Go Down

Topic: Extend Pulse from Op-Amp/Photodiode combo? (Read 14628 times) previous topic - next topic

cjdelphi


I wonder Cjdelphi... If you have ANY IDEA of the photodiode current available in that circuit or the gain of the Op-Amp?.
Stretching the pulse is trivial.. A half monostable from a 40106 or a 4584 will serve nicely..
It is possible too, to use a 4069UB to do all of it with 3 sections connected as a linear amplifier 40 DB of gain can be easily realized and the remaining 3 sections can clean up and delay the pulse. The down side is that the 4069 isn't very fast because it is unbuffered..
I would suggest that in the future you supply schematics to back up your rather unusual? claims.

Doc


Amazing ignorance.

Saying this can't be done without using an ic ... and then like a pack of starving dogs attack me for suggesting otherwise lmao... i say wild more like an annoying poodle.


nickgammon

I would like both of you to stick to the technical discussion without calling each other names. Thank you.
Please post technical questions on the forum, not by personal message. Thanks!

More info: http://www.gammon.com.au/electronics

cjdelphi

#17
Oct 13, 2013, 04:39 am Last Edit: Oct 13, 2013, 05:26 am by Nick Gammon Reason: 1
First stage amplifies the tiny signal from the ir detector, this gets amplified via the transistor attached image 1.

Second stage (providing a square wave is applied) and some extra (maybe) components to filter it, then lastly extending the pulse.

Pretty straight forward by the looks of it...


Moderator edit: Non-technical comment removed, <sigh> (Nick Gammon)

runaway_pancake

#18
Oct 13, 2013, 05:26 am Last Edit: Oct 13, 2013, 05:28 am by Nick Gammon Reason: 1
What you attached here is a lot different than what you posted about

Another way might be to use a dalington transistor pair or 2 transistors biased to amplify the signal then feed it through a schmitt trigger to clean the waveform up .


How is the "photodiode" supposed to interface with the AC amp you've attached?
I think that you ought to be able to explain your proposals beyond, "right, well here's a copy&paste of something or other; so make sure you get it all right, simple."

karl101 has left the building.


Moderator edit: Non-technical comment removed, <sigh> (Nick Gammon)
"Who is like unto the beast? who is able to make war with him?"
When all else fails, check your wiring!

cjdelphi

Am i to understand correctly, you have no "voltage" to be amplified?   (what's the opamp for then?)

cjdelphi



That's the method i'd use to amplify (via a darlington)

there are many ways of doing this job using just a few discrete components, but demanding someone build one!

jremington

#21
Oct 13, 2013, 07:01 am Last Edit: Oct 13, 2013, 07:18 am by jremington Reason: 1
The OP's original circuit relied on the photodiode working in "photovoltaic mode", i.e. it generates a voltage. But the op amp isn't properly biased, as previously pointed out.

Since the OP originally wanted to use a photodiode with a single-supply op amp, for the record here are conventional dual-supply op amp circuits that use proper biasing and a small feedback capacitor for circuit stability:  
For single-supply, there needs to be an offset applied to the Vin + terminal, for example this positive pulse version:

Any 5V op amp will work in this latter circuit (and most others as well).  However the feedback capacitor needs to be chosen judiciously if speed is important. There are lots of design notes dealing with that choice. Without the feedback cap, the circuit will probably oscillate.

Since op amps often come four to a package, here is a convenient pulse stretcher circuit that will work fine with the positive pulse version, or an extra op amp could be used to invert the negative pulse produced by the other:
.

polymorph



That is a bad way to bias a transistor circuit, and the voltage gain is less than one.

Can we get back to solving the OP problem?

I think several people have given simple answers. Changing the amplifier configuration as jremington suggests, or use a phototransistor and a monostable.
Steve Greenfield AE7HD
Drawing Schematics: tinyurl.com/23mo9pf - tinyurl.com/o97ysyx - https://tinyurl.com/Technote8
Multitasking: forum.arduino.cc/index.php?topic=223286.0
gammon.com.au/blink - gammon.com.au/serial - gammon.com.au/interrupts

karl101

Hello,

Well, that has been a vigorous discussion. Having read the various posts, here is my proposed circuit for time stretching, It should give a 500ms pulse.



Any thoughts?
Karl

runaway_pancake


Any thoughts?


The one-shot timer is triggered on the falling edge of the input (pin 2).
For the output to return low, trigger must return high.

"Who is like unto the beast? who is able to make war with him?"
When all else fails, check your wiring!

karl101


The one-shot timer is triggered on the falling edge of the input (pin 2).
For the output to return low, trigger must return high.


So to fix that I put the photodiode in the other way round?

dc42

From your own reply #3:


Thanks for the replies.
I didn't know about the interrupts, a test shows it works rather well and saves me having to add a 555 timer


So why do you still want to stretch the pulse? Provided the pulse is at least a few microseconds long, you don't need to stretch it, if you use it to trigger an interrupt.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

polymorph

Karl101, the photodiode must go the other way round in order to work. Think of it like a tiny solar cell, it generates voltage when light hits it. But the cathode becomes the positive source.
Steve Greenfield AE7HD
Drawing Schematics: tinyurl.com/23mo9pf - tinyurl.com/o97ysyx - https://tinyurl.com/Technote8
Multitasking: forum.arduino.cc/index.php?topic=223286.0
gammon.com.au/blink - gammon.com.au/serial - gammon.com.au/interrupts

karl101


So why do you still want to stretch the pulse? Provided the pulse is at least a few microseconds long, you don't need to stretch it, if you use it to trigger an interrupt.


Strange as it may seem, I actually want to learn some electronics, its much more interesting than writing web applications in PHP/Python all day. And perhaps in around five years I too could join in a squabble on how to extend a pulse...  :-)

Karl.

jremington

#29
Oct 14, 2013, 07:40 pm Last Edit: Oct 14, 2013, 09:16 pm by jremington Reason: 1
In addition to turning the photodiode around, you need a positive bias on the positive input of the op amp in the circuit posted (response #23). For example, you could use a resistive voltage divider of 2 x 10K ohm resistors and a 10 uF filter capacitor to set the bias at Vcc/2.

If you want to learn about how these circuits work without having to wire everything up, I strongly recommend using LTSpice to simulate them. It is free and pretty easy to use, with lots of tutorials and a good forum. Of course, it is just a simulation and real life can intrude, but a lot of silly mistakes can be fixed before the actual wiring begins.

Here, I've simulated the photodiode response with a current source, and changed the time constant so that the output pulse is about 5 ms. I've attached the simulator source file (NE555_mono.asc) if you want to try it out.

Go Up