CTR AL811 User manual

1
Upgrade Kit AL811 and AL811H amplifier
Information in this modification manual is copyright 2021 by W8JI and CTR Engineering. I understand people like to
share things, but please remember we have several weeks’ labor composing this document. Please consider a
PayPal gift to [e-mail] when using these instructions if you do your own kit.
Thank you for purchasing this kit (and/or modification manual). These kits and this technical manual are intended
to help fellow Hams understand failures and reduce expensive damage. This technical instruction manual and kits
bring 811 series amplifiers up to latest revisions. With basic hand tools and soldering skills anyone should be able
to upgrade or service their own amplifier.
Please order via email links on https://www.ctrengineeringinc.com/ameritron-811-811h-upgrade-
kits/
or email order query only to [email protected]
This modification manual is also the result of feedback, questions, and suggestions by fellow Hams. More
suggestions are always welcome. The overriding philosophy is everything reasonably possible should be done to
prevent expensive or annoying damage, especially to expensive radios. An amplifier should work as well as
possible for the cost.
These kits come in two basic forms with one add-on:
(Power Supply rebuild) 811KS
1.) four better sized higher-voltage much longer-life 5000 hours 105c rated electrolytics
2.) four improved bleeder/equalizer resistors
3.) one large 6A grid meter protection diode
4.) four 499K 1% 1kV rated two-watt temperature stable meter multiplier resistors. These resistors also correct a
slight meter error
(Protection Kit) 811KP
1.) three application tested GDT tubes
2.) one large 6A meter protection diode
3.) one 100k 2-watt resistor
4.) two 3.9V 5-watt Zener diodes
5.) 10-ohm 9-watt CCS 10kV rated fault resistor
Add on part:
811R200K
200-ohm 25-watt non-inductive TO-220 case load resistor kit with hardware
Overview
I designed the AL811 series many years ago. It was intended to be a minimal cost reasonable power level SSB and
CW only amplifier. The initial design request was for just two tubes, but that obviously would have been
inadequate. As a result, I added a third tube as a bare minimum cost amplifier, and pushed to have a better but
still not too expensive four-tube amplifier. The better version became the AL811H.
The initial schematic circulating for these amplifiers was hand drawn on a drafting table by pencil around 1990!
This instruction manual and the 811 kits are the result of direct feedback from servicing several dozens of
amplifiers. Read these instructions thoroughly before you start working on your amp. Make sure you understand
every step to avoid errors or omissions.
AL811 Stability
Tubes with long thin grid leads and large internal structures sometimes have significant anode-to-cathode
feedthrough. This can destabilize amplifiers at upper HF. 811 and 572 tubes are particularly bad about this.

