Impedance across photodiode

[tubacn]

Hi,

I would like to measure the series resistance of a photodiode. Could you please help me on this matter. I bias the photodiode in the forward direction (0.3-0.5 V) and try to analyse the data but I am skeptical about this method as we do not know whether such high currents flowing may damage the optical properties of the photodiode or not.

Thanks in advance!

tubacn

The series resitance of a photodiode in not as important as the shunt resistance ..

It is common industrial practice to put a 10mV reverse bias on the photodiode and measure current ..
The ratio between the bias voltage and current determine the shunt resistance value ..

Regards,
IanP

 

[BigBoss]

What will you use this photodiode ?? RF Transmission or simple ON/OFF or low frequency switching..??

 

[tubacn]

I will use it as a photon detector! I have to determine the series resistance...

Thanks for your reply...

What will you use this photodiode ?? RF Transmission or simple ON/OFF or low frequency switching..??

 

[E Kafeman]

Forward bias results in low sensitivity. Reverse bias is the common way to use a photodiode. The diode can then be seen as a current source where the impedance is high and of less interest. Often is a active voltage feedback used to compensate for ambient light and improve dynamic response. Less useable as photon detector.

 

[FvM]

The series resistance of a diode (also a photodiode) can be understood as a deviation from ideal diode behaviour, that mainly shows at higher currents. You can derive it from the measured current/voltage characteristic. Of course you should keep maximum ratings for the measurement, and you would want to aplly pulse measurements at higher currents to avoid a faulty measurement due to temperature dependant diode forward voltage. A diode charateristic measurement would rather use (pulsed) constant current than voltage.

It's most likely easier to refer to manufacturer data, if available. You should also consider, that a lumped series resistance + ideal diode + capcitance equivalent circuit is a simplification, that won't be able to model the high frequency behaviour of a photodiode very well.

 

[tubacn]

Hi E Kafeman,

Thanks for the info but I am rahter interested in measuring the series resistance. The operation of a photodiode is no problem, I am trying to determine this resistance an preferably at voltages of typical diode operation, namely at reverse bias. Any idea?

Thanks

---------- Post added at 10:04 ---------- Previous post was at 10:02 ----------

Hi FvM

Thanks for the detailed and very useful info! Is there anyway that I can measure the series resistance at voltages of normal operation (reverse bias). Or at least near zero bias? Any idea/suggestion would be grately appreciated!

Thanks, regards

The series resistance of a diode (also a photodiode) can be understood as a deviation from ideal diode behaviour, that mainly shows at higher currents. You can derive it from the measured current/voltage characteristic. Of course you should keep maximum ratings for the measurement, and you would want to aplly pulse measurements at higher currents to avoid a faulty measurement due to temperature dependant diode forward voltage. A diode charateristic measurement would rather use (pulsed) constant current than voltage.

It's most likely easier to refer to manufacturer data, if available. You should also consider, that a lumped series resistance + ideal diode + capcitance equivalent circuit is a simplification, that won't be able to model the high frequency behaviour of a photodiode very well.

 

[KerimF]

Hi E Kafeman,

Thanks for the info but I am rahter interested in measuring the series resistance. The operation of a photodiode is no problem, I am trying to determine this resistance an preferably at voltages of typical diode operation, namely at reverse bias. Any idea?

Thanks

---------- Post added at 10:04 ---------- Previous post was at 10:02 ----------

Hi FvM

Thanks for the detailed and very useful info! Is there anyway that I can measure the series resistance at voltages of normal operation (reverse bias). Or at least near zero bias? Any idea/suggestion would be grately appreciated!

Thanks, regards

I think you are aware that there are 'static' and 'dynamic' resistances.

For any bias point (on the V-I curve), the ratio V/I could be called as the static resistance (sometimes DC). It is not contant along the curve for non-linear devices as diodes.

On the other hand, at any bias point we may define a dynamic resistance by assuming that the slope V versus I (dV/dI) is rather constant around it. So for a small AC signal the device is equivalent to this dynamic resistance and not the previous DC one.

For non linear devices, both the 'static' and 'dynamic' resistances change with the bias point on V-I curve.

I assume we are talking here about DC bias and low frequency signals, otherwise the device internal capacitance(s) and inductance(s) would play also an important role in determining the circuit response.

