How to measure short times, low current, low ringing on oscilloscope? UGB?

[hakunamatatarr]

Hi everybody,

first of all: I am an absolute noob regarding electronics and especially HF techniques. So if I explain something weird please just ask! I hope you can help me!

I am trying to measure short voltage pulses of solar cells which are excited by a 1 ns laser pulse. So electrically, there is a 0.8 - 1.3 V background voltage with a roughly 0.02 V pulse on top. The rise takes about 10 - 500 ns and the following decay 1 - 10 us. I am especially interested in the rise.
The solar cell is contacted in a measurement box which can be connected to a BNC cable which goes directly to an oscilloscope (LeCroy Waverunner HRO 66Zi 600MHz analog bandwidth).
My problem is that I see ringing/oscillations thatmake it hard to see the rise properly. The main frequency is at 23 MHz if that gives you any information. I understood so far that the ringing is caused by reflections due to the impedance mismatch between the 50 Ohm BNC and the 1 MOhm scope. Indeed, if I change the entrance of the scope to 50 Ohm the ringing disappears. However, I cannot measure with 50 Ohm because I need to measure close to open-circuit conditions, so I want to draw the least amount of current from the device as possible; ideally even with a higher impedance than 1 MOhm.
I could reduce the ringing already by choosing a shorter BNC and avoid adapters which were installed before but now I am stuck.

Is there any way to reduce it even further?
I read something about a unity gain buffer / voltage follower, but I do not understand very well if that would help.
Otherwise, people mention that for high frequencies one should use an active probe which already has a high impedance?

(The impedance of the solar cells changes for every sample and the light intensity, so I cannot say much about that).

I would be thankful for any idea!
Thanks!

Attached is an example of a pulse I would get, once in linear and once in logarithmic time scale.

[bdunham7]

Not an easy task you have set out for yourself!  Making a measurement like this illustrated the idea that the probe and scope become part of the circuit, so you either have to minimize that or figure out how to deal with it in your calculations.  In this case, it isn't initially clear to me that your ringing is actually due to reflections and there are other possible causes.  Could you tell us:

The size of the solar cell and exactly how it is connected to the initial BNC connector or cable?

The length and type of BNC cable that you are using to connect it to the scope?

What your actual measured results are with the equipment you have?  Screenshots of the scope would be best, including some with the pulse and rising edge spread out over the whole screen, so 5-10ns/div.

How you manage the offset?  Your pulse is ~20msV, so that would imply a scope setting of 5 or 10mV/div--but the 1.2V starting voltage would push that off the screen.  So do you add an offset somehow, use AC-coupling, what?

Also, if you can, the theory behind why a 1ns laser pulse yields a much longer, spread-out electrical pulse.  Is that simply due to the electrical characteristics of the panel, such as capacitance and inductance of the leads, or something else?  Also, why 1.2V?  That implies 2 cells or something other than a typical silicon cell.  The laser hits both?

[KT88]

To test for reflections of the cable you could use different leghts. It would create different ringing frequencies due to different propagation times.
The challenge I see is to match the impedance of the solar panel to the cable. A 10X probe would already help although it comes with more noise due to the 10X smaller signal voltage. A slight improvement would be achieved through averaging. As the frequency limit for passive 10X probes is 500MHz ( I'm not aware of faster ones at least). This would not show a very steep edge, unfortunately.
An active probe is the best way if you don't intend to build your own probe...

[hakunamatatarr]

Thank you both for the fast answer!

The size of the solar cell and exactly how it is connected to the initial BNC connector or cable?
Also, if you can, the theory behind why a 1ns laser pulse yields a much longer, spread-out electrical pulse.  Is that simply due to the electrical characteristics of the panel, such as capacitance and inductance of the leads, or something else?  Also, why 1.2V?  That implies 2 cells or something other than a typical silicon cell.  The laser hits both?
About the solar cell: It is actually not a Si cell but a perovskite solar cell which we produce in-house. It is a very small pixel with the size of 4 x 4 mm2 and produces directly a voltage of 1.2 V with this light intensity (comparable to the amount of sun light a cell would get).
Regarding the time difference in laser and voltage pulse: The perovskite absorbs the light and creates an internal voltage. But the charge carriers need some time to reach the electrodes, depending on their transfer through the different layers. This is exactly what we want to study, the faster the better (more or less).
I am not sure about the connection to the BNC. There is definitely some internal wiring because one can choose between 4 cells with a rotary switch (which obviously doesn't help). I will ask tomorrow about more info.

