Variable frequency 555 circuit

Does something here help?

Not really practical, sorry.

The 555 is basically two comparators and an SR latch. The output toggles low when the high threshold input is reached, and high when the input drops below the low threshold.

The charge/discharge time to reach the high or low thresholds is totally dependant on the external components.

The only internal "variable" is by changing the bias voltage on the resistor divider that the comparators use for the thresholds; the upper comparator point in the divider is brought out on pin 5.

That allows some change in frequency but at a guess possibly 4:1 or so, not 1000:1
The control voltage is not linear.

555 internal schematic:




To actually use a 555 over that 1000:1 range, you would need to select different external components using a bank of analog switches, each giving that 4:1 range, so six or seven ranges, each range individually calibrated and with logic to switch between them. It's very unlikely it would ever be long term stable, even if you could get it to work initially.


To get the type of frequency range with complex external switching using analog parts takes such as a purpose made voltage controlled oscillator or such as the VCO designs for modular music synthesisers.


A single MCU could do the whole lot far more accurately with minimal components.

And, as you want to control it via an internet connection, you would need an MCU to do that. The audio could be generated directly in the same device.

What accuracy do you need of waveform parameter control,
freq, duty cycle, etc... Do you need burst sine, arbitrary, tri, saw....?

Something like this, one chip :


Regards, Dana.

If you have a pot =>100k the max/min ratio value on tap may exceed 200:1: then a simple RC logic Astable Osc will give this f ratio. The are 2 types: a) Schmitt inverter b) dual inverter with 2 R’s, 1C.
The a) method can also be a CMOS OpAmp with single or bipolar supplies using any R ratio hysteresis (+in) and self biased (-in) RC.

Search _____ and you will find it easily.

You have mentioned variable frequency. Do you also need a variable duty cycle, or is a fixed 50/50 good enough?

Once the 55 circuit is configured for your application, you can gate the output on and off by driving the Reset input. Note that when the Reset is released, the first half-cycle always will be positive, and it always will be approx 40% longer than the succeeding half-cycles. If this is an issue, there is a way to reduce this error.

ak

Fixed 50 percent duty cycle would be fine. I'm looking to keep the frequency at a very accurate and stable value (an earlier prototype circuit did mostly what I wanted with a potentiometer, but I am wanting to hold very specific frequency output values, selectable with digital accuracy levels.)

danadak said:

What accuracy do you need of waveform parameter control,
freq, duty cycle, etc... Do you need burst sine, arbitrary, tri, saw....?

Something like this, one chip :

Test Bed DDS and WaveDAC burst tool

Occasionally I need to generate custom wave forms and burst a known number of cycles into a DUT. This is a single chip solution, board I typically I use is $ 10. You just drag and drop components onto schematic, wire internally with a wire wizard tool and out to pins, right click and config...

www.edaboard.com

Regards, Dana.

Click to expand...

Just simple, stable, highly-accurate square wave frequency generation.

jonj44 said:

Just simple, stable, highly-accurate square wave frequency generation.

Click to expand...

So not a 555 then - use a PIC micro-controller or similar, and use an external crystal as it's clock source.

The DDS24 used in post #4 will have a resolution of 1.43 Hz using internal clock of
24 Mhz. If DDS32 used internal then resolution is 5.6 mili Hz. Note the accuracy is
dependent on xtal used with part, versus using the internal clock.

jonj44 said:

Just simple, stable, highly-accurate square wave frequency generation.

Click to expand...

Define accurate. At 22000Hz and 20Hz.

Are you expecting to get resolution of 20, 21, ... 21999, 22000
Or 20, 22, .. 20000, 22000 (10% step increase with each move)
Or
20.000, 20.001, 20.002 ... 21999.998, 21999.999, 22000.000
Or something else?

jonj44 said:

Hello all,

I seem to recall having a schematic for one roughly 20 years ago, but there have been several hard drive crashes and at least one plumbing "flood" since then.. and I can't seem to re-locate it online.

My goal is to externally trigger square wave pulses (of variable frequency say from 20-22000 Hz) with a 555 timer circuit using a function generator/laptop/tablet/etc. Most of the schematics I have found show 555 pins TRIG 2 & 6 joined together for astable square waves.