I noticed you are content to work near the zero volt bias so the 'static' and 'dynamic' resistances are alike. And your photodiode is enclosed in a dark box... Could a general purpose opamp (which I only have) help us measure this relatively high resistance? If not... there must be another way to do it :wink:

 

[E Kafeman]

A reverse photodiode is a current source, not a resistance. If you want something to measure resistance as a value of light is it better to select a CDS cell.
Most photodiodes reverse dark current can be found in data sheet and from that can a resistance for biasing be calculated. Typical resistor values for biasing in total darkness is around 50k-1M. If ambient light occur increases current in the diode and then must bias resistance be reduced for optimal dynamic range. That is why bias is wide variable in many designs.
Especially for low light conditions can also temperature be a important factor that affects photodiode current. Cooling a photodiode reduce noise and lowers internal current leakage. Link to more information: **broken link removed**

 

[biff44]

I agree with the commentors. Your thread title is right, but your method is wrong. Since the diode is reverse biased during normal operation, measuring the "series resitance" would be meaningless.

You really want to measure the thevenin equivalent, which is the effective parallel resistance when it is reverse biased.

 

[tubacn]

Hi,

First of all, thanks for all your comments. I should explain the problem in more detail. First of all, I do not have any datasheet as it is a home made photodiode. My problems is that the response of one is a bit different than the other in terms of impedance match. Otherwise they seem to be identical in terms of a dark current. I have to determine this impedance difference and in fact I saw that in the forward bias there is a considerable difference in the I-V curve of the two! However, I am not allowed to flow such high currents under normal conditions for other diodes than these two test diodes. Is there any way I can do this analysis near zero bias or at the bias voltage of normal operation. I know that the shunt resistance is dominant at the reverse direction, therefore I think it is impossible to determine the series resistance at high reverse bias voltages. Do all manufacturers measure the series resistance at forward biases or is there any other method that you can suggest? I read about RC time constants but this can be only checked with a preamplifier stage, or?

Thanks and regards

I agree with the commentors. Your thread title is right, but your method is wrong. Since the diode is reverse biased during normal operation, measuring the "series resitance" would be meaningless.

You really want to measure the thevenin equivalent, which is the effective parallel resistance when it is reverse biased.

 

[E Kafeman]

A photodiode can much be compared with a bipolar transistor and as you maybe know, no manufacturer specifies CE forward V/I in terms of a resistance for a bipolar transistor and neither for a photodiode. For a photodiode is base current of a transistor corresponding to amount of photons that activates electrons in base layer of photodiode.
In forward situation should a photodiode behave as a diode with a small emk in serial due to that ambient light cause a internal charging but it is of less interest, as it not gives any information about how the diode behaves in the situation it is intended to be used for but measuring the voltage is a simple functional check as a broken diode will show nothing. There is no need measure with more then a few mA as it is a small signal diode with small internal chip that not can handle high current.
You can measure any active component at low voltage and low current bias, and the result will be reliable for just that situation but not for other bias levels as it most likely is nonlinear curves. When biasing a photodiode are also amount of ambient lightness and wavelength bias factors, just as base current is it for a bipolar transistor. To be able to characterize a photodiode must therefore ambient light be possible to control and quantify, except for one endpoint in this measurement curve and that is when amount of light is zero as we then can measure "darkness current".

 

[FvM]

You couldn't clarify yet, if the observed differences in diode I-V characteristics are actually related to a diode series resistance. In a first order, you have an exponential diode characteristic. According to the diode area and other parameters (e.g. semiconductor doping) you'll observe different saturation currents in the basic diode characteristic. As already said, the ideal diode characteristic will be modified by a series resistance, which can be imagined as resistance of the diode contacts. But depending on the utilized diode current range, you won't necessarily see the series resistance.

 

[tubacn]

First of all, many thanks for the useful comments...

I checked the IV curves in both forward and reverse bias conditions. I can see the difference between two photodiodes only above approx. 0.4 V of forward bias. Is there any other way to see this difference; for instance can a CV measurement useful?

Thanks in advance...

You couldn't clarify yet, if the observed differences in diode I-V characteristics are actually related to a diode series resistance. In a first order, you have an exponential diode characteristic. According to the diode area and other parameters (e.g. semiconductor doping) you'll observe different saturation currents in the basic diode characteristic. As already said, the ideal diode characteristic will be modified by a series resistance, which can be imagined as resistance of the diode contacts. But depending on the utilized diode current range, you won't necessarily see the series resistance.