Formatting transmitting tubes

and associated questions

You need to format or "burn in" old transmitting tubes that have been on the shelf for some time in order to avoid arcing and the risk of destroying them. You can find descriptions and procedures on how to do this on several places on the Internet. SM5BSZ among others. This is my way of dealing with this process.

I use the old HAIR-rule: Heater - Anode - Idle - RF . Start with heater voltage only, then apply anode voltage, after that run idle current in the tube and finally apply RF to the tube. This rule has been successfully used on many different tubes over the years; 4CX250B, 4CX350A, GS23b, TH308, TH328, TH338, YD1277 etc.

Formatting process

Take the following steps in order to format transmitting tubes that have been on the shelf for a long time:

  1. Run the tube with heater voltage only for at least 24 hours. You may start with lower heater voltage than specified for a few hours (may be disputed due to the risk of poisoning the cathode). NOTE! During this and all following operations, full cooling air flow shall be applied to the tube, even to the cathode area as specified/recommended by the tube manufacturer.
  2. Apply then HV to the tube with series resistor (SM6EHY recommends 100 k). I use different resistors depending on the tube. 33 kohm is a good start. Start with a low voltage, say 500 V and work your way up in 300 to 500 V increments until you have reached the voltage level you will use in normal operation. Stay at each level for half an hour to an hour. For tetrodes this step shall be done without screen voltage applied.
  3. Connect the high voltage line without the current limiting resistor to the tube. For tetrodes also apply screen voltage, you may use a reduced value (say 70 % of normal). Start again at 500 V anode voltage and adjust the bias to a small anode current (say 50 mA). Let the tube run for an hour. Increase the anode current to say 100 mA (depending on the tube in question). Run for half an hour. Reduce the anode current to zero again and increase the anode voltage by 300 to 500 V. Adjust the bias for 50 mA anode current and run for half an hour. Increase the anode current to 100 mA again. Continue this procedure until you are close to your desired operating anode voltage. Increase the anode current to a higher value and let the tube run. The anode of the tube shall be warm. NOTE! DO NOT touch the anode to check. Do not even think about it. Measure the outlet air temperature instead. The outlet air temperature shall increase considerably, but make sure you do not over heat the tube. I usually go up to 50 degrees C of air outlet temperature. I measure the air outlet temperature by the use of the outdoor sensor of a regular indoor/outdoor digital thermometer. 
  4. Decrease the anode voltage again to a moderate value, say 60 % of the desired operating voltage. Adjust the bias to a low but proper anode current. Apply a small amount of RF, increasing the anode current slightly. Run for a few minutes. Increase the RF drive somewhat to give an slightly hinger anode current. The RF may be pulsed or why not send "TEST DE YOUR CALL" in CW. Increase the anode voltage in moderate steps, say 100 to 200 V t a time and run for some time. Repeat until you are close to your desired operating voltage. This may take a few hours to get there. Monitor the outlet cooling air temperature again in order to have a check on the anode temperature. Keep it at about 50 deg C. Run the tube extensively at this reduced power level on for instance local rag chaw QSO's for some time, the more the better.
  5. You should now be ready to run the tube at full data.
HSP series resistor exterior  Limiting resistor box with MHV connectors.

HSP series resistor interior Inside the limiting resistor box.

Post formatting considerations

Even after a well performed formatting process occasional arcing can occur in a transmitting tube. Some people claim that a transmitting tube will adopt to the actual working voltage and by that give occasional arcs even at the normal operating voltage. I have not been able to confirm this thesis, but some unexpected arcs may be explained by this idea. In order to limit the consequences from such an arc some precaution can be taken.

I have had arcing several times in different types of tubes and PA's but have not yet had one tube destroyed from the arcing. The three phase high voltage power supply I am using has about 10 uF capacitance and an transformer capable of delivering  >4.5 kVA (about 45 kg of iron). Fuses on both the incoming three phase (3 x 400 V) and on the outgoing high voltage line. The incoming fuses have never opened up, but that have the fuses on the high voltage line. In the high voltage line is also a low inductance current limiting resistor of 100 ohm with a power rating of 100 W.