Is this an internal IC connection or just a common way of externally wiring the 555 for astable mode?

Ideally, I would like to keep my output voltage fairly close to the maximum 15 VDC input (and I'm wondering how duty cycle would impact this goal). Also, I am looking to software control a 555 square wave oscillator, not "set" a hardware output frequency.

Click to expand...

Hi,

There are DSP modules capable of highly stable and accurately set frequencies from below 1Hz to above 10MHz. If that does not help nothing will. You control it by sending a digital word that defines the frequency. The word is 32 bits and would be sent by a microcontroller or something like that.

The 555 is stable but not as stable as a crystal which is how those DSP things are controlled.
Also, if you use the 'control' input you lose more stability from power supply voltage changes. The 555 is immune to power supply voltage variations but the control input voids that.

Without knowing exctly how you want to control the ferquency source one of these modules may meet your needs.
https://www.ebay.co.uk/itm/256574812230?itmmeta=01J3CQWMR6SBW0596NS7E5KRY3&hash=item3bbd0cf046

Les

MrAl said:

Hi,

There are DSP modules capable of highly stable and accurately set frequencies from below 1Hz to above 10MHz. If that does not help nothing will. You control it by sending a digital word that defines the frequency. The word is 32 bits and would be sent by a microcontroller or something like that.

The 555 is stable but not as stable as a crystal which is how those DSP things are controlled.
Also, if you use the 'control' input you lose more stability from power supply voltage changes. The 555 is immune to power supply voltage variations but the control input voids that.

Click to expand...

That's rather an optimistic claim

A 555 is fine for what it does, but it's certainly got no claim of being 'stable' in any sensible meaning of the word, it's just a crude RC oscillator, with not even any kind of temperature compensation.

Timing error for the bipolar part :



Regards, Dana.

Until the OP gives us his definition of "accurate", a 555 may be good enough... or it may not be.

Nigel Goodwin said:

That's rather an optimistic claim

A 555 is fine for what it does, but it's certainly got no claim of being 'stable' in any sensible meaning of the word, it's just a crude RC oscillator, with not even any kind of temperature compensation.

Click to expand...

Hi there Nigel,

Not sure what you mean here by optimistic. The board I was talking about is a DSP frequency generator made just for that. It's at least as accurate as the crystal, and it uses digital techniques to create the output wave. That's how they get something like 0.01Hz resolution even up to 10MHz and beyond. They are amazing devices. The chip on board does most of the work, and they use a 7th order bandpass filter to filter the square wave into a sine wave. So we get square and sine, but I do not think there is any triangle.

If you have never seen one of these things I'll provide a link, and you are going to be amazed
The fact that you have to use a 32 bit word to program it says something big about the resolution. There are almost 4295 million possible settings, meaning high resolution even at high output frequencies.
I have to admit though I do not remember all the specs. They are very impressive though, I was totally surprised, especially after using those 8038 sine chips and the like, and hunting for the 25MHz version that I never found because they don't make them anymore.

It's called this but you can get it with a dev board too:

High Speed AD9850 DDS Signal Generator​

danadak said:

Timing error for the bipolar part :

View attachment 146467

Regards, Dana.

Click to expand...

Hi,

The drift with supply is only applicable when you do not use the control input pin for anything other than a 0.01uf to ground.

MrAl said:

Hi there Nigel,

Not sure what you mean here by optimistic.

Click to expand...

As in the part I highlighted, you suggested the 555 was a 'stable' chip - I'm fully aware of the AD9850, and have a few of them.

Nigel Goodwin said:

As in the part I highlighted, you suggested the 555 was a 'stable' chip - I'm fully aware of the AD9850, and have a few of them.

Click to expand...

The 555 is plenty stable enough for many thousands of products you can find them in - even recently designed products like our little coffee grinder.

On the other hand, some simple crystal- controlled circuits are not nearly stable enough in some projects and additional steps need to be taken.

As I said above, stable is a vague specification and we need a more clearly defined specification from the OP - it's his project, his budget and his failure to accept.