2
The three-tube 811 model is not neutralized. As with two unneutralized 572B tubes, three 811A tubes are
marginally stable at upper HF. The AL811 should have a 200-ohm cathode swamping resistor, or even lower
swamping resistance. Kit PN 811R200K has one chassis mount RF-rated 25W resistor, one 0.01uF coupling
capacitor, and necessary hardware. In addition to standard hand tools, you will need a 0.113 to 0.125” (1/8”)
drill to add one or two holes for 4-40 machine screw clearance.
AL811H
The four-tube model is neutralized. Four 811 tubes, regardless of layout, become unstable someplace above 15
meters and behave poorly on ten and fifteen meters when not neutralized. Any swamping resistance is optional;
it just makes neutralization less critical and closer matches drive level to 100W exciters. With bias changes the
swamping is not required.
NOTE: I find many amplifiers that have NOT been properly neutralized from the factory. This is not from a variation
in tubes. The only time tubes might require neutralization is when 572B’s are swapped in. To equipment operators,
poor neutralization is most easily noticed by a fairly large 10-meter input SWR variation as the amplifier is tuned.
A second neutralization indicator is exciter power varying as the amplifier is tuned past 10M grid current peak.
A third and by far least dependable identification method is observed by plate current dip not coinciding with
maximum output power or maximum grid current.
Neutralization
I neutralize 811’s by activating the relay system with an external low voltage dc supply of about 14-18 volts
positive applied to where meter lamp positives connect. The antenna relays are activated by grounding the
RLY jack. Feed an antenna analyzer into the RF input or output jacks, and listen with a radio on the other port.
With the amplifier on 10 or 15 meters adjust the analyzer and receiver to the same frequency. Adjust the
PLATE and LOAD for maximum signal. Now you can safely vary flapper plate positioning for minimum signal
level. The best flapper setting is the compromise between 10 and 15M nulls, with 10M having priority.
Caution: Do not neutralize before grounding grids directly to chassis at sockets! The amplifier must be
functioning normally with grids properly grounded to neutralize.
Arc Major Concerns
There have been dozens of instances of hard faults in tubes damaging radios. This has been blamed on many
things by many people, including ALC voltage or relay line spikes. The actual damage to radio RF sections is from
gas or debris arcs in tubes, or from tube anodes warped so badly by excessive heat the anode contacts the tube
grid. Radio damage comes from the tube arc energy making it out the RADIO coax connector, through that coax,
and into the radio. I’m 100% positive of this! While this happens infrequently, radio damage can be anything from
minor to catastrophic.
Tube arcs range from minor flashovers that clear the fault low energy to hard solid faults that dump full high
voltage on the tube cathode and grid. In light faults, with a directly grounded grid, the grid is able to divert most of
the arc energy harmlessly to ground. Only the most sensitive radios are damaged by minor flashovers.
Warning: No matter what you see other places, do NOT run long grid leads
to a common ground. The longer and thinner grid lead path to the chassis,
the more feedthrough and less stable the amplifier will become. Grids
should be directly grounded to the chassis with the shortest possible leads.

3
With hard faults an ionized path occurs through and around the grid. This transfers a great deal of energy to the
tube filament. These arcs can be devastating to amplifier components and to the radio. Hard anode-to-filament
arcs produce a very sharp edge high voltage spike on the filament. The filament, if unprotected, reaches well over
1000 volts in microseconds. This arc is very much like a miniature lightning surge. The filament choke has
significant common mode series impedance. Along with bypass capacitors this forms a low-pass, only passing arc
frequencies below ~100 kHz through the filament winding to the bias and relay circuit. It takes a very hard
sustained arc to damage this path.
The second more troublesome path is through parallel 0.01 µF filament coupling capacitors to the tuned input. The
tuned input circuit, being a low-pass, rings. This stretches the pulse out into a several mS long series of lower-level
oscillations below the band selected. The pulse can then make it through the tuned input much like a brief high-
power transmitter of a few hundred watts into the radio. Worst case peak voltage level is an arc while on ten
meters, while worst case pulse length is on 160 meters.
The red line is the primary damaging high frequency arc path. The path goes through tubes, out through the
filament coupling capacitors. It then travels through the tuned input circuits to the RF input, and that is where your
radio is connected. The arc path is NOT out through the relay control line; a relay line isolation device will not
prevent or reduce arc damage. The arc path is NOT out through ALC; using or not using ALC will have no effect on
radio damage.
This kit does the following:
1.) Adds two “150 volt” Gas Discharge Tubes from filament directly to chassis at the tubes. The Littelfuse gas
discharge tubes (GDT) surge arrester devices protect personnel and equipment from damaging high
Figure 1 primary arc path

