I am working on a project involving detection of pulse. I'm currently using infrared emitter and a photodiode detector but I have been having problems with this. I get very little variation in amplitude and tried amplifying this signal...but I'm not sure I'm doing the right thing. Please any help would be greatly appreciated...
Thanks in advance

 

I am working on a project involving detection of pulse. I'm currently using infrared emitter and a photodiode detector but I have been having problems with this. I get very little variation in amplitude and tried amplifying this signal...but I'm not sure I'm doing the right thing. Please any help would be greatly appreciated...
Thanks in advance

By far, the best way to use a photodiode is with a transimpedance amplifier.

http://www.newport.com/Tutorial/139634/1033/catalog.aspx

 

It would help if you could share some very necessary information with us. We need to see the schematic of the circuit, and the devices you are trying to use (manufacturer's part numbers).

 

Another method of interfacing to a phototransistor is to use a current mirror composed of two transistors. This is a transimpedance technique in a manner of speaking. The advantage of the current mirror is that the phototransistor is not permitted to saturate. In fact the voltage Vce of the phototransistor is held constant. If you draw the load line of the phototransistor connected into a current mirror on the Vce versus Ic curve for the phototransistor, you will notice the improving effect that it has the responsiveness and gain of the phototransistor.

hgmjr

 

Steveb
Thanks for the 'transimpedance amplifier' idea...it was very helpful. But i still have problems seeing a very significant change in the amplitude of the waveform, please do you have any idea how i could improve this?.
Thanks

 

Hi beenthere
the devices I'm trying to use are these....
IR emitter SFH 485
photodiode detector SFH 203FA
here is the schematic of my new circuit involving transimpedance amplifier


Thanks

 

Here is a link to an article with a simulator attached - http://www.ecircuitcenter.com/Circuits/opitov/opitov.htm

 

Try lowering the gain by significantly reducing the value of that 5 MΩ feedback resistor, as (with the 69 pF cap) it will roll off at about 470 Hz, and this will limit your maximum amplitude with the input pulse widths you're using. Try a 100 kΩ for starters. Without the 68 pF cap you'll get significant gain peaking due to a pole being introduced by the parasitic junction capacitance of the photodiode. This intrinsically limits the maximum gain of such a photodiode amp; the usual workaround is to follow the TI amp with another gain stage.

I haven't checked the spec of the emitter LED and PD, so make sure they're fast enough too.

Bonnie Baker knows her onions. Also Bob Pease. hgmjr's current mirror technique intrigues me, I must look into that, it may well save the day at some point in the future...

 

What are your power supply voltages? What is the output voltage of your op amp?
It will only go negative in the presence of light. Do you have ambient light screened out?

 

Try lowering the gain by significantly reducing the value of that 5 MΩ feedback resistor, as (with the 69 pF cap) it will roll off at about 470 Hz, and this will limit your maximum amplitude with the input pulse widths you're using. Try a 100 kΩ for starters. Without the 68 pF cap you'll get significant gain peaking due to a pole being introduced by the parasitic junction capacitance of the photodiode. This intrinsically limits the maximum gain of such a photodiode amp; the usual workaround is to follow the TI amp with another gain stage.

I haven't checked the spec of the emitter LED and PD, so make sure they're fast enough too.

Bonnie Baker knows her onions. Also Bob Pease. hgmjr's current mirror technique intrigues me, I must look into that, it may well save the day at some point in the future...

Greetings Darren,

If you decide to look deeper into this technique, I will be happy to walk you through the details and answer any questions that arise.

hgmjr

 

Belated thanks hgmjr, I'll take you up on that in the next few days (got some holiday coming up; snowed under until then). Much appreciated.

There's another technique I've used sucessfully in the past where the PD parasitic capacitance is buffered from the op-amp input by a common-base transistor. There's still the Miller capacitance of the transistor to cope with, but this is much smaller if the transistor is chosen well, and consequently the gain of the TI amp can be substantially increased. I can't find any references to this technique (I've only ever seen it one one white paper, not alas in my collection) so I'll have to work it out from scratch. When I do, I'll be sure to share.

 

Belated thanks hgmjr, I'll take you up on that in the next few days (got some holiday coming up; snowed under until then). Much appreciated.

There's another technique I've used sucessfully in the past where the PD parasitic capacitance is buffered from the op-amp input by a common-base transistor. There's still the Miller capacitance of the transistor to cope with, but this is much smaller if the transistor is chosen well, and consequently the gain of the TI amp can be substantially increased. I can't find any references to this technique (I've only ever seen it one one white paper, not alas in my collection) so I'll have to work it out from scratch. When I do, I'll be sure to share.

I designed a CB amplifier for basically this same app for another guy recently (it might have been on another forum), and offered to post it, but he never asked for it. I am now 2500 miles from home, and the design is on my home computer.

Strictly speaking, the CB amp has no Miller capacitance. C-B capacitance is still an issue, but it is not multiplied by the gain as it is in a C-E amp.