The length and type of BNC cable that you are using to connect it to the scope?
The length is about 15 cm. We cannot go shorter just because of the measurement setup. I don't know what different types of BNCs there are... Could you explain me more about this? I am pretty sure that it has 50 Ohm, but otherwise no idea. We actually tried with different cables. I attached a figure comparing a BNC and a LEMO cable with roughly the same length. I can ask tomorrow as well.

What your actual measured results are with the equipment you have?  Screenshots of the scope would be best, including some with the pulse and rising edge spread out over the whole screen, so 5-10ns/div.
The figures that I attached are measured with the equipment we have currently setup. Sorry, I don't have any screenshots right now; again, I will update tomorrow.

How you manage the offset?  Your pulse is ~20ms, so that would imply a scope setting of 5 or 10mV/div--but the 1.2V starting voltage would push that off the screen.  So do you add an offset somehow, use AC-coupling, what?
Yes, I used scope settings of 5 mV/div typically. I just offset 0 V with the position control to -1.2 V. Is there something wrong with that?

In this case, it isn't initially clear to me that your ringing is actually due to reflections and there are other possible causes.
I can try this tomorrow by using cables with different lengths as KT88 suggested. But why do you doubt that?

Btw. as trigger we use a photodiode that also gets hit by a small part of the laser. It shows a pulse amplitude of about 45 mV.

Could you also explain me easily why an active probe would be better? Where is the difference?

[bdunham7]

I am not sure about the connection to the BNC. There is definitely some internal wiring because one can choose between 4 cells with a rotary switch (which obviously doesn't help). I will ask tomorrow about more info.

The length is about 15 cm. We cannot go shorter just because of the measurement setup. I don't know what different types of BNCs there are... Could you explain me more about this? I am pretty sure that it has 50 Ohm, but otherwise no idea. We actually tried with different cables. I attached a figure comparing a BNC and a LEMO cable with roughly the same length. I can ask tomorrow as well...

I can try this tomorrow by using cables with different lengths as KT88 suggested. But why do you doubt that?

I'm not sure how you are using a LEMO cable, but sticking with BNC (assuming the actual cable is a suitable 50-ohm type like RG-58) a 15cm cable gives you about 0.75ns each way.  So we'd have to see your ringing to see if that might correlate.  Otherwise, the inductance of the wiring at the cells combined with any capacitance in the system might cause ringing as well.  And some small amount of ringing won't destroy your measurements--what you have doesn't look that terrible at first glance.

Yes, I used scope settings of 5 mV/div typically. I just offset 0 V with the position control to -1.2 V. Is there something wrong with that?

No, not at all.  If your scope and/or probe support that, it's the way to go.

Could you also explain me easily why an active probe would be better? Where is the difference?

One of your problems is that it is impossible to match impedances passively anywhere in your system without causing other problems.  An active probe will have 'high enough' input impedance and very low input capacitance--and capacitance is where you have issues with high frequencies and fast edges.  If your wires are very short, you don't need to worry about matching impedances like you do with anything long enough to be a transmission line.  So the active probe will have the high impedance/low capacitance input and amplifier/buffer/offset source right there at the test point, eliminating all of those issues with impedance matching and reflections.  Then it will have a 50-ohm buffered output that can drive a transmission line and termination (scope set to 50R) without any distortion or losses.  One issue to look for, though, is that most active probes are 10X attenuation, but your signal is so small that you really should have a 1X probe.  They do exist, and there are even some inexpensive AC-coupled versions that might work, given the timeframe of the pulse you are looking at.