4
voltage transients induced by lightning, inductive switching, or electrostatic discharge. This limits
transient to about 350 volts absolute peak. Any substitute GDT’s should be vetted for performance.
GDT’s do not clamp. GDT’s ionize at a certain voltage, becoming a near-short once ionized
2.) Instructs you to directly ground grids in older version that had grid equalizing resistors. This is a critical
update. The grids should always be directly grounded. THIS IS A MUST DO in any grounded grid amplifier
that floats grids from chassis.
3.) Increases or adds bias. Bias improves tube life and efficiency without noticeably hurting IMD with 3.9 volt
5 watt Zener bias diodes. If your amplifier has a string of bias diodes, you add only ONE of these diodes in
series with the white center tap. If your amplifier does not have a string of bias diodes on the input board,
two diodes are used. The goal is to be around 10-20mA quiescent current per tube. This would be 30-
60mA zero signal keyed idle current in the AL811, and 40-80mA idle in the AL811H
4.) Add a higher surge rated negative rail clamp diode. This diode protects the grid and plate meters and
shunts. The original diode is a small 1N4001 series diode. The kit diode is much stronger and fails far less
often. This reduces the aggravation of protection diode failures
5.) Add a 100k resistor from filament to ground, or to a place I provided in later AL811’s that Ameritron for
some reason never used. This prevents a weak “popping” noise in your receiver from intermittent gas
tube ionization with some GDT’s and tubes
6.) Supplies an additional GDT if you want to further clamp the RF input line to the input circuit
7.) You should remove MOV’s if they are defective or if they are readily accessible. MOV’s do not add
anything worthwhile in protection and MOV’s are failure prone. They are too far downstream and
beyond the choke to do any good for your exciter. If MOV’s fail, they can cause problems
The AL811 three-tube model normally does not require dropping the back panel unless it has defective MOV’s to
service. The 811H back panel almost always has to be dropped. This is not a big deal if your particular unit was
wired correctly from the factory. It is a little more work if the transformer wires have been cut too short.
Always make sure the amplifier is unplugged and the capacitors are discharged before starting
this modification procedure. Never attempt to make any repair on the amp with it powered up
and the cover disconnect overridden. There are LETHAL VOLTAGES present inside the amplifier.
In all cases be careful and work slowly.
811PK
All common grounded grid amplifiers should have grids directly grounded with the shortest possible leads.
811H amplifiers with the terminal strips and resistors, as shown below in figures 2 and 3, must be upgraded
for safety.! Use of grid resistors, long grid leads, or grid chokes increase chances internal tube arcs might to
continue to the filament and out to the radio. A hard tube short will almost always damage or open the grid
resistors, causing operational issues.

5
Figure 2 remove resistors
Figure 3 remove resistors
If you plan on dropping the back panel, do this shaft step first. The AL-811 three-tube does not require
dropping the back panel just to install GDT’s or do most changes.
GDTs in the AL811 three tube are installed as shown here:

6
Figure 4 GDTs remove outer tube socket add lug(s) and GDT AL 811 only
Remove the input switch shaft by ONLY removing the very front screw of the coupling! The front screw will be
a small hex or Philips’s screw. Remember to remove the front screw ONLY! If you remove the rear screw you
will have to re-index the rear switch wafer. Refer to Fig. 4.
Figure 5 band switch shaft
Remove the side rails on each side. (Fig 6)
Remove the screws and pull the rear panel off the shaft. Do not damage or rotate the blue plastic input switch
on the input circuit board. Once you have the panel back and off the shaft lay the panel down as in Fig. 6. You
can now remove the shaft from the chassis.

7
Figure 6 remove siderails
To remove the tube chassis and work underneath it freely, the back panel must be dropped. Wiring in
amplifiers unfortunately is sometimes inconsistent. Properly wired the back panel will drop just like Fig. 7, only
removing small screws in side rails and the rear bottom panel lip. This is a properly wired panel, which saves
you considerable time. Correctly wired panels have long transformer leads that allow rear panels to drop
without removing any wiring, with the exception of output coax:
Figure 7 properly wired back panel dropped
Note: Incorrectly wired panels will not drop without unsoldering multiple transformer wires, and perhaps coax
and other leads. It’s advisable to take a picture of the panel with wires attached to aid in reassembly of the
panel. Be careful unsoldering leads. Leads are accessible through a plate on the back of the back panel. Use a
large iron, and have a solder sucker or solder wick handy! Use a large hot solder-wetted tip on the wire ends
that stick through the boards, so you can get on and off the solder joint quickly. Do not linger and overheat the

8
printed circuit board foil while waiting for the wire to melt free. You do not want to lift foil pads off the boards
from excess heat. Heat the wire end with a wetted tip flat against the wire, do not heat the foil.
This is an incorrectly wired panel. The transformer leads are too short to allow the panel to fully lay down and
must be unsoldered for input board or tube chassis service. Figure 8.
Figure 8 wires too short
After rear panel removal, remove any MOV’s. They will be located on the circuit board right down by the fan.
MOV’s are sometimes blue. They will be present along with capacitors. Refer to Figs. 9, 10, & 11 to help
identify MOV’s.
Figure 9 Capacitors and MOVs

