ad: AlphaRF-1

SB-220 C4 Overheating

Discussion in 'Amateur Radio Amplifiers' started by NE6I, Oct 18, 2014.

Thread Status:
Not open for further replies.
ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: L-MFJ
ad: Left-3
ad: FBNews-1
ad: Subscribe
  1. AC0OB

    AC0OB Platinum Subscriber Platinum Subscriber QRZ Page

    This is the circuit I have used to replace the chokes:

    View attachment Twin 3-500Z.pdf

    I would also replace the Zener with a 7.5V 10 Watt stud mount or two 7.5V 10 Watts in Parallel, and as Sue suggested, put a ! Amp slow Blow fuse in series with the filament transformer Center Tap.



    AC0OB - A Place where Thermionic Emitters Rule!
    Last edited: Oct 24, 2014
  2. WA7PRC

    WA7PRC Ham Member QRZ Page

    This presumes that hard-grounding grids causes tube arcs. Tube arcs are caused by a gassy tubes, not direct-grounding of grids. And the only way the tube can oscillate is if there is sufficient feedback (in phase) to overcome losses. The only way that is going to happen is if the grids are high enough above ground and the input and/or output circuits are tuned at/close to the tube's resonant frequency. Over the decades, MANY 3-500Z amplifiers have used hard-grounded grids as designed with ZERO issues.

    A 1mH choke bypassed by 600 pF (total each) of capacitance puts the grids at essentially RF ground, with nearly zero negative feedback at RF (XC = 76Ω @ 3.5 MHz, and 9Ω @ 29 MHz). The 1mH RFC is essentially open at RF (> 20KΩ @ 3.5 MHz). Hard-grounding the grids is the only way they should conduct to ground. Anything else looks like a fuse. In the event of a tube arc, an open fuse in the grid path potentially allows a large pulse of current to reach the rig driving the amplifier via the cathode. The widely-acknowledged BEST insurance is a glitch resistor in series with +HV, a meter protection diode from -HV to chassis, and properly sized fusing in the AC mains. Tom W8JI shows you why direct-grounding grids is correct (link).

    I'll agree about nichrome wire being voodoo. But, after 40 years of hard and nearly daily use, the original 47Ω/2W Carbon Composition resistors in my SB-220's parasitic suppressors though close to 100Ω, were still (barely) usable. I replaced them with new Allen-Bradley parts (they should outlast me).

    Depending on HV, 5.1V is typically the correct voltage to obtain the most linear operation. The W7RY QSK board includes electronic adjustable bias regulation and switching. Without QSK, Tony W4ZT's board is equivalent to the W7RY board. After W4ZT passed, K4POZ has continued producing his board.

    vy 73,
    Bryan WA7PRC
    Last edited: Oct 24, 2014
  3. NE6I

    NE6I Ham Member QRZ Page

    Just testing to see if anyone was awake. :)
  4. AC0OB

    AC0OB Platinum Subscriber Platinum Subscriber QRZ Page

    I have to agree that there are a number of approaches and think we have to agree there no perfect solution to flashovers.

    I prefer an energy sink in the grid circuit to absorb some of the ionization energy at the grid during flashover. I think this was possibly what Heath had in mind with the choke.

    As far as the zener voltage, I use a 10 volt zener because I run my Henry at 3,500 volts loaded. Since the SB220 has a lower voltage power supply, the 5.1 volt zener is probably about right.

    I know Ameritron uses 7.5 or 15 volts (two zeners in series) bias for 3-500Z's depending on operating mode.

    And oh yes, a glitch resistor in series from the HV B+ to the plate choke (with a 1 Amp FB fuse) can be a component saver as well.:rolleyes:



    AC0OB - A Place where Thermionic Emitters Rule!
    Last edited: Oct 24, 2014
  5. WA7PRC

    WA7PRC Ham Member QRZ Page

    The two chokes (1mH, 24Ω DCR) provide a DC path to chassis, and lift the grids above RF ground (XL = 22K at 3.5 MHz, 176K at 28 MHz). 600pF of bypassing per tube (XC = 76Ω at 3.5 MHz, 9Ω at 28 MHz) is what keeps the grids close to RF ground. When a gassy tube arcs over, the chokes' fine wire function as fuses.
  6. AF6LJ

    AF6LJ Ham Member QRZ Page

    Heath did not directly ground the grids for two reasons, it lowers IMD by three to six DB and it raises the drive requirements. Other amplifiers of the day the Swan Mark II and the Drake L-4B ground the grids directly.
    This has been hashed time and again here in this forum.....
  7. KA9JLM

    KA9JLM Ham Member QRZ Page

    That Amp could still have a bad tube.

