Musical jacobs ladder mark 2

[boylesg]

I am at the stage where my circuitry is stable and robust, but it seems the tv flyback transformers are failing some how.

They seem to work well for a short while with long arcs but gradually they become erratic and fail.

By touching the leads I can still draw an arc with them but it is much smaller than it was.

Any ideas as to what is failing inside the tv flyback transformers?

 

[BradtheRad]

Have you felt it get overheated? This could cause varnish to break down and have reduced insulating capability.

Have you heard buzzing or snapping sounds within? It could mean arcing (sparking) is occurring. Once it starts, it can deteriorate coil performance.

 

[betwixt]

Bear in mind that the transformer probably contains a rectifier which could have gone short/open circuit because you are drawing far more current than they are normally rated to carry and also that the transformers internal wiring has to withstand the same voltage that you see as the arc. Consider the spacing of the wires inside the transformer compared to the distance the spark has to jump outside it.

On a more musical note (forgive the pun) how are you applying the music to the arc? Have you considered that if the audio positive and negative cycles both drive the arc, you are doubling the frequency?

Brian.

 

[boylesg]

Bear in mind that the transformer probably contains a rectifier which could have gone short/open circuit because you are drawing far more current than they are normally rated to carry and also that the transformers internal wiring has to withstand the same voltage that you see as the arc. Consider the spacing of the wires inside the transformer compared to the distance the spark has to jump outside it.

On a more musical note (forgive the pun) how are you applying the music to the arc? Have you considered that if the audio positive and negative cycles both drive the arc, you are doubling the frequency?

Brian.

I am applying the music via the 555 ctrl pin. The ipod output goes through class A amplifier that amplifies the signal about x80. The sound is coming out a little distorted - it sounds like there might be some signal clipping going on.

I think I will try my opamp based amplifier as I have a pot on that with which I can adjust the voltage gain.

Doubling the frequency? I am not sure I get that. The oscillator runs at about 100kHz without audio. I would have thought the the audio signal would reduce the frequency so that an audible sound is produced in the arc. But perhaps my understanding of how the modulation is happening is not right.

I am going to experiment with using the tesla coil I am currently winding like a flyback transformer, i.e. run it not at resonance. If it works I will have HV AC and I can audio modulate it with the same circuit.

I tried winding my own flyback secondary the other day. But it is a messy job applying epoxy resin and when I extracted it from the former the capacitance was 1.6uF which means it is useless for audio modulation.

Without many turns on top of each other I would expect the capacitiance of a tesla secondary to be very low, which means it can run at very high frequency (beyond hearing range) as I understand it.

Tesla secondaries are far more practical and convenient to produce even if they are rather time consuming.

 

[betwixt]

I would imagine there is a lot of clipping going on!

The quality issue is more likely due to the fact that you are converting amplitude variations into frequency variations so what you hear is a kind of FM though a narrow bandwidth filter and therefore some semblance of amplitude being recovered. As an experiment, you could try feeding music into the reset pin instead. It would give an interrupted arc with very high distortion but might sound better!

My concern over doubling the frequency is not relevant, if you had been generating the arc by directly steppng up the audio to high voltage, you would get a 'flash' on both positive and negative half cycles so the audio tones would double the original frequency.

As a side issue, while experimenting, have you considered applying the technology to make electrostatic loudspeakers? I know someone who used a similar HV generator to provide a polarizing voltage to speaker electrodes. One electrode was sheet aluminum, the other was "heat shrink" secondary glazing (stuff from 3M you tape to a window frame then heat to shink it and make it pull tight) coated with graphite paint. It worked fantastically.

Brian.

 

[boylesg]

I would imagine there is a lot of clipping going on!
.

Problem was that I was trying to measure the peak voltage from the ipod with a multimeter. I thought I was getting of the order of tens of mV which would give a lot of room for voltage gain with a 12V supply. I was trying to bump the peak audio voltage up to a volt or two below the supply. But with a multimeter who the hell knows what I was really measuring.

It would be a whole lot easier to just have an adjustable gain opamp based amplifier. I built one with a LM358 opamp while I was playing around with speakers - I might give it a try with the HV arc.

The quality issue is more likely due to the fact that you are converting amplitude variations into frequency variations so what you hear is a kind of FM though a narrow bandwidth filter and therefore some semblance of amplitude being recovered. As an experiment, you could try feeding music into the reset pin instead. It would give an interrupted arc with very high distortion but might sound better!

My concern over doubling the frequency is not relevant, if you had been generating the arc by directly steppng up the audio to high voltage, you would get a 'flash' on both positive and negative half cycles so the audio tones would double the original frequency.

As a side issue, while experimenting, have you considered applying the technology to make electrostatic loudspeakers? I know someone who used a similar HV generator to provide a polarizing voltage to speaker electrodes. One electrode was sheet aluminum, the other was "heat shrink" secondary glazing (stuff from 3M you tape to a window frame then heat to shink it and make it pull tight) coated with graphite paint. It worked fantastically.

Brian.

Will have to read up a bit more about what you can do with the reset pin.

Electrostatic speakers huh. Haven't come across that on the net yet.

What do you think about my idea of using a tesla coil more like a flyback transformer? I have read that you get flyback transformer size arcs if you don't have the things tuned properly.

- - - Updated - - -

The sound coming from the arc was slightly scratchy, particularly with the drum beats, and that means clipping doesn't it?

Apart from the slight scratchyness, the sound is also a little tinny I think you would describe it as. I presume that is the narrow band filter distortion you are talking about.

- - - Updated - - -

The 555 reset pin is rather interesting.

When I attempt a proper SSTC I could potentially use the 555 reset pin as an interuptor, with a little additional circuitry. Unfortunately the mosfet gate drivers I have dont have an enable pin so I can't interface an interuptor at my mosfet gate drivers as steve ward does.

Was thinking tonight I could also put some suitable transistors in CE series with the inputs or outputs of my TC4422s. The bases of those transistors would then effectively become enable pins for my TC4422s.

 

[betwixt]

The sound coming from the arc was slightly scratchy, particularly with the drum beats, and that means clipping doesn't it?

Apart from the slight scratchyness, the sound is also a little tinny I think you would describe it as. I presume that is the narrow band filter distortion you are talking about.

Well, the real problem is that you want a change in amplitude from it so produces 'normal' audio but from a spark rather than a loudspeaker. By using the control pin of a 555 you are changing the frequency, not the amplitude! Varying the control pin voltage makes the 555 oscillator change frequency but it's output amplitude stays almost constant. Try connecting a loudspeaker to the 555s output pin to hear what you are actually sending to the arc generator.

The reset pin will change amplitude but only as an off/on control, it starts and stops the 555 oscillator so again you will hear distored sound but proably much louder and you may preserve some of the tone content of it as well.

Good electrostatic loudspeakers will beat most conventional loudpseakers with ease. Because they respond to instantaneous voltage rather than changes in magnetic field they can have a frequency response theoretically down to DC and because they have no cone or cone support their diaphragm mass can be less than a conventional speaker which gives better HF response too. If you look at commercially made ones you will note they are extremely expensive, mostly because of the high voltages and safety issues in a consumer environment. I used to have electrostatic headphones which were amazing quality but unfortunately their power unit burned out beyond repair. I wish I had kept the headphones themselves, with todays technology I could have resurrected them again. Electrostatic technology isn't far removed from what you are doing, that's why I mentioned the topic.

Brian.