High Gid Current

Discussion in 'Amateur Radio Amplifiers' started by NE7X, Jun 5, 2011.

Thread Status:
Not open for further replies.
ad: L-HROutlet
ad: l-rl
ad: L-MFJ
ad: Subscribe
ad: Left-2
ad: Left-3
  1. NE7X

    NE7X Ham Member QRZ Page

    The problem is not the tubes.....I have test them before, and again just now. I have a Heny 2K-4 that is working. I removed the tubes from the 2K-4 and inserted them into the 2K Classis, same issue. I installed the 2K Classic tubes into the 2K-4, tubes works perfect.

    We can rule out tubes.

    Thomas NE7X...
  2. NE7X

    NE7X Ham Member QRZ Page

    What is puzzling me, why do I see 300ma of Ig on the meter when the amp is keyed, no drive.
    Also when I inserted an external MA meter in series with the BIAS zener, I saw the exalt same reading as the front panel Ig meter, 300ma.
    W8JI said having the external meter in series with the zener is showing Ic current, not Ig.
    Please look at the schematic, this is how its wired. Why is the Ig meter showing Ic current?

    73s, Thomas NE7X…
  3. W9GB

    W9GB Ham Member QRZ Page

    T -

    You mentioned that this Henry was a bunch of parts that you had to reassemble.
    You have a good eye (and other Henry amps to compare):

    1. Any "non-standard" or odd parts? -- such as different tube sockets?
    2. Evidence of previous burnt areas of amplifier? (carbonization tracks)

    In previous amplifiers on the bench --- the tough ones to troubleshoot over past 30 years:
    These are the ones that can cuase you to talk to yourself ..

    1. Cracked resistor (hairline invisible crack) in bias supply resistor that tested OK -- but went OPEN when in circuit (current or heating).
    2. Break in a solid hookup wire. The insulation was perfect -- the butt weld (at factory) from end of one wire coil to next coil had failed !

    After looking at the schematic -- I would suspect one of the resistors (you already changed bad diode)

    Last edited: Jun 10, 2011
  4. W8JI

    W8JI Ham Member QRZ Page


    I'm assuming the amp amplifies?

    Either way should work the same, but connection to the low side of R109 (at the diode) gives more protection for meters.

    I assume that was D106?

    Cannot be the push button. It would not cause it to do what it is doing.

    Good, then that means the tubes are GOOD. You only have a metering problem.

    This all implies the shunt for the plate current meter is open or not connected. That would be R116, which is 0.2 ohms.

    That is NORMAL. It should happen.

    With F101 out you should see exactly what you report. F101 does not fuse anything except the meter, and fuses cannot protect meters very well. That proves that area is OK.

    This tells me something is wired wrong around R116 or R118. Something is making R116 appear to NOT be in circuit when you connect to the top of R109, and it makes the B- or negative end of C112 appear to be grounded when you move the wire you are moving.

    Wire it as shown on the schematic.

    1.) Measure from the positive post of the plate current meter to D1 cathode on the RF deck. Should be zero ohms.

    2.) Measure from the positive terminal of the plate meter to both ends of F101. One end of F101 should be about .2 ohms or so, the other end of F101 slightly higher.

    Unplug the Cinch Jones plug on the cable to RF deck.

    3.) Measure positive post of plate meter to chassis. Should be about 0.2 ohms with F101 in, or with F101 out.

    4.) With F101 back in, measure C112 negative lead to chassis but try it with meter probes BOTH directions. The highest reading should be 150 ohms or more.

    Something in those 4 steps might be wrong. If it is, what step is wrong???

    73 Tom
  5. NE7X

    NE7X Ham Member QRZ Page

    #1 - zero ohms
    #2 - .2 ohms to both sides of the fuse. Wouldn't it be the same ohm value no matter what side of the fuse I measure?
    With 8 pin octal plug to RF deck unplugged
    #3 - .2 ohms
    #4 - 170/55 ohms (reverse meter leads)

    Both my Henry 2K-4 and 3K-A, when I key, no drive, Ig = zero & Ip = 100ma
    For the 2K Classic, when I key, no drive, Ig = 300ma & Ip =160ma

    I think there is something wrong in the BIAS circuit, not the metering circuit. The reason I say this, the red of the tubes during idle on the 2K Classic look to be a lot brighter hotter red, where the 2K-4 and 3K-A tubes have more of a light pink rose red color. Something tells me the meters on the 2K Classic are telling the truth and this is why the tubes look a more fire red.

    The guy I got the 2K Classic from spent lots of time in the meter circuit and I also totally 100% disassembled it and wired it back together. There isn’t very many parts in the meter circuit. Unless the schematic diagram is wrong, I feel the metering circuit is good.

