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-2
ad: Subscribe
ad: Left-3
ad: MessiPaoloni-1
ad: L-MFJ
ad: K5AB-Elect-1
  1. KM1H

    KM1H Ham Member QRZ Page

    Not me BUT I was on the Design TEAM of 3 products of the 60's. The still much sought after HRO-500, the beautiful NCL-2000 which is still popular, and the desktop KW transceiver NCX-1000 which came out shortly before National went under the first time due to crooked high up management.

    I was responsible for some changes to Amp Supply products as a consultant....and then they went under since Denny couldnt keep it in his pants.....again.....see Dentron for the first pass.

    Carl
     
  2. KD2ACO

    KD2ACO Platinum Subscriber Platinum Subscriber Life Member QRZ Page

    It's also a possibility that a post can change since the like was administered.

    Regardless, the suppressor in my AL-80B looks like this:

    P1010368.JPG

    P1010370.JPG

    The caps are 100pf/500v silver micas.

    The resistor is pretty big and resembles 5W Yageo MOX ones I have in a drawer. Does that look about right, Carl?

    http://www.yageo.com/NewPortal/yageodocoutput?fileName=/pdf/throughhole/Yageo_LR_RSF_2013.pdf

    Here's a picture to give you an idea of the layout. The resistor spec sheet is rather sparse.

    P1010362.JPG

    click any pic for bigger size
     
    Last edited: Jul 14, 2019
  3. KM1H

    KM1H Ham Member QRZ Page

    Yep, 1/4 of what is used in the AL-811H and AL-572 and less turns of the coil.


    Hmmm, he must have decided my ancient design was better at some point or someone figured it out for him.:eek: However I left that design behind years ago. Yes, those are 5W MOX and I used to use the generic ones from Mouser to give some overhead above the old 2W and should not run hot even on 10M. I also used to use 3W in parallel before low cost 5W came out and misspoke earlier. The Ohmite OY wasnt available yet.

    I also use only one resistor size/value for all amps (coils vary) and back to 47 Ohms 2W for the Ohmite OY which can handle a lot of short term overload. When I service an amp I dont expect it back for something simple like a cooked resistor, reliability is cheap. Plus the single resistor and coil is so much neater, especially in a multi tube amp.

    Hope I didnt confuse you!

    Carl
     
    Last edited: Jul 14, 2019
  4. KD2ACO

    KD2ACO Platinum Subscriber Platinum Subscriber Life Member QRZ Page

    Not at all.

    I'm curious to see if we gave Jeremy enough info to go on. He's doing what I don't have the tools or brain to do, but it's interesting! :p
     
    KM1H likes this.
  5. KD2ACO

    KD2ACO Platinum Subscriber Platinum Subscriber Life Member QRZ Page

    For me the big question here is, "Does the steeper filter mitigate the issue of directly grounding the grid?"

    I can see the charm of having the grid strapped to the chassis with a foil sheet if the tube decides to arc... no mica caps and itty bitty resistors to explode. I'm just curious if that suppressor makes things better or worse in this case.

    It would be neat to see modeling results.

    [/MORBID CURIOSITY]
     
  6. K9AXN

    K9AXN Premium Subscriber QRZ Page

    Calculate the inductance of a 1/4 X .010 inch ribbon 1.5 inches long. Now calculate the inductance of a .05 inch wire 1.5 long wire. Any surprises?

    Tom W8JI and myself have gotten into the bare knuckle ring a time or two over the years but I have to give him some Kudos for courage, writing skills, and developing a good deal of interesting material. I frequently disagree with Tom but in the larger scheme of things, he has contributed more to the Ham experience than the sarcastic complainers. Some have never written any technical material, cringing at the thought of suffering a mistake or two --- there's a name for that. I would prefer to read possibly flawed material than listen to hate filled diatribes slandering others. It's easy to stand aside while this takes place thinking that it's not your business --- there is also a name for that behavior --- you see it in the big city all the time.

    G-Day Jim
     
    N2EY and KA0HCP like this.
  7. KM1H

    KM1H Ham Member QRZ Page

    What steeper filter?

    If you are 100% sure of your skills at building a SB-220 from a kit then the direct grounded grids and increased gain may be OK. Many do not use 10/15 M so never find out the truth. Personally I see no real benefit after decades of SB-220/221/HL-2200 service/repairs and hundreds of mods to 6M. If 1200W isnt enough buy another amp:D. I dont waste my time writing articles that cant be commented to and dumb stuff lives on forever on the Internet, just read some of them from posters on here. I do my writing on forums where they are peer reviewed and sometimes something good comes of it. I have no problem with making a mistake and readily admit it; I grew up many decades ago:p. I do have a problem with those who just make snide sarcastic comments to cover up their own poor skills on here and in "articles".:mad:

    Eye candy charm only, no proven benefits and only an issue after several decades of trouble free performance and thousands still in use without it. Gee Whiz!