9
Figure 10 capacitors
Figure 11 MOVs
Remove MOV’s, do not remove capacitors.
Generation I board:

10
Figure 12 Generation I input board
The Generation I board does NOT have the 24V bias point for the 100K resistor. In these amplifiers the 100K 2-
watt resistor goes from anyplace on the filament system to chassis ground.
Gen II board:
Look for the small capacitor near dual relays (Fig. 13) and remove it, if installed. The string of 6 large forward
connected rectifier diodes adds about 4 volts of bias.
Figure 13 Gen II input board

11
Gen III board around 2012
This board das GDT pads and does not have the problem relay line capacitor near the relays.
Note: this generation is after I recommended omitting all MOVs. Production boards may or may not have
MOVs
Figure 14 Gen III and later input boards
This is the ideal working position for all back panels. This is a correctly wired back panel. The back panel easily
slips backwards off the switch drive without any wire removal. This is a two-relay board (the two black relays
to the right side). Notice the string of rectifier diodes used for bias running along the black filament choke.
Figure 15 All units rear panel down

12
The rear of the input switch shaft is exposed. The sharp shaft edges should be filed or sanded to break the
edge into a taper or radius. This significantly helps slipping the shaft into the rear switch. A few minutes
tapering the edges can prevent switch damage and prevent expensive broken switches.
Figure 16 All units break or radius shaft rear edge
Remove the tubes, the terminal strips, and lugs in these locations, and disconnect the two orange 1000 pF
high-voltage blocking capacitors that go between the RF parasitic board and plate choke. Unsolder these
capacitors at the end that solders to circuit board above the Plate Tune capacitor. See figure 17
Figure 17 811H only unsolder and unscrew

13
Flip the unit carefully on its side and remove the following screws:
Figure 18 811H remove bottom screws
Lay amplifier down and maneuver tube chassis to reach all sockets. Depending on wiring, the filament leads or
red HV choke lead could require removal. The filament leads are supposed to be loosely twisted as in Fig 18.
The optional 200-ohm 25-watt resistor requires drilling a #4 clearance hole, deburring the hole, and using
heatsink compound and proper hardware (4-40 machine screw with lock washer and nut) to mount the
resistor. The resistor connects across the neutralizing capacitor. Fig. 19 Ground the tube grids directly using
solder lugs provided in the kit and solder the GDT’s in Fig. 18.
Figure 19 add GDTs AL811H only (811 was fig 4)
Check all the solder connections carefully. Inspect the plate choke. If the plate choke has loose or overlapped
wires it needs to be put back in a single tight wrap and “re-glued”. Q-dope can be made from a mix of lacquer
thinner or Xylene and pure white Styrofoam “peanuts” used for packing. Dissolve the white packing peanuts

14
or other pure Styrofoam (do not use anti-static peanuts) in thinner until it is a thick paste. Paint the paste on
the coil to hold the windings in position. Allow time to dry. A dot of Gorilla Glue is fairly good end-glue at
winding ends.
Figure 20 reglue choke all units
If your amplifier has tan or sand color bleeder resistors they should be changed. The sand or tan bleeders
have proven to be failure prone. When they fail they ruin all four electrolytic filter capacitors. A 50k to 100K
higher reliability resistor of suitable wattage must be used. I include suitable resistors and superior filter
capacitors in the 811KS kit.
The small meter protection diode on the very center far right below the ferrite bead gets changed to a larger
diode that is supplied in the kit with all kits.
Figure 21 Replace sand color bleeders. Replace diode.

15
Figure 22 Metering and power supply
D16 can be mounted any physical way you like so long as the diode anode is at the center pad
and the striped cathode end is toward the rectifier bank and filter capacitor bank negative. This
diode allows the any arc energy to get from the chassis to the capacitor bank negative,
harmlessly bypassing meters. There is plenty of room to solder. The large rectifier diode in this
kit replaces D16. D16 is located near the tank coil. The kit diode has thick leads. It can lay-solder
on top of the board. It does not need to fit through holes. Refer to Figures 21, 22, and 23.
Figure 23 Meter and HV Component Locations