    I hope you marked them on the swaps.

    You will know when you fire it up sparky.

    I like Fuses.

    Good Luck on your project.
  8. KM1H

    KM1H Ham Member QRZ Page

    I presumed nothing, higher gain can increase the chance of instability.

    On 10/12M very few SB-220's are fully neutralized, on 10/12M high value suppressor resistors can not always stop a parasitic forming. On 10/12M with a gassy tube it doesnt take much to ignite that gas.

    On 10/12M with all 3 conditions met it is a formula for disaster.

    Over the decades (going on 4.5 of them) I dont know of any commercial ham amp with 3-500's that hasnt had an arc event. Of course my business is servicing dead amps so I do get to see the bad and seldom the good.

    Of the hundreds of 3-500 amps Ive repaired or converted to 6M since the SB-220 came on the market in a December 1969 QST ad, after several delays and then even longer until shipments started, Ive seen only a handful of suppressor resistors that were even within 20% of 47 Ohms and all others were much higher particularly if they had substantial 20-10M use. When the planets are aligned and the sun gods look at earth with benevolence even high ones will prevent a parasitic from starting. On 80-40M most anything works and 10/12M is the most critical.

    I wouldnt point anyone to that site unless they know how to seperate facts from pure BS.

    Potentially is correct but Ive heard of little to substantiate that as even grounded grids and other type of tube amps have been blamed, most often dear JI's Ameritrons acting up. A typical SB-220 arc happens even with the exciter off or not even connected. Other times it happens during a slow relay closure and full RF drive applied; Ive had that happen several times but never had exciter damage

    There is a fairly wide range of good bias voltages and what fits at 2500V wont at 1800V. The easist way to replace the easily damaged zener, even 50W wont survive a major arc is to replace it with a string of reverse connected 1N5408, or even 6A10, 10A10 with a 10K uF or more electric across them to give voltage stability.You can select the number in series for your desired idle current

    I was introduced to the surge resistor way back in 1964 with the NCL-2000 when I worked on it at National; it was specified in the RCA 8122 tube spec documents so it has been around even longer Im sure. Ive been championing them even since and see some amps have included them.

    Using a standard 250V glass fuse in the HV line is a recipe for disaster as it maintains a plasma arc after the fuse vaporizes, explodes the glass and starts its dance looking for a ground. You have to see one to understand it. A real HV fuse can be used or made. Installing a SB fuse in the filament CT wont stop most of the damage, if any, since it is so slow. A surge resistor that limits the current until the breaker on the rear panel opens is a basic defense that works and the resistor should be able to absorb the energy and survive.

    Ive measured around 3dB in several tries. The L4B uses a very low DC resistance RFC at the grid pin to ground; it does little for IMD and adding a 10 Ohm resistor in series with the drive at the filament pins will provide the needed NFB.
    Direct grounding the grids AND THEN adding that resistor brings the gain back to original; Ive nothing against doing it that way.

    Last edited: Oct 24, 2014
  9. AC0OB

    AC0OB Platinum Subscriber Platinum Subscriber QRZ Page

    I think the OP wants to get his rig running ASAP.

    I too, have a 1 mH choke if he needs one. Email good on QRZ.

    Maybe another thread debating the merits of or the critique of various changes or upgrades of the SB-220/221 could be opened.



    AC0OB - A Place where Thermionic Emitters Rule!
  10. KA9JLM

    KA9JLM Ham Member QRZ Page

    That is great that you offer Phil. Is your Avatar hosed for a reason ?

    Nice post Carl.

    I wonder what the last band operated was when the smoke got out.

    Keeping some things original is a nice thing if you can. After all it lasted a few years.

    Have Fun Everyone. Discharge those caps and the fun last longer.
Thread Status:
Not open for further replies.

Share This Page