HOW MUCH CURRENT DOES A 555 NEED ?

[Simon]

How much current above the output current does the 555 timer need ? can i run a 555 timer off a 100mA regulator if the load is only a few mA ? I think the issue may be the shoot through on switchover ? or do modern 555 chips have that sorted now ?

[chrome]

RTFM

Seriously it's right in there.

[SeanB]

You need a current capacity of 300mA or so, to handle the spikes. You must have minimum 100uf across pins 1,8 minimum. If the supply is limited in source capacity as big as possible, 470/1000/1500/2200uF are all usable. Otherwise poor performance on switching, and you can have some odd oscillations on the edges.

[Simon]

thing is I have cooked this up and we are having a company make 2 for us so I don't have room for messing (although I have got the subcontractor to put some through hole resistor pads in so that I can change them if needed).

From the datasheet I found a graph showing that the IC supply is no more than 10mA and I should not need more than  a few mA so I think our subcontractor was over fussing in wanting to put a TO-220 part in.

[Simon]

You need a current capacity of 300mA or so, to handle the spikes. You must have minimum 100uf across pins 1,8 minimum. If the supply is limited in source capacity as big as possible, 470/1000/1500/2200uF are all usable. Otherwise poor performance on switching, and you can have some odd oscillations on the edges.

The "other" subcontractor we used before who's design i was asked to base mine from used a 78L12 (100mA) and only 10uF apparently it worked but then I have not investigated it myself.

[Simon]

IT WILL BE THE SE555. trouble is I am uncertain of how much current I need and there is a 12k load too for the output circuitry that I may or may not use.

[westfw]

The TI SE555 datasheet says a max Supply current of 12mA (@15V) or 5mA (@5V), so I'd think a 78L style regulator would be plenty.

I don't know how the  hypothetical switching transients would affect this; I've never used a 555 in a situation where that behavior was critical, or where the power supply didn't already have significant bypassing.  A 1000uF cap for the 555 sounds ridiculous, especially for a 12k load.  (Does anyone have a pointer to a more technical analysis of this issue?  I can't figure out a search string that yields anything other than datasheets or "noise maker" circuits.)

[Simon]

I think there is a mix of practices used for the old bipolar designs and lack of specific info on the modern variants.

[JVR]

Why still use a 555?

A SOT23-6 PIC10F is cheaper, smaller and a hell of a lot more versatile.

[Simon]

Why still use a 555?

A SOT23-6 PIC10F is cheaper, smaller and a hell of a lot more versatile.

Tell my boss that, I tried to point out the advantages of MCU's but he kept saying something about bugs. Then he went to see an electronics subcontractor and they convinced him. This is a prototype only, if it becomes production i will make them use an MCU

[JVR]

Horrible way of doing business that, but I know the type of boss your referring to.

Just build a 10F circuit on your lonesome, and show him how it kicks a 555 in the arse on every conceivable level, from cost and size, to the accuracy of the waveforms and on the fly adjustablity.

Ok, your only going to get 256 steps, but thats plenty in most 555 usage senarios

[Simon]

WELL AS WE ARE DOING 50, 75 AND 100% RESOLUTION IS NOT A PROBLEM. YES WE HAVE BACKWARD CUSTOMERS TOO WHO WANT THE £30 SWITCH INSTEAD OF  A £2 POT

[westfw]

Just build a 10F circuit on your lonesome, and show him how it kicks a 555 in the arse on every conceivable level, from cost and size

Well, up till you want it triggered by up to 18V, and output up to 15V at 100mA or something.
Seriously, there are things that a modern microcontroller doesn't do very well.  I made a significant effort toward figuring out how to do a microcontroller-based replacement for vintage door chime mechanisms.  These electro-mechanical wonders typically are triggered by, and run from, 24VAC transformers, and end up driving 24V solenoids to bash the chime itself.  (http://www.eurekamodern.com/vintage_door_chimes_repairing_mechanisms.htm )  Trying to figure out how to make a "drop-in" replacement based on a microcontroller was VERY frustrating...

[Simon]

Yes the problem with MCU's is that you have a voltage shift, but you see a lot of what we do is 24V too. You can easily get mosfet drivers that will take a 0-5V in and provide up to a 18V totem pole output with shoot through protection etc and decent current capability to act as the output driver/buffer. Some things like something else I'm looking at now has an low switched input and a pull up resistor so no need to go anywhere near VCC and worry about level shifting, just bung a 2N2222 or 2N7000 and pull the receiving controller (brushless motor) to ground.

I've just found that they could do with a general purpose PWM driver so I'll make one based on an ATtiny13 or 12F615 and see if they'll buy it off me. Apparently the people that make the fans can't supply a relatively cheap PWM driver.