It seams like if you have a current limiting series resistor and well fused you can limit the devastation from an arc in the transmitter tube.

Heater voltage considerations

A transmitter tube is sensitive to the applied heater voltage. You should always follow the recommendations of the manufacturer closely. Running a tube on a too high heater voltage will shorten the life of the tube. Running the tube on a too low heater voltage can poisoning the cathode. This is especially true for the Thoriated Tungsten cathode type.

Running a transmitting tube on UHF close to it's upper frequency limit may cause back-heating of the control grid and by that emission from the grid. This can be harmful to the tube. In order to avoid back-heating you can slightly lower the heater voltage. Many manufacturer of tubes have clear recommendations of how much to lower the heater voltage.

One way of finding the optimum heater voltage for a tube is to run the tube with a standing current (I.E. Ia =100 mA) and vary the heater voltage while monitoring the change of the standing current. You will find that the standing current will vary with the heater voltage in an non-linear way. By finding the "knee voltage" and using a heater voltage just above the knee you will get the most operating time out of the tube without sacrifice performance.

It is an good idea to limit the rush-in current when applying heater voltage to a cold tube. Connect a series resistor on the primary side of your heater transformer and short it out after 5 to 10 seconds of heater time. Using a variable resistor in the primary circuit of the heater transformer gives the possibility to adjust the heater voltage. If the heater transformer is used for generating other voltages as well it may not be possible to vary the resistance in the primary circuit. In this case you need to have the resistors in the secondary circuit of the heater transformer. Always let the tube pre-heat the specified time before applying anode current and RF to the tube.

A too high heater voltage can induce arcing in the tube. This may be a effect of back-heating.

Transmitter tube cooling considerations

Adequate cooling of a transmitting tube is essential. Always follow the manufactures recommendations about the required amount of cooling air. More cooling air than recommended is just fine. But, how to know how much fan is enough?

In the data sheet of many transmitting tubes you can find graphs on back pressure versus air flow as well as dissipated power. The same can be found on many fans. By checking these against each other you can find out if the fan is enough for the application.

There are three major fan types; radial, diagonal and axial. The radial fans have a 90 degree angle between air inlet and outlet and gives in general the best back pressure. The axial fans have a 0 degree angle between air inlet and outlet and does not give a high back pressure. The diagonal fan
have a 0 degree angle between air inlet and outlet but a different angle on the rotor blades compared to the axial fan. The diagonal fan gives in general a better back pressure than the axial fan.

It is also very easy to measure back pressure by using a non opaque plastic tube. Form the plastic tube to a U-shape and mount it on a stand. Fill the U with water so you get the two water surfaces at a convenient hight. Attach one end of the plastic tube to the high pressure side of the tube. Measure the hight difference of the two water surfaces. This measure in mm is the back pressure in mmVp.

You can find many different units for both back pressure and air flow and it is not always easy to convert between them. Here is a small guide to these conversions:

10 mmVp = 1 mbar = 100 Pa = 0.39 inches of water

1 m3/min = 60 m3/h = 1000 l/min = 35.3 cfm

For my 23 cm PA I am using the below combination of fan and tube. For 12 V of fan voltage (at low speed dashed curves) I measure 15 mmVp that corresponds to an air flow of 150 m3/h and at 11 V (full speed solid curve) 28 mmVp corresponding to ~200

Bosch fan   Air flow versus back pressure for Bosch BPA 12 V radial fan (part number 0 130 007 804)

 TH338 cooling  Air flow versus back pressure for Thompson TH338 triode

A simple way of measuring back pressure is to use this back pressure gauge. It uses physical constants to measure the back pressure and is by this self calibrating.

Back pressure gauge  Back pressure gauge

Connect the end of the plastic tube of the back pressure gauge to the back pressure test outlet of the pressurized anode compartment and read the difference between the two water surfaces with the fan running. This is the back pressure in mmVp or in inches of water if you prefer that.

Back pressure test outlet Back pressure test outlet


SM5BSZ home page:

About tubes / valves in general:

About "getters":


[1] Care and feeding of POWER GRID TUBES, Laboratory staff , Varian, Eimac Division, 1967, Library of Congress No. 67-30070

Updated July 27th, 2009.