    What about grounds? Could there be some chassis or circuit ground potential return difference somewhere? What other circuits would come into play to cause high grid current on idle? Could there be some cathode leakage being injected into the BIAS circuit? I know you said to get parasitic out of my mind, however maybe not a parasitic, maybe some sort of interaction feedback loop due to capacity coupled or lead dressing? I keep thinking about a possible environment issue. Maybe add more .01 caps in different spots.

    Like W9GB’s comments, need to look outside the normal stuff for something very unusually and totally unexpected. I think the obvious stuff has been looked at already.

    73s, Thomas NE7X…
  6. NE7X

    NE7X Ham Member QRZ Page

    The guy I purchased the amp from said he replaced D1 BIAS zener. I only checked to see if it was shorted/open, however tonight I removed it to look at the markings. He replaced it with a 1N3311B zener, wihich is a 12V zener, The schematic calls for a 10V. How critical is this voltage value?

    Thomas NE7X...
  7. W1QJ

    W1QJ Ham Member QRZ Page

    Tom, the only difference the 2 volts would make in the zener diode he put in would be that the amp would idle lower. For example, if you have 3500 volts on the plate the amp might idle at say 200ma with the 10v zener. With the 12v zener instead it might idle at 160ma. That would be the only difference. The difference it would make also would be that the anodes would get a brighter red quicker with the 10v zener than with the 12v zener. Other than that, there still should be no grid current flow when the amp is at idle. W9GB may be on to something. Let's say for argument sake that the meter is correct and when you are in idle the grid current really is 300ma and the Plate current is 160ma. Well if that is indeed true then something is causing the 300ma grid current and that would be your problem. SO assuming the metering is OK what would cause the grid current to rise whille the plate current is normal. You said you tested the tubes so lets rule them out. If it was a parasitic like you thought then there would be an increase in plate current as well, but there is not, so yhat is out. What the meters indicate if true, one would think you have a bad tube saying perhaps a grid to filament short. SO we need to look at the schematic and see why here might be a current floe between the grid and filament. Could it be there is a problem with the filament transformer? How about an imbalance or a short of the windings on one side of the filament transformer to the center tap? I would have to look at the whole schematic again to see how some type of "noise" voltage could be getting to the grid through the filament(cathode) in the case of the 3-500. Since I never seen a problem like this I could not say off hand. But assuming the meters are reading correctly then yoiu are surely looking in the wrong area for the problem. It might up under the RF deck in the tube socket area? Something is getting in between the cathode and filament to cause that current flow. Oh, and only in the keyed mode of the amp. A bad tube whould show it in standby indicating it was the tube. It seems as Tom W8JI was also intent on the metering circuit as well. Tom, if your still in here, you look someplace else and see what else might be causing an actual grid current flow in idle of 300ma. Tom EX have you listened to your signal, is there a hum on your signal? Turn on a seperate RX and see if a hum is on the signal. Lou
  8. W8JI

    W8JI Ham Member QRZ Page

    I have given up on QRZ. I just spent over an hour typing a detailed post twice, and lost it twice part way near the end.

    I am all done except to say this. Sorry it is so brief.

    1.) Tom does NOT have a zener problem unless he wired it like Henry shows. Henry drew the zener backwards.

    2.) He does not have a filament to chassis problem, if he did the grid meter would read backwards and he would have no RF output.

    3.) He does NOT have an oscillation problem.

    4.) Measuring from filament pins to chassis and zener to chassis with RF deck unplugged should be open.

    5.) Measuring from junction of R117 and 118 to F101 should be 0.2 ohms.

    6.) Measuring from junction of R117 and 118 to positive of plate meter should be 0 ohms.

    7.) Measuring from chassis to F101 should be .4 ohms, RF deck plugged an unplugged.

    8.) Measuring from the positive end of C112 to chassis should be high resistance with slow charge to bleeder value.

    9.) Measuring negative of grid meter to chassis should be zero ohms.

    Finally I would recheck D106.

    I wrote a post telling Tom how to use a power supply and current limiting resistor to test his metering, but it got lost.

    I'm out of time now.

    73 Tom
  9. W1QJ

    W1QJ Ham Member QRZ Page

    NE7X, Tom, Stupid question here. Surronding the paremeter of the schematic there is a solid line with circles with numbers in them. When someone did rewiring did any of those points where the circles with numbers in them get connected together?
  10. AF6LJ

    AF6LJ Ham Member QRZ Page

    Tom That's why I use MS word or similar to type my longer posts.
    This seems to be a complaint with the newer version of V-Bulletin.
    Maybe Fred can fix it.
    It would be a shame to loose you as you are an asset to the forums.
Thread Status:
Not open for further replies.

Share This Page

ad: k1jek