Edit:  Just to be clear, I was explaining what an active probe would accomplish, not specifically recommending one for your application.  Noise may be an issue, for example.  What you are trying to do is fairly challenging based on the signal level and bandwidth required.

[jmelson]

"Solar cells" are terrible for measuring fast pulses.  What you want is a small-area PIN diode, and set it up to be reverse-biased (photoconductive mode).  You do need to set the scope input to match the cable impedance.  If you can't get enough signal that way, then you need to use a transconductance amplifier.  Smaller photodiodes have less capacitance, and reverse biasing them reduces capaciance by at least an order of magnitude.
Jon

[KT88]

To my understanding it's about charactzerizing the solar cell - not the laser.
For that purpose measuring the voltage is not wrong. The problem is the impedance matching as bdunham7 is explaining....

[jonpaul]

Bonjour: Your setup and question indicate the wrong sensor, and poor transmission sustem.

A matched Zo system is reqiuired, Zo source = Zo cable = Zo load 50 Ohm (or 75 etc)

Use scope on 50 Ohm.


Exactly which  generic "solar cell" is used?

You need a fast rise photodetector from USA firms  like EGG, Untied Detector, or Japanese Hamamatsu.

These have special low capacitance and controlled area diodes. Some requote DC bias. Most have 50 Ohm BNC terminations. Expect t6op pay $30..500 each.

An generic low cost "solar Cell" has inherent high capacitance and current is proportion into a short circuit, not high Zo.

Finally study the transmission of fast signals as in Tektronix Probe Circuits Concepts books.

Nuclear physics photodetectors are run at much faster speeds than you seek,


Bon Chance,

Jon

[KT88]

How would you characterize a specific solar cell with another PD?

[jonpaul]

Reccomend UDT OSI PIN diode high speed high sensitivity

PIN-10D

https://docs.rs-online.com/0aa7/0900766b8137f701.pdf

Jon

[KT88]

The solar cell is the DUT, not the laser...

[jonpaul]

spec on solar cell DUT?

Normally the output current is proportional to light intensity,over,surface into a,short,circuit

Perhaps OP needs a transconductance amp, current >>>voltage converter?

Jon

[Marco]

Why not just put your expensive scope into 50 Ohm mode?

I'd personally do this kind of measurement with a capacitively coupled transimpedance amplifier so you don't have to compensate for solar cell capacitance, but I'm not a solar cell scientist (and not following a consensus is rarely a good idea in science, so just keep it fully 50 Ohm and measure the capacitance and compensate for the RC time).

[jmelson]

The solar cell is the DUT, not the laser...

OK, so why are we testing solar cells with 1 ns laser pulses, then?
Seems like the wrong test.
Jon

[Marco]

From a quick google I would guess it's a "time of flight" measurement (with a special meaning for solar cell science) :

https://aip.scitation.org/doi/am-pdf/10.1063/1.4948344

[Electro Fan]

Any chance something in here might be useful?

[jonpaul]

To the OP:

Both photodiodes and photomultipliers for fast light pulse measurement were developed decades ago for nuclear research and are readily avcailble.
Photodiodes for nanosecond pulses are specially made with low shunt C and Zo 50 Ohm, and usually a BNC connections.
Any low cost commercial "solar cell" will not work.

I suggest to first study the equiv circuit and physics of photo diodes to learn more.

Kind Regards,

Jon

[Marco]

They are characterising the solar cell, not the original light pulse. Solar cells have huge transit time, which is part of what is being measured. The light pulse kicks lose electrons, the shape of the resulting current pulse gives information about the solar cell.

The standard way of doing these measurements is with a fully 50 Ohm load, so all they have to do is use coax and put his scope input to 50 Ohm mode (which it does support).

[hakunamatatarr]

Thanks everybody already for the answers!

