Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Photodiode falltime problem

Status
Not open for further replies.
P

pooya_BA

Junior Member level 2
Joined
Mar 24, 2021
Messages
23
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
194
Hello
I have a problem with the photodiode. I bought the SFH 2704 photodiode and connected it directly to the oscilloscope circuit without an electronic circuit.
The fall time of this piece was very, very long, about 380 microseconds, while in the information of this piece, the fall time was 67 nanoseconds.
What do you think is the problem? And how can I solve this problem?

The diagram below shows the output of the photodiode with a pulse of 160 microseconds.
This pulse is taken with a commercial PDA10A2 photodiode and is exactly equal to the pulse, meaning that the light source has no low time.
1a.jpg
a2.jpg
 

Hi,

Obviously capacitive load.
The photodiode includes capacitance, the wiring and last but not least the scope probe.

If you want to improve timing then use a TIA.

Klaus
 
KlausST said:
Hi,

Obviously capacitive load.
The photodiode includes capacitance, the wiring and last but not least the scope probe.

If you want to improve timing then use a TIA.

Klaus
Click to expand...
Thank you very much
Explain more if possible.
 

Hi,

I don't know what you want me to explain.
A capacitor, a scope probe, a TIA, why a TIA improves fall time?

Everything already is explained. Do a forum search, do an internet search first.

If then still are questions you are welcome to ask a detailed question. Please refer to documents you have read, so we can discuss about it.

Klaus
 
pooya_BA said:
The fall time of this piece was very, very long, about 380 microseconds, while in the information of this piece, the fall time was 67 nanoseconds.
Click to expand...
Did you notice that the datasheet specifies RL = 50 Ω and operation with bias voltage of 5V? Your waveform looks rather like Mohm load and photovoltaic mode (zero bias). Try with a medium load resistor, e.g. 10 kohm if you don't want to go for a TIA.
 

Not sure I understand. Just terminate amp that is inside part with 50 ohms, that
should fix you Tr, Tf problem you are experiencing.


Regards, Dana.
 
What's the time constant of a 1Mohm, 7pF probe?

Now loading the diode with 50 ohms may not be
anything like what you want to do for the circuit
task (certainly will reduce photons-to-voltage
output, though current may be unaffected).

The transimpedance amp could improve that
with lower input capacitance, and close-in gain.
You can transform impedance along the way.

Depending on what follows the detector, a (say)
1Kohm load and a 1pF-input-capacitance gain
block might be a suitable compromise. You'd
want to play with that a bit.
 

His probe 2.2RC rise time for 1 V is < his observed Tfall by a factor of > 10. If
he is using 1 X


Regards, Dana.
pooya_BA said:
Hello
What would you suggest to me if I wanted to use amplifiers?
So far I have found that I can use transimpedance amplifiers for my work.
Explain to me if I can follow the parameters that are suitable for my photodiode and the wind in selecting the amplifier.
Click to expand...

The suggestion was the fact you stated your measurements were on a PDA10A2 that
has a TIA in it, so I thought thats what you are using....my error.

Regards, Dana.
 
dick_freebird said:
What's the time constant of a 1Mohm, 7pF probe?

Now loading the diode with 50 ohms may not be
anything like what you want to do for the circuit
task (certainly will reduce photons-to-voltage
output, though current may be unaffected).

The transimpedance amp could improve that
with lower input capacitance, and close-in gain.
You can transform impedance along the way.

Depending on what follows the detector, a (say)
1Kohm load and a 1pF-input-capacitance gain
block might be a suitable compromise. You'd
want to play with that a bit.
Click to expand...
Hello, I appreciate your help.
Currently, I am away for a few days. I will test it afterward and let you know the result. I have a question for you : Do you think it is better to use OTA or TIA amplifier? I have got seperate suggestions to use both of them but I can not make a decission which one is suitable.
I am sending you names of some amplifiers that I have in access. Please let me know which is more suitable for SFH 2704 photodiode.
 

Attachments

  • Analog_Devices-AD845KNZ-datasheet.pdf
    92.4 KB · Views: 78
  • CA3130MZ-Intersil.pdf
    142.9 KB · Views: 59
  • lf351.pdf
    616.1 KB · Views: 54
  • opa354.pdf
    1 MB · Views: 75

OTA converts input voltage difference to current.

Photodiode puts out current in proportion to
illumination, to turn that into a large voltage
wants the transimpedance amp (TIA).

If you had already converted current to voltage
(such as a biased detector and burden resistor)
then the OTA might be suitable.
 
dick_freebird said:
OTA converts input voltage difference to current.

Photodiode puts out current in proportion to
illumination, to turn that into a large voltage
wants the transimpedance amp (TIA).

If you had already converted current to voltage
(such as a biased detector and burden resistor)
then the OTA might be suitable.
Click to expand...

Hi there, what do you think about using OPA354?
 

Attachments

  • opa354.pdf
    1 MB · Views: 62

you need a low DC load resistance to blead off the stored charge to make the falling edge quicker.

its the exact same thing as if you were using a schottky diode detector as an RF detector, you need to set the oscilloscope to 50 ohms input.
 

Hi,
biff44 said:
you need a low DC load resistance to blead off the stored charge to make the falling edge quicker.
Click to expand...
What circuit are you talking about?
A TIA?
A TIA does not need a resistor as low as 50 Ohms to get fast edges.

Klaus
 

You need a low resistor if you have a high
(in relative terms, to the diode self-C) shunt
capacitance.

A TIA fixes the shunt C by its minuscule input
capacitance, at least in ROIC designs. What
a piece-part TIA burdened by few-kV HBM
ESD protection and package might impose,
has to be sorted.
 

hi
I also want to use a second amplifier and increase the output of the first amplifier so that I can have an output of about 2 volts. Do you think I can use AD 8042 friends?
Here I put the simulation for you.
 

Attachments

  • AD8042.pdf
    420.2 KB · Views: 51
  • opa354.rar
    972 bytes · Views: 53

Hi,

Yes, it will work.
But I'm not sure whether you are satisfied with the result.
We don't know what you expect. No specification, no test conditions, no requirements.

Klaus
 
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top