TWT amplifier standby growl/noise?

[CopperCone]

Kind of curious, I noticed some of my TWT amplifier units, they tend to growl/make tesla coil noises when they are in standby mode.

When I actually turn the tube on, they work fine.

Is this common in high voltage equipment? It almost sounds like arcing, but I don't see any signs of arcing. Kind of like a plasma globe noise. It's kind of unnerving (maybe this is a good thing not to be comfortable around them).

I had a peak inside of one when it was working/during calibration, but there are no obvious sources of this noise.

is there some kind of 'natural' arcing that occurs inside of a TWT when it is heated but helix current is low/off? the noise seems proportional to helix current. I can't look inside of the tube because it is inside of a metal enclosure

[4CX35000]

I'm not sure what your using the TWT on, but HT (High Voltage) circuits tend to be simple affairs from experience (High Powered Transmitters) which use a three phase supply with a 415 volt (Three Phase) transformer to whatever the HT output is. I think the highest I have come across was 20kv. The three phase output of the transformer is passed through a rectifier and large capacitor smoothing and little else other than simple protection as the voltage and power is simply too high to bother with regulation, so instead regulation is done on the bias stages of the amplifiers were using a buck and boost type circuit and this increase and decrease DC Bias voltage and therefore gain of the amplifier as required.

Flashovers tend to be obvious with smoke and smell, followed by failure on the equipment I have worked on, but on low power equipment it can be the case it would eventually burn out the area of flashover. So I suspect that is not the case.

To me it seems as if something else is still running in the background and your hearing eddy currents vibrating a transformer and the noise is resonating through the chassis of the equipment or a switch mode type PSU no longer under load and your hearing the high pitched noise they can often make when no longer under load.

[CopperCone]

well, it sounds different then a typical transformer hum. The only thing I have experience comparing it to is a plasma globe, and the noise goes away once the TWT power is enabled.

I would imagine this means that the noise is worse under a no-load condition. Which is different then what I am used to with a regular transformer, I found that kind of equipment actually makes more noise when it is pushing current through it.

Also, this equipment is vacuum tube based, so I don't think there is switching power supply circuits in it.

[4CX35000]

well, it sounds different then a typical transformer hum. The only thing I have experience comparing it to is a plasma globe, and the noise goes away once the TWT power is enabled.

I would imagine this means that the noise is worse under a no-load condition. Which is different then what I am used to with a regular transformer, I found that kind of equipment actually makes more noise when it is pushing current through it.

Also, this equipment is vacuum tube based, so I don't think there is switching power supply circuits in it.

Whose the manufacturer ?
What is it used for ?
What year do you think it was manufactured ?
What is method of powering everything up ?.

Without the above being answered it is difficult to know. But to explain further.

Some modern equipment uses switching power supplies on the post tube stages such as the oscillators, pre amps and so on. More efficient and easier to repair and replace faulting stages when something goes wrong by removing the faulting module and insert replacement. I understand modern TV transmitters work in this way and the tube stage is a quick swap device now using a single small IOT device (Induction Output Tube) for the power amplifier output stage. These amplifiers are a world away from the Klystrons and Tetrode equipment I worked with which required several hours downtime and tuning before use.

To me it seems as if the HT circuits are powered when you activate the TWT ?. If so it could be a spark gap flashing over until the HT power supply is loaded by the TWT and the voltage drops below the spark gap flash over level. A spark gap is a method of controlling high voltages of HT circuits and those things can sound like static cracking noise for DC supply. Some are enclosed devices as in the data sheet below and others can be two ball close together or two wires close to each. Or even a corona ring.

English Electric - Spark Gap data sheet
https://www.e2v.com/resources/account/download-datasheet/1032

https://en.wikipedia.org/wiki/Corona_ring

[CopperCone]

HP and logimetrics units do this. Only the HP one that works at a higher potential does this (the C band unit) the logimetrics is unknown, solid state, and S band.

i noticed the HP units < C band run at ~1kV, while the C band unit runs at ~2kV.

If I figure out the potentials of the logimetics units, which runs at 2-4GHz (s band), maybe we can say this phenomena only happens at potentials <<1kV. , but right now it seems common to a logimetrics unit, which runs 2-4ghz, and a HP tube unit, that runs 4-8ghz.

they both have line power and rf power switches. the HP one is well documented 49X series, made in the 60's or something, and the logimetirics I have NO idea. Can't find shit about it. Not sure if i want to take it apart.

I also have some hughes units that I need to take a look at.


I did not see a spark gap in the HP unit that I took apart... only some vacuum tubes, vacuum tube relays and vacuum tube voltage reference, plus a few transistors.

Ima look at the schematics (one of my HP S band amplifiers burned up recently.. unsure of where the smoke came from), so I will look for damage and try to see if I find evidence of a spark gap. I think pot for helix current was bad and I was running it in over drive for a while, then I put in the wrong sized fuse 

very much hope that the TWT is ok in that poor guy  , hopefully the testequipment reaper does not show up.

[4CX35000]

HP and logimetrics units do this. Only the HP one that works at a higher potential does this (the C band unit) the logimetrics is unknown, solid state, and S band.

i noticed the HP units < C band run at ~1kV, while the C band unit runs at ~2kV.

If I figure out the potentials of the logimetics units, which runs at 2-4GHz (s band), maybe we can say this phenomena only happens at potentials <<1kV. , but right now it seems common to a logimetrics unit, which runs 2-4ghz, and a HP tube unit, that runs 4-8ghz.

they both have line power and rf power switches. the HP one is well documented 49X series, made in the 60's or something, and the logimetirics I have NO idea. Can't find shit about it. Not sure if i want to take it apart.

I also have some hughes units that I need to take a look at.


I did not see a spark gap in the HP unit that I took apart... only some vacuum tubes, vacuum tube relays and vacuum tube voltage reference, plus a few transistors.

Ima look at the schematics (one of my HP S band amplifiers burned up recently.. unsure of where the smoke came from), so I will look for damage and try to see if I find evidence of a spark gap. I think pot for helix current was bad and I was running it in over drive for a while, then I put in the wrong sized fuse 

very much hope that the TWT is ok in that poor guy  , hopefully the testequipment reaper does not show up.

Looking at various websites it appears as if the HP is a low power amplifier for microwave (SHF) use, are these for lab experiments ?
Not come across HP amplifier before, but it could be the cooling fan your hearing buzzing away inside. Being honest there tends to be very little to go wrong inside and the big finisher for these things is component obsolence such as the TWT will no longer be manufactured. Even in communications were TWT was popular, these type of amplifiers were taken out of service well over 20 years ago as solid state is more reliable and cheaper.

[CopperCone]

yes, the HP units are 1 watt versions. I thought they were a good start for certain experiments. I am more interested in electrophysics then communications systems.

In one of them I found a open circuit resistor that was causing a problem. 

I also thought it might be harder to break in the lab environment, since the output is a tube. (due to error, not time). I thought an argument could be made that the tube devices are better suited for not breaking if mistreated, but the solid state devices are better if they are designed into an installation (or the experiment is more carefully planned).