Innovation in Photonics, Materials, and Sensor Research with the VHFLI Lock-in Amplifier – Webinar Q&A

Are slow measurements, weak signals, or overly complex setups holding back your research? That’s exactly what we set out to address in our recent webinar. We introduced the VHFLI Lock-in Amplifier: the successor to the HF2LI and the newest addition to the Zurich Instruments lock-in amplifier portfolio.

If you want to watch the webinar recording, you can find it on our YouTube channel at this link.

In a conversational Q&A format, we explored:

• Market positioning & budget flexibility: The VHFLI is designed to cater to the needs of a very diverse set of customers and their applications, with a product that can be tailored to their needs and budgets. The base version is priced very close to the price of the base HF2LI.
   
• Portfolio sweet spot: Frequency-wise, the VHFLI covers the DC - 50 MHz / 200 MHz frequency range, sitting between the MFLI and the GHFLI, a range that sees many different applications in photonics, surface science, material science, and sensor development. Beyond frequency, the VHFLI brings technology that was never present at these frequencies, like differential voltage inputs and outputs, current inputs, impressive noise performance, and a very large maximum measurement bandwidth.
   
• Hardware & signal integrity: The VHFLI is equipped with a set of inputs and outputs ideal, for instance, for many different types of sensors: differential voltage output and input, an integrated current input with a powerful transimpedance amplifier -- times two because of the two channels, plus two high-quality auxiliary inputs. With the multi-demodulator option, the VHFLI gets 8 demodulators, so all these signals can be measured in parallel, with demodulators still available for harmonics analysis.
   
• Ultra-low noise performance: Our engineers have spent considerable time ensuring the best analog performance possible. One of the outcomes is a very flat noise floor of around 3 nV/√Hz, starting from just a few tens of kHz. In addition, the VHFLI has demodulators with low-pass filters with a minimum bandwidth of just a few mHz, to remove noise when the signal SNR is very low to start with.
   
• Speed & measurement bandwidth: The VHFLI has a minimum time constant of only 14 ns, allowing the demodulators to capture events that are only a few tens of nanoseconds in duration, a very powerful tool in photonics, MEMS, and material development. Crucially, a very small time constant is quite useless if you cannot capture enough data to make use of it; that's why the VHFLI is capable of triggered acquisition with a data rate up to 25 MSa/s.
   
• LabOne software & workflow: The LabOne instrument control software offers a way to analyse a signal from as many points of view as possible. It integrates a Scope, a Spectrum Analyser, a data plotter, and automation tools like the Sweeper and the Data Acquisition Module. Together with onboard functionality like multiple demodulators, PID controllers, and, quite soon, a dual Boxcar averager, this greatly reduces the need for many instruments in most labs, especially because all the tools can be run in parallel.
   
• The Timeline Module: The Timeline Module is the latest innovation available from April to all VHFLI, GHFLI, and SHFLI users. It simplifies setting up experiments with complex timing: the Timeline Module editor makes it easy to properly time and coordinate pulse generation and triggered acquisition together with continuous signals and measurements. It also allows coordinating other instruments in the experiment through the generation and acquisition of triggers, making the lock-in amplifier effectively the timing hub for the entire experiment.

The webinar concluded with a live demo showcasing the VHFLI's capabilities in the LabOne user interface, including a hands-on look at the Timeline Module in action.

Below, you'll find answers to all the questions submitted by attendees during the session.

Q1: If you have the short time constant of 14 ns, then what is the range of delay times you have on the instrument?

14 nanoseconds is the minimum time constant, but it's not fixed; it can go all the way up to 21 seconds. So if you need to integrate for a long time, you can do that. 14 ns is simply the smallest value available when you need to capture very fast transients.

If the question refers to the Timeline Module and its delay block, that's a different mechanism. It is not bound to the time constant of the demodulator and can be extremely variable in its length, from tens of nanoseconds, as shown in the live demo, to milliseconds and seconds. The two things are very much decoupled.

Q2: Can we get a live demo of this instrument?