I guess I need to clarify a few things:
As KT88 and Marco mentioned already a few times, the solar cell is the DUT. The measurement technique is called transient photovoltage (TPV). There is a continuous-wave laser shining on the solar cell to provide the background voltage and then the pulsed 1ns laser to provide the small signal pulse that I am interested in. One wants to measure at specific cw light intensities and will get different voltages depending on the quality of the cell and then one adjusts the pulsed laser intensity to get a pulse of about 20 mV. Typically, people look at the decay of this pulse but I want to study the rise. We do not apply any external voltage/electric field, this makes it different from a time-of-flight experiment. The whole measurement is done in the hope to understand the physics better that happen in the solar cell (recombination, diffusion, ...).
We cannot use the 50 Ohm impedance because this would extract a lot of current from the solar cell but we want to test it at an open-circuit situation, the physics change in between these cases. That's why I suggested the unity gain buffer (UGB) because it seems that it can have a very high impedance at the entrance (generating the open-circuit situation) and a 50 Ohm exit, allowing to use the scope with 50 Ohm as well. If we could put the UGB close to the sample we could avoid most of the reflections? And the UGB apparently does not change the voltage signal, making it possible to measure the actual voltage values which is important as well (or at least be able to calculate it easily).

I'm not sure how you are using a LEMO cable, but sticking with BNC (assuming the actual cable is a suitable 50-ohm type like RG-58) a 15cm cable gives you about 0.75ns each way.  So we'd have to see your ringing to see if that might correlate.  Otherwise, the inductance of the wiring at the cells combined with any capacitance in the system might cause ringing as well.  And some small amount of ringing won't destroy your measurements--what you have doesn't look that terrible at first glance.

I am using a LEMO cable just by inserting two LEMO to BNC adapters. I was quite wrong about the length though... The LEMO cable is 30 cm and the shortest BNC is 50 cm. I tried the same measurement with different BNC cables lengths and attached the scope screenshots below. At least one of the cables (the 100 cm one) was an RG-58, about the others I don't have any information. The frequency and amplitude of the oscillations does change. I also attached an image where I blocked the pulsed laser to just see the background with very small oscillations. Not sure if this give you any info.
In general, the ringing doesn't look too bad in the images and we could probably live with this but it would be nice to be able to reduce it further.

Edit:  Just to be clear, I was explaining what an active probe would accomplish, not specifically recommending one for your application.  Noise may be an issue, for example.  What you are trying to do is fairly challenging based on the signal level and bandwidth required.

But could you maybe recommend one? :D I understand that this would be probably better than using a BNC or maybe SMA cable (I heard that these are supposed to be good for HF) for us?


Regarding the measurement setup, I realized that I didn't explain it very well and that probably a lot of problems come from the whole sample box. I am still not able to explain it properly because I don't know the exact wiring. But I made some pictures, so that you get an better idea. Each substrate of our DUT has 4 pixels on it (each with size 4x4 mm2) and is placed on some pins for electrical contacting (see Measurement setup 3.jpg), all having a common ground. The pins are mounted on a PCB which is wired to a 14 pin plug (Measurement setup 4.jpg). From there, a cable (not sure what kind of cable) goes to another box which is made to choose the pixel under test (just 1 of the 4 is illuminated) with a rotary switch (Measurement setup 2.jpg). In principle, one could apply a voltage to the cell here ("wire" connection) but we just measure and connect to the "sense" the cable (here BNC-to-Lemo and Lemo cable) which goes to the scope (Measurement setup 1.jpg).
It is clear that all these electrical elements cannot help. So we thought about making a new box in which the substrates lies still in the pins, but the pins from each pixel are directly wired to a SMA connector which can go directly to the scope. Does this make sense?
We cannot get rid of the box because we need to measure under inert atmosphere (N2). And the problem is that this will take some weeks to be built...

[Marco]

With an open circuit you will suffer severe RC effects, isn't it enough to be open circuit at DC and low impedance at high frequency?

[hakunamatatarr]

With an open circuit you will suffer severe RC effects, isn't it enough to be open circuit at DC and low impedance at high frequency?

To be honest, I didn't know this was an option. How would that work?
I am also not sure if this would change the physics, I need to think about this.

[Marco]

For a low impedance AC measurement you use capacitive coupling to the coax. So with a 3db point at 100 kHz you would use a ~30 nF blocking capacitor to the coax, then put the scope into 50 Ohm mode.