16
For easy soldering, pre-form D16 leads as shown. Also pre-tin them:
Figure 24 D16 form leads
Zener Diode Bias
There is no question that using as much positive cathode bias as possible, within limits of IMD,
makes the tubes run cooler. These graphs show bias means more than drive power to tube
dissipation:
Figure 25 Drive vs. Dissipation for various conditions

17
Kit 811KP includes two Zener bias diodes. These diodes are conservatively rated when installed
in the airstream of the fan. The 811KP Zener diodes connect in series with the transformer
white center tap lead. The banded or striped diode end connects toward the transformer Fig.
27, since they are Zener diodes. Each Zener diode is rated at 5-watts and adds 3.9 volts of bias.
They are safe to 1.3 amperes continuous duty when in open air.
I suggest you use one Zener diode in addition to the six rectifiers used by Ameritron for a total
of about 7.3 volts, or two Zener diodes in series if your amplifier does not have the bias rectifier
string.
There will not be any noticeable increase in IMD from the additional bias, and the tubes will run
a lot cooler. Please look at this dissipation graph. Anything over 65W is over the tube rating.
The diodes can be soldered in by cutting the white wire near the fan. Strip both ends, slide large
heat shrink over the white wire well away from the diode(s). Make hook leads in the diodes.
Crimp and solder them to make connections. Use either one or two diodes as required.
Figure 26 preparing and connecting Zener’s
This kit also includes a small 100K 2–3-watt metal oxide resistor. There are multiple points this
resistor can be connected; without a great deal of operational change. This resistor can go
from any tube pin to chassis or from any point on any filament wire to ground. You can even
attach it to the Zener mod as shown, since this is a convenient place. This resistor is necessary
to keep GDT’s from “ticking” or “clicking” on receive on some bands with some tubes. The 100K
resistor’s position on the Zener is meaningless so long as the other lead can reach a ground
point. The resistor can also go underneath the amplifier tube chassis, or on the input board.

18
In order my preferred places would be Fig. 28 or 29 and lastly just to chassis depending on the
board’s generation.
1.) Between a +24V pick point and the filament center tap or filament lead (shown below in fig
28 and 29)
2). Near the filament CT (white lead) that is accessible, such as on the Zener’s
3.) Between a filament pin and ground on a tube socket
This would be a suitable alternative resistor location on the Zener’s prepped for installation
when the resistor is not in the preferred location. (Remember one Zener for amplifiers that
have a string of bias diodes.)
Figure 27 Zener's with 100k (avoid dropping back panel or early Gen I board)
There was an intention in revised boards to supply +24 to +30 Vdc as a standby bias offset
reference. There is a pick point for this reference on the later boards. The 100K 2W resistor
connects between that 24+ point and any nearby easiest accessible point of the filament
winding.
The tap point I provided for bias or accessory use can be identified by looking for two small
rectifier diodes near one of two small electrolytic capacitors. If you have the back panel down
Warning: Ameritron has some transformers that are out-of-spec on the LV control
winding. The LV control winding was specified to produce 12-14 Vdc output from a
capacitor input filter system. Due to transformer errors, so later unit can have DC
voltages almost 50% beyond the specified voltage. I am trying to get an idea how
many bad transformers are out in the field. Please contact me if you have one.

19
and can access these points, they are a slightly better point for the 100K resistor. This is NOT a
mandatory point; it was just an intended production point when we were having problems with
some early Chinese early 811 tubes.
Figure 28 100K 2W resistor location A
Early two-relay board with +24V point for resistor
Figure 29 100K 2-watt resistor location B
Figure 30 Zener location. Note Zener band is toward the transformer, diode is in the cooling fan
air stream. This unit has the 100K resistor elsewhere:

20
Figure 30 Zener location in fan airstream
811KP Surge Kit Glitch Resistor
The large resistor in the surge kit is a special 10-ohm 10kV+ surge rated resistor. It replaces
the small ferrite bead wound with wire.
It adds an additional ten-ohms of surge resistance over the ~8 ohms of the RF plate choke
and ~4 ohms of the filter capacitors. This resistor provides almost 50% reduction in worst
case current surge.
The small ferrite core choke must be removed to install the surge resistor, since that
resistor’s inductance replaces the ferrite choke in RF function.