Absolutely! We have several ways of doing this. We have a lot of technical salespeople and application scientists who are eager to talk to customers, learn about their challenges and applications, and give live demos. You can reach out to us for an online live demo, going through the VHFLI, and even doing some specific tests. We are also going to be at many conferences with the VHFLI throughout the year; check our Events page to find the conference or trade show closer to you, and you'll experience the instrument in action. 
Our application scientists are also often traveling and will be carrying the VHFLI around. Don't hesitate to contact us to arrange a demo. 

Q3: I'd like to know what is the maximum current output of the AUX outputs of the VHFLI?

It's 100 milliamps, similar to what you have on our HF2LI and UHFLI lock-in amplifiers.

Q4: How do you suppress the noise due to external circuits and cables?

The principle of a lock-in amplifier is exactly to measure a signal at a specific frequency, to isolate it from anything else but the reference frequency. When you have a setup with many different components, there might be interference coming from external circuits, but that hopefully is not synchronous to the signal you're measuring. For example, one typical noise source is the 50 Hz or 60 Hz disturbance from the power lines. Without a lock-in amplifier, these signals would leak into the signal you actually want to measure. But by locking in at a specific frequency, the lock-in filters out everything else and isolates your signal from the interference of other components, circuits, or cables. This is exactly why you should use a lock-in amplifier: to isolate your signal from external interference and maximize your signal-to-noise ratio.

Q5: Is quadrature control and PID control possible?

Absolutely. You can measure the amplitude and the phase, but you can also look at X and Y from the demodulators, and all of these signals can be fed into a PID controller.

Q6: Would this amplifier be compatible with SPAD and time taggers, for example, in quantum interferometry?

This is a harder question to answer because we don't necessarily know all the specs from other vendors. In general, we provide input and output specifications for our lock-in amplifiers, and the VHFLI is no exception. If you're using photodiodes, make sure the output doesn't exceed the input's safety range. If you don't see a clear specification on the website, please reach out. We can provide it to you, and our application scientists can discuss your needs and the rest of the equipment you have, so we can figure out the compatibility together.

Q7: The LabOne GUI shown in the demo looks more user-friendly than I have seen in the past. Is this more visual GUI available for the other higher frequency lock-in modules?

Yes, indeed. We initially launched the new LabOne graphical user interface on the MFLI, so it's available on that instrument. It's available on the VHFLI, and since January, with the latest release of our LabOne software, it's also available for the GHFLI and SHFLI, our gigahertz-range lock-in amplifiers.

Q8: I am using a HF2LI as a low-noise scope. For the VHFLI, could you elaborate on what the scope/probe module can do?
(1) How many inputs can be captured simultaneously? (2) Can the trigger point be moved in time (i.e., pre-samples)? Finally, is there a chance to get any of this done with the HF2LI?

The VHFLI scope has two channels. You can configure either of those channels to whatever input you want to track -- whether it's a signal input, current input, or auxiliary input -- and those work in parallel. Regarding pre-triggering, it is not yet available, but it's something that is on our roadmap. 
Unfortunately, due to the hardware architecture differences, these specific improvements cannot be back-ported to the HF2LI.

Q9: Will the timeline module, and particularly pulse, trigger, and transient capture functionality, be available on the UHFLI and on other higher frequency lock-in amplifier units?

Unfortunately, the timeline module cannot be ported to the UHFLI. It's based on the new architecture that you can find on the VHFLI, but also on the GHFLI and SHFLI. The timeline module will be available starting from the end of April with our next LabOne release for the VHFLI, the GHFLI, and the SHFLI.

Q10: What is the accuracy of the demodulator? Can we lock to a frequency with an accuracy of 5 millihertz?

Accuracy and precision are two important concepts here. When it comes to accuracy, it refers to accuracy with respect to a very accurate frequency reference. In this case, you would need to use an external frequency reference, like a GPS reference or a rubidium atomic clock reference. Out of the box, this is not something we offer with the VHFLI or our other lock-in amplifiers, with the exception of the rubidium-atom-optioned UHFLI.

When it comes to precision, it depends on the circumstances. The minimum resolution in terms of frequency that we can have with the VHFLI is 6 µHz. So 5 millihertz is considerably larger — in general, that's a precision that can be attainable in many situations with PID controllers.