Or to go even lower than 50 Ohm, directly capacitively couple into a transimpedance amplifier and only afterwards have coax (with a much larger capacitor and it will take a while to charge). But 50 Ohm seems low enough for time of flight, so probably here too.

[hakunamatatarr]

But this would distort my pulse shape, no?
I think it is better to stay at open circuit and try to minimize the RC effects

[jonpaul]

https://www.davmar.org/TE/TekConcepts/TekProbeCircuits.pdf

[bdunham7]

In general, the ringing doesn't look too bad in the images and we could probably live with this but it would be nice to be able to reduce it further....

If we could put the UGB close to the sample we could avoid most of the reflections?

But could you maybe recommend one? I understand that this would be probably better than using a BNC or maybe SMA cable (I heard that these are supposed to be good for HF) for us?

It is clear that all these electrical elements cannot help. So we thought about making a new box in which the substrates lies still in the pins, but the pins from each pixel are directly wired to a SMA connector which can go directly to the scope.

Thank you for describing your experiment in detail.  I'll try to give you some ideas.

For what you have, the signal actually doesn't look all that bad, but I have no way of really telling how accurate your results might be.  If you are looking for relative results, perhaps it is OK.  But I think your precision, especially on the issues of time/decay, are probably not very good on an absolute basis.  You are wanting to test the response of your cells, but I think what you are looking at is largely the response of your circuitry.

It does not appear to me that what you are seeing is reflections in the BNC cable due to the impedance mismatch at the scope.  There may indeed be such reflections, but a 50cm cable implies a round-trip time of about 5ns.  I think what you are seeing is from a complex mess of LRC from your wiring between the samples and the probe box and the probe switching box itself.  Different BNC cables will change the LRC equation by about 1pF/cm, so they will have an effect too.  There are ways to try and deal with this passively, with either capacitive coupling as Marco suggested or making an ad hoc 20X probe using a 950R resistor, but I think this will both load your cell in ways you don't want (an issue for your physics calculations) and also drive your signal down into the noise.

A 1X active probe and a unity gain buffer would be effectively the same thing.  I can't really recommend a probe because in your setup I think you need to do it differently, but as an example the ancient Tek P6201--1X, 900MHz, 300uV_RMS noise--would be a performance level to shoot for.  Your issues include the need for low loading, low noise and high bandwidth and the usual answer to that question would be "pick any two".  You can discuss it with an applications engineer from LeCroy, or perhaps someone with more knowledge will chime in here, but I think these will all prove difficult as you would need to run the probe all the way to the sample to get a good result.  Otherwise you will still be looking at the response of your wiring in addition to that of the cell.  What you need is to have your input amplifier/buffer right at the cell itself, and that implies that you would want four buffers, then you can run SMA-ended cables out to your scope.  I would eliminate the probe switchbox and just connect the cables directly to individual scope channels.  You could use a LeCroy AP034 or similar probe and one cell to make a calibration fixture so you could compare the result of your setup with a more established, commercial, calibrated result.

For the UGB, I would suggest first trying a low-cost capacitively coupled FET-buffer.  Presuming you don't need a DC response (and if you do, that can be handled separately with some care) this will allow a relatively simple circuit to solve all of your impedance matching issues.  These are inexpensive and easy to build so that they could be integrated into your sensor array.  They are also available on eBay in completed form if you want to be testing something in a few days.  There are probably many other solutions as well, including some that might give you some gain rather than just being a buffer. 

https://cdn.instructables.com/ORIG/FQZ/1QZP/IPJTFO82/FQZ1QZPIPJTFO82.pdf

https://www.ebay.com/itm/332393676082

Edit:  You also might want to look at something like the OPA855-Q1 which would allow you to make an isolated wideband transimpedance amplifier with a reasonably high input impedance (>> the cell impedance), 10X gain and decent bandwidth.   This is would require more design work, but it is a solution that seems to be designed with applications similar to yours in mind.

https://www.ti.com/lit/ds/symlink/opa855-q1.pdf?ts=1645209368066&ref_url=https%253A%252F%252Fwww.mouser.com%252F