time control for transistor switch.

[birt]

i am looking for a circuit that makes a transistor "turn on" for a period of time set by a pot.

so i have a control voltage wich i can feed directly to the transistor, but than it will stay turned on as long as the voltage is there. what i want is the time it's on to be independent of the time it gets the control voltage. so if i put the voltage on a momentary, it doesn't matter how long i push the momentary, the time it's on just starts the moment i push the switch and stops after the set time. it will only work again if i let the switch go and than push it again.

my guess is a need some kind of multivibrator circuit?

[R.G.]

I don't understand exactly how you want this to act. Probably I've just not had enough coffee yet.

But it's quite easy to get a specified on-time. The CMOS 555 time will do this for you really well. You arrange for something to trigger the timer, and its output goes high for a time that's independent of what triggered it.

[birt]

i'm going to try to explain this the best way i can (it's in these situations my english is not good enough :p)

what i want is a transistor or chip or whatever that switches and opens a path to ground. it has to stay open for a period of time set by a pot and it should do this when i feed a voltage to it. the open-time must be completely independant from the time there is voltage fed to this circuit.

i just read the datasheet of the TLC555 and as far as i understand what it sais, i guess this is a good start for what i need.
so i connect the voltage i feed to it to the trigger input, the output probably goes to the base of a transistor wich will make a path to ground when it gets the voltage provided by the timer? this is as far as i understand it, i hope someone can give me a little more insight how to hook up a pot to set the time.

[Nasse]

I think I used 555 "retriggerable monostable" confiquration, copied it somewhere, worked fine, I needed it stay "on" for fixed delay after signal level had dropped. I think you can do non retrigger too with 555. I know there are easier and simpler methods and 555 can be "noisy" in gtr pedal because it takes huge current but for precise and stable pot adjust control it worked fine

[R.G.]

No apologies for your English are needed.

That helps a lot. So you want to open a switch to ground for a set time. The set time is to be controlled by a pot. The timer is kicked off when a control voltage starts the timer. The on-time of the timer is independent of the time the control voltage is on, so that a momentary switch for instance might do the control.

The CMOS 555's work this way: if you set it up as a monostable timer (that is, "one shot"), it triggers on a negative-going edge of the trigger input. Once triggered, it's pretty much independent of the control voltage as long as the control voltage is back to the high condition before the pulse-time is done. So if you have some way to feed it a negative-goind pulse, the 555 will do the proper triggering. If you only have a positive going pulse, you'll need to invert it to use the 555. A momentary switch to ground shunting a pullup resistor to the trigger input would do this directly.

The output of the 555 is positive going, so when you trigger it, the output goes high and stays high for the timer period. If you want this to open a path to ground for the timer period, you will need to feed an inverter before feeding a transistor to ground. This inverter could be just an NPN transistor. So you could connect a resistor of about 2K to 10K to the base of an NPN transistor with its emitter at ground. The collector of the transistor is connected to the + supply. A second NPN transistor has its emitter grounded, and its base connected to the collector of the first NPN. The collector of this second transistor is then opened from ground with the timer is on, connected to ground when the timer is off.

To control the timing with a pot:
The timing is set by a single resistor to V+ and a capacitor to ground. The resistor and cap are connected with resistor to +, resistor to cap, and cap to ground. The resistor-to-cap junction is connected to both the discharge and threshold pins of the 555. The time it fires is T = 1.1*R*C. If you make your pot be the resistor, plus some fixed resistor in series with it to set a minimum time, then the operation is as you have described. Max pot setting = max time, min pot setting = min time.

It is important that the trigger pin rise before the timer times out, otherwise it will start a second timing cycle.

See http://cache.national.com/ds/LM/LMC555.pdf

Specifically the CMOS 555 is much quieter than the ordinary bipolar 555, and does not pull huge current spikes like the bipolar one does.

[birt]

ok i tried to make a very quick drawing of what you just said:


1- is this drawing correct?
2- if i want to use for instance a nine volts control voltage to do this switching instead of the momentary i put the same npn inverter as the one i use after the 555 before the trigger input?
3- is the time 1,1*...ohm*...pf=...ms?

[Seljer]

might be a bit of help, heres a nice page on some 555 circuits
http://home.cogeco.ca/~rpaisley4/LM555.html#2
with handy little calculators for the time/frequency and stuff like that

[birt]

thanks for that page. an 1M pot with a 1k resistor in series and a 0.47uF cap would give me a max time of about 0,52 seconds wich is basicly my goal.

[birt]

can someone please correct my drawing if necesary?

[R.G.]

Your drawing is basically correct, give what information you had.

I think this works.



I added a few filips that you'll need. The inverter on the front end will fire the timer any time the control input is over about 1V. If this is too sensitive, put several diodes in series at the input. Each diode adds about 0.5V to the triggering voltage.

The grounding to keep timer clicks out of the audio is important. I've shown the way to start. You can still get clicks if you route signal wires near the timer. At least this way the timer ground current pulses will not be traveling through the signal ground and causing clicks that way.

Your audio signal may or may not have a DC voltage on it. The two 10uF NP caps will keep the slight DC offset voltage that comes from the NPN transistor switch from appearing on the signal line. The 4.7K in series with the NPN will prevent the NPN from trying to kill a low impedance source driving the signal line.

Signals bigger than about 0.7V will forward bias the collector-base junction of Q2 and be distorted. Single coil guitar signals should be OK as they sit. If you want to switch humbucker level signals or boosted signals, the output can be changed to put a larger reverse bias on Q2's base or it can be changed for a JFET with a little rearrangement.

[birt]

this is great. on my paper version of the drawing (everything i draw is on paper first) i already added caps on the signal path  to block the DC and also added an input and output buffer/gain stage.

thank you very very much for your time, your explanation and your drawing!

bert