Q11: What is the best way to characterize the signal recovery performance of a noisy pulsed signal with a limited number of cycles with a lock-in amplifier on an MFLI?

For our MFLI lock-in amplifier, we have the Boxcar Averager option, which is very well suited for the measurement of noisy pulsed signals, especially with low-duty-cycle pulsed signals. With the boxcar averager, you're basically measuring only where the pulse is present and ignoring everything around it that contains only noise. This allows you to generally obtain a much better signal-to-noise ratio compared to a standard lock-in measurement.

The Boxcar Averager is available on our MFLI and UHFLI lock-in amplifiers at the moment, but in the future, it will also be available on the VHFLI lock-in amplifier, so your pulsed measurements will also be possible with the new instrument.

Q12: How can we validate that a detected weak signal is real and not an artifact of the measurement chain?

Starting with the lock-in amplifier end of the chain: we go to great lengths to make sure our signals are as clean as possible. Of course, spurious signals are impossible to completely remove, but we suppress them as much as possible and characterize them, so it's fairly easy to see where they are, and they are very narrow in frequency, in addition to being very low in amplitude. In general, you can slightly move around in frequency: if it's a spurious signal, it will disappear, whereas a real signal is usually much broader compared to a spurious signal from the measurement instrument.

If it's coming from an external source in the measurement chain, we recommend doing a "dry run", i.e., removing the real signal and seeing if you still measure something. If you do, it's likely an artifact. The good thing about lock-in amplifiers is that you are working with periodic signals, and disturbers tend not to be periodic, unless they are known sources like 50/60 Hz line noise.

Another important aspect is to look at your signal from different perspectives: in the Scope, in the frequency domain with a Spectrum Analyzer, etc. This is possible with the VHFLI because we have a full toolset of signal analysis tools that can be run in parallel and simultaneously, giving you more perspective about your signal and helping you determine whether it's real or an artifact.

Q13: Does this device act as a modulator and then demodulate the previously modulated signal? Or does it just implement the demodulation stage? 

The VHFLI can do both. It features signal outputs that can generate stimulus signals (e.g., a sine wave at a chosen frequency and amplitude), which you can use to modulate or excite your device under test. At the same time, it demodulates the incoming signal at the reference frequency to extract amplitude and phase information. So, while many users employ it purely for demodulation (using an external modulation source), the VHFLI is also fully capable of providing the excitation signal, making it an all-in-one solution for modulation and demodulation.

Q14: Could this be used with remote data collection for LIDAR data collection?

In principle, yes. The VHFLI's high-frequency demodulation capability and fast time constants make it well-suited for extracting modulated signals from optical returns, which is at the core of many LIDAR techniques (e.g., amplitude-modulated continuous-wave LIDAR). Additionally, the VHFLI can be fully controlled remotely via our LabOne APIs (Python, MATLAB, C, .NET, LabVIEW), enabling integration into automated and remote data-collection pipelines. That said, the specifics depend heavily on your LIDAR architecture, detector type, and signal characteristics. We'd encourage you to get in touch with our application scientists to discuss your setup in detail.

Q15: I have a question regarding data acquisition with the Zurich Instruments lock-in amplifier (MATLAB LabOne API). In a pump-probe experiment, every time I call poll(), it returns a sequence of samples instead of a single value. Is there a recommended way to directly acquire only one demodulated sample so that the acquisition time corresponds to my integration time?

The poll() function is designed to return all samples that have been accumulated since the last call, so getting multiple samples is expected behavior. One approach to keep using the poll() function would be to adjust its duration argument to zero or a small value and execute it. Then, you can simply keep the last sample in the returned data and drop the rest: this sample corresponds to the demodulated signal at the time of poll() execution.
Alternatively, you can use the Data Acquisition module (also known as the DAQ module), which allows you to define precisely timed acquisitions triggered by an external event or a software trigger. This gives you much finer control over what is captured and when. We recommend reaching out to our support team with details about your specific setup. We'd be happy to help you optimize your acquisition workflow.