability problems described here are expected when the AL811 system is analysed. You can see he has added a shunt 200R resistance R100 at the cathode to try and improve things. Earlier in the thread I mentioned that the 30L1 might need a 300R AC coupled shunt resistance at the cathode to get K>1 everywhere. This would make it unconditionally stable (on the system model at least). However, this mod probably isn't 'needed' on the 30L1 because I spent some time today exploring the implications of the fact that K dips just below 1.0 on the 30L1 at the top of the HF band and a bit beyond this. It gets down to something like 0.91.

    The results of the analysis showed that if I was allowed to present any reflection coefficient I liked at the input of the 30L1 the most negative resistance I could generate at the output was a couple of ohms negative. This probably isn't going to cause issues because the ESR of the rest of the tank system after the tubes will normally offset this. Normally the ESR of the tank components themselves will be in the milliohms range but when the pi tank is adjusted to 'create' the equivalent of a large inductance down at HF the ESR will creep up a lot. So it might be the case that only the unlucky few will ever find this potential instability in the 30L1. So there probably isn't any point adding the 300R shunt to the 30L1. The other thing is that I'd guess that a lot of these old 30L1 amps will have had the tubes replaced more than once. So it might not be fair to assess the stability of this amplifier based on samples that have had newer and possibly inferior tubes fitted over the years.

    But contrast this to the AL811. Assuming the schematic in the manual is correct, there is no way this design would have got through any kind of formal (engineering) design review for stability. K steeply nosedives below 1 (towards zero!) only part way through the HF band and the system model predicts that net negative resistances of -20R are easily achievable at the amplifier output when varying the phase angle of a deliberately mismatched source input impedance. So I'm not surprised to see random instability issues on standby and band aid fixes like R100 in the link above.

    Note that the 200R shunt resistor R100 doesn't do much for K in the AL811. It nudges it in the right direction but not much more.
    Last edited: Jul 12, 2019
    KM1H likes this.
  7. G0HZU

    G0HZU QRZ Member

    I went back to look at the AL811H system model again this evening and I noticed it still has the (1 wire) 60nH grid inductance that was originally suggested by K9AXN. If I put in the same value for a genuine 2 grid wire RCA-811A tube rather than a 1 wire Chinese version this amplifier is much improved. The neutralisation range extends to the top of the HF band now and K is up close to 1 almost everywhere. So I'm going to guess this amplifier was designed with the RCA-811A tubes in mind rather than the later 1 grid wire tubes from China.

    I still can't explain the K factor response of the 3 tube AL811 though as it has grounded grids and no (classic) neutralisation circuit. Even with the lower RCA-811A grid inductance the K factor collapses close to zero part way through the HF band. If I add the same neutralisation circuit as the 4 tube AL811H then K factor looks to be at least as good as the AL811H response.
  8. N2EY

    N2EY XML Subscriber QRZ Page

    You're welcome!

    Some of the projects presented are really good, some are really...questionable, most are OK for the time.

    IMHO, the most amazing is the LWM-3, which appeared in GE Ham News Vol 16 No. 6 and Vol 17 No. 1 (last issue of 1961 and first of 1962). The LWM-3 was a pre-WARC 80-10 SSB transceiver inspired by the Collins KWM-2, and quite a project!

    73 de Jim, N2EY
  9. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    I haven't caught up with this entire thread, but noticed this comment.

    The AL-811H's shipped for years now have all four tube grids directly grounded:
  10. G0HZU

    G0HZU QRZ Member

    Thanks. So it looks like I was working from an old AL811H schematic as this mod is dated in 2011. I think those mod changes to ground the grids directly would actually improve K for the AL811H because the lead inductance of the 10nF grid caps would be removed. This is possible because the AL811H uses a classic transformer/capacitor neutralisation network.

    However, I'm not so sure about all this if these amplifiers use Chinese tubes. If the newer (fragile) Chinese variants of the 811A tube also have greater internal grid inductance it must cause a few quality control headaches for the designers of the newer amplifiers. It may also cause issues with stability if these newer (inferior) tubes get fitted to classic old designs that were designed with the original RCA-811A tubes.
    Last edited: Jul 13, 2019

Share This Page