With the first handheld oscilloscope to achieve 500 MHz at a 5 GS/s
Electronics technicians and engineers face a number of challenges when troubleshooting high speed modern electronic components. They constantly work around high voltage and current. Many of the plants they work in are extremely dusty and humid, which creates a hostile environment for sensitive troubleshooting instruments like bench oscilloscopes. Sometimes they have to climb a ladder on the side of a machine, which limits the weight and size of the equipment they carry.
Still, technicians often need wide bandwidth oscilloscopes to deliver a high degree of accuracy and detail, and the response speed to measure rise times in the tens of nanoseconds. We recently heard from a customer, Leigh Copp, Chief Engineer and Business Unit Manager for the Linamar Advanced Systems Group, who had been addressing those needs with the 200 MHz Fluke 190 Series ScopeMeter® Test Tool for lower frequency applications and then turning to a bench oscilloscope for applications that required higher frequency and/or faster rise time measurements.
Catching nanosecond events
Among his many projects, Copp periodically gets calls to diagnose problems with the electronic controls on large inverters in induction heat treating systems. Those inverters have many parallel or series devices, including Insulated Gate Bipolar Transistors (IGBTs) and metal - oxide - semiconductor field-effect transistors (MOSFETs) that operate at very high frequencies and run in short cycles.
“In power electronics I want to see cause and effect,” says Copp. “If the inverter fails, faults, trips, or blows up a device, I want to be able to see in the millisecond range what the current and voltage were doing at the exact instant the event occurred.”
The output resonant frequency on inverter controls is relatively high, and these controls operate on very high power, so troubleshooters have to watch the switching overlap between opposing power semiconductor devices. Accurate rise time readings, identifying parasitic oscillations, and understanding what those mean is crucial to pinpointing the problem.
The digital pulses on some of those applications produce rise times in the tens of nanoseconds and require a scope that has a rise time in the low nanosecond range to accurately view those wave forms. Those applications also require a fast sampling rate and enough bandwidth to capture the rise time in a single shot. In the past that typically meant bringing out a bench scope that had at least 500 MHz bandwidth.
Streamlining differential measurements
Ideas for an ideal scope
Leigh Copp’s experience with ScopeMeter™ test tools goes way back to the beginning of the product. He was customer #51 in Canada for the Fluke 97 ScopeMeter when it first came out in the early 1990s. He’s spent a lot of time since then with that model and several upgrades. So with that kind of history, we decided to ask him what his ideal portable scope would include. Here is what he said:
Memory: lots and lots - “If I had to pick one feature that I would like to see in a ScopeMeter portable oscilloscope, it’s more acquisition memory. I’d like to see at least 10 times what’s there now; 100 times the current memory would not be at all out of the question. I know you have to balance size, thermal considerations, battery life, and cost, but I’d like to see as much memory as you could shoehorn in there. This is because if I’m trying to get really high resolution, then I crank up the time base to highest resolution and record it with as many samples as possible.”
Four channels - “Having 500 MHz bandwidth is huge benefit. Now I just need four channels so I can use it for most of my applications.”
Easier user interface - “The Fluke ScopeMeter 190 series user interface has more of a DMM-like interface, which makes sense because there are probably a lot more electricians in the field using them than engineers. However, I would love to see a bit more intuitive user interface with more of a traditional ‘scope’ look and feel.
It’s not surprising that when Copp needed more than 200 MHz bandwidth, he would pull out a 500 MHz or 1.5 GHz bench scope and use differential probes. The obvious drawback was the fact that bench scopes are designed for the bench, so they aren’t sealed against dust and moisture and they aren’t exactly easy to lug around. All channels of most bench scopes are tied to a single ground, which makes it extremely difficult to safely take floating measurements.
To compensate for that, Copp used differential probes with the bench scope, but those probes had only 100 MHz bandwidth. “I have a 500 MHz scope but the differential probes are only 100 MHz, so I would only use the bench scope in that particular scenario for the additional record length,” says Copp.
When Copp needed to take advantage of the higher bandwidth available with the bench scope to measure rise time, for example, he would use 250 or 500 MHz single-ended probes and put the Channel A probe on the positive test point and the Channel B probe on the negative test point and then set the scope on the Ch1 - Ch2 math function. The sum of Channel A minus Channel B provided the floating voltage. However, that process immediately reduced the four channel scope to a two channel scope. It also had the potential to reduce the accuracy of the readings due to noise picked up in each probe, which introduced an additional degree of uncertainty to the results.
When Copp learned that Fluke had a 500 MHz ScopeMeter - the 190-502 500 MHz - with a 5 GS/s sample rate, he saw it as potentially a much quicker and more convenient solution for accurately troubleshooting his induction heating inverters. So he put it to the test.
The first difference he saw was that ScopeMeter 190-502 has 500 MHz capabilities at the probe tip and two independent floating inputs, up to 1000V so it can take differential measurements in one shot. “Having a 5 GS/s sampling rate to nail it the first time is really important,” says Copp. “And my two channel scope remains a two channel scope because I have an isolated ground on each channel. That means I can measure across two switching devices simultaneously.”
Analyzing the need for 500 MHz
Copp uses the 500 MHz 190-502 ScopeMeter oscilloscope to troubleshoot some of his inverters that require the higher fidelity measurement of a 500 MHz scope. However, the need for high bandwidth wasn’t entirely obvious. “The fastest inverter I have is running at 400 KHz, so some might say I only need a 10 MHz scope,” says Copp. “That would be fine if I just want to see that the inverter is turning on and off and not see how it’s turning on and off. We’re switching a square wave; a perfect square wave is the infinite sum of odd harmonics of the base frequency. So if the inverter is running at 400 KHz there are harmonics well into the hundreds of MHz to make it a square wave. That’s why you need wide bandwidth to capture these rapid switching transients.”
The bandwidth also affects the rise time. Copp needs a scope with a rise time that is orders of magnitude faster than the rise time of the device under test. His inverters typically have rise times ranging from microseconds down to tens of nanoseconds so the 0.7 ns rise time of the ScopeMeter 190-502 meets that requirement as well.
Although Copp doesn’t see the 190-502 completely replacing the need for bench scopes in the field, he does see its high resolution measurements, isolated ground safety features, and convenient form factor as major factors in reducing the frequency of having to drag out an expensive bench scope. He also sees the 190-502 as a real time saver in a couple of ways. First, he can carry the 190-502 in his tool bag, rather than have to drive back to the office for the bench scope. It is also much faster to set up than a bench scope and has its own built-in isolated ground on each channel for increased safety. And it is IP-51 rated to be dust and drip proof, which makes it compatible with both the indoor and outdoor environments.
At Fluke we’re always interested in hearing suggestions from our customers about what more they need and want in our tools. So if you have suggestions for what you’d like to see in the next ScopeMeter, please let us know.