    After all the AL-80 family and AL-82 still have arcing events and the poorly amp educated just blame the tube.....never the amp design by so called Mr Perfect who had to add the "glitch" resistor many years later and chose an expensive overkill to prove his whatever maybe.

    That resistor goes back to the 1963 National NCL-2000 and was ignored for decades by just about all amp companies. It was even missing in the low production rather temperamental Hallicrafters SR-2000 that used the same finals although its use was clearly mentioned in the RCA spec sheet....a DUH moment for a sarcastic troll who likes to give "likes" with no true idea of why.
    https://frank.pocnet.net/sheets/079/8/8122.pdf

    Id rather have the stock itty bitty choke, not a resistor, help save a tube if the arc is minor which is almost always from anode released gas. Show me any benefit of a copper strap in ANY gas case! It only makes component destruction more intense. OTOH the glitch resistor in place of RFC-2 in the SB-220 has dual benefits as Ive written many times. And for those with a selective memory and reading comprehension issues here it is again: It is a current limiter (entry level I squared R math) that acts to reduce the the destructive surge until the SB-220 breakers open or a fuse opens in other amps. The Heath breakers are slow and get worse with age. And the other benefit is to kill the oscillation contribution of that choke due to the cavity effect of the RF enclosure. Note that choke or enclosure style design is no longer used....Another DUH event for a few here Im sure.

    Off for some lunch, berry picking in the yard, and lawn mowing. Unlike others that threaten to run away....but dont.... I Shall Return. Stay tuned.

    Carl
     
  8. G0HZU

    G0HZU QRZ Member

    I've been watching the cricket all day so I've only just had a quick look at modelling the AL80B circuit. This is a single 3-500Z amplifier. So the first big difference will be a halving of the overall tube capacitance compared to a dual 3-500Z amplifier. So I'd expect the range of frequencies that it is likely to oscillate at to go up by sqrt(2) = 1.4. So I'd expect maybe 170MHz typical.

    I put together a quick amplifier model including the interconnection strays and the first plot below shows the negative resistance at the output when the input is terminated with the worst possible source impedance (i.e. the source impedance that causes the most negative resistance at the output)

    The solid lines show the response with the suppressor and the dashed line show the response without it. So it looks like it will typically oscillate at around 175MHz (without a suppressor) but this assumes a typical tank setting when tuned on the 20m HF band.

    The next two plots show the series equivalent resistance and inductance of the AL80B RLC suppressor vs two classic RL suppressors. The plots show that the AL80B suppressor can produce lower ESR at 28MHz compared the classic 'suppressor A' plot. Or, if the suppressor A inductance is adjusted to give the suppressor B plot , the ESR at 28MHz is now the same but the AL80B suppressor gives a steeper response in resistance vs frequency. Note that I did include a guess at the lead/package inductances for the caps and the resistor in these suppressor models.

    Based on this simple system model, it looks like the amplifier has a reasonable margin against instability. The only thing that seems to threaten this is a variation in tube transconductance. It the tube transconductance creeps up to 15mA/V or beyond then K becomes a bit marginal.

    I hope the plots are interesting. Don't take them too seriously because this is purely based on the datasheet parameters for this tube and I think this is getting increasingly risky for a single tube amplifier like this because typical instability would probably happen up at 175MHz or so. Trying to model stuff up at 175MHz puts strain on the very basic tube model I have here. I think a much better VHF model of the tube would be needed here! I've also had to use the HT choke model from the SB-220 as I don't know how to model the AL80B HT choke.

    Based on the datasheet based system model, this amp (with the RLC suppressor) looks OK to me.
     

    Attached Files:

    Last edited: Jul 14, 2019
    KM1H, KA0HCP and KD2ACO like this.
  9. KM1H

    KM1H Ham Member QRZ Page

    Is that what we call a "rug rat" over here? AKA curtain climber and cookie crumbler.

    Thanks for that info Jeremy.

    So that seems to justify my 30+ year old suppressor design using just a spectrum analyzer and a 50's era GDO.
    It also shows how useless it is to try to measure L with some hand held meter at 1 kHz or so and the model fits your earlier guess pretty close. Congratulations again.

    I learned that lesson in the early 80's from a couple of real engineers when a Sr Engr Aide on a CATV amp project.
     
  10. W1QJ

    W1QJ Ham Member QRZ Page

    Modeling of the AL-80b from only looking at a schematic probably would require some guessing as to the inductance of several components including lead lengths. Seems that a very critical inductance is the grid structure to ground inductance of the tube and grounding scheme of the grids. That said, if you estimated this, what grounding methods did you assume if you haven’t seen the actual grounding scheme? Assuming what % of error from actual versus estimate would make a significant difference in the vhf frequency of oscillation? 170 versus say 175mhz? More or less? How would difference in the two affect the the building of the perfect parasitic suppressor?
     
    W2VW likes this.

Share This Page

ad: UR5CDX-1