EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: Carabus on November 04, 2021, 09:15:58 pm
-
Hey.
I made a flyback driver with 555 timer that works great. I want to play music on it. My best result is music plays fine, but when there is no music signal the arc is still present with high-frequency noise.
How would I make it so that when the music stops arc stops as well? My thought is that pin5 picks up noise when the signal is low. I tried to puta pull-down resistor, but the arc is smaller but still there producing sound I don't like.
My audio gets amplified and output goes through 100nf capacitor to pin 5 of the NE555. I want pin5 to be at 0v when there is no music.
Maybe some sort of filter?
Any ideas?
-
Part of the problem is probably that your driver frequency is too low, producing too much audible high frequency sound. Iirrc, commercial H-iFi ionic tweeters used around 27MHz as their fundamental frequency.
Unfortunately Pin 5 of the 555 simply leads to a tap on the internal resistor chain that sets the thresholds of the comparators, yes it allows PWM, but only over a limited duty cycle range. You will need to come up with another strategy to disable the oscillator when there are musical silences - one that detects actual silences, rather than just zero crossings in low frequency audio signals.
A video/audio of the results you have achieved so far would be interesting.
-
Maybe something like 2kOhm resistor from the signal to capacitor would work. When the signal is low but noisy, resistor could drop the voltage down and there would be 0v at the pin5
-
If I remember rightly, I don't have time to look it up on the data sheet at the moment, pin 5 provides pulse position modulation. The length of the pulse stays the same, but the time between pulses, varies depending on the voltage on pin 5. It's similar to PWM, but the frequency varies, as well as the duty cycle.
There's no straightforward way to do what you want. If you bias it so pin 5 is at zero, when the input signal is zero, the signal will be half wave rectified, which will completely mangle the music. You need to add another circuit which looks at the peak voltage of the music signal and pulls pin 4 (reset) low, if it's under a certain level.
-
Ofcourse I will post video tomorrow.
-
I agree.
-
Maybe p mosfet would work for that?
If the voltage of the signal is above threshold of the mosfet it turns on and doesn't pull down reset pin to ground. And when voltage is lower misfet would be off making path from reset to ground.
-
If I remember rightly, I don't have time to look it up on the data sheet at the moment, pin 5 provides pulse position modulation. The length of the pulse stays the same, but the time between pulses, varies depending on the voltage on pin 5. It's similar to PWM, but the frequency varies, as well as the duty cycle.
Ah yes, you're right, the 555 only manages PWM if configured as a monostable and fed from an external clock source (Section 9.2.2 / Fig 18 of the TI NE555 datasheet)*. With a single 555 in astable mode it will only do PPM (section 9.2.3).
It might be worth adding a second 555 timer to achieve PWM modulation, and maybe improved audio quality(?).
EDIT: * https://www.ti.com/lit/gpn/ne555 (https://www.ti.com/lit/gpn/ne555)
-
Maybe p mosfet would work for that?
If the voltage of the signal is above threshold of the mosfet it turns on and doesn't pull down reset pin to ground. And when voltage is lower misfet would be off making path from reset to ground.
Hard to say. Using the reset pin is the logical way of stopping oscillation, but whatever circuit you use to control it must have a time-constant that is longer than the wavelength of lowest audio frequency.
-
If I remember rightly, I don't have time to look it up on the data sheet at the moment, pin 5 provides pulse position modulation. The length of the pulse stays the same, but the time between pulses, varies depending on the voltage on pin 5. It's similar to PWM, but the frequency varies, as well as the duty cycle.
Ah yes, you're right, the 555 only manages PWM if configured as a monostable and fed from an external clock source (Section 9.2.2 / Fig 18 of the TI NE555 datasheet)*. With a single 555 in astable mode it will only do PPM (section 9.2.3).
It might be worth adding a second 555 timer to achieve PWM modulation, and maybe improved audio quality(?).
EDIT: * https://www.ti.com/lit/gpn/ne555 (https://www.ti.com/lit/gpn/ne555)
Right, so the negative pulse length stays constant and the on time varies with the control voltage.
I'm not sure if it's worth going to the trouble of adding another 555 timer. An arc is never going to produce excellent sound quality.
I've done a quick simulation with a triangle wave converted to PPM by a 555 and going into a low pass filter. It doesn't look too bad. I set R1 and R2 so the output duty cycle is 50%, when the control voltage is half the supply and set the triangle wave to go from 2V to 8V, with the 555 run off 10V. It seems to be more linear, when the input signal is at lower voltages, as the capacitor charging curve is straighter there. Biasing the control voltage signal closer to the negative rail would make it more linear, at the expense of less modulation, so not so much sound. A triangle wave generator and comparator would give a linear PWM signal, but it probably isn't worth the effort.
(https://www.eevblog.com/forum/beginners/pin-5-of-555-timer/?action=dlattach;attach=1316378;image)
-
Part of the problem is probably that your driver frequency is too low, producing too much audible high frequency sound. Iirrc, commercial H-iFi ionic tweeters used around 27MHz as their fundamental frequency.
Unfortunately Pin 5 of the 555 simply leads to a tap on the internal resistor chain that sets the thresholds of the comparators, yes it allows PWM, but only over a limited duty cycle range. You will need to come up with another strategy to disable the oscillator when there are musical silences - one that detects actual silences, rather than just zero crossings in low frequency audio signals.
A video/audio of the results you have achieved so far would be interesting.
I am happy to share my progress on this project.
https://www.youtube.com/watch?v=KiMtt7rQssY&ab_channel=ZygimantasT (https://www.youtube.com/watch?v=KiMtt7rQssY&ab_channel=ZygimantasT)
-
Hey, that's not bad for such a simple circuit, especially as you are trying to do it with a short direct arc and flyback rather than the more 'diffuse' discharge from the top of a tesla coil.
Looking at "tesla music" clips on YouTube, the source music tends to be fairly carefully chosen for bold tones and simplicity (techno?). franzolielectronics looks like the guy you want to be trying to emulate though... https://www.youtube.com/channel/UCLsUeooHSJlE_5xL4BPQIOQ (https://www.youtube.com/channel/UCLsUeooHSJlE_5xL4BPQIOQ)
From Zero999's simulation, the KISS approach seems to be the best way to go, together with trying to optimise the discharge electrodes maybe.
-
I think I will add Zero-cross detector that franzoli uses in his flyback driver.
-
I will try making a tune with triangle waves, then amplifying it before 555. As my results go the music sounds much better when the melody is simple and one note is played at a time.
-
It seems to go quiet between notes, in the video.
How about applying the AC coupled rectified audio to pin 4? That way pin 4 is only pulled-up, when there's a signal.
(https://www.eevblog.com/forum/beginners/pin-5-of-555-timer/?action=dlattach;attach=1316654;image)
-
The thing is when audio is coming from my phone the noise in between notes is low meaning no arc, but when it's from my computer the noise is quite significant